RAINWATER MANAGEMENT FOR DISASTER MITIGATION AND SUSTAINABLE DEVELOPMENT, MOMBASA, 4- 8 DECEMBER 2006

PAPERS AND ABSTRACTS

Compiled by: KRA/GHARP SECRETARIAT C/O Kenya Rainwater Association P.O. Box 10742-00100, Nairobi, Kenya Tel/Fax: 254 (0)20 2710657 E-mail: gharp@wananchi.com Website: www.gharainwater.org

JULY 2007

2

TABLE OF CONTENTS

3

1 Introduction The 10th Southern and Eastern Africa Rainwater Network (SearNet) International Conference was organized and by Kenya Rainwater Association (KRA) in collaboration with the SearNet Secretariat under the theme “Rainwater Management for Disaster Mitigation and Sustainable Development” with an aim of facilitating experience sharing in rainwater harvesting through presentation of technical papers, posters on planning, research and development activities. The conference was held at Sand and Sun Hotel, Mombasa, Kenya on 4-8th December 2006 and brought together 47 participants from diverse professions in Kenya, Uganda… This report brings together all the proceedings of the Conference.

1.1 Welcome Remarks

1.2 Opening Remarks/Speech

1.3 Participants’ Expectations

1.4 Conference Objectives

4

2 Thematic Papers and Abstracts

2.1 Institutional Capacity Building for Rainwater Management

KEYNOTE ADDRESS ON INSTITUTIONAL CAPACITY AND PARTNERSHIP BUILDING FOR RAINWATER MANAGEMENT By Prof. Nick G. Wanjohi, Ph.D, EBS, Vice Chancellor, Jomo Kenyatta University of Agriculture and Technology (JKUAT)

LADIES AND GENTLEMEN, I am indeed pleased to participate in this 10th International SearNet Conference whose theme ‘rainwater management for disaster mitigation and sustainable development’ is critical today as we continue to suffer massive damages due to floods and yet water supply is insufficient for agriculture, drinking and sanitation, and the environment. This can be attributed to lack of capacity for sustainable management and use of the available rainwater. INTRODUCTON The harvesting of rainwater has been practiced for centuries in different parts of the world but has received little attention until recently. Water is a vital component for any form of development and the knowledge of harnessing and management of water resources need to be expanded at all levels. The current upsurge of interest derives from a growing recognition that human life is threatened in many places either because the quantity of water available is insufficient to meet expanding needs or because the quality of available water is detrimental to health. The demand of water is increasing rapidly on account of growing population, food insecurity and the need to produce more (especially in the marginal rainfall areas), the spread of irrigation, the growth in industry and the expansion of urban areas. At the same time, new problems have arisen because of the breakdown of delivery systems, pollution of existing sources and conflict between users. The growing crisis over water has focused attention on ways to make greater use of rainfall and runoff. Before introducing new technologies it is necessary to find out what is already in existence. In many countries the land users have rainwater harvesting technologies and innovations that have not been documented. Such technologies need to be identified and disseminated to other areas of similar environment. There are many promising indigenous water-harvesting systems used by farmers in the sub-Saharan Africa (SSA). Most prominent are runoff-farming systems e.g. contour ridges, Zai pits earth and stone bunds, vegetation strips and micro-basins (Ngigi 2003). In addition, Rainwater harvesting systems for supplementary irrigation and conservation tillage are becoming popular. Despite such revelations, most farmers in SSA experience food insecurity because of low production. In order to increase food production, much attention should be given to rainwater harvesting, utilization and management. Such cannot be accomplished without building the human capacity and infrastructure that would enable efficient transfer of technology (Rockstrom, et al., 2001). INSTITUTIONAL CAPACITY AND PARTNERSHIP BUILDING Institutional capacity building is defined as provision of technical and material assistance designed to strengthen one or more elements of organizational effectiveness. The elements of organizational effectiveness include governance, management capacity, human resources, financial resources service delivery, external relationships and sustainability (Barbara, 2002; SAFFIRE, 2002). There are three components of sustainability that include managerial, technical and financial or resource sustainability. It is a process of developing and strengthening skills, instincts, abilities, processes and resources that organizations and communities need to survive, adapt and thrive in the fast-changing world. The process involves value added instruction, the training of trainers, activities with multiplier effects, and networking. It involves both institutional capacity building, as well as human capacity building. It ensures the creation of an enabling environment with appropriate policy and legal frameworks; institutional development,

5

including community participation; and human resources development and strengthening of managerial systems (Crowder, 1996). Capacity refers to organization’s ability to achieve its mission effectively and to sustain itself over the long term. Capacity also refers to the skills and capabilities of individuals. Capacity building activities are designed to improve an organization’s ability to achieve its mission or a person’s ability to define own goals (Graham, 2002). PROGRAMMES IN INSTITUTIONAL CAPACITY BUILDING There is a wide range of capacity building approaches that may include peer-to peer learning, facilitated organizational development, training and academic study, research, publishing and grant making. Capacity building also takes place across organizations, within communities, in whole geographic areas, within and across sectors. It involves individuals and groups of individuals, organizations and groups of organizations within the same field or sector (UNESCO-IHE 2006). Institutional capacity building can be divided into three components tam can be summarized as follows: Bilateral programmes with universities: These are aimed at developing indigenous education, training and research capacity by means of staff development. The training of staff in education management and research, including short duration specialized training courses to professionals working in the country is part of the bilateral programme. Curriculum development or restructuring, upgrading of laboratories and library and development of education and training materials and tools are basic changes that have to be implemented to ensure programme continuation. In-house capacity building: These are designed to provide technical assistance to a government agency or to an in-house training centre of a ministry. A development partner in collaboration with a local university or an international consulting firm delivers a total package of services to the recipient government agency. I some cases the local university is also strengthened. Such project may last between 1 and 3 years. Partnership and Networking: In today’s development, there is much emphasis on networking of individuals and organizations in the same profession nationally, regionally or globally. In the recent times, many networks have formed to become knowledge centers. In the area of water harvesting and management, many of the centers function as the ‘hub’ of regional networks. They become complementary partners in the global network. Each of these knowledge centers has a different focus that can contribute to specific knowledge base of the network. The capacity building through this process is enhanced by development of conventional as well as innovative applications of information and communication technologies. CAPACITY BUILDING AGENTS Capacity building agents may come in different ways such as: Management consultants who provide expertise, coaching, training and referrals. Mostly these are profit-making. Management support organizations that provide consulting, training, resources, research, referrals and other services for nonprofits. Grantmakers that may include foundations and government organizations, which often get involved in capacity building either through their grants or sometimes by offering training, consulting and resources. Researchers play an important role in capacity building in identifying issues and trends, and building knowledge for nonprofits and other capacity builders to use.

6

Universities and other academic centres provide formal training and certification opportunities for individuals. They also conduct research and often have resource centres (online and on-site) for nonprofit organizations. Intermediaries and umbrella organizations with multiple grantees or chapters usually conduct their own capacity building activities that respond to specific organizational priorities and needs. Technology firms and other service providers who often play capacity building roles. National and international organizations, membership organizations, coalitions, think tanks, research institutions that are part of the nonprofit infrastructure and seek to make systematic improvements in capacity building (Philbin, 1996). CONCLUSION Human resource capacity building encompasses aspects of awareness creation, education and training, attitude change, confidence building, participation in decision-making and action. A critical goal of HRD is that of maximizing people’s potential to contribute to development by participating fully in all its activities. Through capacity building, individuals and groups are empowered to expand their abilities to more fully participate in the development process. As people get more involved in directing and controlling the process of change that they themselves are bringing about, then the knowledge, skills, attitudes and behavior they require also change. Education and competency-based training are essential for building the human resource capacity required to improve productivity and to manage natural resources for sustainable development. Systematic capacity building requires a supportive and enabling policy environment and a realistic investment in both formal and non-formal education. Policies that create, strengthen and support HRD systems should be a high priority for developing countries, donors and technical assistance agencies. Capacity building efforts should focus on institutional strengthening, including the design of new organizational structures to improve the “goodness of fit” between the policy contexts for sustainable development and enacting institutions in both the public and private sectors. A multiplier effect can be achieved if strong linkages among education institutions, NGOs, research organizations, public and private extension services are fostered. This approach recognizes that there are multiple sources of technology development and dissemination and that integrated institutional network capacity building is required. REFERENCES Barbara, E. 2002. Institutional Development Sanitation Connection. Water and sanitation Program (WSP). www.sanicon.net/titles/topicintro Crowder, L.V. 1996. Human Resource and Institutional Capacity Building through Agricultural education. FAO Research, Extension and Training Division. Rome Graham, C. 2002. Strengthening Institutional Capacity in Poor Countries; Shoring up Institutions, reducing global Poverty. www.bookings.edu/comm/policybriefs Ngigi S.N. 2003. Rainwater harvesting for improved food security: Promising technologies in the Greater Horn of Africa. Thomas, D.B., Mutunga, K. and Mburu, D.M. (eds). Kenya Rainwater Association, Nairobi. Philbin, A. 1996.Ccapacity Building and Organizational Effectiveness. Alliance for Nonprofit Management. Ford foundation. www.allianceonline.org/about/capacity_building Rockstrom, J. 2001. Green water security for food makers of tomorrow; Windows of opportunity in drought-prone savannahs. Water Science and technology, 43 (4): 71-78 SAFFIRE, 2002. Institutional Capacity Building. Strategic Alliance for freshwater information, resources and education. www.water-saffire.org/institutional-capacity UNESCO-IHE, 2006. Institutional Capacity Building. UNESCO-IHE projects. www.unesco.ihe.org/projects/icb.htm

7

Towards widespread rainwater harvesting in Ethiopia: ERHA’s experience as a successful Rainwater Harvesting Capacity Centre (RHCC)

By: Ephraim Alamerew; P.O. Box 27671/1000, Addis Ababa, Ethiopia; e-mail: erha@ethionet.et

Abstract Access to safe and adequate water supply is an important condition for livelihood improvement and sustainable development of the rural people in Ethiopia and other African countries. Unfortunately, a large number of rural communities in arid and semi-arid areas of Ethiopia are deprived of permanent water sources. Meanwhile, there has been a growing global recognition that rainwater harvesting (RWH), offers a viable alternative to complement water supply solutions in areas facing water scarcity. Notwithstanding the greater magnitude of RWH interventions Ethiopia has seen in the past decades, most of these RWH programmes had suffered from problems such as lack of training, skilled labour, competent local management, community ownership and lack of adapting technologies to local conditions.

On the other hand, promotion of effective and efficient utilisation requires facilitation of an enabling environment. Institutional strengthening aimed at triggering emergence of capable focal organisations to promote RWH causes, their collaborative initiatives and partnership building, among other things, constitutes an important aspect of such an enabling environment. This, in turn, facilitates wider public involvement, enhanced awareness and skills of professionals and end-users, focusing research and training initiatives as a basis for knowledge building as well as networking, information availability, accessibility and dissemination on the different issues of RWH.

Based on experiences of ERHA-RAIN partnership over the last few years, this paper depicts the advantages of strengthening institutional capacities of focal organisations and the tremendous role they could play in spearheading RWH initiatives for widespread dissemination. The paper provides tangible accounts of how the capacitation and collaborative support of RAIN1 enabled to build ERHA as a national Rainwater Harvesting Capacity Centre (RHCC) and successfully demonstrated possibility of tackling the hitherto challenging constraints hindering effective promotion of RWH in Ethiopia.

The paper, thus, provides insights into the magnitude of the problem, the partnership building, the local initiatives and the processes involved in implementing RWH activities; stakeholders involved, means of stakeholders’ involvement and roles they play(ed); impacts of the implemented activities; sustainability issues and long-term commitment of stakeholders and target communities; costs involved and lessons learned. Conference participants shall appreciate the originality and innovative aspects of the approach through analysing the different features of a pilot programme carried out by ERHA-RAIN Partnership in Ethiopia as well as what specific roles ERHA plays as RHCC in working with the other implementing partners. The paper finally reflects on the ‘Bigger Picture’ – how this new approach promises realization of widespread RWH replication possibilities in Ethiopia and BEYOND.

1.0 Introduction 1.1 Contextual Background: Water Issues in Ethiopian Ethiopia has nine major rivers and twelve big lakes. Apart from its endowment with big rivers and major tributaries, however, there is hardly any perennial flow in areas below 1,500m altitude. Seven of the river basins are trans-boundary that carry away over 95% of the annual runoff. Only four basins, located on the western part of the country, alone constitute 83% of the country’s surface water potential, while other areas produce very low surface runoff. Despite the country’s higher annual renewable freshwater potential estimated at 122 billion m3, only less than 5% remains in the country with disproportionate distribution over its different parts (UN, 2006).i

Challenges to life and wellbeing: Ethiopia is largely dependent on the agricultural sector, which provides 86 percent of the country’s employment and 57% of its GDP. Rainfed crop cultivation is the principal activity and is practised over an estimated 23% of potentially arable land area in the country. Recurrence of severe droughts, which prevails at 3-4 years’ cycles cause significant damages to the livelihoods of rural households whose earnings rely on agriculture. While estimates show enormous irrigation potential, the country’s socio-economic conditions remain so feeble to allow developing beyond a mere 8% of the available potential. Given the meagre economic status of the country coupled with the rapidly growing population, prospective large- and medium-scale irrigation schemes, if at all realised, are unlikely to ensure food security. Drought is a frequent natural disaster in Ethiopia.

8

Recent observations have shown that the frequency of droughts have increased over the last few decades. The recurrent droughts affects up to over 20% of the total population in the country in serious cases (bad years) and on average, about 8% of the Ethiopian population is under a permanent threat of drought in any year. Overall, it is estimated that nearly 52% of the population is below the national poverty line,1 with poverty in urban and rural areas estimated at 58% and 48% respectively.i

The situation in terms of access to water supply and sanitation in Ethiopia is also very poor. According to the recently released WWDR2 of the United Nations, only 31% of Ethiopians have access to safe water and 10% to proper sanitation facilities, with relatively higher service coverage in urban areas than in rural areas (74.4% and 23.1% respectively).i To make matters worse, about 25 - 40% of water installations in rural areas are not functional at any given time.ii,i Consequent incidences of diseases related to unsafe water supply and inadequate sanitation are very high. The major causes of morbidity among patients include respiratory infections, malaria, skin infections, diarrhoeal diseases and intestinal parasitic infections. Diarrhoea, the most prevalent water-related disease, accounts for 46% of the under-five child mortality rate. The five illnesses mentioned above account for over 63% of all reported cases of child morbidity. The situation gets worse during dry periods, as water carriers (usually women and girls) have to walk longer distances for even smaller quantities of poor quality water, which makes women and girls, to be particularly vulnerable to water-borne and water-related diseases as they more frequently come into contact with contaminated water.

Several relevant studies have indicated that efforts of the government and other stakeholders to redress water shortages problems are mainly hindered due to lack of resources, intensity and extensity of recurrent droughts, higher rates of population growth and inadequate capacities of local administrations …etc. Apart from these drawbacks, situations of water resources in certain agro-ecological regions are further constrained by their natural limitations involving lack of dependable sources of water in terms of both quantity and quality to address the different water supply needs of the people in those areas. Furthermore, such other factors as those associated with debilitating socio-economic conditions of the people present impediments to use advanced technologies that are very expensive and sophisticated to be managed at community level. Besides, even if provided, such solutions involving expensive initial investments and costly operational inputs as well as sophisticated technologies requiring higher levels of skill are unsustainable in terms of technical, economic, social, environmental and institutional aspects.

1.2 The Rainwater Harvesting Alternative and the Ethiopian experience Rainwater harvesting provides a simple, low cost alternative to meet the growing demand for safe water that has been practiced for hundreds of years. It requires low capital and operation costs compared to conventional systems in drought prone and mountainous areas. Besides access to safe water, rainwater harvesting yields numerous environmental, social and economic benefits to justify that it can significantly contribute towards poverty alleviation and sustainable development.

For Ethiopia, much of whose surface waters are carried away across the borders by transboundary rivers, the issue of emphasising the RWH option to augment the available water sources to meet the water supply needs of its people becomes a necessity and timely in light of the looming water scarcity. Based on the current trend of population growth, Ethiopia will have nearly 120 million people by the year 2025, and the per capita water availability will drop to about 947 m3/person/year (UNEP/IETC, 1998).iii According to Falkenmark's definition of water scarcity (Falkenmark et al 1990)iv, this situation puts Ethiopia among the 8 African countries facing water scarcity in the next decade.iii The above facts, among other things, strongly support the need in Ethiopia to emphatically focus on sustainable management of the available water resources along with the identification of appropriate freshwater augmentation technologies such as rainwater harvesting and their promotion to complement available water supply sources.

In Ethiopia, RWH has been practiced over many centuries. However, this traditional practice was almost forgotten in modern days’ water sector development endeavours due to the attention and preferences given towards exclusive use of modern technologies. The recent revival of RWH is triggered by problems such as overexploitation and dropping groundwater levels as well as natural and socio-economic limitations to fully utilise high-tech water supply facilities.

During the past decade, the RWH alternative has been increasingly promoted by many NGOs and through government-initiated interventions in many parts of Ethiopia. Nevertheless, several RWH projects initiated 1 National poverty line is deemed appropriate for a country by its authorities. For this reason, the national poverty line should not be used for comparison between other countries as it varies significantly (Human Development Report, 2005).

9

in the country couldn’t enable realisation of anticipated results. The discouraging experiences in promoting RWH were caused by various constraints in implementing the RWH projects. These included, among other factors, inadequate public awareness and sensitisation of local communities, lack of adequate knowledge and skill in management of RWH schemes, and insufficient involvement of communities in planning and implementation processes. The lack of facilitation for establishment/strengthening of CBOs and absence of local ownership are also important attributing factors for the failures. In addition, the various efforts significantly lacked supports of research information, for example, in terms of identifying indigenous knowledge and best practices in RWH, improving traditional practices and/or adapting new technologies to local conditions … etc, which constitute among the critical inputs for a successful intervention.

1.3 The Role of Institutions for effective promotion of rainwater harvesting From the vantage point of ensuring long-term solutions, effective promotion of rainwater harvesting (RWH) and utilisation, would be indispensable in reversing the ever-worsening water shortage problems. Nevertheless, effective promotion of the RWH alternative and its efficient utilisation requires attainment of a mass transformation towards inclusive understanding of the looming problems and the potentials held in RWH. This involves facilitation for wider involvement of the public and decision-makers towards conscientious actions at all levels. Yet, realising such facilitation of a far-fetched task to ensure effective promotion of rainwater harvesting would be unthinkable in the absence of a focused and capable organisation working on promotion of RWH.

For any development initiative to be sustainable, to flourish and to be widely replicable, existence of effective and supportive institutions is essential. Institutions comprise any agreement between people, whether formal or informal, and include laws, contracts, and regulations sector organisations educational establishments, NGOs and communities themselves (Gould and Nissen-Petersen, 1999).v Organised public initiatives in the form of Civil Society Organisations (CSOs) or Non-Governmental Organisations (NGOs) are, therefore, meant to give the necessary institutional basis for the development initiatives. To this end, CSOs/NGOs established for the purpose promoting RWH, inter alia, solicit and facilitate a wide-spread public involvement and local community initiatives as a basis for the long-term attainment of a mass transformation towards effective utilisation of the rainwater resource.

Some of the roles and/actions anticipated by CSOs/NGOs effective promotion of rainwater harvesting include:

Conduct public awareness raising campaigns and/or reorientation of relevant managers at all administrative tiers of the government and aid agencies towards ‘positive reconstruction’ of any biases or wrong perceptions concerning the merits of RWH;

Facilitate knowledge building in the conceptual, practical skills and management of RWH activities through supporting research, public discussion/dialogue forums and use of other communication methods;

Provide training (i.e. improvement) and demonstrations to enhance the knowledge base on RWH concepts and practical skills along with support for consultative mechanisms and knowledge/information exchange activities;

Assist in establishment of community-based organisations (CBOs) and their technical and managerial capacitation so that they will play a leading role in the installation, management and maintenance of RWH facilities in their respective localities;

Initiating and/or strengthening local and international networking and partnerships; Promoting local networking of interest groups, professionals, researchers and public servants to

consolidate alliance, initiate popular movement and harmonious inter-sectoral interactions towards mainstreaming the RWH alternative, its effective promotion and sustainable development;

Introduction of up-to-date concepts and techniques, such as participatory appraisal, stakeholder analysis, environmental assessment, impact assessments (technical, economic, social, environmental … etc) into promotion of RWH;

Advocating/supporting the reorientation of public water authorities towards adopting a facilitating and regulatory rather than a ‘do everything’ role in the promotion of RWH;

The success of programmes and services intended to advance a wide-spread promotion of RWH heavily depends on the resources, skills and technical expertise of the responsible institutions. Thus, institutional development and capacity-building constitute the most essential determinants for effective promotion of

10

RWH. These institutions need to be appropriately structured and provided with apposite governance and administrative framework which favour effectiveness and efficiency. 1.4 Ethiopian Rainwater Harvesting Association (ERHA) The inception and establishment of the Ethiopian Rainwater Harvesting Association (ERHA) is part of the regional movement traversing the entire Southern and Eastern African countries, which in turn followed the global response to the ever deteriorating trends of water resources and the looming challenges ahead. It was established in 1999 through efforts of concerned individual citizens who recognised the challenges of water shortage at global and locally within the country. They initiated establishment of ERHA cognising the potentials held in tapping the rainwater resource to address the problem and the tremendous role civil society organisations (CSOs) could play in spearheading rainwater harvesting initiatives at a wider scale within the society. ERHA is, thus, an instrument of achieving the collective aspirations of its constituency, which is promotion of rainwater harvesting to address shortage of water for domestic supply, food production and environmental needs.

The Ethiopian Rainwater Harvesting Association (ERHA) is a non-governmental and non- profit-motivated national organisation. Accordingly, ERHA’s Vision is to see that “Water & food security of all Ethiopians are ensured for sustainable development and perpetuity of their livelihood”; and its Mission is “to facilitate promotion of rainwater harvesting in Ethiopia through advocacy, networking, research and capacity building towards developing, adapting and disseminating rainwater harvesting”.

ERHA’s organisational development and operational endeavours are guided by its five-year strategic plan (2003-8), which has set six major strategic objectives encompassing various activities envisaged to be accomplished in compliance with the action steps described within the strategic framework. ERHA’s strategic objectives are: institutional development, policy advocacy, promotion of RWH technologies, addressing socio-economic issues affecting promotion of RWH, development of information management & communications systems for promotion of RWH, as well as strengthening networking and partnership.

2.0 ERHA’s Experience in Institutional Strengthening and Partnership Building 2.1 The need for institutional strengthening It is widely believed that failures of many water development initiatives in the developing world can be attributed to systematic deficiencies in the relevant institutions responsible for addressing the required institutional functions (Alaerts et al., 1997)vi. Strengthening relevant institutions, through the necessary institutional capacity building interventions, is therefore vital to ensure long-term sustainability of the development initiatives. According to Austin et al (1987)vii, “institutional development refers to the organisation, management, financing, staffing, training, operation and maintenance of the relevant systems and facilities.

The role of focal organisations to promote and support wider promotion of rainwater harvesting (RWH) is indispensable. Successes in terms of attaining a mass transformation towards wide-spread adoption of the RWH alternative could only be realised through appropriate institutional development at all levels. It is also unlikely that widespread replications of appropriate RWH technologies and experiences of successful community-based strategies to be achieved in the absence of supportive institutions at higher levels. To ensure this crucial support, it would be essential to convince and win supports of senior government and aid officials. This requires demonstrating the benefits of using appropriate RWH systems and their community management approaches as reliable options for survival and sustainable development. Unless organisational capacities of relevant CSOs/NGOs like the RWH associations are enhanced to optimal competence levels, their effectiveness in promoting rainwater harvesting would be more difficult.

Unfortunately, most funding partners decline from providing necessary supports to address the needs for organisational capacity building of local organisations in developing countries. The usual interest of such funding partners is focused towards investing on infrastructure developments (hardware), and thus, risking the eventuality of unsustainable intervention. Nevertheless, the need for organisational capacity building in water development has been recognised since several decades in the recent past. Capacity building is among the seven areas for action within national water strategies identified by the UN Secretary-General for the 1990s. Similarly, two of the guiding principles identified at an international post-Water Decade Conference in New Delhi in 1990 were related to capacity building, put as: “strong institutions are essential for sustainable development; and capacity building is necessary to make community management

11

effective”. Most recently, the World Food Summit, Rome, 1996, has also touched upon aspects of capacity building in its conclusions as “the implications of neglecting food security can be serious, and investment in water infrastructure, continued reform of supporting institutions, and an enabling environment are necessary to improve food production.”viii

2.2 Partnerships amongst Key Actors: the need and benefits The water crisis and lack of financial resources available to address needy community groups compel most local CSOs/NGOs involved with the water sector development into partnership with other more resourceful organisations of similar objectives. Such partnerships are usually started as cooperative arrangements of mutual interest, centred on defined actions and aimed at mobilising resources and reduction of expenses. These collaborative arrangements include strengthening of alliances and networks between NGO's at various levels; implementation of action-oriented projects focused on varying scales of infrastructures and services close to target communities. These arrangements are often based on democratic and transparent rules within contractual arrangements of obligations and accountability of participants.ix

Partnership initiatives are emanating from mutual interests of the partnering organisations, which have common objectives to be achieved. Funding organisations (e.g. international NGOs or bilateral agencies) are interested to ensure that their supports reach the rural poor. Traditionally, such funding agencies are usually engaged with local governments. However, the records of government-delivered services, and the new appreciation of water scarcity and value, have led to a reappraisal of potential actors and their roles. This aspect partly explains funding agencies’ interest towards building of alliances and partnerships with a wide range of stakeholders, notably CSOs/NGOs, that has become a theme familiar within development cooperation for water-related activity, as in other areas.viii

In this regard, local NGOs and their international counterparts have attracted considerable attention in the recent past because of their relative effectiveness in reaching the poor and their knowledge and experience of working closely with communities. They also have a reputation, in many cases deserved, of achieving much with little, and their methods have therefore attracted attention for cost-efficiency reasons. Because of the pioneering role they have played in demonstrating the practicability of user participation in the management of all kinds of community development interventions, NGOs are now regarded as part of the mainstream in water development co-operation.Error! Bookmark not defined. There are many NGOs in developing countries, known for their successful implementation of development projects that transformed lives of needy communities. On the other hand, many international NGOs and bilateral agencies from the North have been recognising that their effectiveness in reaching intended targets are often realised through partnership with local CSOs/NGOs in the South. However, the size of contributions by these local CSOs/NGOs in the South are proportionately small, and not all are equipped to operate effectively without external supports. This is mainly because of their feeble organisational capacities and the inescapable dual concern of ensuring sustenance of organisational functions while striving to remain operational with their services. From a wider perspective, capacity-building and institutional development supports provided to CSOs/NGOs are considered essential for sustainability of water development initiatives.

1.3 ERHA’s Experience in Partnership Building

Strategic Approach to Partnership Building: Being both the result and part of the prevailing development challenges in Ethiopia and most other developing countries, ERHA shares the different constraints faced by development organisations in the country. These include inadequate capacity in terms of material and technical resources needed to run an organisation charged with a grand mission of promoting RWH as a means for alleviating the challenges of water shortage. During ERHA’s strategic plan development, the need for building partnership with other organisations of mutual interests was identified as one of the critical issues. Accordingly, as one of its strategic objectives, ERHA was required to be proactive in seeking partners and clearly establish areas of partnership. It was also indicated that such partnerships might include implementing joint projects/programmes in research, training, awareness-raising and demonstration activities aimed at promoting improved rainwater harvesting and utilisation at all levels. Accordingly, ERHA has since been striving to establish and define its collaboration and partnership with NGOs, learning institutions and other development organisations that are interested in advancing promotion of rainwater harvesting one way or another.

ERHA’s partnership building experience:

12

ERHA’s establishment and its emergence as an active agent of RWH promotion in Ethiopia could be taken as a success story or an aspect of a regional cooperation in development. In fact, it might worth indicating here that each partnership engagement should be seen not only as a process forming some kind of continuum in organisational development but also very much linked to each other in a sort of ‘chain reaction’. That is, partnership linkages of an organisation involves processes of facilitating for other partnering initiatives to emerge like forming buds on a tree and growing into larger branches and so on. In this regard, ERHA’s experience in forging partnership relations with different entities in the region during the past few years provides a highlight on how partnership building initiatives have mutually reinforcing phenomena where every initiative paves the way for the next, yet stronger and resourceful partnership engagement. This could be seen from the partnership experience that ERHA traversed since its initial stage of formation – i.e. the major partnership engagements entered with RELMA, SEARNET, GHARP and others, as discussed below.

RELMA (Regional Land Management Unit) of SIDA (Sweden International Development Agency) had been a close supporter of ERHA since its inception and continued until end of 2002. As of January 2003, RELMA secured initiation of partnership with GWP, which paved the way for establishment of the SEARNET (Southern and Eastern Africa Rainwater Network) Secretariat. Then RELMA transferred its RWH promotion activities to the SEARNET, to which ERHA has been one of the founders. ERHA’s partnership with SEARNET was on a par with that of RELMA, and has enabled ERHA to significantly improve its Secretariat facilities and services as well as execution of some operational activities. In fact, ERHA’s partnership with SEARNET has been a major factor in facilitating its latest such initiative to result in a successful partnering with RAIN and UN-HABITAT.

ERHA was one of the founders of the Greater Horn of Africa Rainwater Partnership (GHARP), which was formally established in March 2001, through a financial grant from USAID. ERHA has gained its part of the partnership’s institutional strengthening support as was also realised by GHARP secretariat and the rest of member associations.

Currently, ERHA’s close supporter is RAIN (Rainwater harvesting Implementation Network). RAIN started its RWH programme in Ethiopia in 2005 and selected ERHA as its Rainwater Harvesting Capacity Centre (RHCC). Four local implementing partners were also identified: AFD, ASE, ERSHA and Water Action. Being the designated RHCC in Ethiopia, ERHA supported and monitored these four NGOs, and documented and evaluated the rainwater harvesting projects. In 2005 and 2006, RAIN’s financial and technical support gave ERHA a solid base to perform its activities. Both parties have expressed their intentions to continue this partnership in order to promote and mainstream RWH in and beyond Ethiopia. It would be worth mentioning that SEARNET has played a significant role in encouraging the partnership from the start.

In 2005, ERHA established a partnership with UN-HABITAT to implement a pilot project entitled "Promoting Rainwater Harvesting to Augment Sustainable Water Management in Urban Areas". The pilot project was aimed at demonstrating the potential benefits of rainwater harvesting to augment water supply services in urban areas. UN-HABITAT found ERHA’s pilot project to go in line with its own WDM strategies and the two organisations established a partnership to implement the project. The accomplishment has been so encouraging that UN-HABITAT and ERHA are currently working on extending their partnership at a wider scale. Here again, ERHA’s being a close partner of SEARNET has contributed in the initiation of the partnership.

Implications to ERHA’s Institutional Strengthening: Currently, ERHA has many stakeholders and partners in and beyond Ethiopia, including Ethiopian governmental organisations, institutions and authorities, research organisations, international NGOs, international funding parties such as UN agencies and bilateral agencies, and local non-governmental and community-based organisations. Such a wider partnership linkage of ERHA has, on the one hand, enhanced public image in terms of its credibility to attract more partners. Secondly, the partnership engagements imply availability of resources for its active involvement in organisational development and operational aspects of realising its overall mission and strategic objectives thereto. Thirdly, it expands learning opportunities, which are vital for a young organisation like ERHA. Finally, it enhances the organisational resourcefulness in a way that guarantees its sustainability as a people’s institution.

2.4 Challenges involved with Institutional Development and Partnership Building

13

As far as CSOs/NGOs partnership relations with foreign donors are concerned, local CSOs/NGOs have continued to rely on foreign assistance for the bulk of their funds and the funding system has remained short-term and results-oriented, or ‘donor-driven’.

In fact most donor agencies recognise the needs of CSOs/NGOs in developing countries for resource supports in order to realise their organisational sustenance and operational objectives. However, supports provided for realisation of such needs come with an increasing demand for accountability. Donors usually want to see their scarce resources used more efficiently and to have more visible impact on poverty reduction. In the process, less attention is given towards allocation of resources for capacity building support in the partnership arrangement. This usually undermines the important aspect of empowering partner CSOs/NGOs to ensure sustainability of local development initiatives.

Meanwhile, there is also a growing awareness among progressive-minded donors of the need for CSOs/NGOs to overcome their dependence on traditional development financing and become more financially independent and sustainable.

3.0 The RAIN Partnership in Ethiopia 3.1 Rainwater Harvesting Implementation Network (RAIN)

Establishment: RAIN (Rainwater harvesting Implementation Network) is an international network based in Amsterdam, The Netherlands. RAIN started in December 2003 under the name “Global Rainwater Harvesting Collective”. It has been renamed as off October 2004, as RAIN (Rainwater harvesting Implementation Network). RAIN is hosted by the RAIN Foundation that is registered in the Netherlands. The RAIN Foundation is administered by a Board and is advised by an International Advisory Committee (IAC) consisting of internationally acknowledged experts in the water sector. Its programme management unit is located in Amsterdam, the Netherlands.

Objectives: RAIN aims to increase access to water for vulnerable sections of society in developing countries, women and children in particular, by collecting and storing rainwater in water tanks and wells. RAIN’s rainwater harvesting programme focuses on field implementation of small-scale rainwater harvesting projects and knowledge exchange on a global scale.

Operational Strategies and Approach: RAIN provides funds for implementation of small-scale rainwater harvesting projects through local organisations. RAIN is presently implementing rainwater harvesting projects in Nepal, Senegal and Ethiopia. It intends to expand its activities in Asia and Sub-Saharan Africa in the near future.

RAIN’s approach in each target country is to select a strong local NGO as Rainwater Harvesting Capacity Centre (RHCC). This RHCC is trained, equipped and supported by RAIN to coordinate the national or regional RAIN programme. Local NGOs are in turn trained, guided and supported by the RHCC during the implementation of RWH projects. In addition, RAIN programme facilitates a global exchange of knowledge on rainwater harvesting between its partners and other interested organisations.

RAIN also ensures integration of rainwater harvesting in water and sanitation programmes as well as educational and health programmes through partnerships with different local organisations. Since its establishment, RAIN has facilitated constructions of several rainwater storage tanks through local partner organisations in different countries.

Partnerships: Principal donors of the RAIN Foundation include PLAN Netherlands, EU, Aqua4All and others. It is an active member of UNEP’s global Rainwater Partnership in which all leading rainwater harvesting organisations, including SEARNET, are represented.2

3.2 Formation of the RAIN Partnership in Ethiopia

2 For more information or to contact RAIN:

RAIN Foundation, Donker Curtiusstraat 7-523, 1051 JL Amsterdam, The Netherlands. T: +31 (0)20 58 18 270; F: +31 (0)20 68 66 251; E-mail: info@rainfoundation.org; Web site: www.rainfoundation.org

14

In September 2004, RAIN conducted a scoping mission to Ethiopia in order to select organisations to be invited for the pilot phase of the RAIN programme in Ethiopia. This scoping mission led to the selection of partners for implementation of rainwater harvesting (RWH) projects involving four Ethiopian NGOs: Action for Development (AFD), Agri-Service Ethiopia (ASE), Ethiopian Rural Self Help Association (ERSHA), and Water Action (WACT). The Ethiopian Rainwater Harvesting Association (ERHA) was also selected as Rainwater Harvesting Capacity Centre (RHCC) for this pilot programme.

In December 2004, RAIN and ERHA signed a contract defining their roles in the pilot phase of the RAIN programme in Ethiopia. The four implementation NGOs prepared proposals, which were reviewed by RAIN. Upon approval of the proposals, agreements were signed between the four partners and RAIN. The role of ERHA was mentioned in each of the contracts and thus agreed upon by the implementing NGOs. In order to establish the framework of the project formulation, implementation process and subsequent communications, RAIN provided the RHCC and the implementing partners with a prior prepared Management procedures, guidelines, checklists and formats as well as Monitoring guidelines and parameters. The partnership thus formed was named as “RAIN Partnership in Ethiopia”.

3.3 Partners’ Involvement and Roles The division of roles among the different partners is in line with the RAIN Partnership Approach, involving the various parties as described above, have been elaborated as follows:

RAIN: RAIN’s main tasks are:

• to build the capacity of the RHCCs through operational (technical) and institutional strengthening assistance;

• to manage and share knowledge and lessons learned in its whole network and between all RHCCs;

• to ensure continuation of the RAIN Partnership programme by allocating financial and technical resources (in collaboration with the RHCC and other partners);

• to oversee all RAIN global activities, both technically and financially; • to influence policymakers (in collaboration with partners and RHCCs) in order to facilitate

instigation of favourable policies towards widespread RWH implementation; • to engage in fundraising; matching proposals from within the RAIN network with available

funds and interested donors; and, • to explore, develop and refine appropriate technologies related to all aspects of rainwater

harvesting.

ERHA (RHCC): ERHA, as RAIN’s RHCC in Ethiopia, has the following responsibilities during the implementation of RWH projects by implementing NGOs:

• Provide local capacity building at local NGO level and at community level (awareness creation sessions, CBO’s meetings, Training of Trainers);

• Monitoring, guidance, documentation & evaluation of the projects implemented by implementing partner NGOs;

• Facilitating networking, information communication & knowledge-sharing with relevant stakeholders on RWH issues;

• Production of periodical monitoring and progress reports (covering both financial and technical performances) to RAIN and, as appropriate, sharing with partners and other key stakeholders; and,

• Although the proposal for the year 2005 implementation were reviewed by RAIN (since these projects were the first RAIN projects in Ethiopia), in the forthcoming project rounds, ERHA will gather and review proposals and send the proposal with its advice for approval to RAIN. This way, RAIN can concentrate on its responsibilities as mentioned above. Advantage of this approach is that the local RHCC, with more knowledge about the local conditions and

15

circumstances, reviews the proposals, advises RAIN which proposals are ‘approvable’, and provides training, guidance and support during and beyond the implementation of RWH projects.

Implementing NGOs: The local partner NGOs are responsible for implementing their projects according to the approved Implementation of their respective projects as outlined in approved proposals – i.e. in accordance with the activity plans and budget within the given timeframe;

• Provide the necessary field supervision and monitoring of the project implementation according to the indicators outlined in RAIN monitoring format;

• Secure verification of the proposed technical designs of the rainwater harvesting structures with the RHCC (ERHA);

• Actively support and collaborate in training, monitoring, documentation and evaluation activities carried out by the RHCC.

• Deliver a technically sound RWH structures as per the modifications made by the RHCC &/or RAIN (if any);

• Submit technical & financial reports within the agreed timeframe after termination of the project;

• Communicate any alterations in the approved proposal and project plan and secure agreement from RAIN directly or through the RHCC;

3.4 Accomplishments of the RAIN Partnership in Ethiopia The RAIN Partners’ programme in Ethiopia was fully operational in 2005; and a total of eleven RWH systems were constructed by four implementing NGOs. These include:

• AFD: 3 tanks of 50,000 litres in 3 villages of about 60 households each (Oromiya Regional State, Borana Zone, Dirre District);

• ASE: 2 tanks of 25,000 litres at a school with 650 children, 25 teachers and families; 2 tanks of 25,000 litres at 2 health posts each serving 10 patients daily & personnel (Oromiya Regional State, North Shewa Zone, Bereh Aleltu District);

• ERSHA: 2 tanks (115,300 litres) at a school with 616 pupils (SNNP Regional State, Gamo Gofa Zone, Kutcha District);

• WaterAction: 2 tanks of 40,000 litres in a community of 750 people (SNNP Regional State, Alaba District)

These tanks, with a total harvesting capacity of about 445,300 litres, are now daily benefiting about 3,000 people. The different designs and RWH systems include: above-ground ferro-cement, reinforced brick-cement, and brick-cement tanks as well as underground reinforced cement tanks; rooftop RWH as well as surface run-off harvesting systems. The projects enabled to realise various improvement on existing conditions as described under impacts below.

The constraints encountered during implementations of the activities include rising prices of materials and transportation, which were managed to cover by contingency budgets.

The anticipated implementations for 2006, however, were not realised due to lack of funding as a result of the different factors beyond RAIN’s control that influenced delivery of funds that were pledged earlier. Consequently, the available fund was used to run the RHCC.

3.5 Impact of the RAIN Partnership activities

Social impacts: • Improved health and less diseases, especially water-borne diseases, through access to safe water,

better and cleaner sanitation and improved personal hygiene, and awareness concerning these issues; • Improved education: children can spend more time on education – especially girls (often responsible

for fetching water, and staying away from schools due to poor sanitation facilities);

16

• More social contacts: since it is not needed to walk for hours fetching water, women can spend their time on other things such as income-generating activities or getting involved in village groups or committees.

Environmental impacts: • Overflow water from the RWH tanks is used for watering vegetable gardens and trees at schools and

health posts and/or groundwater recharge; • The groundwater level believed to recover when collected rainwater is used instead of groundwater,.

Economic impacts: • During construction, some (un)skilled members of the community are hired to work. • Although some part of the local labour input was performed as local contribution, all workers also

received varying payments that they used to meet their immediate needs. • More time available for income-generating activities, and thence, improvement in household

income; • When enough water is available, it can be used for (drip-)irrigation or livestock keeping, thereby

increasing agricultural yields and incomes, and thus improving food security; • Due to improvements in health and education, income-generating activities can be performed more

efficiently.

The scale of impact at each project site is local. However, RAIN is planning to expand its programme in Ethiopia through ERHA where more local NGOs will be involved to implement RWH projects in order to ensure up-scaling of the RAIN programme in Ethiopia. In that case, the sum of all local impacts can become a reality at national level and even beyond.

According to the assessment made during the evaluation of RAIN Programme in Ethiopia in 2005, local users told about their experiences and expectations:

• Two women, Diama Kanchara and Gelmo Guracha, from the Ola Dube Oda village told about their hardship due to the water scarcity: “We had to leave our children with neighbours to go searching for water even when our children are sick. We often are away more than six hours, during which our young children miss our caring and breast feeding. Everybody in our community is excited about the RWH tanks.”

• Ato Mohamed Shafi, a 70-year old man with 8 children and 17 grandchildren, told ERHA: “I consider myself to be a fortunate person to see the rainwater harvesting tanks being constructed in my village to benefit me.”

• The 35-year old mother Berida Shukura explained that she considered the day when the construction of the rooftop RWH tanks started as her re-birth day. “A bright future is to come as I will be relieved from the tedious and risky task of water fetching.”

3.6 Sustainability and long-term commitment In order to ensure sustainability of the RWH structures, all tank designs were reviewed by ERHA and RAIN. The RWH systems are designed to be very durable; duration of operation is estimated to be at least 20 years, provided that the systems are well-maintained. Occasional repairs might be necessary and should be possible with local materials and skills. Nevertheless, ensuring continued attendance of the maintenance aspect is very important. The only way to ensure maintenance and management after the project termination is by fostering local ownership. To ensure this, local water management committees have been established at each project site. These committees have been trained before the projects started, and have been – in collaboration with the implementing NGOs – responsible for the local management of the projects. During the projects, these committees facilitated optimal local participation and involvement. They also managed the local contributions to the project, both in kind and cash. Moreover, the water committee manages accounts from small water fees. The level of such a fee is determined by collective decisions of users depending on its affordability to the users. The fees are saved for eventual maintenance and possibly extension of the RWH structures.

Members of these committees were trained on all the management issues involved with RWH systems, such as maintenance of the RWH structures, water quality management, and financial management. All

17

these activities enlarged the local commitment and local ownership. RAIN, ERHA and the implementing NGOs are therefore convinced that the sustainability on the long-term is ensured. In case of unforeseen problems, the local water management committees can directly contact ERHA or the implementing NGOs to get their assistance.

During the activities described in this local action, ERHA has gained substantial experience and knowledge about managing the RWH programme implementation. Besides technical and institutional experience, ERHA also gained experience in fundraising, considered crucial to success. RAIN and ERHA together wrote several project proposals including the European Union Water Facility (EUWF) for 2006, Aqua4all, Italian Embassy in Addis Ababa, ReSource Award 2007, the Lee Project … etc. Although a few of these (e.g. EUWF 2006) couldn’t succeed, some of them are still being considered.

Evaluation of the RAIN programme in Ethiopia was carried out in December 2005. During this evaluation, both the activities of the implementing NGOs and those of ERHA have been assessed. Based on this evaluation, recommendations have been given to the RAIN programme in Ethiopia to improve in a few necessary aspects. There is a plan to deploy the experience and knowledge of ERHA throughout the entire RAIN network, e.g. in the RAIN West Africa Programme. ERHA and RAIN consider this kind of South-South exchange between RHCCs from different countries to be very valuable and crucial as it facilitates for mutual learning and knowledge transfer.

3.7 Originality and innovative ideas There are three levels at which the RAIN Partnership in Ethiopia considers as original and innovative. Firstly, at construction level, local traditional knowledge has been taken into account in order to optimally suit local conditions in order to construct cheaper, and yet, reliable RWH tanks. ERHA and RAIN work together in knowledge management and transfer through collecting local and worldwide experiences that can be used to improve RWH.

Secondly, at project level, local water management committees are established and their members trained before the projects commence in collaboration with the implementing NGO, which is responsible for local management. This has enhanced communities’ involvement during the project implementation in terms of optimising the local contribution, enhanced sense of ownership and assuming responsibilities for management and maintenance after the project termination.

Thirdly, at programme level, RAIN’s innovative approach of capacitating a local organisation (in this case ERHA) to become Rainwater Harvesting Capacity Centre (RHCC), managing a RAIN programme that is being implemented by local NGOs and CBOs, which were four organisations in 2005, and expected to be more in the years ahead.

The activities in Ethiopia in 2005 mainly included innovation at programme and project levels. Since some RWH programmes in Ethiopia implemented before were not as successful as planned, RAIN and ERHA focussed on addressing the underlying causes: lack of training, skilled labour, local management, local ownership and lack of adaptation to local conditions.

ERHA has a proven track record and it is therefore considered capable of performing the local management of the RAIN programme in Ethiopia. Enthused by the emerging needs for ensuring optimal quality of harvested rainwater in storages, ERHA has written a Rainwater Quality Manual in collaboration with RAIN, in which the measures that need to be taken to maintain the rainwater quality are clearly described. This manual will be distributed among the RAIN partners, both in the field and internationally.

3.8 Costs involved About 42,000 euro project costs for the total storage capacity of about 445,300 litres (material, transportation, labour, communication, supervision and local administration costs – i.e. including management costs of implementing organisations). About 18,000 euro support for ERHA to become RHCC (capacity building: salaries, office rent, office equipment, training costs, communication costs etc.). The total cost of the first project round of the RAIN programme in Ethiopia, including RHCC costs, was 60,000 euro.

In close collaboration with ERHA, the implementing organisations tried to lower the costs of the projects. However, due to rising prices of materials and transportation, this proved to be hard. During next project rounds, construction costs will likely be lower, by adopting simpler designs. In collaboration with ERHA and other partners such as the Practica Foundation, interesting opportunities were identified to decrease the

18

costs. These opportunities will be assessed and tested during future project rounds. Moreover, ERHA and RAIN anticipate introducing the use of micro-finance structures in order to increase the local contribution.

The partner NGOs have their own programme activities with management systems in place, and have good relationships and experiences with target communities. The RAIN Partnership programme benefits from this through reduced management costs, more efficiency and complimentarity effects of other ongoing activities (for example aimed at sanitation, hygiene, water management, food security).

4.0 Lessons Learned

4.1 Suitability and potentials of RWH The main lesson learned from the experience of the RAIN Partnership programmes in Ethiopia and others (e.g. Nepal and Senegal) is that RWH can and should be regarded as suitable and highly potential contributor to addressing water supply issues. RWH is a very simple, affordable and decentralised option for answering the need of millions of people for access to safe water. For scattered rural villages and households like in rural Ethiopia and other African countries, demonstrating decentralised and affordable water supply options of varying sizes through RWH has a good chance of being easily adapted by other rural communities and scaled-up to other areas of comparable settings.

4.2 Practicality of RAIN’s approach The RAIN approach, illustrated in the RAIN Partners programme in Ethiopia, contains the main elements that are needed for global implementation of RWH structures:

• Focus on local grass-root management of projects and RWH structures; • Establishment of local capacity to manage RWH programmes in different countries through the

capacitation of RHCCs; • Knowledge management and transfer: facilitating South-South exchange of knowledge and lessons

learned, combined with knowledge from Northern and Southern partners. This element ensures continuous innovation, adaptation and improvement of technical designs.

The design of interventions with inherent flexibility and keenness to address local problems through local actions triggers a learning process. In the RAIN approach, RHCCs gather the lessons learned and innovative local solutions. Through its extensive network, RAIN and the RHCCs share the gained knowledge on different levels (national, regional, global) in order to facilitate widespread and worldwide replication.

19

2.2 Rainwater Harvesting Center Construction at Saint Andre College Kigali-Rwanda

Dr Umaru Garba Wali, Department of Civil Engineering, National University of Rwanda, Butare & Eric A S, Department of Technolog, Kigali University of Science and Technology, Kigali, Rwanda. Abstract

The problem of insufficient water supply in Kigali town and Rwanda in general can not be over emphasized, regardless of available sufficient rainfall ranging from 800 – 1200mm annually. The Rwanda Rain Water Harvesting Association (RRWA) one of the members of the South and East African Rainwater harvesting network (SearNet) is looking for the possibility of constructing a Rainwater Harvesting Centre in the country, which will serve as an enlightment, propagation and advocating center for rainwater harvesting for sustainable water supply in Rwanda. This center can be constructed in Saint Andre College one of the important educational institutions in Rwanda located in Kigali. The Kigali water supply shortage is also affecting the college, which is having enough potential for rainwater harvesting (i.e. roofing of about 11340 m2 suitable for rainwater harvesting). The College is also facing a serious runoff problem, which is imposing a conflict situation between the college authority and its neighbours. The construction of the rain center if accomplished would serve as a tool for advocating rainwater harvesting, supplement water supply of the College thereby reducing stress on Kigali water supply network, resolving conflict, and improving aesthetic condition of the place. The RRWA has long time experience in rainwater harvesting and have enough human resource to accomplish the project if adequate funding is secured.

1.0 Introduction 1.1. Back ground and problem Everyday water is becoming an increasingly scarce resource in Kigali the capital of Rwanda and other towns within the country. Research shows that the Kigali water supply system (by Eletrogaz) has the capacity to provide only about 15,000 m3 which is about 37.5% of the total daily water requirement (40,000 m3). However, Rwanda has sufficient rainfall ranging from 800-1200 mm annually, depending on the region, which makes it suitable for sustainable rainwater harvesting. In view of the above reason stated above the Rwanda Rain Water Harvesting Association (RRWA) one of the members of the East and South African Rainwater harvesting network (SearNet) is looking for the possibility of constructing a Rainwater Harvesting Centre in the country, which will serve as an enlightenment, propagation and advocating center for rainwater harvesting for sustainable water supply in Rwanda. The rainwater harvesting center has been agreed to be constructed in Saint Andre College one of the important educational institutions in Rwanda Saint Andre College is located at Nyamirambo district of Kigali town and was opened on 1957. Since then the college has produce many student. Currently the college population is 1021 people including 950 students. The population is expected to increase by about 200 at the year 2015. This would increase the stress on the Kigali water supply network. However, the Collage is having enough potential for rainwater harvesting (i.e. roofing of about 11340 m2 suitable for rainwater harvesting). Considering the roofing’s runoff coefficient of 0.9 an mount of 12,247 m3 may be collected annually. This would be equal to about 76% annual water requirement of the school if the daily water requirement is taken as 40 liter per capita per day. The College is also facing a serious runoff problem from the street and within it territory, which is imposing a conflict situation between the college authority and its neighbours. The roofing area is only about 10% of the College’s territory. The management of Saint Andre College has shown the interest and has given approval for the construction of the rain center in there college. The RRWA has long time experience in rainwater harvesting and have enough human resource to accomplish the project if funding is secured. If accomplished the construction of the rainwater harvesting center in Saint Andre College Kigali, would serve the following purposes:

i. Provide a rainwater harvesting knowledge based center for RRWA that can be use as a model and for the propagation and advocation rainwater harvesting for sustainable water supply.

20

ii. Supplement the insufficient water supply of the College and reduce the stress on the Kigali water supply network, which is currently not having the enough capacity;

iii. Solved the conflict between the school management and the neighbours, who are requesting the College authority to find way of stopping the runoff within the territory of the College.

iv. Impact knowledge of rainwater harvesting to students and teachers of the College, who may later apply the principle in there various town;

v. Improved aesthetic condition of the college and neighbouring houses. 1.2 Scope of the project The project was design to include roof-top and land surface rainwater harvesting system. All the two systems would be comprised of collection, conveyance, storage and distribution systems. In case of the roof top it would be comprise of up the gutters, downspouts and pipes, tanks and distribution taps. While the land surface system would be made of up open channels, a collection and pipe distribution system for irrigation purpose. In general the project would include the following:

i. Determining the contributing catchments areas (roofs and land surface); ii. Estimating amount of expected runoff;

iii. Sizing collection, conveyance, storage and distribution systems as well as constructing them; iv. Printing pamphlets for advocating rainwater harvesting.

2.0 Methodology 2.1. Data acquisitions Simple measurement would be use to determine the roofing areas. Topographic survey is to be use to draw the land map of the college. And satellite image would be use in determining the total area that contributes runoff to the territory of the College. Rainfall data would be obtained from Kigali meteostation and Ration method would be applied in estimating the runoff amount. The College population is to be use for estimating the water demand. 2.2. Design of structures: The gutters, downspouts, pipes and open channels would be sized using continuity and manning equations with discharge determined by using rational equation. Tanks would be sized using average monthly rainfall and water demand. 2.3. Construction of structures:

• Gutters and downspouts: Gutters and downspouts would be constructed using steel sheets bend to form welded and painted with oil paint this is due to the fact that many of the roofs are very long and wide. However, were applicable PVC materials may used. An appropriate design shall be made to support gutters in place.

• Pipe network: PVC pipes of appropriate size and metallic support are to be use for conveyance and distribution networks.

• Storage tanks: Storage tank would be constructed using ferroconcrete materials.

• Open Channels: Concrete or Riprap channels would be constructed and properly networked to convey land surface runoff to a constructed pond.

• Constructed Pond: A very suitable sited has been identified in the College for construction of a pond to which all surface runoff can be diverted. Pond’s dimension about 15x10x1.5 m3

21

2.4. Materials requirement: The project would require the construction of 2 km of gutters’ length, seven km of 100mm PVC pipes, minimum of 12 tanks and about four km of open channel. The same materials are expected to be use for gutters and downspout. This calls for need in welding material. 2.5. Human Resources Requirement: The human resources requirement would include:

i. Carpenters and welders for gutters and downspouts constructions; ii. Brick liners for tanks construction;

iii. Labourers for digging and lining of channels and pond; iv. Plumbers for pipes network construction; v. Inspection of the work by the designers and external inspectors (SearNet officials) for quality

control. 3. Expected Outcome: The major expected outcome of the project is a rainwater harvesting center that would be use by the Rwanda Rainwater Harvesting Association for propagation and avocation of rainwater harvesting principle for sustainable water supply in Rwanda. However, at it location the rain center would supplement water supply, resolved conflict and serve aesthetic purposes. 4. Constructions timing: It was estimated that the whole project would take six (6) month after the securing of adequate funding. 5. Financial Implication: The project is estimated to cost Million Rwandese Franc which is equivalent to about ………US dollars or Euros. These include References Orodi J. Odhiambo, Alex R. Oduor & Maimbo M Malesu. Impacts of Rainwater Harvesting. A Case study of rainwater harvesting for domestic, livestock, environmental and agricultural use in Kusa. Patricia W. Waterfall. Rainwater harvesting for landscape use. College of Agricultural and Life Sciences, University of Arizona. Second edition, October 2004. Rainwater harvesting supply from the sky. A water conservation guide by Water Conservation Office City of Albuquerque, 1995. The Texas Manual on Rainwater Harvesting. Texas Water Development Board in cooperation with, Chris Brown Consulting, Jan Gerston Consulting Stephen Colley/Architecture, Dr. Hari J. Krishna, P.E., Contract Manager. Third Edition, 2005. Austin, Texas. Ven Te Chow, David R. Maidment and Larry W. Mays. Applied Hydrology. International edition, Civil Engineering Sseries. McGraw-HILL, 1988. Will Critchley, Klaus Siegert, C. Chapman. A manual for the design and construction of water harvesting schemes for plant production. FAO, Rome, 1991.

22

2.3 The Nakuru Sustainable Water and Sanitation Program: A Short

History and Lessons Learned Author: John Boot, Sustain Africa Foundation, johnboot@shaw.ca The Nakuru Sustainable Water and Sanitation Program provides an excellent example of the evolution of partnerships and how these partnerships have the potential to be structured into an institutional structure that will ensure sustainability. The Nakuru Program was initiated by the late John Mbugua. He realized that the supply of potable water was the key first step in alleviating poverty of the rural poor and that rainwater harvesting was a very effective way to supply water. This paper is dedicated to John who strived long and hard with an open mind to find the best solutions for helping the rural poor of Africa.

The Nakuru Situation

As in most of Africa it is a common sight to see women hauling water from local streams and ponds to their homes in the Nakuru countryside. This situation is the result of many factors, some of them unique to the region. First, there are only a few opportunities for shallow wells. The town of Nakuru and a few of the large estates have deep bore holes but these are generally not available for the rural communities. Second, the ground water in low lying areas near Lake Nakuru has very high levels of fluoride making the water unsuitable for consumption. There have been attempts to deliver water to some rural communities via pipelines but these have generally failed due to lack of proper maintenance. Fortunately the area has a reasonable amount of rainfall with dry periods lasting only about 3 months. For this reason the use of water tanks, collecting the water off the roof of homes, was seen as a practical solution for the supply of water. It was found that the best solution was to build a 10,000 liter cement tank, which can provide a family unit of about 10 a supply of potable water under normal weather patterns.

23

A Brief History John Mbugua sought out financial partners to help achieve his goal of helping the rural poor. His first partnership was with the local Anglican and Catholic Churches. These churches supplied funding for building tanks and formed the basis for the Nakuru Program. Initially this partnership worked well but after a few years the church funds were no longer available. To continue the building of tanks, John Mbugua found a second partner: the local Nakuru Rotary Club. Rotary was very effective in that it has a system of linking rich donor clubs in North America, Japan and Europe to recipient clubs in areas of need. These projects also have access to funds from the Rotary Foundation which finances projects throughout the world. Rotary Club of Summerland was the initial donor club to partner with the Nakuru Club and together a more systematic approach to building water tanks was structured. The most important element of the new program was to structure Rotary Community Corps (RCC’s) or Common Interest Groups. Within these RCC’s “Women Groups” were empowered to manage the allocation of tanks. There was a very high demand for water tanks from the start. To ensure commitment beneficiary families were asked to pay a deposit of about $70, supply manual labour and sand, as well as plant 100 trees and install a latrine. Artisans were trained in tank building and trainers were paid from the deposit. The Rotary Clubs supplied the funding for the cement, rebar and hardware (about $350 US$). A single family often could not afford the deposit and so groups were created that pooled their money to qualify for a tank. These groups signed a contract to ensure that until each member was allocated a tank they would be communally owned and the water shared. This required a contract amongst the group and strict guidelines on how to manage the process. Over time (less than 4 years) it became clear that the system of RCC’s and women’s groups provided an excellent channel to deliver training and additional programs to help alleviate poverty. Table banking (micro finance) was introduced and training was supplied systematically. Training programs included HIV/AIDS education, sanitation procedures, water harvesting techniques and better agricultural practices. By 2006 the program had grown into a complex system of 22 RCC’s and over 800 water tanks were built. To ensure that the system would be sustainable an RCC manager, a team of trainers and artisans were formally structured into a team to manage the program. This was a critical step since it was unrealistic to expect the volunteers of the local Rotary Club to manage such a large program. The Nakuru Program had now become well known in the Rotary community and was recognized as one of the most effective systems to alleviate poverty. Recently the Canadian International Development Agency (CIDA) agreed to partner with Rotary to finance the program over a three year period and build an additional 1,200 tanks. This partnership required an even more sophisticated approach since priorities of CIDA, such as minimizing environmental impact, were now a necessary part of the program.

24

The Evolution of Partnerships The Nakuru project provides some interesting insights into how a water supply program has grown and gained greater stability and continuity. Stage One – Non Specific Funding Partnerships The first stage of the program is financed by an agency, in this case a church group, that had a mission to help the poor but did not necessarily have a mandate to develop a systematic program for water management. With church, or church based organizations, it is not entirely unexpected that over time funding for a water project is not maintained. From the time of the early Christians the management of water systems has always been seen as a responsibility of the state or municipal type of organization. Donor churches from North America or Europe would rarely contemplate becoming involved with water supply projects in their home communities. There is a somewhat unique situation in Africa where church groups have taken an expanded role to compensate for lack of alternate sources of funding. However, given the fact that the mission of a church is not to manage such projects over the long term, the partnership may ultimately be non-sustainable. Stage Two – Short Term “Project” Specific Funding Partner In the Nakuru Water Tank Project stage two started when the Rotary Clubs of Nakuru and Summerland took over the project. At this stage the project became more systematic and funding had no other purpose than the stated purpose of providing potable water for the rural poor. The donors have no religious affiliations and are apolitical ensuring water systems were delivered equitably. An RCC structure was put in place, contracts were made and complementary programs were implemented such as table banking. A limitation of the funding during this period was the need to reapply annually for grants. This resulted in funding being somewhat unpredictable making it difficult to make long term plans or hire employees to manage the project.

Stage Three – Long Term “Program” Funding Partner At this stage the project was restructured into a long term “program” with the assistance of another partner – the Canadian International Development Agency (CIDA). This partnership made it possible to develop a long term funding strategy that had at least a three year horizon. At this time the organization that manages the project became formally organized and contract employees now have the confidence of long term funding. Stage Four – A Permanent “Market Driven” System The ultimate goal of programs designed to alleviate poverty is to make sure that the program has an “end point” where the recipients can now enter the “real” economy. In the Nakuru Model, once the full allocation of water tanks is built in the area, there is no need for an organization to manage the water

RCC

RCC

RCC

RCC

RCC

Nakuru Rotary Club

25

systems since the tanks are now privately owned. A private enterprise system of trained artisans would be contracted by the owners to maintain and build additional tanks. Ideally the system of table banking would be supported and a specific micro finance organization formed to ensure continued access of capital. This organization could also maintain the system of trainers and training programs to ensure long term economic growth. These programs can, and should be, self supporting and sustainable over the long term. Maintaining the System to Achieve Environmental and Economic Sustainability The system of RCC’s and women’s groups provides organizations such as Rotary a very effective way to deliver sustainable programs. Having access to potable water is only the first step. Upgrading agricultural capacity through rainwater harvesting techniques and irrigation is important to help families support themselves. An additional focus of future programs is to help reduce energy consumption, primarily the use of firewood for cooking. This can be done in a variety of ways including better cooking facilities. Ideally a local Rotary Club would continue to maintain the RCC structure as an extension of their service. Rotary provides a simple philosophy: “Service Above Self”, and important guidelines such as the Four Way Test (see appendix) that help the women’s groups within the RCC maintain its focus and purpose. “Franchising” of the Nakuru Model The Nakuru model of using a system of RCC’s has proven to be very effective in delivering access to potable water and helping communities to alleviate poverty. Documenting the system and designing effective training programs will make it possible to duplicate the model in a way similar to how fast food franchises have expand systematically. Using the Rotary system of Clubs provides an excellent base to deliver the model systematically throughout Africa.

New Opportunities The “Off the Grid” Potential In the case of family sized water tanks, where water is collected from a specified space on the family property (such as off the roof of a home), there is no long term a requirement for a formal institutional structure; i.e. a Municipal Water Department. This opportunity to sustain a livelihood “Off the Grid” is in some ways a return to an independent lifestyle of years ago. The popularity of cell phones in Africa illustrates the benefits an unwired, off the grid, systems. This concept is also now possible with electricity with solar panels, self generators and energy efficient lighting and appliances. To achieve an off the grid model for rural development there is a need for a program to initiate the process. The Nakuru System has provided the basis for such a development. By structuring Common Interest Groups, helping finance an important first step (access to potable water), structuring a microfinance program and providing systematic training the rural communities now have an opportunity to evolve “Off the Grid”.

RCC

RCC

RCC

RCC

RCC

RCC

RCC

NakuruRotary Club

RCC

RCC

RCC

RCC

RCC

Navaish R.C.

RCC

RCC

RCC

RCC

RCC

KisumuRotary Club

26

Conclusions

Based on the Nakuru experience to date, a number of important conclusions can be drawn:

Improved access to potable water is a critical early step in achieving sustainable development;

It is critical to equip Common Interest Groups (or Rotary Community Corps) with a structure and management tools to ensure success;

Micro Financing provided through this structure proved very effective and expanded the economic

potential of participating families and communities;

A systematic program of training is important to ensure continued growth;

Selection of partners and a clear understanding of a partner’s limitations are critical components to achieving the programs ultimate goals;

Rain Water harvesting technologies provides the basis of developing an independent “Off the

Grid” development strategy that is not dependent on a formal institutional structure.

Appendix

The Four Way Test • Is it the Truth? • Is it Fair to all concerned? • Will it build Goodwill and Better Friendships? • Will it be Beneficial to all concerned?

27

2.4 Institutional Capacity and Partnership Building for Rainwater Management:

The cases of Kigezi Diocese Water and Sanitation Programme, Rakai Women Groups and Community Groups involved in the Domestic Roofwater Harvesting Pilot Project in Uganda Alex Lwakuba and Gloria Karungi Kamugasha, Uganda Rainwater Association (URWA), Plot 15, Stretcher Road, Ntinda, P.O Box 34209, Kampala, Uganda, Tel: 256-41-285654, 256-312-276766, Email: urwa@infocom.co.ug Abstract

The Government of Uganda cherishes partnerships with Non-governmental Organisations (NGOs), Community-Based Organisations (CBOs) and donors in pursuing sustainable development. Indeed, effective collaboration, capacity and partnership building among stakeholders are vital linchpins for successful rainwater management programmes. Partnerships encourage cross fertilisation of ideas, stimulate innovation and promote cost effectiveness in rainwater management. These lessons and more are a result of URWA’s interaction with Kigezi Diocese Water and Sanitation Programme, Rakai women groups as well as the community groups involved in the Domestic Roofwater Harvesting (DRWH) pilot project in Sheema county, Bushenyi district and Bukanga and Isingiro counties of Mbarara district, herein, presented as case studies to demonstrate the benefits associated with institutional capacity and partnership building for rainwater management. The paper starts with a general country background and presents a situational analysis of the Hydrological cycle in the country, the need for government’s partnership with other stakeholders and hence, the origin of URWA, the vision, mission and activities of URWA, interaction of URWA with Kigezi, Rakai and community groups in the districts of Mbarara and Bushenyi, the impact of URWA’s partnerships with these groups in rainwater management, lessons learnt, challenges met and suggested solutions to sustainable partnership building in rainwater management. The paper concludes that only through establishing effective and formidable partnerships, can URWA be assured of vibrant, conspicuous and sustainable rainwater management programmes. Key words: Partnership, Capacity, Build, Rainwater Management. 1.0 Introduction In Uganda (the ‘Pearl of Africa’) only about 12% of the total population of 24million, approximately (National Population and Housing Census of 2002) have access to clean, tap or piped water. This percentage decreases to as low as 3.8% in the rural areas where about 96.2% of the rural population get their water from other available sources like boreholes, unprotected springs, hand-dug shallow wells to mention but a few. These sources dry up sometimes during periods of drought and some of them are dirty, salty and unsafe for human consumption, which leaves Rainwater harvesting as the only available, viable option to remedy this scenario. In her effort to eradicate poverty and improve upon the livelihoods of her citizens, Uganda formulated a National Water Policy and an Action Plan in 1999 after establishing a Water Statute in 1995, to guide water resources management and development in the country. The overall policy objective of Government with respect to water resources management is to manage and develop the water resources of Uganda in an integrated and sustainable manner, so as to secure and provide water of adequate quantity and quality for all social and economic needs of the present and future generations with the full participation of all stakeholders. The directorate of Water Development (DWD) is responsible for the overall implementation of the policy, while the District Local Government, NGOs and the private sector are responsible for provision of water services and maintenance of facilities besides making capital cost contributions. Although the policy regards water for domestic use as top priority over water for production and emphasises groundwater and surface water abstraction for domestic water supply, there is now an increasing attention on rainfall as a source of water supply. For the matter, major Government projects now include and component of rainwater harvesting for institutions like schools and health facilities.

28

Inherent within the National Water Policy (1999) is a call for capacity and partnership building to implement the policy effectively. Consequently, URWA and other NGOs, CBOs involved with issues of water management in the country are legally grounded and are crucial in the policy implementation. Similarly, URWA cherishes partnership building and operates through and or by others. Over the years, URWA has learnt that effective collaboration and capacity building among stakeholders are vital lynchpins for successful and sustainable rainwater management programmes. URWA has further observed that partnerships promote synergy in implementation of activities which in turn encourages cross fertilisation of ideas, stimulates innovation and promotes cost effective and efficiency in rainwater management. URWA has learnt these lessons through interaction with Kigezi Diocese Water and Sanitation Programme, Rakai women groups as well as the community groups involved in the Domestic Roofwater Harvesting (DRWH) pilot project in Sheema county-Bushenyi district, Bukanga and Isingiro counties of Mbarara district. This paper therefore, presents URWA’s experience of interacting with the above CBOs as case studies demonstrating the benefits associated with Institutional Capacity and Partnership Building for rainwater management. 2.0 Country Background 2.1 Geographical Location Uganda is located in the great lakes region of Africa and shares Lake Victoria with Kenya, Tanzania and lakes Albert and Edward with the Democratic Republic of Congo (DRC). Within its boundaries, are lakes such as Kyoga, George and Businia. The country has two main rivers; Kagera and the Nile and it is endowed with many other smaller streams, which drain into wetlands, lakes or form tributaries and sub-tributaries of the major rivers. 2.2 Land Size The country is divided into 56 districts. The districts are further sub divided into 167 counties, 893 sub-counties, 4517 parishes and 39,692 villages. The country occupies an area of 241,038 km2 3 of which 43,941km2 is open water and swamps and 197,097km2 is land. It has the largest fresh water lake-Victoria in Africa (second in the world) and a complex system of rivers that drain into it. And out of it, drains River Nile, the longest river in Africa.

2.3 Population The population is currently estimated at 24.7 million4 of which 87% live in rural areas. Females constitute 51% of population and 49% males. The annual population growth rate is 3.4%. 2.4 Climate Uganda is warm with plenty of sunshine moderated by the relatively high altitude in most parts of the country, like the eastern with Elgon mountain ranges, and north- western highlands. Mean annual temperatures range between 16º C in the south -western highlands, to 25º C in the northwest, but in the north- east, temperatures exceed 30º C for 254 days per year. Uganda has an average rainfall of 1200mm per year with a minimum of 500mm in the semi-arid north-eastern region and a maximum of over 2300mm in Lake Victoria. Roughly, the southern half of the country receives bimodal rainfall with the first rains falling in March–May and the second rains falling in September–November; the Northern half of the country receives unimodal rainfall between April and October/November with a marked decrease in June and July. The low rainfall belt is the flat plain running across from the northeast to the southwest of the country and is referred to as the ‘cattle corridor’. The districts that are located partly or wholly in this cattle corridor are Sembabule, Mpigi, Mbarara, Rakai, Kiboga, Nakasongola, Soroti, Pallisa, Kumi, Katakwi, Moroto, Masaka, Mubende, Luwero, and Nakapiripirit. 2.5 Situational Analysis

3 Uganda statistical Abstract, 2003 4 Projections of the 2002 National Population and Housing Census

29

The above climatic situation has changed greatly. Rainfall is no longer reliable; it is usually poorly distributed and quite erratic. Abnormal prolonged droughts, crop failure and livestock death are now more common and the situation is worse in the cattle corridor. The impact of this unpredictable hydrological water cycle is far-reaching. Declining water levels of lakes and rivers have not only directly affected power generation and the manufacturing industry but also the quantity and quality of domestic water supply. This then calls for affected people to turn to all available natural resources and harness these resources, which makes rainwater the most viable option. The Government of Uganda in its rural water strategy and investment plan covering the years 2001–2015, preference has been given to point sources such as protected springs, hand-pump equipped shallow and deep wells and gravity flow schemes as water supply technology options for rural communities. Rainwater harvesting systems will only be considered for communities, where there are no other viable water options. However, experience over the years has shown that there is great potential for Rainwater harvesting due to a number of reasons which include: i. Favourable rainfall pattern ensures that rainwater is readily accessible for most part of the year

ii. The replacement of traditional roofing (thatch) with impervious materials (e.g. corrugated iron sheets and tiles)

iii. The fact that the user owns, maintains, and controls the system without the need to rely on other community members.

iv. Suitable roof and gutter materials and storage tanks (cisterns) are common products in Uganda.

2.6 Genesis of Uganda Rainwater Association (URWA) Against this background, Uganda Rainwater Association (URWA) was formed in 1999 as an alternative action to raise the profile of Rainwater harvesting (RWH) in the country. URWA’s mandate is to promote rainwater harvesting, management and utilisation for domestic use, agricultural production and environmental conservation in Uganda 2.7 Statement of objectives The Vision of URWA is “to contribute to the improved quality of life of people in Uganda through effective management and utilization of rainwater”. The mission of URWA is “to promote sustainable rainwater management as an option for water supply for domestic, agricultural production and environment conservation as a means of improving the quality of life of the beneficiaries in Uganda” The objectives of URWA are to:

• Promote networking and collaboration between actors in RWH in Uganda,

• Develop strategic partnership with key sector actors to ensure successful promotion of the appropriate RW management systems being developed

• Advocate for stronger commitment to RWH in the water, agricultural, and environment sectors.

• Strengthen the capacity of members to deliver services (implement RWH activities).

• Mobilise resources for rainwater harvesting activities.

URWA’s strength lies in its ability to support communities to improve their socio-economic situation through mobilisation, information, skills an experience sharing. As part of its objectives to build and strengthen capacity of members to deliver services, URWA has worked with a number of groups among which are the Rakai Women Groups and Kigezi Diocese Water and Sanitation Programme. This paper briefly gives the experiences of URWA’s interaction with the two groups. 3.0 Cases of Partnership Building for Rainwater Management 3.1 Kigezi Diocese Water and Sanitation Programme (KDWSP)

30

Besides spiritual aspects, Kigezi Diocese is also involved in development work particularly water provision for its people through the Kigezi Diocese Water and Sanitation Programme (KDWSP). Kigezi Diocese Water and Sanitation Programme (KDWSP) is a community water supply and sanitation programme serving the needs of the predominantly rural population of Kabale district about 617,501 Ugandans5. The programme was initiated in 1983 through contact between the late Bishop Festo Kivengere and Tear Fund. KDWSP works with communities on the basis of prioritised demand and self help. The key programme activities are community mobilisation, water supply and hygiene promotion, capacity building to Community Based Organisations (CBOs) and advocacy. It also addresses wider needs such as income poverty, food security, agriculture and soil conservation either directly or through linkages to other Diocesan programmes or those of other NGOs or Government. Problem Kigezi is a hilly area with mountains ranging in heights from 1800 to 2000 metres above sea level and has a variety of valleys including V-shaped valleys, interlocking spurs and plateaus characterised by heavy rainfall (average 1000mm) spread over three wet seasons. Common water sources in the area include low-lying springs and rivers/streams. Even with the protection of springs and construction of gravity flow schemes KDWSP could not meet its objective of reducing distance to water points and time spent in water hauling. It is against this technology that the programme adopted rainwater harvesting. Under this KDWSP constructs 420litre capacity rainwater jars, 4000litre capacity Ferro-cement rainwater tanks at households and 15,000-20,000litre capacity brick masonry tanks at institutions. 3.2 Collaboration with Uganda Rainwater Association (URWA) The programme, through its membership and involvement in URWA learnt about the training of women groups in tank construction, which had taken place in Rakai district. An interested group; Buyanga Women’s Group with membership of 25 in Kamwezi sub-county was identified by the programme to be taken to Rakai for a study tour. Subsequent to the study tour, they expressed a need to take up the technology and were trained by the programme masons in tank construction. Activities included study tours and training. 3.3 Impact of the URWA/KDWSP Partnership

1) Once capacity was built, the women’s group trained Nzungu and Bishaki women’s groups comprising of 19 and 26members respectively, in Muko sub-county. The group also trained Kantare women’s group comprising of 16members.

2) The Muko sub-county groups together with the Kamwezi group later joined together to train Kagaana group with 110members. Of the 110 trained, 64 have Ferro-cement tanks while 46 have rainwater jars at their households.

3) Bishaki women’s group also trained Kakore women’s group in Hamurwa sub-county operating under the Two Wings Agro-forestry (NGO).

4) The Kamwezi group trained groups outside the district in Ntungamo and Rukungiri under Ankole Diocese programmes.

5) Another group in Nangara sub-county also trained groups in Bushenyi district and one in Kagarama (Hamurambi women’s group with 80members.

6) As a result of the successful trainings, the Catholic Diocese also approached KDWSP for training in jar construction.

3.4 Rakai Women Groups Rakai district lies in the South-western part of Uganda about 200km SW of Kampala and has four counties (Kabula, Kakuto, Kyotera and Kooki) and 22 Sub-counties. The major problem in Kooki County, where the women groups are based, is lack of clean water. According to Ugandan government and World Health Organisation standards, the highly mineralised groundwater in this area is unsafe for human consumption. Local people are forced to use contaminated surface water and as a result water borne diseases are very common. Against this background, the Regional Land Management Unit (RELMA) and Rakai district administration agreed to explore rainwater harvesting as a means to providing safe water to the communities in Kooki County. As a result of this Katuntu Twekembe and Bakyala Kwekulakulanya women groups were formed and trained in tank construction and took up the technology.

5 Population projections per district as at December 31st 1999

31

The problem/Need From the inset, the groups were brought together by a common problem of lack of clean and safe water. The people (women in particular) had to walk long distances to ponds, where they drew dirty water. All the women focused on the shared vision of improved living standards and clean drinking water. Bakyala Kwekulakulanya and Katuntu Twekembe women groups had been existent since 1992 and 1995 respectively and were formed specifically for home improvement (construction of pit latrines, drying racks and general home hygiene), improving household incomes through a revolving fund, helping the disabled and orphans and improvement of women’s status through money-making activities especially baking. 3.5 Collaboration with Uganda Rainwater Association (URWA) Once the water need was identified in Kooki County, RELMA, through URWA sponsored six Kenyan women and a technician to train the women in rainwater tank construction in 1997. The training lasted for a period of two weeks. Members of Bakyala Kwekulakulanya and Katuntu Twekembe were trained in two technologies that is construction of a 2000litre jar and a 3500litre Ferro-cement tank. After the training, the two women groups started a programme to build tanks for each of the households. The District Water Officer assisted with follow up and quality assurance of the tanks constructed. Activities included training and technical support and study tours. 3.6 Impact of partnership between URWA and the Rakai Women Groups

1) The two groups trained in Rakai district later trained another group Basooka Kwavula women’s group which took up rainwater tank construction and a Technical Report6 published in 2002 revealed that there were 26women groups in the district engaged in rainwater tank construction.

2) They were responsible for training the Kigezi Diocese Community women groups and other groups in other districts including Rukungiri, Kisoro and recently in Mpigi, Busia and Tororo districts

3) They are contracted to construct rainwater tanks by the district and to train other groups at a fee. 4) The groups not only stopped at domestic RWH but also diversified to include water for

agricultural production and livestock production besides carrying out other activities like farming, piggery, poultry farming, handcrafts, jam making, coffee growing etc as a result of increased incomes from tank construction.

5) Each of the groups to date has a code of conduct and management committee and a disciplinary committee enforces the bylaws. For instance when a member is contracted to construct a rainwater system, the proceeds are brought and shared equally among the group members.

6) By early 20007, a total of 54jars and 87tanks had been constructed with many more under construction. To date, all group members have one or two rainwater tanks at home depending on the initial size.

3.7 Community-led Domestic Roofwater Harvesting (DRWH) Pilot Project in Mbarara and Bushenyi districts The Domestic Rainwater Harvesting pilot project was implemented in the districts of Mbarara covering 11 sub counties of Bukanga and Isingiro Counties8 and Bushenyi covering 3 sub-counties of Bugongi, Shuuku and Kitagata in Sheema South between June 2004 and December 2005 by Ankole Diocese and by Agency for Cooperation and Research in Development (ACORD) respectively. This was the first time that Government of Uganda through the Directorate of Water Development (DWD) invested resources into subsidizing individual water sources at household level. As part of the overall national strategy development for DRWH from a study commissioned by DWD in 2003, it was recommended that a pilot programme be implemented to yield information on the viability of different proposed approaches. The objective of the pilot programme was therefore to test the recommended community led rainwater harvesting approaches from the study before large-scale application by the government. Specifically the project was designed to:

• Test a community-led approach to rainwater harvesting and document results

6 Empowering Rural Communities: Rainwater Harvesting by Women Groups in Rakai District, Uganda (Technical Report No. 29 7 Empowering Rural Communities: Rainwater Harvesting by Women Groups in Rakai District, Uganda (Technical Report No. 29) 8 The counties are Birere, Masha, Kabingo, Nyakitunda, Kikagate in Isingiro County and Ngarama, Kashumba, Rugaaga, Endinzi and Mbaare in Bukanga County.

32

• Test implementation of government co-funded community water supply programmes through NGOs

• Empower communities in the pilot project areas with clean and safe water technologies through installation of domestic Roofwater harvesting systems.

Problem/Need Bukanga and Isingiro are the most water stressed counties in Mbarara district. The two counties lie in the dry belt, and thus receive an annual rainfall of between 1000 to 750mm9. The main social activity of the people is livestock keeping. They have the lowest sanitation and safe water coverage in the district. The current safe water coverage in the area is about 26% only compared to the district average of 84%10. Shallow well potential is almost non-existent; boreholes have a low potential and the few that have been drilled cannot meet the large demand for water. 27%11 of the boreholes reportedly have a high level mineralization (iron and manganese) so the quality of water from them is objectionable. Sheema County in Bushenyi district on the other hand is characterised by hills and valleys and most people live on top of the hills. The water points that people are depending upon are springs; shallow wells, ponds and streams are found in the valleys hence it is cumbersome for people to collect water down in the valleys. The official safe water coverage is 77% 12but because of the terrain the benefits are not fully reaped. Sheema was chosen because it offered the opportunity for testing approaches for an area with high iron roof coverage and relatively affluent communities. Collaboration with Uganda Rainwater Association (URWA) Uganda Rainwater Association (URWA) was at the forefront in championing for the pilot project through a good working relationship and goodwill existent between URWA and DWD. As a result URWA was contracted as the overall overseer and Technical advisor and was responsible for;

• Soliciting and evaluating proposals from prospective implementing NGOs in the project districts; • Monitoring and evaluating the pilot project as to whether it was meeting the set targets and

outputs; and • Carrying out regular technical backstopping for the project.

URWA worked in partnership with Uganda Water and Sanitation NGO Network (UWASNET) through which the funds were channelled, WaterAid Uganda the fund manager and the DWD; the project donor. Activities carried out under the project The project made use of the decentralised government structures that is district, sub-county and parish and village level structures in implementation and worked with communities using community structures (groups) where existent and forming new ones where non-existent. Specific strategies used in the pilot included;

• Community mobilisation • Community sensitisation through exposure visits and workshops • Technology demonstration and hands-on training • Rainwater harvesting system production for selected beneficiary households (cost-shared).

Impact of the DRWH pilot project

1) Community empowerment especially for the groups involved in the pilot project; the groups were empowered in budgeting for themselves, procurement of materials for RWH systems and record keeping. Records of contributions from group members and how much they have used on each rainwater system were kept by each group to ease documentation and some groups even had bye-laws drafted to guide the operation of the groups.

2) Skills transfer through training of masons; over 30 masons were trained in both districts by the end of the project. It is hoped that as resource persons in RWH within their communities, they

9 A concept note for Piloting of Rainwater Harvesting for Domestic Water Supply 10 Water and Sanitation Sector Performance Report 2005 11 A concept note for Piloting of Rainwater Harvesting for Domestic Water Supply 12 Water and Sanitation Sector Performance Report 2005

33

will use the skills acquired to earn a living as the demand for RWH systems grows in the project areas.

3) 385 rainwater tanks were constructed at the end of the project; 108 and 277 in Bushenyi and Mbarara districts respectively. Only 49 of these were demonstration tanks, the rest were cost-shared by the groups. Interesting to note is the fact that most people paid more to construct larger tank capacities than the 5m3 and 6m3 set by the project for the two districts; some as large as 12m3 and 20m3.

4) Demand for RWH systems; by the end of the project, Ankole diocese and ACORD reported that other people had organised themselves into groups and written demanding for support to also construct RWH systems.

5) Increased awareness of RWH and its benefits through the initial sensitisations carried out by the implementing NGOs in the districts and demonstrations carried out in the initial stages of the project coupled with the exposure visits to areas where RWH activities have been successfully implemented (‘seeing believing’).

6) Strengthened partnership between URWA and the implementing NGOs through working together. As a result, due to the professionalism and competence exhibited by ACORD during the pilot, URWA will partner with ACORD to implement the project promoting Domestic Roofwater Harvesting for HIV AIDS/TB affected households to be implemented in Masaka and Rakai districts with support from the World Bank.

7) Strengthened institutional capacity for the groups and implementing NGOs especially Ankole Diocese, which had never implemented a RWH project of such magnitude. Ankole diocese was able to improve its documentation skills through constant mentoring and was also able to mainstream its financial reporting guidelines to suit the project hence strengthened capacity.

4.0 Lessons Learnt 1) Community involvement to enhances project ownership and sustainability 2) Group action/cohesion (particularly existent groups) is more sustainable in the implementation of

rainwater harvesting initiatives because individuals do not have to shoulder the costs on their own, which would make replication difficult. This is evidenced in the pilot project where individuals were able to add more funds to the subsidy given to construct larger capacity tanks ranging from 8m3, 10m3, 11m3, 12m3 and 20m3.

3) With some support communities and groups are willing to take up the technology due to its associated benefits of convenience, reduced burden of water hauling and reduced incidence of water-borne diseases.

4) Domestic roofwater harvesting is easily promoted where other alternative water sources are not available or where people have to move long distances to the point water source.

5) Study/exposure visits should be encouraged as a means of experience sharing and for learning purposes.

6) Timely disbursement of funds is imperative if activities are to continue as planned and project benefits/outputs maximised. Delays in disbursement of funds demoralises the communities at times leading to withdrawal from the project.

7) Goodwill from the local leadership is vital for project success. Once the support of the local leadership has been gained, there are less implementation ‘roadblocks’ and they can even monitor activities when NGO staff are not in the field.

5.0 Challenges in capacity and partnership building for Rainwater Management 1) Documentation: a lot of information has been generated especially by the Kigezi Diocese

projects, which deserves to be documented for learning purposes. The Rakai experiences were documented through SIDA/RELMA but the KDWSP experiences are yet to be documented. A proposal submitted to SearNet in 2005 was not successful and efforts to seek alternative funding have been unsuccessful.

2) Support: more support is needed for groups to train other groups and also support some marginalized groups in the communities

3) Partnership sustainability: it is difficult to maintain partnerships especially after the project close because there are usually hardly any funds allocated for the same.

4) Follow up issues: it is not easy to follow up progress after the project has ended due to financial constraints and thus there is difficulty in knowing whether more was done after the project close.

6.0 Suggested solutions/Remedies

34

1) Country Associations should seek funding and budget for documentation of experiences so as to share them with others. Associations should also invest in documentation tools like still and photo cameras.

2) Associations should also come up with clear mechanisms for supporting local NGOs/CBOs for instance through providing seed funds which can be used to support other households which are not able to contribute all the tank costs

3) On monitoring and follow up issues, associations should build organisational capacity to monitor RWH activities in their areas of operation or even use NGOs and local CBOs who in the project areas to monitor progress. An example of this is seen in the URWA /ACORD partnership; where URWA used ACORD to implement and monitor progress during the pilot project.

4) On sustainability of partnerships, associations should devise cost-effective strategies/mechanisms with low financial input including email, telephone, fax etc to keep partners informed about what is going on even when a project has ended.

7.0 Conclusion Generally, the major constraint experienced by most NGOs and even Government Institutions is lack of inadequate capacity-financial, human, material etc. However, the key solution to the challenge is to work with the support of or through others. Therefore, only through establishing effective and formidable partnerships can URWA and other SearNet member countries be assured of vibrant, conspicuous and sustainable rainwater management programmes.

35

2.5 ANALYSIS OF Social, Economic and Institutional Issues Affecting Utilization of Rainwater Harvesting Technology, Eastern Tigray, Ethiopia

Abadi Teklehaimanot, Agricultural Technical and Vocational Education and Training College, P.O.Box 372, Mekele, Ethiopia. abadi372@yahoo.com and Tesfaye Beshah, PhD, Haramaya University, Department of Rural Development & Agricultural Extension, P.O.Box 290, Haramaya University, Ethiopia. btesfaye2003@yahoo.com Abstract The main factors constraining agriculture and rural development are low productivity and output, the performance of rain fed agriculture aggravated by intermittent drought and famine arising from the vagarious of nature. The poverty and environmental degradation demands constant efforts to improve the effectiveness of agriculture and natural resource management (NRM). Therefore, rainwater-harvesting technology (RWHT) plays important role in reducing crop failure and soil erosion. This study examines the analysis of utilization of rainwater harvesting technology. The study is motivated by the belief that the constraints of the low productivity leads to poverty and famine cannot be overcome by simply concentrating on the rain fed agriculture therefore the rainwater issue needs to be addressed as well. In the course of this study primary data were collected from 201 households out of which 101 were users of RWHT while the rest 100 sampled households were engaged only in rainfed agriculture. In this study multistage sampling technique was employed. In the first stage two wereda were selected purposefully and six Tabias, selected randomly. Households in the sample Tabia were stratified as user and nonuser of RWHT. From the stratified households sample respondents were selected using probability proportional to size method. Descriptive statistics such as mean, standard deviation and percentage were used to describe sampled respondents in terms of some desirable variables. A binary logit model was also used to analyze the determinants of the utilization of the RWHT. Fourteen variables were included in the model of which eight were found significant at (P<0.10) probability level. Extension contacts, training, animal product income, market distance, location, cash availability, farmland size and input were found to be highly important variables influencing utilization of RWHT. Additionally 18 items were selected and 16 of them were analyzed using attitude scale (1-5). As a result of this, RWHT demand of labour, cost, land, skill and knowledge were found to be highly important items related to utilization of RWHT. The item RWHT take large area & increase cost shows significance difference at (P<0.05) and (P<0.10) probability level. Moreover, the grand mean for both categories were found to be 3.46, which shows favorable attitude. The scale statistics mean score of the sampled household calculated of 69.1 percent indicated that there were favorable attitude towards RWHT. The plausible explanation implies that for both group of the users and nonusers there may be some thresholds influencing RWHT utilization as the result of the favorable attitude. The forgoing discussion has revealed that RWHT activity, which includes trapezoidal pond and percolated pond, is widely undertaken in the study area. Households involved in those activities could benefit more if they got favorable environment for utilization the RWHT. The main bottlenecks that hamper the development of RWHT activities required by the farmer include knowledge, capital, raw material and access to market. The finding of this study indicates that the social economical psychological conditions of farmers differ from farmer to farmer. Moreover, the existence of difference among farmers implied by difference in perception, opinion attitude and decision in the allocation of the scarce resources and utilization of the technology are among the major finding of this study. Key Words: Rain Water Harvesting, Utilization, Tigray, Ethiopia.

1. Introduction Ethiopia has a vast water resource potential yet only one percent of the estimated annual surface water of the 110 billion cubic meters is used for irrigation and hydropower (Alamneh, 2003). The agricultural growth rate of the country is low as compared to the rate of the population growth of three percent. Consequently, the country’s agriculture becomes highly dependency on rain-fed agriculture. As land pressure rises, more and more marginal areas in the world are being used for agriculture. Much of this land is located in the arid or semi-arid belts where rainfall is irregular and much of the precious water is soon lost as surface runoff. Recent droughts have highlighted the risks to human beings and livestock, which occur when, rains fail. While irrigation may be the most obvious response to drought, it has proved costly and can only benefit a fortunate few. Nowadays, there is increasing interest in the country in the low cost alternative-generally referred to as ‘water harvesting’. The latter refers to a practice of inducing, collecting, storing and conserving local surface runoff for agricultural production (Nigigi, 2003).

36

The study was carried out in Tigray, northern Ethiopia (Figure 1). Most of the hydrographic 87 percent of Tigray is associated with the Mediterranean hydrological regime. Only 13 percent of the total area of 54572.6 square kilometers of the region drains to the Indian Ocean. There are western and eastern hydrographic regions and three main river basins; the Tekeze, Mereb, and Afar. As a result of ITCZ movements, there are two main seasons in Tigray. Kiremt long rainy seasons, mid June unto mid September and Hagay long dry period; October to may but, in addition, some areas also have two less pronounced seasons: Kiwie dry and low temperature, october – January and Belgi, low amount of rainfall in eastern part of the region, February-May (Hunting, 1976a). About 77 percent of Tigray falls within a slope range of between 0-8 percent. Steeper areas than this are found around northeastern part of the escarpment in Atsbi-wenberta, Erob and at the foot of the escarpment of Ofla, Alaje and Welkait-Tsegede.

2. OBJECTIVES OF THE STUDY

The overall objective of this study was to provide the regional development practitioners, decision-makers and other stakeholders information on factors that affect utilization of rainwater harvesting technology that is given a very high attention in the region. Specific objectives of the study were: 1. To identify the socio-psychological factors influencing utilization of rainwater harvesting technologies. 2. To analyze the institutional, technical and economic factors affecting utilization of rainwater harvesting

technologies. 3.0 Theoretical and Analytical Frameworks Theoretical framework used in this study mainly comes from the classical diffusion of innovation schools (Rogers, 1995), but also recent insights on the human behaviour (Leeuwis, 2004), among others. Concepts and theoretical under-pinning from these sources were captured as an analytical framework to guide the study at the field level (Uphoff, 1986; Ervin and Ervin, 1982; Ostram, 1990 Critichley and Sieger, 1991, Tesfaye, 2003).

Laelay_MaychoAsgedo_Tsembela

Welkayet

Tegedi

Kafta_Humera

Kola_Temben Tselemet

Medebay_ZanaNaeder_Adet

Laelay_AdiyaboTahetay_Adiyabo

Tahetay_MaychoTahetay_Keraro

Mereb_Lehe

Enderta

Wefla

Samre Abergele

Raya_Azebo

Alamata

Alaja

Hentalo_Wajerat

Ende_Mehoni

Hawzen Wereilehe

Dega_Temben

Adewa

Entecho

Saesie_Tsaeda_Amba

Wukero Atsbi_Wenberta

Golomeheda Gentafeshum

Erob

Fig 1: Map of Administrative Weredas of Tigray Region, Ethiopia

The study area

37

3.1 Research Methodology • Sampling Design

The study employed a multi-stage stratified sampling design. A sampling frame of the study was drawn from two Weredas selected purposively from eastern Tigray, out of nine Weredas in the zone. The purpose of selection is a wide practice of rain water harvesting among farmers. The selected Weredas are Atsbi-Wenberta and Wukro. Farmers households in the two Weredas were stratified according to their participation in the rain water harvesting practices. Those who use RWHT were designated as “users” whereas those who do not use any improved rain water harvesting practices were categorized as “non-users.” Finally, 101 and 100 heads of farmers households were selected respectively from users and non-users, using probability proportional to size technique.

• Data Collection and Analysis The study employed both qualitative and quantitative data. Quantitative data was collected using structured questionnaire that was duly tested and thoroughly improved. In addition, a likert type attitude scale of 1-5 was developing to assess the underlying attitude of the community on rain water harvesting technology. Secondary data were collected from relevant sources, such as reports, socioeconomic survey documents of the area, maps, books and Non Governmental Organizations (NGO). Data was analyzed using descriptive statistics and a binary logit model. Qualitative data was used to specify contexts of the study and enrich information generated from quantitative data analysis.

• Variables in the model Dependent variable of the model: The dependent variable for logit analysis was RWHT utilization, which is dichotomous. It is represented in the model by (1) for those farmers who are users of micro pond (trapezoid and percolated) RWHT practiced and (0) otherwise for those farmers who are not using any RWH practice. Independent variable: Based on literature review and researchers personal experience, the following factors, which are expected to influence the RWH practices, are presented with their operationalization. Continues variables: Education, Labour, market distance, non-farm income, extension contact, animal product and farm size. Discrete variables: Fertilizer input, land security, experience in RWHT, Slope of the farm, land location of the RWHT, rainwater retention, belief of the farmers, credit, cash for down payment, training, soil type, assistance, type of crop grown, RWHT use & purpose. The variables used in the model were tested for multi-colinearity. Accordingly, variance inflation factor (VIF) for continuous variables and contingency co-efficient for qualitative variables were tested. 4.0 Results and Discussion 4.1 Characteristics of Sampled households Among the sampled households, 58.7 percent were males, while the remaining 41.3 percent were females. The average family size of the sampled households was about 6.5; the largest family size being 11 and the smallest being three. The average number of economically active family members 15-64 years of age was 2.74 for user and 2.23 for non-user of RWHT. The age structure of the sample households showed that the average age of the users farmers was about 45.95 years whereas that of non-users was 42.95 years (Table 1). Table 1: Mean of some socio-economic variables Description Total Users Non-users Significance Family size (numbers) 6.5 6.9 6.0 ns Age 44 45.9 42.9 * Active labour force (numbers) 2.5 2.74 2.23 ** Education (years) 2.08 2.37 1.80 ns Total land owned (ha) 1.08 1.19 0.97 ** Total cultivated land (ha) 0.62 0.68 0.58 *

38

Total livestock owned (TLU) 2.08 2.54 1.56 ** *** Significant at 1%, ** 5%, and * 10% probability level

The educational status of the sampled households indicates that 45.4 percent of the users farmers were illiterate, while the remaining 54.6 percent were literate. On the other hand, 60 percent and 40 percent of the non-users farmers were illiterate and literate, respectively. The total average educational level was 2.08 years of schooling with standard deviation of 2.59. The mean educational level of users and nonusers was 2.37 and 1.80 with standard deviation of 2.61 and 2.54 respectively. The average land size of sample households was 1.08 ha of which users of RWHT and nonusers own on average 1.19 and 0.97 ha respectively (Table 1). With regard to land use patterns, from the total land holdings sampled households allocated on average 0.62 ha cultivable land, 0.05ha forest land, 0.06 ha grazing area, 0.24 ha homestead and 0.05 ha perennial crop land. The T-tests indicate the mean difference of the perennial cropland holding is significant at (P<0.01) probability level. Wheat barley and pulses are the principal crop grown in the area, which ranked first, second and third respectively. Among, of the respondents 44.2 percent reported to have decreasing trend of production during the last 10 years. Out of a total agricultural production in the study area livestock and beekeeping contributed to 25 percent of household income. The mean size of the TLU of the sampled farmers was 2.08 with standard deviation of 3.57. The user farmers on the average had 2.54 TLU, while the non-users had 1.56 TLU which was significant at (P<0.01) (Table 1). The major livestock problem in the study area was lack of grazing. Moreover, 50.5 percent and 68 percent of users and nonuser farmers encountered oxen shortage, respectively. About 30.3 percent of the sample farmers apply the improved inputs, 40.6 percent being those who are user in the RWHT activities while the other 20 percent are nonusers. The major inputs applied by the farmers are chemical fertilizers 30.3 percent, improved seeds three percent, herbicides and pesticides 0.05 percent. Among the sampled farmers, 66.2 percent applied manure on their farm. The average amount of DAP and Urea fertilizer used by the sampled farmers was 15.26 kg and 12.87 kg respectively. On average user farmers applied 19.19 kg and 16.56 kg DAP and Urea. With regard to pesticide and improved seeds the mean amount used by the sampled farmers was 0.53 liter and 22.28 kg, respectively. About 86 farm households 42.8 percent reported that they were engaged in non-farm activities besides farming. The mean income from non-farm activities was found to be 114.4 Birr per year. The average number of extension contact for users and nonusers of RWHT was 7.78 and 2.38 with standard deviation of 12.38 and 3.97 respectfully, these shows statistically significant difference at (P<0.01) probability level. About 72.3 percent of RWHT users and 37 percent of non-users farmers took part in extension program. The chi-square test indicated that there was significant difference at (P<0.01) probability level in the participation of the extension service between the users and non-users sampled households. The average distance of farmer training and extension center was found to be 3.19 km and 2.79 km with standard deviation of 2.90 and 4.40 for users and non-users of RWHT. Furthermore; in 2005 production year, 63.7 percent of the sampled households benefited from the visits while the development agents did not visit 36.3 percent of the farmers field or their home (for details of socio-economic characteristics of the sampled farmers, see Abadi (2006). 4.2 Factors Affecting Utilization of RWHT Utilization of RWUT is affected by institutional, psychological, social, economic and physical factors. These factors are briefly outlined below (for details, see Abadi (2006).

Institutional factors: Organizational cultures, hierarchy of decision making which ranges from the lowest (Tabia) level to Wereda and Region. In some situation zonal level also provides technical backstopping. In this connection availability of manpower at each level and work experiences are very crucial for the implementation of RWHT.

39

Psychological factors: In order to understand farmers’ reaction to RWHT understanding their attitude toward the practice plays an important role. Thus, eighteen items were developed and administered to 201 sampled households. A Likert scale was used in this procedure. The items were assessed by experts in the field for their construct validity. Farmer’s perception of the moisture status, location and topography of their farm land and their attitude towards RWHT technology are those partial indicators of the utilization of the technology. List of items used were; RWHT is important to secure food, RWHT is appropriate technology, RWHT is profitable technology, RWHT can help improve livelihood, RWHT increases yield, RWHT is labour demanding, RWHT sustains production, RWH structures & design are easy to implement, RWHT cause animal health problem, Rain-fed agriculture is sufficient to produce enough food, RWHT increase cost, RWHT demand much knowledge & skills, possible to sustain production with out RWHT, RWHT take large area, RWHT cause human health problem (malaria) and indigenous knowledge is superior to new RWHT. Reliability analysis of items considered in the scale was carried out where Alpha (α) was 0.598. Therefore, in order to maintain optimum trade-off between brevity and reliability, items that increased the Alpha value were dropped from the scale. Accordingly, two items that increased and Alpha value were dropped from the list. For the subsequent analysis, the remaining 16 items were grouped into users and non-users group. The response of the users and nonusers for item stating “RWHT increase cost” and “RWHT take a large area” shows a significant difference at (P<0.01) and (P<0.05) probability level respectively. The grand mean 3.46 represents favorable attitude towards RWHT and it lies between favorable attitude towards RWHT which lies between neutral and agree. Generally, the analysis revealed that using 16-likert item show favorable attitude towards the RWHT. Both group the users and nonusers believe that RWHT increase cost, require managerial skill and knowledge as well as wasteland for RWHT structure construction, which is unfavorable attitude for both users and nonusers of the RWHT. This shows that farmers have low perceived control over the technology. However, they mostly believe that they can acquire the knowledge and skill the technology demands. The nonuser farmers explain that almost 95 percent of them had a positive attitude to the practice of RWHT. Taking their resources and implementing capacity vis-à-vis the perceived reality, about 57 percent of them planed to practice the RWHT. Believe of users sample household on utilization of RWHT indicate that all farmers perceive RWHT positively. Sampled households participation in the RWHT based on voluntary and compulsory basis were 86 and 14 percent respectively. Nevertheless, those who practiced irrigation using RWHT were only 58 percent. The worth of covering RWHT structure cost, investing from their own were 43 percent of the sampled farmers, and the rest were not interested to invest. With respect to the worth of getting assistance for the RWHT structure, about 57 percent were willing. Social factors: The sampled households have reported various use of the water harvested using RWHT. These include, irrigation (58%), domestic use by family (27%), sanitation (13%), and for animals (2%). The linkages of RWHT with watershed management were more indirect than envisaged. However, women’s access to rainwater have social benefit such as improved health, income and saved time to undertake the social role and production activities. This may power and opportunity to use their land to FHH's with out entering share cropping with male-headed household. Moreover, utilization of rainwater for family, sanitation and irrigation increased as the rainwater availability is near by to their residence unlike other distant water sources. Economic factors: Sampled farmers explain that, the constraints for implementing the current RWHT structure were finance (60 percent), Know-how (20 percent), technical (13 percent), and labour (7 percent), respectively. As compared to rainfed agriculture and livestock, respondents believed that RWHT was better in terms of generating income that avoids risk & uncertainties, but rainfed is preferred in terms of labor requirements. Farmers reported a number of problems associated with RWHT activities, lack of capital, technology, lack of skill, labour, and market problem.

40

Figure 5: Percolated pond (Photo Abadi T.H, 2005)

Physical factors: Physical factors of RWHT such as site selection have serious repercussions on the performance of the RWH structures and thereby on farmers’ decisions. For instance, ponds whose surface is not covered with cement, compacted-clay or plastic were found to be ineffective because they lose water in few months. Farmer’s preferences with respect structures leans towards eala (percolated pond) 54 percent, horoyo (trapezoidal pond with plastic) 27 percent, baska (rectangular pond) 12 percent, gidib (dam water) five percent, and degdag (trapezoidal pond clay blanketed) two percent. In the six Tabias of the study 21% of the RWH structures were blanketed with plastic, seven percent cemented (Figure 3), 43 percent clay blanketed (Figure 2), 29 percent percolated pond type . All constructed ponds have trapezoidal cross section and square plan while the percolated ponds have the shape of the land escape. Seepage water loss from ponds was identified as one of the critical issues. This was overcome through the provision of cement, compacted-clay lining and/or installation of plastic sheeting. It was found that farmers do not believe that clay-lined ponds will hold water unless they are covered with plastic.

Many structures hold water for some period during the past seasons, after the on set of the main rainy season in July. Few ponds hold water only till September and others till October. Few ponds that hold water for longer period last till January. Variations in the design were observed in the field, which is derived because of the physical factors. Inefficient utilization of the plastic was observed which resulted from poor shaping of the ponds, irregular top width and lack of beams and over sizing of the plastic for the pond dimensions. Variation in water holding capacity also observed. This emanates due to variation in the construction quality of the pond. Most of the clay blanketed and cemented pond are considered by farmers as ineffective in retaining water after the rainy season. Ponds closer to the houses receive good attention and follow up from the households who own them, especially from the women. Those in the middle of the arable holding receive the same attention as the arable crops and this is neither effective for the vegetable production nor an efficient utilization of the water stored. Sampled households opinion regarding similarity of RWH designs, 62 percent and 41 percent respectively replied that the trapezoidal micro pond and percolated type pond as suitable for the locations selected (Figures 4 and 5). Corresponding figures for those who replied the locations were unsuitable 38 percent and 59 percent.

Figure 4: Trapezoidal pond blanketed with plastic sheet (Photo Ababi T.H, 2005)

Figure 3: Trapezoidal pond with cement (Photo Abadi T.H, 2005)

Figure 2: Clay blanketed Trapezoidal pond (Photo Abadi T.H, 2005

41

Almost all irrigated plots are located near the ponds, both upslope and down the slope. Most farmers were abstracting irrigation water for utilization directly from the ponds using old jerry and oil cans and applying the water either directly to the plants or via unlined furrows and basins. Few farmers do use tridle pump and drip irrigation. Some farmers have already developed their own labor saving devices using such devices as rope lining. It should be noted that there were a number of incidents of drowning of people and animals in the ponds that were reported by farmers in the study area. Excavated infertile sub soil from pond area has been both deposited on agricultural land and also heaped near to the pond. This has on the one hand occupied rather scarce agricultural land and rendered unproductive by covering good topsoil. All of the ponds are uncovered and which aggravates incidences of malaria in the residential areas. Obviously, these practices should be accounted for when benefits derived from RWHT is calculated. Results of Econometric model Logit model was used to analyze the determinants of farmers RWHT utilization. The farm household either utilizes or not used RWHT. The variable to show utilization of RWH activity was used as a binary dependent variable, taking a value 1 indicating the farmer is utilizing at least one or more micro pond activities and, 0 other wise. Fourteen explanatory variables (seven continuous and seven dummy) were included in the model. Prior to running the logistic regression analysis, both the continuous and discrete explanatory variables were checked for the existence of multi-co linearity using Variance Inflation Factor (VIF) and contingency coefficients, respectively. It was apparent that there is no strong association among the variables, for this reason, all of the explanatory variables were included in the final analysis. Then, the Maximum Likelihood method of Estimation (MLE) was used to elicit the parameter estimates of the binominal logistic regression model. Out of the fourteen explanatory variables hypothesized to influence utilization of RWHT activity in the study area, eight were found to be significant at less than or equal to ten percent probability level. The model results show that the logistic regression model correctly predicted 148 of 201, or 78.4 percent of the sample households. The sensitivity (correctly predicted Rainwater harvesting users) and the specificity (correctly predicted non-users) of the logit model are 79.7 and 76.8 percent, respectively. The significant explanatory variables included: Market distance, input, cash availability, location, training, animal product income, extension contacts and farmland size. Each of these variables is briefly discussed below (also see, Abadi (2006). Distance from market center: The variable is significant at (P< 0. 05) and related negatively with the farmers desire to involve in the RWHT activity. The odds ratio (0.938) indicates that under constant assumption the utilization of RWHT decrease by a factor of (0.938) as the distance of the homestead from the market center increases by 1 km distance. Input: Availability of input is highly important when farmers are ready to adopt new technology. Inputs are positively and significantly related to the utilization of RWHT (P< 0. 10). The positive relationship shows that the odds ratio in favor on the probability of utilization of the RWHT increases by a factor of (2.214) as availability of input increases by one unit. Cash availability: Finance have positively related to the utilization of RWHT by farmers (P< 0. 05). Other things being constant, the decision to use RWHT increases by a factor of (5.139), as availability of cash increases by one unit.

Table 4. Parameter estimates of the logistic regression model (n=201)

Explanatory variable

Estimated coefficients B S.E.

Wald statistics Sig.

Exp(B)

Man Equivalence 0.144 .168 .739 0.390 1.155

42

Animal Product 0.001 .000 4..221 0.040** 1.001 Farm Land 0.256 .152 2.852 0.091 * 1.292 Education -0.004 .090 0.002 0.964 0.996 Extension Contact 0.119 .044 7..302 0.007*** 1.126 Market Distance -0.063 .032 3.895 0.048** 0.939 Non-Farm income 0.000 .000 0.027 0.869 1.000 Training 1.575 .491 10.312 0.001*** 4.831 Location 0.951 .471 4.074 0.044** 2.588 Input 0.795 .459 3.000 0.083 * 2.214 Credit 0.246 .522 .222 0.637 1.279 Belief on RWHT 0.102 .547 .035 0.852 1.108 Land Security 0.225 .538 .174 0.676 1.252 Cash Availability 1.637 .495 10.946 0.001*** 5.139 Constant -3.920 1.052 13.884 0.000*** 0.020 Chi-square 76.12 -2 log likelihood 128.38 Count R2 78.4 Sensitivity 79.7 Specificity 76.8 Number of cases 201 *** Significant at 1%, ** 5%, and * 10% probability level

Location: Location of RWH structures correlates positively and significantly at (P< 0. 05) with utilization of the technology. The odds ratio in favor of participating in RWHT activities increases by a factor 2.588 when the farm location is suitable for RWHT. Training: The model result indicates that it affects the decision of farmers to participate in RWHT practices positively and significantly at (P<0.01). The odds ratio of utilization RWHT by a farmer increases by a factor of 4.831 as member of a household is trained in the given rain water harvesting technology. Animal and honeybee product income: The variable is significant at (P< 0.01). As the animal product and honeybee income increases by one unit, the utilization of RWHT increases by 1.001. Extension contact: The result indicates that it affects decision of farmers positively and significantly at (P<0.001). The odds ratio (1.126) indicates the utilization of RWHT increases by a factor of (1.126). Farm size: Farm size was positively related to the utilization of RWHT and significant at (P< 0.10). The odds ratio of 1.292 for availability of farm size implies that, other things being constant, the decision to use RWHT increases by a factor of 1.292 as farm size increases by one unit. Conclusion and implications The result of the descriptive studies show that users on average have large farm size, better adult equivalent of active labour force, educational status, labour used for farm, use of input, resource categories in the better off, TLU, oxen ownership, land tenure in terms of years operated of users of RWHT by far exceeds that of the nonusers of RWHT. The attitude scale result indicates that RWHT demand of labour, cost, skill and knowledge found to be highly important items affecting RWHT. The econometric result show that, training, market distance, farm size, location of the farm land, extension contact, income from animal product, cash availability had a positive and significant influence on the utilization of the RWHT. The finding of this study implies that even if they operate under or less similar conditions the social, psychological and economic performance differ from farmer to farmer. This implies that difference in perception, opinion; attitude and decision are among the major finding of this study. Therefore, this study underlines the needs for understanding social, economic, institutional, psychological and physical-technical factors that influences farmers decision-making in relation to utilization of rainwater harvesting technology.

43

REFERENCES Abadi, T., 2006. Analysis of Social Economic and Institutional Issues Affecting Utilization of Rainwater Harvesting Technology, Eastern Tigray , Ehiopia. MSc Thesis School of Graduate Studies of Alemaya University. 162p Alemneh, D., 2003. Integrated National Resources Management to Enhance Food Security: The case of community based approach in Ethiopia, FAO, Rome, Italy .44p. Berhanu, G., 1998. The Economics of Soil Conservation Investments in the Tigray Region of Ethiopia. PhD Dissertation, Michigan State University, U.S.A. Central Statistical Authority (CSA), 1995. The 1994 Population and Housing Census of Ethiopia. Critchley, W. and C. Sieger, 1991. A Manual for the Design and Construction of Water Harvesting Schemes for Plant Production. FAO, Rome, Italy. Ervin, C.A and D.E. Ervin, 1982, Factors Affecting the Use of Soil Conservation Practices: Hypotheses, evidence, and policy implication. 58 (3), 277-292. Hunting, 1976. Tigray Development Study. The Government of Ethiopia, Hunting Technical Service Ltd, Addis Ababa. Leeuwis. C., 2004. Communication for Rural Innovation. Rethinking Agricultural Extension. Blackwell Science Ltd. Oxford, UK. 381 p. Ostram, E., 1990. Governing the Commons: The evolution of institutions for collective action. Cambridge University press, Cambridge, UK. Mitiku, H., Eyasu, Y., and Girmay, T., 2001. Land Tenure and Plot size Determination Issues in Small-scale Development in Tigray, Northern Ethiopia, and Paper Presented for the Workshop on Current Issues on Land Tenure in Ethiopia, Addis Ababa. Roger, E.M (1995). Diffusion of Innovation. Third Edition. The Free University Press, New York. Tesfaye, B., 2003. Understanding Farmers: Explaining soil & water conservation in Konso, Wolayta and Wollo, Ethiopia. PhD Dissertation. Wageningen University and Research Centre. 245 p. Uphoff, N, 1986. Local Institutional Development: An analytical sourcebook with cases. Westhort ford: Kumarian press.

44

2.6 Institutional Capacity and Partnership Building for Rainwater Harvesting (A Zambian Perspective)

Author: Bob Muzyamba – Rural Development Consultant (Secretary General – Zambia Rainwater Harvesting

Association), Impact Management Consulting, 5th Floor, Premium House, Independence Avenue, P.O. Box 38906,

Lusaka. Zambia. Or Zambia Rainwater Harvesting Association, Mulungushi House, Independence Avenue. P.O. Box

50291, Lusaka.Zambia.

INTRODUCTION Basically, rainwater harvesting management is the process (and outcomes) in which organizations – individually and collectively- become active, knowledgeable and goal-oriented actors who take and/or support initiatives to overcoming its challenges. Hence, rainwater harvesting management refers to a strategy to achieve maximum utilization of rainwater as well as inherent capacity building process. Yet, effectively capacity building requires conceptual clarification and common understanding among institutional actors. Therefore the following questions need to be answered: what do I mean by institutional capacity? How can it be developed? More specifically, how can it contribute to the effectiveness and efficiency of rainwater harvesting and management? CONCEPTUAL UNDERSTANDING OF CAPACITY BUILDING The concept of institutional capacity gains transparency by distinguishing institutions from organizations. This distinction is also of strategic significance. Up to now in Zambia, many organizations involved inn rainwater harvesting’s empowerment capacity building efforts had a focus on the organizational aspect. But practice indicates that this approach is too limited to accomplish real transformation: this requires a change of deep structures. Institutional change thus emerges as a constituent in the process of transforming information imbalance; lack of a supportive policy; gender inequality; traditional and cultural traits. Capacity building actors must be aware and responsive of this. A recent scrutiny of structures in Zambia show that, although repeatedly equated, “institutions” and “organizations” refer to different dimensions of social reality. Institutions are “systems of rules shaping behaviour, including the mechanisms for development enforcement”. They are rooted in social interactions and emerge from agreements regarding norms, values, and customs. Organization, on the other hand, refers to the material expressions institutions can take; to forms that legitimate institutions. They are, so to speak, the sites where institutional rules are played out. They might be either adequately or badly functioning. Effective capacity building recognizes that institutions operate (or better: provide sets of rules) different spheres of life (economic, political, social, cultural, legal, technological) and are mostly multi-level. Potential successful rainwater harvesting management capacity and partnership building interventions must be performed at the different – interrelated levels:

• The grassroots or micro-level: constraining institutions for rainwater harvesting at this level are for instance norms and stereotypes that sustain traditional settings within a family, society and labour market.

• The intermediate or meso level contains, among others, all kinds of purposeful, task-oriented intermediate organizations as community based organizations and NGOs that concentrate on improvement of the situation of communities and actively strengthen their position. It also encompasses organizations or programs with different focus and that are indirectly support rainwater harvesting.

• The macro-level contains the whole norms, customs and habits that constitute a society’s cultural, economic and social-political environment. It encompasses the regulatory, policy and legal frameworks created to meet specific needs in particular fields. It includes laws, regulations and various policies.

45

PROBABLE PROCESSES OF DEVELOPING CAPACITY AND PARTNESHIP The water management professional world as a representation of a system of organizations, in Zambia, embraces the (supra) national, regional and local level and consists of a range of different organizations as ministries (such as Agriculture, Housing, Planning and Environment; Economic Affairs), various government and private agencies, private industries, banking and financial institutions, farmers’ unions, rural women’s organizations, farmers’ cooperatives and family farms. Each level and each separate organization has its own institutional rules based on specific needs and interests. Therefore, effective sector institutional capacity that would support rainwater harvesting capacity and partnership building would require a shared understanding and commitment to a number of aspects as well as the capacity to make this a guiding principle for all policies, programmes and activities for each organizational unit at the separate levels. So far, many obstacles stand in the way. An important hindrance, in the case of Zambia, is the high tech competitive culture in water management professional world and a lack of a water policy that supports initiatives such as rainwater harvesting for sustainable development. In this light, an effective institutional capacity and partnership building for rainwater harvesting management should be targeted at the different levels and dimensions. It needs awareness of enabling and constraining factors at each level. Approaches to institutional capacity and partnership building (including new norms, values, attitudes, behaviour as well as organizational changes) distinguish different layers at which it should be targeted. The following are the major layers to be considered: Individuals or groups: The focus is on increasing or strengthening knowledge, understanding, skills, abilities, attitude change and increase of self- actualization through training, awareness raising, for instance, and providing the conditions for their implementation. Capacity and partnership building at this level must not be an isolated action. Rainwater harvesting practitioner must be aware of constraints and opportunities for rainwater harvesting management empowerment at other levels. Organizations: The focus is on changing internal organizational processes, resources, management issues relating to the organizational culture. The broader context must be taken into account in order to identify the constraining and enabling factors for capacity building. Experience in Zambia has learned that key factors in the capacity of organizations to achieve their rainwater harvesting management empowerment goals are the organizations mandate; motivation and knowledgeability of staff and personnel. More specifically, a broadly supported mandate to promote rainwater harvesting management within the organization and their work, commit to and knowledge about rainwater harvesting management issues throughout the organization, rainwater harvesting experts in the staff who function as change agents or as catalysts and alliance partners who understand the issue. The broader context (as the national policy and legal framework) must be taken into account in order to identify the constraining and enabling factors for capacity building. Sector/Network: Capacity building may focus on rural and water policy reform (inclusion of a rainwater harvesting management perspective and targets), improvement in service delivery and increased coordination or cooperation among institutional actors. It may further include the establishment of new institutional actors or finding ways of strengthening those function below par and last but not least strategic budgeting. Many obstacles might stand in the way, such as competing organizational, economic and political priorities. Broader Systems level: Changes should be aroused in water management general policies, programmes, structures, legal frameworks, political commitment and underlying attitudes, values, and norms. Capacity building may concern improvement in service delivery, the establishment of new institutional actors, increased coordination among institutional actors, rainwater harvesting management responsive budget, the creation of appropriate monitoring and accountability structures, the creation of mechanisms to enforce rainwater harvesting management legislation within national legislative frameworks and practices. Increasing rainwater harvesting management sensitivity and commitment among actors involved and facilitating them with appropriate methods and tools to deliver responsive policies; and programmes or projects that support rainwater harvesting promotion.

46

CHALLENGES With a focus on rainwater harvesting management, actors engaged in institutional capacity building need to recognize that “rainwater harvesting is a critical social category. A general problem of rainwater harvesting is the scarcity of funds and adoptable technology, especially in more remote rural areas. Low political “will” and a lack of influence of rainwater harvesting actors in decision-making processes and on agenda-setting is another main limitation. Rainwater harvesting activities are under-represented in policy and political bodies at all levels, especially at national levels. This is also true for sector and interest organizations. Rainwater harvesting practitioners, especially in the rural and peri urban Zambia lack sufficient organizational capacity. In general, rainwater harvesting organizations are short of a strong link with the water resource management movement. This impedes more radical changes towards rainwater harvesting management initiatives. OPPORTUNITIES From the foregoing it appears that institutional capacity building is a multi-actor, multi-dimensional and multi-level activity or intervention while changing the underlying rules of the game is a crucial part of the process. Relevant rainwater harvesting management capacity building efforts should include: Building educational capacity: increase rainwater harvesting practitioners’ access to education, training, information and create rainwater harvesting sensitive education programmes for youth and adults (both women and men) through e.g. adjustment of school books, lessons, curricula; fight technological stereotyping of education and training through e.g. promotion of rainwater harvesting technology into education for high tech-defined professions and vice versa. Continued efforts to change the technological norms, values and power relations entrenched in organizational rules of the game through e.g. encouragement awareness raising, understanding and attitude change; political commitment to rainwater harvesting utilization at all levels; setting clear rainwater harvesting targets; technology budgeting; regular assessment and evaluation of efficacy of rainwater harvesting programs and procedures and improvement of strategies. Co-ordination of rainwater harvesting management policies between involved ministries and governments; developing procedures to overcome obstacles for achieving rainwater harvesting management caused by organizational segmentation of relevant Ministries; promotion of a sector wide approach to advance rainwater harvesting management; inclusion of the issue of work-family divide in the political agenda’s Establishment of progressive rainwater harvesting organizations or strengthening of existing one’s. Powerful organizations have good leadership and are functioning well (democratic, participatory, transparent, accountable) among others resulting in a clear presentation of rainwater harvesting management needs, priorities, views and perspectives and good strategies and skills to give rainwater harvesting a voice and influence in mainstream organizations and gremial. Rainwater harvesting organizations should reflect existing diversity among water resource management organizations. Twinning or building (inter)national, regional networks or local organizations and agencies that work towards similar objectives. Building rainwater harvesting networks and alliances with other progressive groups involved in water issues and agricultural and rural development. For instance progressive rainwater harvesting organizations and networks, new rural and farmers organisations, progressive politicians and policy makers, progressive consumer groups, environmental, animal welfare and nature organizations, etc. in order to better influence the political agenda. Development of procedures that deliberately work towards the inclusion of existing rainwater harvesting organizations or networks in intended or standing rural and agricultural development programs.

47

Enforcement of application of rainwater harvesting principle in Structural Funds and removing regulatory barriers that impede rainwater harvesting development initiatives. Advancement of political and administrative support for the multi-functional water resource mangement development model. Building adequate rainwater harvesting desegregated data base (adjustment of existing data collection methods, including review of technological biased-definitions) so as to generate new knowledge; allocation of resources for research on rainwater harvesting issues is crucial. Creation of independent budget streams through e.g. the establishment of regional rainwater harvesting funds that support grassroots rural rainwater harvesting development initiatives. REFERENCES Bolger J. (2000) Capacity Development: Why, What and How. Capacity Development/ Occasional Paper Series, Vol.1, no.1, May 2000, CIDA, Policy Branch, Hull, Quebec. Lavergne R. and J. Saxby (2001) Capacity Development: Vision and Implications. Capacity Development Occasional Paper Series no 3, January 2001; CIDA Policy Branch. Lavergne. Lusthaus C., M-H. Adrien and M. Perstinger (1999) Capacity Development: Definitions, Issues and Implications for Planning, Monitoring and Evaluation. Universalia Occasional Paper, No. 35, September 1999. Rao A. and M. Friedman (2000) Transforming institutions: history and challenges. An international perspective. In: Institutionalizing Gender Equality: Commitment, Policy and Practice. Amsterdam: KIT/ Royal Tropical Institute, Critical Reviews and Annotated Bibliographies Series, 2000, pp.67-80. Rowlands J. (1995) Empowerment examined. In : Development in Practice 5(2):101-107. SIDA (2000) Sida’s Policy for Capacity Development as a Strategic Question in Development Cooperation. SIDA Methods Development Unit, Stockholm.

48

2.7 Urban and Peri-urban Rainwater Harvesting RAIN: MODEL FOR TANK SIZING ROOF RAINWATER HARVESTING SYSTEM By: E.C. Kipkorir Moi University, School of Environmental Studies, P.O. Box 3900, Eldoret, Kenya, E-mail: emmanuel_kipkorir@yahoo.co.uk or ekipkorir@mu.ac.ke Abstract RAIN model has been developed for tank sizing roof rainwater harvesting system. The model is especially suitable for aiding in matching tank size and horizontal roof area for a specified water demand. The model calculations are based on frequency analysis and simulation analysis. Adequacy and the longest period that the tank is empty are the two main parameters used in the assessment. In the calculations, historic daily rainfall data, horizontal roof area and runoff coefficient defines the water supply curve while the user specifies the water demand curve. Irrigation, domestic and zero grazing unit water requirements through out the year can be used to specify the demand curve. Supplemental water supply from other sources throughout the year such as from well, spring can be accounted for. The model is a menu-driven programme and runs on personal computer with window operating system. The potential of the model has been assessed through application of the model to a case study located in a semi-arid zone. Results indicate that the model is a valuable tool for tank sizing roof rainwater harvesting system. Key words: model, simulation, tank size, rainwater harvesting, water supply. 1.0 Introduction Water shortages are common in both urban and rural areas of many developing countries. Rainfall fluctuations and frequent droughts together with poor water storage facilities mean that water supply is erratic. Poorer communities suffer most, by frequently receiving water that is not of potable quality and by paying a higher price for it. They see their inadequate supply as just one of a whole range of problems, others being connected with housing, food, low income, family relationships. Rainwater harvesting (RWH) systems can provide communities with greater water security. Rainwater harvesting, in simple terms, involves the harvesting of water directly from rainfall, by the construction of collection and storage structures, from which it then can be directed to various uses, ranging from domestic, to watering animals, and to kitchen garden supplemental irrigation (Arnold and Adrian, 1986; Ngigi, 2003). The system can be for a single household or a communal rainwater system using the roof of some large public building as catchment, such as a school or church. For the latter difficulties can arise in allocating responsibilities for maintenance of shared facilities and ensuring that individual users do not take disproportionate amount of water. There are four main techniques for RWH. The first one is rooftop catchment where roofs of houses, stores, green houses and animal shelters are covered with iron, plastic, thatch, reeds or mud and channelling rainwater into storage tanks through gutters. This technique has the best chance where houses already have good roofs and active home improvements are evident. One advantage for the household in collecting rainwater from its roof is that the roof has already been paid for, and so additional investments are limited to gutters and tanks. According to Thomas (2002), roof catchment in developing countries provides safe water for domestic use. However, since during the dry spell debris builds up on roofs, resulting in initial runoff during the first rainfall event being full of sediment and highly turbid. To minimise contamination of the harvested rainwater, flush diverters should be incorporated in the delivery system to divert to waste the first say 25 litres of runoff at the beginning of each rain event. The quality of rainwater stored in the tank improves with time since all sediment particles settle at the bottom leaving clear water. If the water tank is kept dark inside with a good lid cover, then neither growth of algae nor mosquitoes breeding will occur. Cleaning tanks annually improves water quality provided any remaining disturbed sediment is allowed to resettle for several days before the tank is used again. With the best pre-tank separator however, the rate of entry of organic material is so low that such material can be entirely removed by aerobic bacteria action and no treatment is required. The second RWH technique is ground catchment where an earth dam or pan is constructed and runoff from the surrounding modified catchment is channelled into the dam or pan. The third RWH technique is rock

49

catchment where run off from rocky areas is collected and channelled into storage tanks. The fourth is construction of sand dams where a concrete wall is built across a river in a rocky area, water collects in the sand and evaporation is reduced. When rainwater is to be collected from house roofs, the main hydrological question asked is ‘how big should the storage tank be?’ This breaks down into three subsidiary problems: (i) matching the capacity of the tank to the area of the roof; (ii) matching the capacity of the tank to the quantities of water required by its users; and (iii) choosing a tank size that is approximate in terms of costs, resources and construction methods. Some researchers have devoted considerable attention to calculating answers for the first two questions, using computer models to process long runs of rainfall records (Fujimura, 1982). In this paper a robust user-friendly model for tank sizing rooftop rainwater harvesting system to answer the first two questions is presented and its potential demonstrated through application to a case study located in Marigat, Baringo District, a semi-arid region. 2.0 System design The main factors in the design of domestic rooftop RWH system are rainfall distribution and variability, catchment characteristics, water demand and cost. It is difficult to bring these factors together to design the system to meet specific requirements of individual households. Many RWH projects have concentrated on one or two tank sizes and tank technologies for household water supply. This may be mainly explained by the fact that what can be afforded determines most choices of the tank capacity and not hydrology. While there may be good reasons to restrict the variability in tank sizes, it has to be recognised that the most suitable RWH system for a specific household in a specified location may not coincide with the tank size offered. Indeed, this may account in part for general conclusion that RWH systems do not have sufficient reliability for households to consider rainwater as their primary source of water (Michael, 1998). The main issues in RWH system design are reliability of the RWH system to satisfy demand, and flexibility in design method to address specific household requirements. Different households may require different levels of reliability due to the proximity and reliability of alternative water sources. For example, if a household is situated in the proximity of a perennial river, then that household may be willing to accept a lower level of reliability than a household whose nearest alternative source of water is a pan which may dry up and is five kilometres away. Flexibility in design is required in recognition that many RWH systems are installed after the roof are built and the fact that there are many different sizes of roofs. Additionally, the level of demand, which is a function of the number of people in the household and the availability and reliability of alternative water sources, varies between households. There are a number of ways of designing RWH systems. One of the main differences between the alternative methods concerns the way in which the rainfall data is used. Simpler methods require monthly rainfall data instead of daily data, which introduces errors in the design process, namely, what is the impact of the simplified data on the resultant design. More complex design methods make use of daily rainfall data in which case there are no approximations of rainfall data introduced into the design process. Complex methods can also reflect water use behaviour, such as rationing during periods of water scarcity (Michael, 1998). A tank sizing design method in the RAIN model that utilises annual mass inflow curves and subsequently frequency analysis to obtain a tank size for a specified probability of exceedance and a simulation model for flexibility analysis of the selected tank size is described below. 2.1 Mass Inflow Curve A simple way of estimating the storage for a specified probability of exceedance is through the use of a mass inflow curve. This involves first plotting the cumulative inflows and outflows for one year on the same scale. The inflow is calculated as product of horizontal roof area and rainfall corrected by some roof based runoff coefficient and outflow is derived from the water demand. The required tank size for the year is obtained as the greatest difference between the two plotted lines as shown in Figure 1. The above procedure is repeated for all years that rainfall data is available and the derived required tank sizes are subjected to frequency analysis.

50

Figure 1: Tank size determination by mass inflow curve 2.2 Frequency analysis The frequency analysis of data available from a subset of population consists in building and interpreting a probability distribution. The analysis is based on ranking. After ranking the data values for the calculated tank sizes in a descending order and assigning a serial rank number to each value, a plotting position is obtained. The plotting position corresponds with the frequency of exceedance on the probability scale of the probability plot (Figure 2). The Weibull relationship r/(n+1) is used to calculate the plotting position. Where r is the rank number and n is the number of years that rainfall data is available. After the selection of the distribution assumption, the observations are plotted and the theoretical distribution line is drawn. Normal probability distribution function is used, however the data can be transformed using log, square or square root functions. Instead of drawing the theoretical line, a straight line can be fitted through the data points by the methods of least squares (Figure 2). Information on the goodness of fit (R2) is then displayed. If the plotted data is in a reasonable alignment, it may be assumed that the data can be approximated by the assumed distribution. The magnitude of the tank size corresponding to various probabilities and return periods is derived from the probability plot and displayed in a table where a tank size for a specified probability level can be selected.

0

20

40

60

80

Cum

ulat

ive

inflo

w a

nd o

utflo

w (m

³)

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Tank size

Inflow

Outflow

51

Figure 2: Probability plot of tank size with square root transformation. 2.3 Simulation Model A simple way of determining the flexibility available for a specified tank size using different daily rainfall records is by using simulation model as a design tool. 2.3.1 Model components The simulation model is based on a water balance approach (Eqn 1) in which all the inflows and outflows to the storage tank are monitored on a daily basis. The main components of the model are shown in Figure 3. The calculation time step in the model is chosen as one day. The simulation run is for all the days in the rainfall record used.

iiiii CSIVV −−+=+1 (1) where, i is the day number in the rainfall record; Vi+1 is the storage in the tank at the end of day i (m³); Vi is the storage in the tank at the begging of day i (m³); Ii is inflow on day i (m³); Si is spill on day i (m³) and Ci is consumption on day i (m³). Figure 3: Schematic diagram of components of RWH The inflow (Eqn 2) is based on a daily rainfall record Pi (mm), the size of the horizontal roof area A (m²), and the runoff coefficient (f: 1.0 - 0.0) related to the catchment material and the gutter efficiency. It should be noted that a iron sheet roof system with poor guttering has a runoff coefficient very much less than that expected of such an impervious surface due to the inefficient guttering.

ii PfAI ××= (2) Spill occurs when the tank is full and there is excess inflow. This depends on the choice of the size of the tank T (m³). Mathematically this is expressed as (Eqn 3): Si = 0 if (T-Vi) ≥ Ii ; Si = Ii-(T-Vi) if (T-Vi) < Ii (3) Consumption on day i (Ci m³) is a function of supply index (z) and demand Di on day i (m³) computed as the product of number of people in the household and the per capita demand. Mathematically this is expressed as (Eqn 4): Ci = Di if Vi > r×T ; else Ci = z×Di (4) where r is the critical tank level when rationing is started and T is the capacity of the tank being analysed (m3). The per capita demand depends on the availability of alternative water sources. The supply index as shown in Figure 4, is an attempt to incorporate the self-regulating behaviour that is adopted in many households. For example, once the water level in the tank reaches a certain level (say ¼ of the tank) then demand will start to drop from say 30 litres per person per day to 5 litres per person per day when the tank is empty. This would reflect a change of usage from general domestic uses to strictly drinking and cooking

1 1 1

1 1 1

1 1 1

1 1 1

52

requirements. The storage that represents the amount of water in the tank on each day depends on the level of inflows, consumption and spill. Figure 4: Supply index function. 2.3.2 Model outputs For a simulation run with specified water demand, recorded daily rainfall for several years, given capacity of tank under analysis, horizontal roof area for a given roof under study and it’s runoff coefficient, the following results are presented as model outputs: Failure rate, computed as the percentage of days that the tank is empty. The total days in the rainfall record are considered in the calculation. Partial failure rate, computed as the percentage of days when full demand is not met. The total days in the rainfall record are considered in the calculation. The length of the longest period (consecutive days) in the rainfall record, when the tank is dry. Adequacy computed as the mean of (Ci/Di), which is a measure of the degree to which water demand is met. Dependability, computed as the standard deviation of adequacy over the simulation period. Ratio (spill/inflow). Assessment ranges for adequacy and dependability used in the model are shown in Table 1. Table 1: Assessment ranges of performance indicators. Source: Molden and Gates (1990)

Indicator “Good” “Fair” “Poor” Adequacy >0.9 0.8-0.9 <0.8 Dependability <0.1 0.1-0.2 >0.2

3.0 Practical Application The RAIN model has been developed as a planning tool and can be used in designing a RWH system. To demonstrate the potential of the model, as an example in this study, the model has been used to develop a set of results relevant to RWH in Marigat area, Baringo District located in a semi-arid region. The mean annual rainfall and the 95 % confidence band is 654 ± 58 mm for Perkerra Irrigation Scheme rainfall station (Kipkorir, 2002). The population of Marigat was estimated in 1999 as 8,442 in 2,200 households (GOK, 1999). Which translates to a population density of 56 people per Km². In Baringo District, the main roofing materials are grass and iron sheet used by 61% and 37% households respectively. However there is a noted decrease in grass use and increase in iron sheet use as roofing materials. The percentage of house holds in Baringo District receiving water supply from various sources are: 80% surface water, 11% ground

0.00

0.10

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1.00

0.00 0.25 0.50 0.75 1.00

Storage as a fraction of Tank Volume

Supp

ly In

dex

53

water, 8% pipe water and 1% from roof rainwater harvesting (GOK, 1999). Water supply from the two major sources is of low quality and requires more than 10 km walking distance daily by women and girls to fetch the water. In the analysis tank sizes ranging from 2 to 50m³ and horizontal roof catchment area ranging from 10 to 300m² were used with 36 years daily rainfall record for Perkerra Irrigation Scheme rainfall station. The roofing material considered is iron sheet. The per capita demand is considered as 30 litres per day. The number of people in a household is varied from 1 to 9, which is within the size range of households in Marigat. The roof runoff coefficient used is 0.9. The allowable reliability (adequacy) is set at a minimum of 90% (‘good’) and rationing is set to start when the tank falls below the ¼ full level. The roof area was decrease from a maximum of 300m² to the lowest value that resulted in a (spill/inflow) ratio of 0 for each demand. In the simulation the maximum period that the tank is empty is set at 0. The results that meets, the set adequacy, (spill/inflow) ratio and maximum period that the tank is empty, for the six selected demand levels are presented in Figure 5. The minimum horizontal roof area required for rainwater harvesting is estimated as 19m², which is rounded to 20m², Figure 5. Figure 5: Curves for aiding selection of RWH for domestic water tanks for iron sheet roofed roofs in Marigat. Numbers 1, 2, 3, 5, 7 and 9 represents number of people per household and per capita demand is considered as 30 litres per day. The results presented in Figure 5 can be very useful as a planning tool for those interested in tank sizing for households within Marigat area. As an example for a household of 2 people or water demand of 60 liters/day and available horizontal roof area of 100m², the required tank size is estimated as 6.0 m³ from Figure 5. This leaves the flexibility in system design with the implementers who are more likely to know which technology is more desirable in the area, but does provide information to match the tank size with the catchment area for a required reliability. Results indicate that the period when partial failure is experienced is during the months of January, February and March. A second application is in estimating the required tank size for Kipkuikui day primary school with a population of 256 and total horizontal roof area of 616m², in Loboi near Marigat. Assuming a per capita of 1 litre/day for drinking during the day, a guttered horizontal roof area of 275m² results in a tank size of 40m³. The size of the tank can be reduced to 24m³ if the guttered roof is increased to 100%. A third application is in estimating required community tank size for Kipkuikui AIC church with available horizontal roof area of 720m². Using the population density for the division of 56 people/km², command

0

5

10

15

20

25

30

35

40

45

50

20 40 60 80 100 120 140 160 180 200 220 240 260 280 300

Horizontal roof area (m²)

Tank

vol

ume

(m³)

1

2

3

5

7 9

54

area of 4km² and per capita of 5 litre/day for strictly domestic use, the required tank size is estimated as 223 m³. The community to benefit from the tank can apply for funding from for example the Constituent Development Fund (CDF) or Water Services Trust Fund (GOK, 2002). 4.0 Conclusion A user-friendly RAIN software package for tank sizing roof rainwater harvesting system was presented and its potential demonstrated through application to a case study in a semi-arid region. Results indicate that the model is very useful as a planning tool for those interested in tank sizing of rooftop rainwater harvesting system for domestic water supply and can be extended to garden irrigation and zero grazing unit water supply systems. Results indicate that roof RWH system for domestic use is feasible in Marigat a semi-arid area however the current problems of poverty and of human condition in the area currently limits its uptake. Reference Arnold, P and Adrian, C. 1986. Rainwater Harvesting, Intermediate Technology Publications, London. pp:216. Fujimura, F. N. 1982. Rain Water Cistern Systems (conference proceedings). Honolulu: Water Resources Research Centre. GOK. 1999. Census Report, Central Bureau of Statistics, Nairobi, Kenya. GOK. 2002. Water Act 2002. Government Printers. Nairobi, Kenya. Kipkorir, E. C. 2002. Analysis of rainfall climate on Njemps Flats, Baringo District, Kenya. Journal of Arid Environments. 50:445-458. Michael, K. T. 1998. Rainwater harvesting in Laikipia District. Kenya Engineer-September/October. 32-36. Molden, D.J., Gates, T.K. 1990. Performance measures of evaluation of irrigation-water-delivery systems, J. Irrig. and Drain. Eng. 116(6):804-823. Ngigi, S. N. 2003. Rainwater Harvesting for Improved Food Security. Kenya Rainwater Association, Nairobi. pp:266. Thomas, T. 2002. Suitability of rainwater for domestic consumption. URWA Strategic Planning Workshop, Kampala, Uganda.

55

Flexible Plastic Tanks: Developing a Design Protocol for Slums and Refugee Camps

By Duncan Onyango Mbuge, University of Nairobi

Abstract Underground water storage tanks made of concrete are expensive due to fact that they require steel reinforcement and cement, both of which are costly. A possible alternative to underground concrete tanks are plastic lined reservoirs. The installation of the plastic reservoir is simple as it involves excavating the reservoir and then lining it with plastic sheeting. While the cost of installation of the plastic lined reservoir is just one-tenth the cost of a reinforced concrete underground tank, there is no information about its durability and its service life of the common material used for plastic lining (HDPE) to enable would be users to make an informed choice and for engineers to determine the maximum load to achieve a given expected service life. This paper explores the opportunities and challenges that result from using this technology for water storage. A discussion is presented on the possibility of exploiting the flexible nature of HDPE for use in areas where space is a limitation such as slums and refugee camps. The design considerations for this possible application are presented. 1. INTRODUCTION

1.1 The Need for Flexible and Portable Water Storage Vessels Slums and refugee camps present a special challenge to engineers and planners in that they are almost always not properly designed. One big challenge to rainwater harvesting in these settlements is the low height of buildings and inadequate space between buildings for construction of rainwater harvesting and storage structures. Yet there is immense potential for rainwater harvesting in these places due to the vast roof area concentrated in one place as well the fact that there is a high population density that is usually not connected to the regular municipal water system. In refugee camps in Kenya the most popular mode of water supply is by trucking. Once the water is supplied, the problem of storage then presents itself. Also, nomadic communities are faced with the problems of water availability during their migration. A design of water storage vessel in the model of the age old water skin to be carried by animals may alleviate this need.

This paper proposes a design of tank that may contribute in part to the solution of the problem of water supply in slums, for nomads in transit and refugee camps. It seeks to test the feasibility of using HDPE, currently is use for lining water reservoirs, to fabricate flexible tanks that can fit in small and irregular spaces. Despite the maneuverability of this tank, the material used is comparatively weak in strength, deforms with increasing intensity as temperature rises and is subject to degradation on exposure to water and solar radiation. Hence it is necessary to precisely define the limits under which this tank would operate. That is, the maximum amount of water that the tank may carry without deforming beyond allowable limits and the maximum temperature at which it would work. Inevitably then, this paper presents a background of basic principles of elasticity, physical aging and time-temperature based long term properties considered. It is from these that a design protocol has been proposed.

1.2 Objectives

The broad objective of the study was to determine the pertinent parameters that would contribute to the design of an un-reinforced HDPE flexible tank. The specific objectives were:

• To determine the tensile strength of HDPE at different temperatures • To determine the effect of welding on strength of HDPE • To obtain long term creep (relationship of strain with time) for HDPE • To obtain design limits for a flexible HDPE tank modeled as a thin cylindrical shell under

pressure.

56

2. THEORETICAL CONSIDERATIONS

2.1 Long term performance of polymers Polymer materials exhibit time dependent behavior. The stress and strain induced when a load is applied are a function of time. When a plastic material is subjected to a constant load, it deforms continuously (Figure 2.1).

Figure 2.1: Creep curve for plastics, a constant load is applied The initial strain is roughly predicted by its stress-strain modulus. The material will continue to deform slowly with time indefinitely or until rupture or yielding causes failure. The primary region is the early stage of loading when the creep rate decreases rapidly with time. Then it reaches a steady state, which is called the secondary creep stage followed by a rapid increase (tertiary stage) and fracture. This phenomenon of deformation under load with time is called creep (Wu, 2000). Reed (2003) adds that creep is the slow deformation of a material under stress that results in a permanent change in shape. Creep phenomena increase with stress, time and temperature. Creep is determined by a competition between recovery and work-hardening material properties. The term “recovery” describes all the mechanisms through which a material becomes softer and improves its ability to undergo additional deformation while the term “work-hardening” stands for all the processes that make the material more difficult to deform as it is strained. Figure 2.1 is an idealized curve. Some materials do not have secondary stage, while tertiary creep only occurs at high stresses and for ductile materials. All plastics creep to a certain extent. The degree of creep depends on several factors, such as type of plastic, magnitude of load, temperature and time. The standard test method for creep characterization is ASTM D2990. In this test procedure, the dimensional changes that occur during time under a constant static load are measured. Primary creep is characterized by the work-hardening behavior of the material. As the strain increases, softening occurs and it leads to the steady state (secondary creep) in which recovery and hardening processes balance one another. Tertiary creep is the result of micro-structural and mechanical instabilities such as cavities, cracks and grain-boundary separations that result in a local decrease in cross-sectional area and consequent higher stress levels (Reed, 2003). If the applied load is released before the creep rupture occurs, an immediate elastic recovery equal to the elastic deformation, followed by a period of slow recovery is observed (Figure 2.2). The material in most cases does not recover to the original shape and a permanent deformation remains. The magnitude of the permanent deformation depends on length of time, amount of stress applied, and temperature.

57

Figure 2.2: Creep curve with recovery. A constant load is applied at t0 and removed at t1 The creep rupture is basically similar to a creep test with the exception that it is continued until the material fails. Since higher loads are used, creep rates are higher and the material fails in a shorter time (usually terminated in 1000h). This test is useful in establishing a safe envelope inside which a creep test can be conducted. The basic information obtained from the stress rupture test is the time to failure at a given stress. Based on this data, a safe stress can be determined below which it is safe to operate, given the time requirement of the end use application. The construction of the creep rupture envelope is shown in Figure 2.3. Test is conducted under constant stresses and the points of the onset of tertiary stage are connected to form the creep rupture envelope (Wu, 2000).

Figure 2.3: Creep rupture envelope 2.3 Design with plastics Design with plastics can be divided into two categories, design for strength and design for stiffness. The strength of a component is limited by the yield strength and rupture resistance of the material from which it is made. As shown in Figure 2.3, a creep rupture envelope can be obtained from creep test. For an expected life-time, the maximum stress allowed (σA) can be decided from the creep rupture envelope line. Design for stiffness with creep curves proceeds by establishing the maximum strain acceptable εmax, thereby establishing a horizontal line on the creep diagram correspondingly. The expected lifetime tL of the part is also determined, and the maximum stress permissible is found on the creep curve at the intersection of these two lines (Wu, 2000).

58

Figure 2.4: Design criteria by creep curves As shown in Figure 2.4, many combinations of σmax and time will yield this maximum strain. For a desired lifetime tL, however, there is one maximum level σL which satisfies the maximum strain. Design basis selection depends on the specific application. Usually, strain or dimension requirement is more critical, and design for stiffness is favored in this case. If the dimension precision of the component under discussion is not so important compared as strength, design for strength is then used accordingly. For complicated structures, both can be used for design criteria to ensure successful material performance during service time (Wu, 2000). 2.3 Modeling Water Tank as Thin Shell under Pressure The tank was modeled as a thin shell under pressure so as to determine the maximum permissible size of tank for different shapes of tanks. To illustrate the determination of strength of thin shells under pressure, a cylindrical shape of water tank was used as shown in Fig. 2.5. The major advantage of the cylindrical design was seen in the fact that it can be used for really low roofs, can be coiled around irregular spaces and has a reasonable large surface supported by the ground. Fig 2.5: Stresses acting on a cylinder under pressure

Longitudinal stress

Circumferential stress

59

Both longitudinal and circumferential stresses act on the cylinder. The formula for the circumferential stresses σc is:

tc 2Pr

=σ ........................................................................................................................................... (2.1)

The longitudinal stress σl on the other hand is given by:

tlPr

=σ ........................................................................................................................................... (2.2)

Where: • P is the pressure applied on the wall • r is the tank radius • t is the wall thickness

By comparing equations 2.1 and 2.2, it follows then that for equal pressure, tank radius and wall thickness, that the longitudinal stress is the greater value and is the one that is safer to use in tank design when the strength of the material used has been determined by tensile and creep tests at different temperatures. The maximum water pressure, P, for water stored in a vessel is given by the formula: ghP ρ= ......................................................................................................................................... (2.3) Where: ρ is the density of water (1000 kg/m3) g is the force of gravity (9.81 m/s2) h is the height of the water from the bottom of the tank For cylindrical tanks, the height h is equivalent to the diameter or 2r. Hence substituting for P in equation 3.2, we obtain a value of the maximum wall pressure as:

tgr 2

max2ρσ = ................................................................................................................................ (2.4)

From equation 3.4 it can be seen that to minimize the wall stress, the radius has to be limited and the thickness of the material maximized. Hence the ideal tank would maximize on the length and minimize the radius. However, other pertinent limiting factors are the magnitude of the strain over time and recovery at the applied stress, effect of lap welding and the prevailing temperature. These were factored in the design by means of creep curves of the HDPE material. 3.0 RESULTS AND DISCUSION From experimentation it was found for a material 0.8mm thick, using a dumbbell specimen 4mm wide at a temperature of 30ºC that to obtain strains of less than 0.5mm/mm over 30 days, a stress of 2MPa would be taken as the maximum stress. This would also allow the material to undergo recovery and deform only slightly upon loading. The ultimate stress of the material was found to be 14MPa. This gives a safety factor of 7. It was also found that the lap welding done to join pieces of plastic for fabrication reduced the strength by 20% of the ultimate strength. This was reduced on equation 2.4 which then became:

tgr 2

max6.1 ρσ = ............................................................................................................................ (2.5)

From equation 2.5 for a stress value of 2 MPa, density of water of 1000kg/m3 and a wall thickness of 1.2 mm, it was found that the maximum radius of tank allowable would be: 0.45m, giving a total height of 0.9m (twice the radius). Considering a tank of length 3m then the volume of tank would be about 2000 litres (πr2*2).

60

4.0 CONCLUSIONS Because of the distinct advantages cited in the introduction, the fact that this tank may hold 2000 l of water gives promising prospects for adoption. This is because the total amount of material required is about 10m3 which would cost KSH 2500 (USD 35) to construct. However, an increase in thickness and strength of the HDPE would serve to make this even better (see equation 2.5). It is recommended that investigations be done to determine what kind of reinforcement can be done on the material to increase strength.

61

2.8 Agricultural Water Management USING SIMULATION MODELLING TO IMPROVE THE DESIGN AND MANAGEMENT OF FURROW IRRIGATION IN SMALLHOLDER PLOTS

Langat Philip1,2 Raine S.R 2 and Khatri K.L2,

1Ministry of Agriculture, P.O. Box 4, Kabarnet, Kenya Email: philkibet@yahoo.com Tel: +254-53-222108 2 Cooperative Research Centre for Irrigation Futures & National Centre for Engineering in Agriculture, Faculty of Engineering and Surveying, University of Southern Queensland, Toowoomba, 4350, Australia. Email: raine@usq.edu.au Fax: + 61 7 4631 2526

Abstract

Over 70 % of all irrigated land in Sub-Sahara Africa is currently under surface irrigation. However, the performance of surface irrigation in this region is often low resulting in significant deep drainage losses and sustainability concerns. This is of a particular concern for tenant based large scale irrigation scheme located in arid and semi-arid areas which are prone to salinity and drainage problems. Field design and irrigation management practices have a significant impact on performance but have received only limited consideration to date. Externally funded irrigation projects in the region are usually constructed using generic design parameters with limited or no analysis of local soil and operational conditions. Similarly, irrigation design and management guidelines for in-field irrigation management are generally lacking due to the high cost and time involved in obtaining data for traditional evaluations. This paper uses data collected from small-holder irrigation plots in the Tana River Basin to demonstrate the benefits of using the simulation program SIRMOD to evaluate the performance of surface irrigation practices. It also discusses the benefits of simulation modelling for identifying irrigation performance indices and guidelines for improving the design and management practices of small-holder irrigation plots. Keywords: Agricultural water management; furrow irrigation; modeling 1.0 Introduction Over 70% of irrigated land is currently under surface irrigation in Sub-Sahara Africa (SSA) and it seems likely to continue to be widely practised for the foreseeable future. However, the performance of surface irrigation in this region is often low resulting in significant deep drainage losses and sustainability concerns (Gichuki 1996; Kandiah 1997; Kirpich et al. 2000). This is of a particular concern for tenant-based irrigation schemes located in arid and semi-arid areas (Hughes 1984; Maingi & Marsh 2001; Ledec 1987) which are prone to salinity and drainage problems (Bakker et al. 2006). Field design and irrigation management practices have a significant impact on surface irrigation performance (Clemmens 2000; Horst 1998; Raine et al. 1998) but have received only limited consideration to date (Horst 1998). Improper design and management of furrow irrigation systems may result in water wastage, water-logging and losses of fertilisers and pesticides out of root zones. Externally funded irrigation projects in the SSA region are usually constructed using generic design parameters with limited or no analysis of local soil and operational conditions (Horst 1998). Similarly, irrigation design and management guidelines for infield irrigation management are generally lacking due to the high cost and time involved in obtaining and analysing data for traditional evaluations. Furrow irrigation system design and management parameters which can be altered by farmers with little effort and cost include inflow rates and irrigation cut-off time (Walker & Skogerboe 1987; Zerihun et al. 2001). Traditionally, inflow rates and cut-off time are determined at the beginning of the irrigation growing

62

season. Farmers are often guided by previous irrigation experiences in making decisions regarding these parameters and often, where experience is lacking, identification and adoption of appropriate management practices is difficult. It is also common for farmers to make decisions on cut-off based on how long water is made available by the managing statutory authorities. Hence, farmers often irrigate for longer than is necessary. The use of simulation models to develop irrigation guidelines for design and management practices provides opportunities for advisors and farmers to make more informed and timely irrigation decisions. The objective of this paper is to demonstrate the potential to use simulation modelling to improve the performance of surface irrigation practices using data collected from smallholder irrigation plots in the Bura Irrigation Scheme, Kenya. 2.0 Methods and Materials 2.1 Field data The field data used in this evaluation was obtained from Mwatha and Gichuki (2000) who conducted furrow irrigation trials in the Bura Irrigation Scheme, Kenya. The Bura Irrigation Scheme is located in the Tana River Basin and was initially developed in 1979 to settle landless farmers and grow irrigated cotton for export. The Bura area is located at latitude 10o 8’S and longitude 39o 45’ E and has an elevation of 110 m above sea level. The mean annual rainfall and evaporation are 400 and 2490 mm, respectively. The rainfall is bimodal, with long rains occurring in March to May and short rains occurring in November to December. Soils in the Bura area are shallow sandy clay loams and heavy cracking clays overlying saline and alkaline sub-soils of low permeability (Mwatha & Gichuki, 2000). The irrigation water in the Bura Irrigation Scheme is pumped from the Tana River into settling basins and main scheme canals before being siphoned into 0.9 m spaced furrows within the small-holder irrigation fields. Mwatha and Gichuki (2000) reported data for two irrigations (fifth irrigation is considered in this paper) during the 1989 growing season from four irrigated cotton fields (lengths of 275-300 m). Furrow characteristics, soil moisture content and irrigation parameters data were collected from February to October 1989 when cotton was growing. The evaluation data were obtained from four fields with average slopes of 0.09, 0.13, 0.25 and 0.31 % denoted in this paper as 9S, 13S, 25S and 31S, respectively. Within each field there were three inflow rate treatments (1.5, 2.0 and 3.0 L s-1 furrow-1) and data were collected from four furrows in each treatment. Inflow was measured using Parshall flumes spaced at 50 m intervals along the furrows and for the purpose of this analysis it was assumed there was no inflow variability. All data were collected from plots located on the same soil type (Mwatha and Gichuki, 2000). The fifth irrigation had a deficit of 63 mm as measured by the difference in the volumetric soil moisture content taken at 50 m distances along the field before the irrigation and two days after irrigation (Mwatha and Gichuki, 2000). Table 1 shows the measured furrow characteristics while the irrigation inflow and advance parameters for the four fields are given in Table 2.

Table 1: Geometry characteristics for furrows in the Bura Irrigation Scheme (from Mwatha and Gichuki, 2000)

Parameter Value Furrow length 275-300m Furrow spacing 0.9 m Furrow slope 0.05 %- 0.3% Cross-section parabolic

Top-width (T) A T= 2.8y0.62

Wetted perimeter (wp) A wp=2.8y0.65

Area of flow A A=1.48y1.55 A y is the depth of flow

Table 2: Advance parameters for the fifth irrigation event, Bura Irrigation Scheme

(from Mwatha and Gichuki, 2000)

63

Advance parameters Slope (%)

Inflow (L s-1) p r tL (mins)

1.5 12.7 0.49 572 2.0 6.1 0.67 308

0.09 (S9)

3.0 10.2 0.57 345 1.5 12.6 0.56 262 2.0 11.3 0.57 290

0.13 (S13)

3.0 18.3 0.53 177 1.5 13.5 0.56 231 2.0 22.2 0.46 256

0.25 (S25)

3.0 16.2 0.61 110 1.5 16.5 0.55 179 2.0 17.9 0.53 186

0.31 (S25)

3.0 13.4 0.68 90 2.3 Infiltration parameters Infiltration parameters for the furrows were estimated using an inverse solution technique and the software INFILT (McClymont and Smith, 1996). This software is designed to calculate soil infiltration parameters using only inflow and advance data and has been used over a range of soils and situations (Bakker et al., 2006; Khatri and Smith, 2005; Smith et al., 2005; Raine et al., 2005). The soil infiltration characteristics are derived from the advance trajectory assuming a power advance function (Walker and Skogerboe, 1987): x = p(ta)r Equation 1 where ta is the time taken for the water to reach advance distance x, and p and r are fitted advance parameters. INFILT calculates the soil infiltration characteristics from the fitted power curve parameters based on the equivalent furrow infiltration Kostiakov-Lewis equation: z = k(ד)a + ƒo(ד) Equation 2 where z is the cumulative infiltration (in m), ד is the time the water has been applied to the soil (in minutes), fo is the steady state infiltration rate of the soil (in m3/min/m) and k and a are fitted parameters. INFILT uses three or more advance points to determine best fit values for the three infiltration parameters a, k and fo. Where a minimum of four advance points are provide, it is also able to estimate the cross-sectional area of flow term (σyAo) if this term is fixed as an input parameter. However, in this study, the inlet area of flow (Ao) was calculated using the Manning’s equation and the measured furrow geometry (Table 1). 2.4 Simulation The estimated soil infiltration parameters from INFILT, measured inflow rates (Table 2) and furrow geometry (Table 1) were used in the latest version of the surface irrigation model SIRMOD (version 4) (Walker 2001) to reproduce each irrigation event as measured. Calibration of the model for each event was conducted by adjusting the hydraulic resistance term (Manning n) until the simulated advance matched the measured advance at the end of the furrow. After calibration, the model was used to evaluate the performance of a range of different designs (e.g. furrow length) and management (e.g. inflow rate, irrigation period) options. SIRMOD has the capability of modifying the infiltration function based on changes in furrow wetted perimeter (Equation 3) at different inflow rates:

b

WPWP

⎭⎬⎫

⎩⎨⎧

=1

2ψ

Equation 3 where WP1 is the wetted perimeter (in m) for inflow rate Q1, WP2 is the wetted perimeter (in m) for simulated inflow rate Q2 and b is an empirical exponent. In this study, the exponent was assumed to be unity for simplicity although Alvarez (2003) and Mailhol et al. (2005) have indicated that greater benefits may be obtained by measuring it for any particular site or soil type.

64

The performance of a furrow irrigation system can be described by three different, but interacting, indices: application efficiency (Ea), requirement efficiency (Er) and distribution uniformity (DU) (Zerihun et al. 2001). It is commonly assumed that deep percolation and run-off is weighted equally in optimisation of application efficiency. This study evaluated the potential irrigation performance that can be attained under the present design conditions assuming adequate farmer irrigation management practices. Other assumptions included: (a) furrow length of 285 m, (b) 3.0 L s-1 inflow rate is the non-erosive limit in the study area, and (c) the farmer practice is to ensure that the irrigation requirement efficiency is met (i.e. Er ≥ 90%). The infiltration function obtained from the 1.5 L s-1 treatment furrow in each field was used for the performance predictions using SIRMOD. The optimisation involved varying inflow rates (Qo), cut–off time (tco) and the presence or absence of furrow end-dyking. Current farmer management was assumed to be the performance obtained if the irrigation was cut-off when the water advance reached the end of the furrow. However, it should be noted that this may have overstated the existing performance as many farmers cut-off long after the water advance has reached the end of the field to ensure that the root zone soil water is completely recharged. 3.0 Results and discussion 3.1 Effect of inflow rates and cut-off time The application efficiency for the existing design and management practices ranged from 31 to 99%. Distribution uniformity ranged from about 70 to 91% and requirement efficiency from 75 to 99.5%. Field S9 had a low application efficiency (<38%) irrespective of the management strategy which suggests that design changes (in field length) are required to improve the performance of this field. For fields S13, S25 and S31, increasing inflow rate from1.5 to 3.0 L s-1 and optimising cut-off time increased the average application efficiency from 79.4 to 87.5%. However, increasing inflow rate to 3.0 L s-1 and cut-off when the advance reaches the end of the field produced an average application efficiency of 84.5%. Increasing the inflow rate and reducing the cut-off time to equal 90% of the advance time improved the application efficiency to 88.7% across the fields. Introducing furrow end-dyking increased the distribution uniformity by between 0.3 and 5.2% but did not significantly affect application or requirement efficiency. However, the furrow end-dykes were over-topped by the irrigation in some events (e.g. field S31 with inflow rate of 3.0 L s-1). It should also be noted that end-dyking may also cause surface drainage problems under high rainfall conditions. Traditional furrow irrigation design and management commonly attempts to maximise the requirement efficiency (i.e. Er ~ 100%). However, where the distribution uniformity is high, the only implication of a low requirement efficiency (i.e Er < 90%) is that the next irrigation is required to be sooner than originally planned. Lower requirement efficiencies may also offer a greater opportunity to capture and utilise in-season rainfall. Farmers in schemes like Bura, generally have little understanding of either the interval between irrigations or irrigation opportunity time required to satisfy the desired soil-water deficit. Irrigation application is often continued as long as water is available or until it is convenient to be manually switched off. Thus, significant water losses due to excessive drainage and tail water run-off may be experienced. The relationship between inflow rate and cut-off time, and the application or requirement efficiency for the S31 field, is shown in Figure 1. This figure can be used as a decision support aid in irrigation management to provide advice or strategies to reduce water and energy costs and give improved environmental management. For example, a farmer using an inflow rate of 2.0 L s-1 can achieve an application and requirement efficiency of 79 and 98%, respectively by irrigating for 180 min. However, irrigating for longer periods will significantly reduce application efficiency without improving the requirement efficiency. Hence, switching off the irrigation at an appropriate time would not only reduce the volume of tailwater run-off but also the volume of water lost as deep percolation and, thus, save a significant quantity of water.

65

0

20

40

60

80

100

0 60 120 180 240 300

Time to cut-off (min)

Effic

ienc

y (%

Figure 1: Effect of irrigation cut-off time on application () and requirement ( ) efficiencies for a 300 m

field length and water application discharges (∆ =1.5, □ = 2.0 and ◊ = 3.0 L s-1) for field S13.

3.3 Effect of furrow length Significant deep drainage due to surface irrigation has the potential to affect groundwater levels and contribute to salinity in the river basin. On high infiltration soils (e.g. field S9), changes in furrow length may be needed to reduce deep drainage. For the S9 field, optimisation of cut-off time for various furrow lengths was conducted to identify the opportunity to improve performance. Deep percolation ratio is defined as the ratio of volume of irrigation water lost below the root zone to the total volume of water applied.

0

20

40

60

80

100

0 50 100 150 200 250 300Length (m)

Irri

gatio

n pe

rfor

man

ce (%

)

Figure 3: Application efficiency () and deep percolation ratio ( ) as a function of furrow length for fifth event with different inflow rates (∆ = 1.5 ls-1, □ = 2.0 ls-1, ◊ = 3.0 ls-1) for field S9.

Application efficiencies of less than 40% were achieved with furrow lengths of 275-300 m regardless of inflow discharge used. However, reducing the field length to approximately 100 m would improve the application efficiency to greater than 80% and reduce deep percolation ratio to <10% for flow rates ≥2.0 L s-1 (Figure 3). Lower flow rates (i.e. 1.5 L s-1) still have high (≥80%) application efficiencies but the proportion of water lost as deep drainage increases to ≥20% (Figure 3).

66

4.0 Conclusion Alternative irrigation system and management practices were evaluated to identify strategies to improve irrigation performance. A simple decision support aid to improve irrigation performance in the Bura Irrigation Scheme of the Tana River Basin has been demonstrated. This work has shown that farmers, with the assistance of the Tana River Development Authority (TARDA) among others, may obtain performance benefits by optimising irrigation inflow rates and cut-off times. In some cases, field re-design to optimise furrow lengths may be required to achieve satisfactory performance. References Alvarez, RJ 2003, ‘Estimation of advance and infiltration equations in furrow irrigation for untested discharges.’ Agricultural Water Management 60, pp 227-239 Bakker, DM, Plunket, G, & Sherrard, J 2006, 'Application of efficiencies and furrow infiltration functions of irrigations in sugar cane in the Ord River irrigation area of North Western Australia and the Scope of the improvement', Agricultural Water Management, vol. 86, no. 1-2, pp. 162-72. Gichuki, FN 1996, Sustainability concerns in African irrigation, in Pereira, LS, Feddes, RA, Gilly, JR & Lesaffre, B (eds) Sustainability of irrigated Agriculture, Kluwer Academic publishers, Dordrecht. Horst, L 1998, The dilemas of water division: considerations and criteria for irrigation system design. International Water Management Institute (IWMI), Colombo. Khatri, KL & Smith, R 2005, 'Evaluation of methods for determining infiltration parameters from advance data.' Irrgation and drainage, no. 55, pp. 1-16. Ledec, G 1987, 'Effects of Kenya's Bura Irrigation Settlement Project on Biological Diversity and other conservation concerns', Conservation Biology, vol. 1, no. 3, pp. 247-58. Mailhol, JC,Ruelle, P & Povova, Z 2005, ‘ Simulation of furrow irrigation practices: a field scale modelling of water management and crop yield for furrow irrigation’, Irrigation Science 24: pp 37-48 McClymont, DJ & Smith, RJ 1996, 'Infiltrarion parameters from the optimisation on furrow irrigation advance data', Irrigation Science, vol. 17, pp. 15-22. Mwatha, S & Gichuki, FN 2000, 'Evaluation of the Furrow Irrigation System in Bura Scheme', in Gichuki, FN, Mungai, DNGachene, CK & Thomas DB, ‘Land and Water Management in Kenya: towards sustainable land use’, proceedings of the Fourth National Workshop, Wida Highway Motel, Soil and Water Conservation Branch, Ministry of Agriculture and Rural Development & Department of Agricultural Engineering, University of Nairobi. Raine, SR, McClymont, DJ & Smith, RJ, 1998, 'The effect of variable infiltration on design and management guidelines for surface irrigation', Proceedings of ASSSI National Soil conference, Brisbane, 27-29 April. Raine, SR, Purcel, J & Schimidt, E 2005, ‘Improving whole farm and infield irrigation efficiencies using irrimate TM tools restoring the balance’, Proc. Nat. Conf. Irr. Assoc. of Aust. 17th- 19th may Townsville pp5 Smith, RJ, Raine, SR & Minkevich, J 2005, 'Irrigation application efficiency and deep drainage under surface irrigated cotton', Agricultural Water Management, vol. 71, pp. pp 117-30. Walker, WR & Skogerboe, G. V. 1987, Surface irrigation: theory and practice, prentice-Hall, INC. Englewood Cliffs, New Jersey. Walker, WR 2001, SIRMOD II - Surface Irrigation Simulation, Evaluation and Design. User's Guide and Technical Documentation. Utah State University.

67

THE POSITION OF RAINWATER HARVESTING IN IWRM FOR SUSTAINABLE DEVELOPMENT Keynote speech at SEARNET 10th International Conference 4-6 December 2006, Mombasa, Kenya By Jennie Barron1) & Johan Rockstrom2)

1) Research fellow in water management Stockholm Environment Institute (SEI) University of York Heslington York YO10 5DD UK Email: jennie.barron@sei.se

2) Executive Director Stockholm Environment Institute Lilla Nygatan 1 Stockholm SWEDEN Wise and fair water management for multiple development goals is imperative to meet the Millenium Development Goals until 2015. Issues emerging in the global development context are if fresh water limiting potential to food production at different local and regional areas? How do we ensure environmental sustainability with more appropriation of freshwater in the landscape? And how does climate change affect fresh water availability for food production and other uses of ecosystem services. In conventional IWRM strategies, only blue water (i.e., stream flow and freshwater in lakes and wetlands) is considered, which usually only constitute part of the available water in the landscape (e.g., the rainfall). Shifting from the blue water management, and also including the green water flows (evaporation and transpiration from all landuse systems in a catchment, basin) can help in assessing all sources and flows of water, both direct and indirect uses of water. The green-blue water approach thus shifts form addressing ‘water scarcity’ to water use potentials and possible tradeoffs between multiple landuse systems. As an example of a green-blue country level analysis, the flows in respective landuse systems in Kenya is shown (Fig 1.), indicating that the IWRM with blue water approach only focuses on 5% of the total water resource. Also, notably approximately 8% of country rainfall is ‘consumed’ in rainfed agriculture. It is important to note that agro-pastoral systems also occur in other landuse systems, typically in the savannah class. Using a green-blue water IWRM approach presents more development potentials in terms of water availability for food production, health and sanitations and maintenance of ecosystems provision functions. Rainwater harvesting provides a natural entry point through the green-blue framework, to address these multiple development goals. In particular rainwater harvesting strategies can significantly improve water productivity (water use per produced grain or biomass). There is now an extensive collective knowledge on appropriate and inappropriate RWH technologies for different uses and different locations in sub-Sahara Africa. Although water is important for food production, once crop water availability is ensured (through improved infiltration and/or supplemental irrigation), fertilizer should be added to ensure improved yields and improved water productivity. A word of caution is that when development occurs, it is important to monitor and document uptakes, implementation, yields etc, but also potential changes in landscape stream flow and shallow groundwater. The monitoring and documentation help us to advance our joint knowledge base and transfer of experiences on RWH: where it works and why. Monitoring of downstream water flows and groundwater levels helps to ensure that large implementation of RWH does not deplete downstream water users of the resource (whether it is environment, industry, urban or other farmers).

68

At the same time, development cannot wait for researchers to ‘provide’ all the answers. We propose strong linkages between development implementers, stakeholders and researchers using action research processes and participatory methodologies to ensure environmental and socio-economic sustainable development of RWH in catchments and basins.

Figure 1: Using green-blue water assessment for landuse systems of Kenya (From SEI 2005. Sustainable

pathways to attain the Millenium Development Goals: assessing the key roles of water, energy and sanitation. Report prepared for the UN World Summit 14th Sep, 2005, New York,(Available ats pdf at www.sei.se )

69

RAIN WATER HARVESTING FOR CROP PRODUCTION

By Brian Kipkurui Cherutich, Project Co-ordinator; Natural Resource Management, Self Help Development International Kenya (SHDI Kenya), Gilgil Area Based Integrated Food Security and Livelihoods Programme, P.O. Box 2248-20100, Nakuru, Kenya. Tel: +254-51-2212291 Telefax: +254-51-2212304 Email: Kenya@shdi.org brian.kipkurui@shdi.org Abstract Many farmers in Kenya lack access to sufficient rainfall and/or water for irrigation. Due to this, reliance on erratic rainfall subjects them to insufficient and unreliable water for agriculture. In the semi-arid areas, like Gilgil Division of Nakuru District where Self Help Development International Kenya (SHDI Kenya) is implementing an integrated food security and livelihoods Area Based Programme (ABP), it is common to find four out of five seasons ending up with either total crop failure or the harvests are too low even for sustenance. In spite of this, many farmers have not realized the possibility to double or triple crop yields through rain water harvesting. This may be done by utilizing rainfall and practicing agricultural water management through simple technologies. These include; double digging, use of fertility trenches, multi-storey gardening, use of half-moon micro catchments, Mandala gardens and planting pits. Our experience in Gilgil indicates that these technologies can produce good results, hence depicting a great potential for improving crop production in dry areas. The technologies increase crop yields by a combination of moisture conservation and harvesting of runoff from the uncultivated spaces. In addition, soil fertility is maintained since the manure and fertilizer cannot be lost through surface runoff. The multi storey garden for example conserves moisture, manure and fertilizer but does not harvest runoff as the other technologies. Once the crop is grown using either of the technologies, some management practices are needed; keeping the field clear of weeds by maintaining a light cover using a cover crop to provide a firm and compacted catchment. This encourages infiltration, and protects the crop from pests and diseases. To date, utilization of these technologies by some farmers in Gilgil has resulted into effective water conservation, reduced soil erosion, improved soil fertility, environmental conservation, improved crop yields and survival at household levels during drought. 1.0 Introduction About 85% of Kenya’s population make their living from rainfed agriculture, and to a large extent, depend on smallholder, subsistence agriculture for their food security (UNDP/UNSO, 1997). The problem of low food production is further aggravated by limited new land for cultivation, land and environmental degradation, poor infrastructure, political and social crises, bad governance, insecure land tenure, health/diseases outbreak, inadequate knowledge/capacity, and donor dependency syndrome. Thus the ever increasing demand for food and household income has to be achieved through an increase in biomass produced per unit land and unit water (Rockstrom, 1999). In the past, very little attention has been paid to the development of rainfed agriculture in Kenya except for the development of conventional irrigation projects. However, most of these projects have proven to be unnecessary, costly and environmentally unsustainable (Ngigi, 2002d). Only 20% of Kenya’s land surface is suitable for rainfed agriculture. Arid and semi-arid lands occur in the other 80% including parts of Rift Valley where Gilgil Division of Nakuru District lies among other regions (Aklilu & Wekesa, 2001). Many farmers in Gilgil Division, Nakuru District do not have access to sufficient rainfall and or water for irrigation. They rely on erratic rainfall for all their crop production, and thus, are subjected to the recurrent problem of insufficient and unreliable water for agriculture (CIR, 2005). In the semi-arid areas, it is common to find that four out of five seasons end up as either total crop failures or the harvests are too low to break even. Many farmers have not realized that it is possible to double or triple crop yields through rain water harvesting. Utilizing rainfall through rainwater harvesting and practicing agricultural water management through simple technologies leads to better yields (IIRR, 1998). However, rainfall in Gilgil Division is highly variable and unreliable for rainfed agriculture and livestock production (Mwango, 2000). The rainfall is highly erratic, and normally falls as storms, with very high intensity and is spatial and temporal in variability. This nature of rainfall is associated with a very high risk for annual droughts and intra-seasonal dry spells (Rockstrom, 2000). This means that the poor distribution of rainfall, more often than not, leads to crop failure rather than absolute water scarcity due to low cumulative annual rainfall.

70

Unfortunately, most dry spells occur during critical crop growth stages (this explains frequent crop failure and/or low yields), and hence the need for dry spell mitigation by improving water productivity. According to (Benites et al., 1998), rainfall patterns that are erratic, and normally fall as storms, with very high intensity and is spatial and temporal in variability, means that the major cropping systems may not be sustainable. This is evident in Gilgil and this implies persistence low food production leading to food shortage and reliance on food relief as a coping strategy. One of the strategies that may be sustainable and reverse this trend is rainwater harvesting. According to (Evanari et al., 1971; Shanan and Tadmor, 1976; Critchley, 1987; Critchley and Siegert, 1991 and Agarwal and Narain, 1997) rain water harvesting is broadly defined as the collection and concentration of runoff for productive purposes (crop, fodder, pasture or trees production, livestock and domestic water supply, etc.). It has ancient roots and still forms an integral part of many land use systems worldwide. It includes all methods of concentrating, diverting, collecting, storing, utilizing and managing runoff for productive use. Once this water is harvested it may then be utilized using a variety of technologies. These include but are not limited to; double digging, use of fertility trenches, multi-storey gardening, use of half-moon micro catchments, Mandala gardens and planting pits also refferered to as nine seeds in one pit technology (BAC, SARD Training Manual). 2.0 Challenges for Improving Rainfed Agriculture

There is no doubt that the immense challenge of increasing food production in Gilgil in particular and Kenya in general to keep pace with population growth and diminishing water resources requires focus on water productivity in both rainfed and irrigated agriculture. According to Rockstrom and Falkenmark (2000) and LEISA (1998) there seems to be no hydrological limitations, contrary to past pessimism from many development spectra, to doubling, or in many instances even quadrupling, yields of staple food crops in rainfed smallholder agriculture, even in drought prone semi-arid tropical agro-systems like Gilgil Division. SHDI Kenya’s strategies are based on the premise that there are several appropriate technologies and methodologies at hand to enable development towards improved soil and water productivity. This optimistic view is supported by a broad overview of recent projects on sustainable agricultural practices and technologies in Gilgil Division by SHDI Kenya and Ministry of Agriculture, which showed that yield increases as a result of introducing practices such as water harvesting, conservation tillage and drip irrigation. These interventions have produced good results, showing that they have a great potential for improving crop production in dry areas. Double digging, fertility trenches, half-moon micro catchments, Mandala gardens and planting pits technologies increase crop yields by a combination of moisture conservation and harvesting of runoff from the uncultivated spaces. In addition, soil fertility is maintained since the manure and fertilizer cannot be lost through surface runoff. The multi storey garden conserves moisture, manure and fertilizer however, unlike the other technologies does not harvest runoff at the same time. The entry point for SHDI Kenya’s strategy was management of water through provision of a means to bridge the persistent intra-seasonal dry spells. It is now evident that if only crop water access is secured, investments in soil fertility, crop husbandry, and timing of operations, will pay off in terms of substantially increasing soil and water productivity. Nevertheless, considering the benefits that even a simple rain water harvesting system can realize especially during severe dry spells, and low value labour-often the case during dry seasons in remote rural areas, it is likely that investment can be readily afforded and quickly recovered. Recent research results indicate that incorporating rain water harvesting- in-situ, supplemental irrigation (drip irrigation) and conservation tillage- can lead to increased water productivity, crop yields and food security (Rockstrom et al., 2001). Therefore, incorporating rain water harvesting with land husbandry that enhances water infiltration, improves water holding capacity and crop management and can have a great impact on agricultural water management and productivity i.e. the water use efficiency at farm level. This suggests that it is probably time to abandon the largely obsolete distinction between irrigated and rainfed agriculture and instead focus on integrated rainwater management. The overall effect would be increased food production, improved food security and water availability resulting to improved livelihoods. 3.0 Rainwater Harvesting Technologies Rainwater harvesting technologies and systems are either; in-situ rain water conservation like conservation tillage or runoff-based rainwater harvesting systems like storage systems, direct runoff application systems, flood diversion and spreading systems, small external catchment systems and micro-catchment systems. A

71

rainwater harvesting system (Figure 1) except in-situ water conservation should have the following components; runoff producing catchment, runoff collection structures, and runoff storage facility like water pans (Figure 2).

Fig. 1: The principle of runoff-based rainwater harvesting technology

In Gilgil Division the following technologies have been utilized, 3.1 Fanya juu terraces Fanya juu terraces are structures laid and excavated along the contours with soil being thrown up hill on slopes of between 5-30% for both soil conservation and moisture retention. The embankment traps the rainwater giving it more time to infiltrate while the channel acts as a retention ditch. To stabilize the embankment, fodder especially nappier grass is planted on it. 3.2 Double digging

Double digging is a method where the top soil is excavated in two different portions. The top one foot is placed on one side and the second one foot dug to break the hard pan before returning the top soil. This technology enhances crop root development and growth, water infiltration and conserves moisture and nutrients (Figure 3).

3.3 Multi-storey gardens

A multi-storey garden is a bio-intensive kitchen gardening technology that is applied in areas with shortage of land or in places with shortage of water. The garden is constructed by having a mixture of top soil and compost or rotten manure at a ratio of 3:1 placed in a gunny bag. A stone column is placed at the centre of the garden to allow for watering and aeration. In a homestead, it can even be squeezed on a verandah or any other little space available. It is able to carry about 40-70 plants. It is suitable for vegetables (Figure 4).

3.4 Planting pits Planting pits are used in dry areas where the soil is sandy or not fertile. It is another method of intensive kitchen gardening in which small holder farmers can increase crop yields especially maize from a small area, when the compost and manure are applied properly. A planting pit is a hole measuring 60cm x 60cm x 60cm in which compost or other organic fertilizer is mixed with top soil. The distance between the holes is also 60cm. Small holes 20 cm apart are then made for planting the seeds.

3.5 Half-moon Micro-catchments Half-moon micro-catchments are a network of earthen bunds shaped as half-circles with the tips facing upslope and on the contour. They are used in areas of 200-750mm rainfall, deep slopes and low slopes (Hai, 1998). They require even topography. The space between tips of consecutive bunds is used for discharging excess runoff. This excess runoff is then trapped by a series of semi-circular bunds below it. Half moons are helpful to rehabilitate degraded land.

Catchment (Natural surfaces, roads/footpaths, gullies, rills

ephemeral streams, croplands, pasture, hillslopes)

Runoff Conveyance and/or “Storage”

Cropland (Water applied directly or through

irrigation for storage systems)

72

3.6 Mandala Gardens A Mandala garden is a chain of open semi-circular trenches of width 60cm dug up to a depth of 60cm. The open sides of the semi-circles should face the source of surface runoff to channel water into the central pit. The semi-circles progressively increase in diameter with a spacing of 60cm between trenches. The sub-soil is put back and mixed with the top soil and compost or mature manure to make a raised bed along the trenches (Figure 5). 4.0 Conclusion It is envisaged that the results of these technologies will contribute to enhancing food security in Gilgil Division. Farmers’ experiences have proved to be the best learning lessons that have reinforced the need for participatory technology development and dissemination. Therefore, what is needed to address food security is to promote best practices through training on land users’ fields, building the capacity of stakeholders, technical improvements and the allocation of more resources to on-farm trials and land users’ exposure. Collaboration and networking would be paramount to achieve widespread promotion and adoption of these technologies and hence improved water availability, food security and livelihoods.

REFERENCES Agarwal, A. and S. Narain. (1997). Dying Wisdom: The rise, fall and potential of India’s traditional water

harvesting systems. Centre for Science and Environment, Thomson Press Ltd., faridada, India. Akilu, Y. and M. Wekesa. (2001). “Livestock and Livelihoods in Emergencies: Lessons Learnt from the

1999-2001 Emergency Response in the Pastoral Sector in Kenya.” OAU-IBAR, Kenya and Feinstein International Famine Centre, USA.

BAC, 2005. Sustainable Agriculture and Rural Development: A Training Manual (Unpublished) Benites, J., E. Chuma, R. Fowler, J. Kienzle, K. Molapong, J. Manu, I. nyagumbo, K. Steiner and R. van

Veenhuizen (eds.). (1998). Conservation Tillage for Sustainable Agriculture. Proc. International Workshop, Harare, 22-27 June, 1998. Part 1 (Workshop Report). GTZ, Eschborn, Germany. p59

Centre for Independent Research, (2005). Baseline survey report on food security and livelihoods for Gilgil Division, Nakuru District. Self Help Development International Kenya, Nakuru. Kenya

Critchley, W. (1987). Some lessons from Water Harvesting in Sub-Saharan Africa. Report from a Workshop held in Baringo, Kenya on 13-17 October, 1986. The World Bank Eastern and Southern Africa Projects Department. P 58

Critchley, W. and K. Siegert. (1991). Water Harvesting: A Manual for the Design and Construction of Water Harvesting Schemes for Plant Production. FAO, AGL/MISC/17/91. http://www.fao.org/docrep/U3160E00.htm

Evanari, M., L. Shanan and N. H. Tadmor. (1971). The Negev. The Challenge of a Desert. Harvard University Press, Cambridge, Massachusetts, USA.

Hai, M. (1998). Water harvesting: An Illustrative Manual for Development of Micro-catchment Techniques for Crop Production in Dry Areas. RELMA, Technical Handbook No. 16. Signal Press Ltd., Nairobi, Kenya.

IIRR and ACT. (2005). Conservation Agriculture: A manual for farmers and Extension Workers in Africa. International Institute of Rural Reconstruction, Nairobi; African Conservation Tillage Network, Harare.

IIRR, (1998). Sustainable agriculture extension manual for Eastern and Southern Africa. International Institute of Rural Reconstruction, Nairobi, Kenya.

IIRR, Cordaid and Acacia Consultants. (2004). Drought Cycle Management: A toolkit for the drylands of the Greater Horn of Africa. International Institute of Rural Reconstruction, Nairobi; Cordaid, The Hague; and Acacia Consultants, Nairobi.

IIRR. (2002). Managing Dryland Resources- An extension manual for Eastern and Southern Africa. International Institute of Rural Reconstruction, Nairobi, Kenya.

LEISA. (1998). Challenging Water Scarcity. ILEIA Newsletter for the Low External Input and Sustainable Agriculture. 14(1). Editorial.

Mogaka, H., Gichere, S., Davis, R., and Hirji, R., (2006). Climate Variability and Water Resources Degradation in Kenya: Improving Water Resources Development and Management. World Bank Working Paper No. 69.

73

Moi, D. Arap. (2002). Keynote address by the President of Kenya at the Integrated Water Resources Management Conference. Nairobi, March.

Mwango, F. K. (2000). Capacity Building in Integrated Water Resources Project in IGAD Region. An IGAD Publication.

Ngigi, S. N. (2002d). Preliminary evaluation of irrigation development in Kenya. In: Blank, H. G., C. M. Mutero and H. Murray-Rust (eds.). 2002. The changing face of irrigation in Kenya: Opportunities for anticipating change in Eastern and Southern Africa. IWMI, Colombo, Sri lanka. 93-111

Ngigi, S. N., (2003). Rainwater Harvesting for Improved Food Security: Promising Technologies in the Greater Horn of Africa. Greater Horn of Africa Rainwater Partnership (GHARP) and Kenya Rainwater Association (KRA), Nairobi, Kenya.

Rockstrom, J. (1999). On-farm green water estimatesas a tool for increased food production in water scarce regions. Phy. Chem. Earth (B), 24(4): 375-383

Rockstrom, J. (2000). Water Resources Management in Smallholder Farms in Eastern and Southern Africa: An Overview. Phy. Chem. Earth (B), 25(3): 275-283

Rockstrom, J. and M. Falkenmark. (2000). Semi-Arid Crop Production from a Hydrological Perspective: Gap between Potential and Actual Yields. Critical Review Plant Science, 19(4): 319-364

Rockstrom, J., J. Barron and P. Fox. (2001). Water Productivity in Rainfed Agriculture: Challenges and Opportunities for Smallholder Farmers in Drought-prone Tropical Agro-systems. Paper Presented at an IWMI Workshop. Colombo, Sri Lanka. November 12-14, 2001.

Shanan, L. and N. H. Tadmor. (1976). Micro-catchment Systems for Arid Zone Development. Hebrew University of Jerusalem and Ministry of Agriculture, Rehovot, Israel.

UNDP/ UNSO. (1997). Aridity zones and dryland populations: An assessment of population levels in the world’s drylands with particular reference to Africa. UNDP Office to Combat Desertification and Drought (UNSO), New York, USA.

74

Photographs

Figure 2: A newly excavated water pan at a farmer's field

Figure 4: A multi-storey garden utilized for kale production

75

Figure 3: A double dug bed in a farmer's field

Figure 5: A Mandala garden utilized for spinach production

76

CHALLENGES OF RAINWATER HARVESTING IN THE ARID LANDS OF KENYA AND TANZANIA

David M. Mburu, Jomo Kenyatta University of Agriculture and Technology, Biomechanical and environmental Engineering Department, P.O. Box 62000 (GPO 0100) Nairobi, Email: dmmburu@yahoo.co.uk

ABSTRACT

There are severe water related problems such as food shortage, poor sanitation, pollution and diseases affecting many countries, and especially the arid regions. Conventional water supply systems for domestic and industrial use are not adequate in solving the needs of the growing population in many countries. In the agricultural sector there is need to harvest and conserve rainwater where it falls. The potential of rainwater harvesting is not yet fully exploited. The promotion and upscaling of rainwater harvesting (RWH) technologies has become a sensitive matter in the fresh water resources agenda within the millennium development goals (MDGs). There is increasing number of countries facing water scarcity with large number of people living without access to safe water. The arid and semi-arid lands of Kenya and Tanzania are characterized by low and highly variable rainfall that rarely exceeds 800 mm, with most areas receiving 200-350 mm annually. A selected number of rainwater harvesting systems in Kenyan and Tanzanian drylands are presented, evaluating the challenges and possible solutions. Key words: Rainwater harvesting, arid and semi-arid, Kenya, Tanzania. INTRODUCTION The semi-arid areas are characterized by low annual rainfall concentrated to one or two short rainy seasons. The average annual rainfall varies from 400-600 mm in the semi-arid zone (Ngigi 2003). In the dry areas rainfall is highly erratic, and normally falls as intense storms with high spatial and temporal variability. The result is high risk of annual droughts and intra-seasonal dry spells. Due to high intensity rainfall events and poor canopy cover in the dryland, surface runoff is generally in the range of 10-25 % of the total rainfall received. Evaporation losses range in the order of 30-50 %, while deep drainage is 10-30 %. The evapotranspiration, which is the portion of rainfall that is used for production of biomass, is 15-30 %. Water scarcity in the dryland is a result of poor distribution of rainfall, large evaporative demand of the atmosphere, and large losses of water in the water balance (Rockstrom and Jonsson 1999). The risk of meteorological droughts and dry spells during the rainy season is high. This means that the poor distribution of rainfall over time often constitutes crop failure than absolute water scarcity due to low rainfall. (Barron et. al 1999). The development of water harvesting and storage systems would alleviate water scarcity problems and improve the livelihoods of the community. METHODOLOGY Water evaporation loss from the sand surface was determined by monitoring the drop in water table level in a metal cylinder closed at one end and buried in the riverbed. The water level was monitored with a tape measure through a piezometer tube. The sand in the cylinders was graded into fine (0.06-0.2 mm diameter), medium (0.2-0.6 mm diameter) and coarse (0.6-2 mm diameter) sand. The volume of sand transported to Nairobi from a part of Thwake river catchment was estimated by counting the number of lorries transported per day. The dimensions of the lorries were measured and the data on sand transported recorded in cubic metres. Transportation was normally between 4.00 a.m. and 8.00 p.m. The number of lorries carrying sand to Nairobi was counted at Machakos Agricultural show ground for 14 days. Transportation was normally by 8-ton lorries. The other water harvesting structures were visited during field investigations on water harvesting systems in Kitui in Kenya and Makanya village in Tanzania. The information on water use and performance of the systems was obtained from the water users.

77

RESULTS Water harvesting and storage in sand dams This technology has been used in Machakos and Kitui districts in Kenya. A masonry or concrete wall is constructed on a natural rock foundation across a riverbed at a suitably selected site with high potential of sand accumulation (plate 1). The dam wall is constructed in stages of 0.5 m high per season and made to conform to the shape of the riverbed. The reduced evaporation losses at 50 cm depth from the sand surface (Figure 1) made the system suitable for water storage (Mburu 2000).

The sand dams are potential target for scooping of sand by traders, who transport and sell it to the construction industry in towns and Nairobi city. Occasionally there arise conflicts between the sand traders and the local residents who want to retain the sand in the dams to store water. Ngachu (1988) estimated that about 200,000 m3 of sand was transported from Machakos district to Nairobi annually. It was found that in each day, an average 180 lorries of sand were transported from a part of Thwake river catchment to Nairobi, and each lorry carried about 5 m3 of sand. This amounted to 900 m3 sand/day with and annual projection of 328,500 m3. The sand dams are target sites for sand harvesters due to high sand accumulation. It was estimated that about 37% of water in the sand dams was lost through surface evaporation when the sand harvesters scooped sand.

Figure 1. Depth of active evaporation zone from the sand surface in a sand dam

0

5

10

15

20

25

30

35

40

45

50

0 5 10 15 20 25

Time (days)

Dep

th o

f wat

er ta

ble

(cm

) fine sand (0.06-0.2 mm diameter)medium sand (0.2-0.6 mm diameter)coarse sand (0.6-2 mm diameter)

78

Plate 1. Sand dam at full capacity (sand and water) during the rainy season in Kalama Location, Machakos district, Kenya. Water harvesting and storage in rock catchments The technology has been used in Mutomo division of Kitui district in Kenya where isolated areas of large rock outcrops is characteristic of the area. Suitable areas with large continuous rock surface are identified and stone/cement gutters are constructed round the rock surface to collect and direct water to a constructed tank or masonry wall that would form a dam. Such a dam (Kaseva rock catchment dam) has been constructed close to Mutomo divisional headquarters (Plate 2). It is the only reliable source of water serving a population of about 1000 people living within the township. In this system, high runoff is generation on the rock surface and most of it is collected and stored for use. It is easy to maintain good sanitation provided the collecting surface is not polluted. There is potential for expansion through construction of water tanks to increase the volume of water stored and limit evaporation losses. Plate 2. Kaseva rock catchment dam in Mutomo division, Kitui district, Kenya Water harvesting in Charco dams in Same district, Tanzania. Charco dams are common water harvesting structures in Makanya village in Same district in Tanzania. Charco dams are similar to farm ponds or water pans constructed in other parts of Kenya and Tanzania. The depth of excavation depends on the financial strength of the individual farmer. A number of farmers would normally team up and construct a Charco dam for their common use. Plate 3b shows the Charco dam full of water after the excavation was completed (Plate 3a). The period for water storage in the Charco dams depend on the livestock population. One livestock unit consumes an average of 50 litres of water per day. Three indigenous cows or fifteen sheep or goats are equivalent to one livestock unit (Mburu 2000). 3a 3b

79

Plate 3 (a) Excavation of Charco dam with hand tools and (b) Charco dam filled with water after excavation was completed in Makanya village, Same district in Tanzania. The soils in Makanya are highly erodible and cause rapid siltation of the Charco dams (Plate 4a). The dam owners spend much money in disilting the dams after three or four years. The silt accumulation reduces the capacity of the dam to retain adequate water for use during the critical period in the dry season. Silt traps were designed but they were not effective in reducing the silt load entering the dam (Plate 4b). Most of the silt traps could not accommodate the peak runoff rate. This caused some runoff to flow to the sides of the trap making it even more ineffective. The topography of the area was long and gentle slopes that made the catchment area of two or three Charco dams to overlap. This also created difficulty in designing silt traps for a specific Charco dam and also in determining the expected peak runoff rate. 4a. 4b. Plate 4 (a) Volume of silt accumulation in the Charco dam and silt trap designed to reduce silt load into the dam in Makanya village in Same district, in Tanzania. DISCUSSION Sand harvesting and water storage practiced in Machakos and Kitui districts can go together in the same sand dam if only sand collection could be strictly controlled. At Kwa Mbula sand dam in Kalama Location, a portion of the stored sand within a distance of 200 metres from the top end of the dam could be scooped out without significantly affecting the water storage capacity. From the site survey, this section had a storage volume of 25,754 m3 of sand, which could be scooped under controlled management. During the rains, the scooped sand could be. It was assessed that 900 m3 of sand was transported to Nairobi per day from part of Thwake river catchment. If all the lorries were directed to Kwa Mbula sand dam, which had a storage volume of 61,755 m3 of sand, it would take only 2 months to deplete all the sand. The results obtained showed that 0.5 metre was the critical depth from the sand surface, below which there were no significant evaporation losses. This indicated that when the surface layer of sand was continually scooped out, much of the stored water was lost through surface evaporation Water harvesting and storage in the rock catchment dams was a viable system where the geology of an area was suited for the system. Apart from surface evaporation, there was minimal water loss through seepage since the crevices in the rock formation would have been sealed with cement mortar at the time of construction. Good sanitation of the water source could be easily maintained by controlling the animals and people trampling on the runoff collecting area. This had worked well in case of Kaseva rock catchment dam in Mutomo division in Kitui district. Water storage in the Charco dams in Makanya had a significant contribution to the water supply for livestock use. High rate of siltation, seepage and surface evaporation were the major challenges in the system. An alternative technology would be treating the upper catchment with conservation measures and re-vegetation to improve the ground water re-charge. In such case shallow wells could be dug at

80

appropriate places in the lowland and utilize the ground water for both livestock and domestic purposes. This would have to be investigated and assess the possibilities of initiating such a project. Construction of storage tanks would reduce the amount of water loss and improve water security in the dry season. CONCLUSION The water harvesting systems described in this paper had positive contribution to the water supply in the arid areas of Kenya and Tanzania. Before any one of the systems was recommended for a certain area proper local investigations will have to be done. For example sand dams are only suitable in areas where the geological formation is such that it will yield high percent coarse sand. The rock catchment dams would only apply in area with large rock outcrop. Charco dams are not suited to areas with vertisols, which form large cracks during the dry season. The economic viability of the systems needs to be evaluated. A water balance accounting for the systems would be important to investigate and evaluate their efficiency in water harvesting, storage and utilization for different purposes. ACKNOWLEDGEMENTS The author appreciates the support obtained from the Regional Land Management Unit (RELMA) for financial support in carrying out the study. The field extension officers and farmers in Kitui and Machakos in Kenya and Makanya in Tanzania contributed variable information and guide to the field. Their contribution is highly appreciated. REFERENCES Baker B.H. 1954. Geology of the southern Machakos district. Geological survey of Kenya. Report No. 27. Government printer, Nairobi. Barron J., Rockstrom J., and Gichuki F. 1999. Rainwater management for dry spell mitigation in semi-arid Kenya. East African Agriculture and Forestry Journal (199) 65(1), 57-69 Gardener W.R. 1958. Some steady state solutions of the unsaturated moisture flow equation application to evaporation from a water table. Soil Science Journal, 85: 228-232. Hellwig D.H.R 1973. Evaporation of water from sand: The influence of the depth of the water table and the particle size distribution of the sand. Journal of hydrology, the Netherlands, 18 (305-327). Mburu D. M. 2000. The role of sand dams in water supply in arid areas: A case study in Machakos District. Land and water management in Kenya: Towards sustainable land use. In: Proceedings of the fourth National Workshop (15-19 February 1993), Nairobi, Kenya Pages 243-248). Morton J.D. 1958. The storage and abstraction of water conserved in sand rivers in the Gwanda district. Rhodesian Agricultural Journal 55: 399-406. Ngachu F.1988. Harvesting sand. Journal of the Institute of Engineers of Kenya. Kenya Engineer, Jan/Feb, Pp22. Ngigi S. N. 2003. Rainwater harvesting for improved food security: Promoting Technologies in the Greater Horn of Africa edited by Thomas D.B., Mutunga K. and Mburu D.M. Kenya rainwater Association, Nairobi, Kenya. Pp.264. Rockstrom J. and Jonsson L.O 1999. Conservation Tillage Systems for Dryland Farming: On-Farm Research and Extension Experiences. East African Agriculture and Forestry Journal (199) 65(2), 101-114

81

A PROTOTYPE TANK IRRIGATION SCHEME WITH RAINWATER HARVESTING

Macarius Yangyuoru13, Koichi Unami14 and Toshihiko Kawachi15

Abstract

A prototype small-scale gravity-fed tank irrigation scheme has been developed in the coastal savanna agro-ecological zones of Ghana, to scientifically investigate applicability of rainwater harvesting technology for agriculture in semi-arid regions. The main facility is the irrigation tank to harvest rainwater from 1.16 km2 catchment area and to supply water for 1.54 ha command area. Maize is the main crop grown in the command area, where Ethiopian Beds (Broad Beds and Furrows) are formed on Vertisols to facilitate furrow irrigation. Several automatic measuring devices are installed around the site to record various field parameters. Numerical models have been developed to simulate water balance in the reservoir. Based on these quantitative observations and computational analyses, it had shown that supplemental furrow irrigation during dry spell is feasible even in a drought year. Indeed, crop yield data in 2005 had demonstrated complete advantages of the tank irrigation.

Introduction Most of the inhabitants of tropical regions rely on local rainfall and traditionally organized and fragile agricultural systems for their livelihood. Low rainfall, its high variability of onset, and prolonged dry spells during the growing season often adversely affect crop growth and yields (Yangyuoru et al., 2003). According to Sonou (2004), the vagaries of the weather and uneven distribution of rain do not favour systematic and sustained development of the agricultural potentials of these areas, particularly in Africa south of the Sahara. Recent droughts have highlighted the risks to human beings and livestock when rains falter or fail, therefore requiring efficient conservation and management of this resource for successful production. Although irrigation might be the most obvious response to the droughts, there is increasing disillusion with sophisticated irrigation schemes which frequently present alarming social, economic, technical and environmental problems and benefiting only a fortunate few (Rowland, 1993). In Ghana for instance, irrigated agricultural land area is still less than 10,000 ha, representing only 0.23% and 0.073% of the total cultivated agricultural land and total arable land respectively (MOFA, 1991). There is therefore the need to increase irrigated agriculture to feed the ever-increasing population. Reviewing the 22 existing irrigation schemes of the government of Ghana, Nagayo (2004) recommends small-scale and gravity command irrigation schemes for sustainable development and for unlocking the country’s irrigation potential. This is in view of the fact that, high cost of electricity and fuel for lifting water for irrigation is becoming unsustainable as compared to gravity command irrigation. In a region where neither perennial surface water nor groundwater is available, the only viable way to achieve gravity command irrigation is to harvest rainwater (Boers and Ben-Asher, 1982). In this regard a pilot project of rainwater harvesting in semi-arid savanna regions has been initiated at Agricultural Research Centre (ARC)-Kpong, University of Ghana, in collaboration with Graduate School of Agriculture, Kyoto University, Japan, since 2001. This pilot project has constructed a prototype tank irrigation scheme that harvest rainwater for agricultural purposes (Kawachi et al., 2005). Various research activities are on-going, such as to computationally reproduce rainwater harvesting process (Unami et al., 2006), and to determine optimal water management strategy for the scheme as a whole (Unami et al., 2005). Here, focusing on water balance in the irrigation tank, rain to be harvested, evaporation loss, release of excess water from spillway, water withdrawal for irrigation, and soil moisture in crop fields are

13 Agricultural Research Centre, Kpong, College of Agriculture and Consumer Sciences, University of Ghana, P.O.Box LG 68, Accra Ghana, macarius_y@yahoo.com 14 Graduate School of Agriculture, Kyoto University, Kitashirakawa-oiwake-cho, Sakyo-ku, Kyoto, 606-8281 Japan, unami@adm.kais.kyoto-u.ac.jp 15 Graduate School of Agriculture, Kyoto University, Kitashirakawa-oiwake-cho, Sakyo-ku, Kyoto, 606-8281 Japan, tk@adm.kais.kyoto-u.ac.jp

82

comprehensively discussed. Then, showing crop yield data, the advantages of gravity command furrow irrigation are demonstrated. Site description The pilot project site at ARC-Kpong, University of Ghana, is located within the lower Volta basin of the coastal savanna agro-ecological zones of Ghana (Figure 1). The soil of the area, locally referred to as Akuse series (Adu, 1985), is classified as Calcic Vertisol (FAO/UNESCO, 1990) and Typic Calciustert (Soil Survey Staff, 1998). The building complex of the research centre is located at latitude 6o 08’ N, longitude 00o 04’ E and at an altitude of 22 m above mean sea level. The site is in the 1,024 ha extension of ARC-Kpong. The mean air temperature is 27.2oC, with mean maximum and minimum temperatures of 33.3oC and 22.1oC, respectively. The relative humidity for the nighttime to the early hours of the day ranges from 70 to 100% throughout the year. The afternoon relative humidity falls to a range of 20 to 65% during the year. Historical records of precipitation data, consisting of monthly rainfall data during 1955-1981 and during 1994-1995, and daily rainfall data since 1996, were analyzed for the climatic pattern of the project site. The rainfall distribution pattern within a year is bi-modal with mean annual rainfall of 1,167.4 mm (coefficient of variation = 20% and a standard deviation from the mean = 230 mm). Figure 2 shows monthly rainfall distributions in 1997, 2005, and mean values for all the available years. Rainfall in 1997 was abundant (1381.5 mm in total) while 2005 was a drought year (909.0 mm in total). Usually, the major rainy season is from March to mid-July and the minor rainy season from early September to mid-November. About 50% of the total annual rainfall (584 mm) occurs in the major season, and 30% (350 mm) in the minor season. Thus, the remaining 20% (233 mm) occurs during the off-season. There is high intensity of rainfall during the months of May to June and October. Excess rainfall, if harvested in the rainy seasons, would be utilized for the deficit periods from late November to early March. Outline of the tank irrigation scheme and observation facilities The main facility of the scheme is the irrigation tank (micro-dam), whose dam body has been completed in 2003. Direct rainfall onto its catchment area is the only water source for the irrigation tank. The total area of the catchment is 116 ha, and the slope is 1/89 on average. The dam site has no perennial or seasonal streams, but only ephemeral streams fed by sheet flows over the catchment area in the event of torrential rains. Thus harvesting rainwater can be achieved by collecting surface water running down the slope of catchment at a dash, in no expectation of subsurface runoff (intermediate and base runoffs). Excess water of the irrigation tank spills through a spillway, which is designed to release 36.2 m3/s flood discharge. The spillway weir is made of concrete and has a crest of B=30 m wide at EL30.58 m. The full water volume at the full water level EL30.58 m is 13,548 m3. The storage effect above the spillway weir crest does not allow the water level to rise higher than EL31.40 m even when the maximum inflow discharge of 41.0 m3/s occurs. The water once stored in the irrigation tank is exposed to the condition of high evaporation in tropical climate. A steel pipe is laid in the transverse of the dam body to draw-off water from the irrigation tank. The draw-off pipe is equipped with a spindle fitted valve for regulating water withdrawal at its mouth located at EL29.00 m. The withdrawn water is conveyed and distributed to 1.54 ha command area extending below EL27.80 m via open channels only by gravity. The concrete conveyance channel is 216 m long, while the earthen distribution channel is 330 m long. As the water passes through the distribution channel, it is shunted laterally into irrigation furrows progressively at 10 m intervals onto the farmland. A control plot of 0.02 ha is selected outside the command area for yields comparison. A plot of the command area and the control plot are respectively subdivided into four blocks for data collection. Both fields are ridged into Broad Beds and Furrows (BBF), also referred to as Ethiopian Beds in some literature. An improved maize variety, Obatanpa (Badu-Apraku et al., 2006), is selected as the test crop. The test crop is sown at a planting distance of 0.4 m within rows and 0.8 m between rows. The seeding rate is two seeds per hill.

Meteorological data are automatically recorded every hour in data loggers located at a station 2.2 km away from the dam body. Data items includes velocity vector of wind, air temperature, humidity, soil temperature, solar radiation, rainfall using a 0.5 mm tipping-bucket raingauge, and evaporation using a standard A-pan. The water level data of the irrigation tank is collected every 5 minutes. After the

83

completion of dam construction works, three other raingauges with 0.2 mm tipping-buckets are further installed to surround the catchment and are connected to pulse loggers. The hourly-recorded rainfall data confirms that rains in this region are torrential and the duration of a storm is about few hours. The maximum daily precipitation observed since 1996 is 104.5 mm (27th May 1996). Volumetric water content of soil is monitored and recorded every 1 hour in the command area. The sensor is situated at depth of 15 cm from the crest of the landform.

Water balance of irrigation tank The water balance of the irrigation tank is represented by the ordinary differential equation

dV dhA R L Q Idt dt

= = − − − (1)

where t is the time, V is the storage volume, A is the water surface area, h is the water level, R is the inflow discharge into the reservoir, L is the rate of water loss, Q is the discharge of spillway overflow, and I is the flow rate of irrigation water withdrawal. The inflow consists of direct precipitation on the water surface and runoff from the catchment. The water loss includes evaporation from the water surface, transpiration of aquatic plants, utilization by cattle, and seepage. Observation and estimation of water loss Using the data observed from August 2005 through July 2006, water loss from the irrigation tank is investigated. A dry spell is defined as a period satisfying the following conditions. • No rainfall is observed during the period as well as the preceding 3 days. • No irrigation water is withdrawn during the period. • The length of period is not less than 3 days.

In a dry spell, R, Q, and I can be regarded as 0. Thus, if the evaporation E is so dominant in the water loss

that L EA≈ , then dhEdt

≈ − is obtained from (1). Table 1 summarizes water levels of the irrigation tank

in 12 dry spells with solar radiation SR (MJm-2/day), mean air temperature T (C°), dhdt

− (mm/day), and

pan evaporation PANE (mm/day). There is strong correlation between decrease in water level and A-pan evaporation with a correlation coefficient of 0.922. Assuming that evaporation dominantly determines the water loss, the Makkink method (Xu and Singh, 2000) is applied to its estimation. The Makkink method relates evaporation E to meteorological data as

S

L

RE a bH

∆= +

∆ + γ (2)

where a and b are variable parameters to be estimated, ∆ is the slope of the saturation vapor pressure curve, γ is the psychromatic constant, and LH is the latent heat. It is known that 2.5 0.0024LH T= −

and 0.06041

1.05 1.4 Te−∆

=∆ + γ +

are available as practical estimations using the mean air temperature T.

When evaporation data in a specific site are readily available and their relationship to meteorological data is to be identified inversely, the Makkink method is more advantageous than the Penman method because the variable parameters can inclusively represent local characteristics. Regarding water level decrease as evaporation, the least square method identifies the parameter values 1.0433a = and 1.1147b = from the 12 data of Table 1. Estimation of inflow discharge If the water loss L is regarded as resulting from evaporation estimated by the Makkink method, the inflow discharge R during periods without irrigation is determined by

84

dVR L Qdt

= + + (3)

from meteorological data, water level data, water level-storage volume curve of the irrigation tank, and relationship between water level and overflow discharge from spillway. Figure 3 shows the water level-storage volume curve as well as the water level-water surface area curve.

The discharge Q over the spillway weir, where a critical section occurs, is represented by 322

3Q C BH gH⎛ ⎞= ⎜ ⎟

⎝ ⎠ (4)

where H is the overflow depth (water level of irrigation tank minus crest level of spillway weir), g is the gravitational acceleration, and C is the discharge coefficient. There is considerable sediment around upstream side of the spillway weir hindering the flow, and accordingly a small value 0.2C = is taken for the discharge coefficient.

After a dry spell, there were 3 rainfall events from 5th October through 7th October in the beginning of 2005 minor rainy season. The rainwater was successfully harvested, and the storage of the irrigation tank recovered up to its full capacity from 51%. Figure 4 shows the estimation result of the inflow discharge R from 5th October 0:00 through 11th October 0:00, together with the water level, and the water loss estimated as evaporation by the Makkink method. In Table 2, the inflow discharge R, which is equal to the runoff discharge from the catchment, is compared with rainfall observed in the three rain gauges surrounding the catchment. The runoff ratio increases drastically as rainfall accumulates, suggesting that the catchment approaches saturation. Water withdrawal and effect of irrigation The flow rate I of water withdrawal for irrigation purpose is designed to be 20L/s. In a day of irrigation, the valve is open for about 6 hours and withdrawn water is 432m3, which is equivalent to 3.2% of full capacity. Rainfall and volumetric water content of soil observed from 9th September 2005 through 6th January 2006 are depicted in Figure 5. There was no rain after 7th December 2005, and irrigation was performed on 16th December and 24th December. Increase in soil moisture similar as in a rainfall event can be seen. Maize yield from each block of the fields is summarized in Table 3. As is observable from these yield data, maize yields from the command area are significantly higher than the solely rainfed control plot. Conclusions Each term of the water balance equation (1) has now identified, and simulation under any water management strategy can be implemented. As a result, it had shown that irrigation every 3 days during dry spells is feasible even in a drought year like 2005. From the yield results, it is evident that rainwater harvesting is important for increased sustainable agricultural production. This will translate into increase income and livelihood for poor resource rural communities of semi-arid regions. Acknowledgement This study is partially supported by the grants-in-aid for scientific research, No.17688010, made by the Ministry of Education, Culture, Sports, Science and Technology, Japan. References [1] Adu, S. V. (1985): Vertisols and Acrisols in Ghana: Their physical and chemical properties and their

management. Laredec, Ahinsan-Kumasi, Ghana, In Vertisol Management in Africa, IBSRAM Proceedings, IBSRAM, No.9, p.329.

[2] Badu-Apraku, B., Twomasi-Afriyie, S., Sallah, P. Y. K., Haag, W., Asiedu, E., Mardo, K. A., Dapaah, S., and Dzah, B. D. (2006): Registration of ‘Obatanpa GH’ maize, Crop Science, 46, pp.1393-1395.

[3] Boers, Th. M., and Ben-Ashers, J. (1982): A review of rainwater harvesting, Agricultural Water Management, 5, pp.145-158.

[4] FAO/UNESCO (1990): Soil map of the world, revised legend. World soils resources report 60. FAO, Rome, 41p.

85

[5] Kawachi, K., Aoyama S., Yangyuoru, M., Unami, K., Matoh, T., Acquah, D., and Quarshie, S. (2005): An irrigation tank for harvesting rainwater in semi-arid savannah areas, Journal of Rainwater Catchment Systems, 11(1), pp.15-22.

[6] MOFA (Ministry of Agriculture) (2001): Agriculture in Ghana: Facts and Figures, Issued by Policy Planning, Monitoring and Evaluation, Ghana, 30p.

[7] Nagayo, N. (2004): An analysis of the present situation in irrigated agriculture in Ghana and recommendation for its promotion, Report of Small Scale Irrigation Promotion Project (SSIAPP), Ghana, 15p.

[8] Rowland, J. R. J. (1993): Dryland farming in Africa, CTA, Macmillan Press Ltd., London, pp.11-12. [9] Soil Survey Staff (1998): Keys to soil taxonomy. United States Department for Agriculture, Natural

Resources Conservation Service, p.288. [10] Sonou, M. (2004): Changes in irrigation organizations in developing countries: actual impacts, In

Irrigation development and management in Ghana; the way forward, FAO Regional Office for Africa, 10p.

[11] Unami, K., Kawachi, T., and Yangyuoru, M. (2005): Optimal water management in small-scale tank irrigation systems, Energy, 30, pp.1419-1428.

[12] Unami, K., Kawachi, T., Yangyuoru, M., and Ishida, K. (2006): A finite volume scheme for simulation of rainwater harvesting process, Proceedings of the second IASTED International Conference on Advanced Technology in the Environment Field, pp. 124-129.

[13] Xu, C.-Y., and Singh, V.P. (2000): Evaluation and generalization of radiation-based method for calculating evaporation, Hydrological Processes, 14, pp.339-349.

[14] Yangyuoru, M., Kawachi, T., Unami, K., Adiku, S. G. K., Mawunya, F., and Quarshie, S. (2003): Comparison of rainfed and potential yields of maize and cowpea on the Vertisols of Ghana, Journal of Rainwater Catchment Systems, 9(1), pp.7-12.

Table 1: Water level of irrigation tank and evaporation in dry spells

Initial day, final day, and number of days Initial h Final h SR T dhdt

− PANE

21-AUG-05 28-AUG-05 8 30.370 30.324 11.5 25.3 5.75 N.A. 11-SEP-05 23-SEP-05 13 30.279 30.188 16.2 27.0 7.00 6.50 27-OCT-05 31-OCT-05 5 30.565 30.528 18.6 26.6 7.40 5.60 25-DEC-05 06-JAN-06 13 30.455 30.390 13.6 27.0 5.00 3.90 11-JAN-06 31-JAN-06 21 30.369 30.250 14.3 27.8 5.67 4.61 04-FEB-06 12-FEB-06 9 30.232 30.170 15.9 28.7 6.89 5.53 13-APR-06 16-APR-06 4 30.157 30.124 20.1 29.6 8.25 7.00 01-MAY-06 06-MAY-06 6 30.371 30.328 19.3 28.9 7.17 7.21 18-JUN-06 20-JUN-06 3 30.512 30.497 14.3 25.3 5.00 3.58 25-JUN-06 27-JUN-06 3 30.500 30.480 18.1 26.6 6.67 5.58 04-JUL-06 06-JUL-06 3 30.513 30.495 16.8 26.4 6.00 5.00 12-JUL-06 28-JUL-06 17 30.465 30.381 13.5 25.5 4.94 3.96

Table 2: Rainfall and runoff in catchment

Start of rainfall 5th October 18:35 6th October 23:36 7th October 11:43 End of rainfall 5th October 22:25 7th October 01:55 7th October 13:44 Average rainfall depth among 3 raingauges

41.8mm 11.7mm 39.4mm

Standard deviation 1.6mm 2.5mm 14.7mm Total runoff 2,100m3 1,420m3 31,020m3 Total runoff depth 1.8mm 1.2mm 26.7mm Runoff ratio 4.3% 10.3% 67.8%

86

Table 3: Maize yield (grain kg/ha) with water management Command area

(irrigated) Control plot (rainfed)

Block No.1 1,340 502 Block No.2 1,056 357 Block No.3 1,447 384 Block No.4 1,031 527 Average 1,218 442 Standard deviation 207 84 Coefficient of variation 17% 19%

87

Figure 1: Water system of Ghana and location of ARC-Kpong

Figure 2: Monthly rainfall distributions

11

10

9

8

7

6

5

11

10

9

8

7

6

5

10123

10123

ARC-Kpong

Lake Volta

White Volta

Black Volta

Lake Bosomtwi

Keta Lagoon

Gulf of Guinea

0

50

100

150

200

250

300

Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

1997 2005 Mean

88

Figure 3: Water level-storage volume, surface area curves of irrigation tank

Figure 4: Rainwater harvesting process in irrigation tank

Figure 5: Rainfall and soil moisture in minor rainy season 2005

0 10 20 30 40

0 10 20 30 4031.5

31.0

30.5

30.0

29.5

29.0

Storage volume (10 m )

Surface area (10 m )

Wat

er le

vel (

m) H-V curve

H-A cu

rveF.W.L.

3 -3

2 -3

100 80 60 40 20 0

02040

SEP OCT NOV DEC JAN2005 2006V

olum

etri

c w

ater

con

tent

(%

)R

ainfall (mm

)

31.2

31.0

30.8

30.6

30.4

30.2

30.0

12

10

8

6

4

2

0

0

20

10/5 10/6 10/7 10/8 10/9 10/10Time (date of year 2005)

Wat

er le

vel (

EL

m)

Dis

char

ge (

m /s

)

Evaporation (m

m/day)

3 Evaporation

Water level

Inflow discharge

89

Rainwater Harvesting as a coping Strategy for Climate Change GREEN WATER MANAGEMENT AS A COPING STRATEGY AGAINST CLIMATE CHANGE Author: Orodi J. Odhiambo, Email: orodiodhiambo@yahoo.com The development agenda of many developing countries are increasingly being affected by climate related disasters including drought, floods and landslides largely because of the increasing climate variability and the risk associated with it. Rain-fed agriculture which accounts for a significant proportion of subsistence food production in Kenya will become more vulnerable with increasing climate variability and long-term climate change. This may increase the risk of food insecurity in the country. Together with other factors including rapidly growing population, poor management of natural resources and limited use of technologies, climate variability or long-term climate change could worsen the poverty situation in Kenya. Many and diverse impacts that are likely to result from climate change include change on the onset and cessation of rains making farming vulnerable, change in temperature and precipitation leading to new pests and diseases and change in the quality and quantity of water. A combination of approaches and this is only possible if stock is taken of what is being done to address current sources of vulnerability to climate variability and use this to inform long term adaptation strategies. The coping strategies reviewed in this paper address policy matters, water resource management, ecological and ecosystem sustainability through rainwater harvesting technologies. Constructed wetlands, artificial ground water recharge, rooftop and in-situ rainwater harvesting technologies are proven systems that could be adapted, adopted, up-scaled and replicated to cope with climate disasters in a sustainable way. USING ROOF STORAGE RAINWATER SYSTEMS AS A COPING STRATEGY ON DROUGHT RELATED DISASTERS Author: Orodi J. odhiambo, Alex R. Oduor and Maimbo M. Malesu, Email: orodijo@uonbi.ac.ke, aoduor@cgiar.org and mmalesu@cgiar.org Annual seasonal droughts of 2-4 months occur in Kusa limiting access of households to safe drinking water. This compounds the health and socio-economic disasters through increased water borne diseases rated at 10% morbility and 63% mortality and marginalizing economically the resource of rooftop rainwater harvesting 5m3 storage tanks has redressed the trend in 30% of the households owning these systems that harness the 900mm annual rainfall on 80-100m2 individual roof catchments. A study carried in the area through structured questionnaires, group discussions and literature survey revealed that the tanks operated at reliability and satisfaction levels of 44-59% when the guttering system covered 25% of the available roof area and 80-100% for coverage of 100% for daily demand levels of 100 liters. An assured supply of domestic water at homestead level resulted in a state of water security leading to increased use of water per capita thereby improving personal hygiene for the rural community. Morbidity and mortality rates from water vorne diseases reduced from 10% to 9.8% and 63% to 31% respectively for households with rooftop-tank systems. The study showed that well sized roof-tank combinations and appropriate demand managed strategies are effective measures for ameliorating household water supply to mitigate against drought caused health and socio- economic disasters in the area. USE OF RAINWATER HARVESTING RECHARGE TECHNOLOGY AS A GROUNDWATER DEPLETION AND FLOOD MITIGATION MEASURE IN THE CITY OF NAIROBI Author: Emily A. Oyoo and Orodi J. Odhiambo, Email: emittna@yahoo.com and orodiodhiambo@yahoo.com

90

Rapid population growth is accelerating rural-urban migration leading to increased urbanization in African cities. This is causing an increased demand for water and housing leading to increased abstraction of ground water resources through sinking of boreholes and increased runoff coming from buildings and pavements resulting in depletion of groundwater resources and flooding. Rainwater harvesting is muted as a mitigation measure against depletion and flooding through the use of artificial recharge technology. The paper reviews the knowledge and policy gaps in the mainstreaming of official rainwater harvesting recharge in the environmental impact assessment tools in the building and construction and ground water exploitation industries and proposes policy advocacy strategies on maximum permissible runoff from constructed sites and mandatory recharge augmentation for boreholes drilled in areas with favourable geological formation.

iUnited Nations, 2006: Water: a shared responsibility - The United Nations World Water

Development Report 2. World Water Assessment Programme of the United Nations, Case studies (pp 477-479).

iiEUWI/Ethiopia, September 2006: Financing Strategy for the WSS sector of Ethiopia: Executive Report (Draft)

iiiUNEP/IETC, 1998. Sourcebook of Alternative Technologies for Freshwater Augmentation in Africa. IETC Technical Publication Series, Issue 8.

ivFalkenmark, M., J. Lundqvist, and C. Widstand 1990. Water Scarcity - An Ultimate Constraint in Third World Development. Linkoping University, Stockholm.

vGould J. and E. Nissen-Petersen, 1999. RAINWATER CATCHMENT SYSTEMS FOR DOMESTIC SUPPLY: Design, construction and implementation. Intermediate Technology Publications, ITDG, London, UK.

viAlaerts G., F. Hartvelt and J. Warner, 1997. Capacity Building – Beyond the Project Approach. Waterlines, Vol. 15, No. 4, pp 2-5. viiAustin J., H. Otterstetter and F. Rosensweig, 1987. Institutional and Human Resource Development. Waterlines, Vol. 5, No. 4, pp2-5. viiiEuropean Commission, 1998. Towards Sustainable Water Resources Management - A Strategic Approach. Office for Official Publications of the

European Communities, L-2985 Luxembourg (CD-RM: World Environmental Library) ixIDRC, 1996. Water Management in Africa and the Middle East: Challenges. (CD-RM: World Environmental Library)