UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND
A C T A U N I V E R S I T A T I S O U L U E N S I S
University Lecturer Tuomo Glumoff
University Lecturer Santeri Palviainen
Postdoctoral researcher Jani Peräntie
University Lecturer Anne Tuomisto
University Lecturer Veli-Matti Ulvinen
Planning Director Pertti Tikkanen
Professor Jari Juga
University Lecturer Anu Soikkeli
University Lecturer Santeri Palviainen
Publications Editor Kirsti Nurkkala
ISBN 978-952-62-2604-0 (Paperback)ISBN 978-952-62-2605-7 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
OULU 2020
C 747
Faisal Bin Ashraf
RIVER REGIMES AND ENERGY DEMAND INTERACTIONS IN NORDIC RIVERS
UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY
C 747
AC
TAFaisal B
in Ashraf
C747etukansi2.fm Page 1 Tuesday, April 7, 2020 10:40 AM
ACTA UNIVERS ITAT I S OULUENS I SC Te c h n i c a 7 4 7
FAISAL BIN ASHRAF
RIVER REGIMES AND ENERGY DEMAND INTERACTIONS IN NORDIC RIVERS
Academic dissertation to be presented with the assent ofthe Doctoral Training Committee of Technology andNatural Sciences of the University of Oulu for publicdefence in the Arina auditorium (TA105), Linnanmaa, on29 May 2020, at 12 noon
UNIVERSITY OF OULU, OULU 2020
Copyright © 2020Acta Univ. Oul. C 747, 2020
Supervised byProfessor Bjørn KløveAssistant Professor Hannu Marttila Associate Professor Ali Torabi Haghighi
Reviewed byProfessor Tor Haakon BakkenAssociate Professor Andrea Castelletti
ISBN 978-952-62-2604-0 (Paperback)ISBN 978-952-62-2605-7 (PDF)
ISSN 0355-3213 (Printed)ISSN 1796-2226 (Online)
Cover DesignRaimo Ahonen
PUNAMUSTATAMPERE 2020
OpponentAssociate Professor Miroslav Marence
Ashraf, Faisal Bin, River regimes and energy demand interactions in Nordic rivers. University of Oulu Graduate School; University of Oulu, Faculty of TechnologyActa Univ. Oul. C 747, 2020University of Oulu, P.O. Box 8000, FI-90014 University of Oulu, Finland
Abstract
River regimes in Nordic rivers are changing, mainly due to climate change and river regulation.As renewables penetrate the power market, the conventional role of hydropower in energy marketsis also changing. These factors are altering river flows in ways that are not fully understood. Thisthesis addressed these information inadequacies in the ecologically sensitive Nordic region.
To assess the relative effect of regulation practices and climate change, a spatio-temporal studyof two adjacent rivers in Northern Europe, the Kemijoki (regulated) and Tornionjoki (pristine),with similar climate and catchment conditions, was performed. Degree of hydrologic alteration onKemijoki was twice than at Tornionjoki, while regulation and climate change seemed to havesimilar degrees of effect on flow alteration in the Kemijoki (50% each). This was confirmed byanalyzing data from Ounasjoki, a pristine tributary of the Kemijoki.
Short-term changes in Nordic rivers were quantified using discharge datasets from 150 siteswith hourly time-step. The results revealed high levels of hydropeaking in Nordic rivers, withincreases especially in recent few years. This indicates that expanding for renewable energy,increasing the need for load balancing in the energy market, may increase hydropeaking in Nordicrivers.
A method based on wavelet analysis was developed to characterize variability in hydrologictime series data in different periods. It was found that, in winter, sub-daily variations in some largeRegulated Rivers were up to 20 times higher than in Free-Flowing Rivers. For smaller RegulatedRivers with lower levels of regulation, the variation was highest in summer.
We have assessed the impact of intra-day power demand variation on Regulated Rivers flowand potential to comply with instant energy demand, by quantifying the impact of reservoirvolume and hydropower capacity constraints using two new metrics, power market impact andsystem efficiency ratio. The metrics were tested for the Kemijoki, with defined thresholds basedon the natural flow regime (Ounasjoki) and hourly energy price in Finland in 2017, to estimate theimpact of regulation on hourly flow regime at Taivalkoski station. Annual flow regime impact in2013, 2014, and 2015 was estimated to be 74%, 84% and 61%, respectively, while monthly impactvaried from 27% to 100%.
Keywords: dams, hydropeaking, power market, regulation, river regime, sub-daily flow
Ashraf, Faisal Bin, Muuttuvat jokien virtaamat: Ilmaston ja energiamarkkinoiden vaikutusten arviointi pohjoisten jokien muutoksiin. Oulun yliopiston tutkijakoulu; Oulun yliopisto, Teknillinen tiedekuntaActa Univ. Oul. C 747, 2020Oulun yliopisto, PL 8000, 90014 Oulun yliopisto
Tiivistelmä
Pohjoismaisten jokien virtaamat ovat muutoksessa pääasiassa ilmastonmuutoksen ja jokien sään-nöstelyn takia. Säännöstelyn voimakkuuteen vaikuttaa vesivoiman alati muuttuva rooli energia-markkinoilla etenkin uusiutuvien energiamuotojen lisääntyvä osuus jolloin tarvitaan enemmänsäätövoimaa. Muuttuvien lyhyen ja pitkän aikavälin säännöstelykäytäntöjen vaikutuksista joki-en virtaamaolosuhteisiin tarvitaan lisätietoa etenkin ekologialtaan herkistä pohjoisista joista.
Säännöstelykäytäntöjen ja ilmastomuutoksen suhteellisen vaikutuksen arvioimiseksi tehtiintila-ajallinen tutkimus kahdelle ilmasto- ja valuma-alueolosuhteiltaan samankaltaisella pohjois-eurooppalaiselle joelle, joista Kemijoki edustaa säännösteltyä ja Tornionjoki säännöstelemätön-tä jokea. Säännöstellyllä joella havaittiin 100 % korkeampi hydrologinen muutos verrattaessasäännöstelemättömään. Joen säännöstellyllä ja ilmastomuutoksella näyttää olevan samanlainenvaikutus Kemijoen virtaaman muutoksiin (molemmat 50 %). Tulos vahvistettiin analysoimallatietoja Kemijoen säännöstelemättömän Ounasjoen sivuhaaran aineistolla.
Lyhytaikaisia muutoksia pohjoismaisissa joissa määritettiin kvantitatiivisesti käyttämällä 150joen laajoja tuntikohtaisia virtaamatietoja. Tulosten mukaan lyhytaikaissäännöstely on näidenjokien tapauksessa korkealla tasolla, ja se on lisääntynyt viimeisen vuosikymmenen aikana. Syy-nä voi olla uusiutuvien energialähteiden lisääntyminen energiamarkkinoilla, sillä markkinoillatarvitaan enemmän säätöenergiana toimivaa vesivoimaa tasapainottamaan sähköntuotannonvaihteluita.
Työssä kehitettiin myös wavelet- aikasarja-analyysiin perustuva menetelmä hydrologisenaikasarjatiedon vaihtelevuuden karakterisoimiseksi eri ajanjaksoina. Tulosten mukaan joidenkinsäännösteltyjen jokien päivittäiset virtaamavaihtelut olivat etenkin talvisin jopa 20 kertaa suu-remmat kuin säännöstelemättömien jokien. Pienemmissä säännöstellyissä joissa muutokset oli-vat suurimpia kesäaikana.
Päivittäisen energian kysynnän vaihtelun vaikutusta arvioitiin kahdella uudella indeksillä(energiamarkkinavaikutus ja järjestelmän tehokkuussuhde). Indekseja testattiin Kemijoelle mää-ritetyillä kynnysarvoilla, jotka perustuivat säännöstelemättömään Ounasjokeen ja tuntikohtai-seen energian hintaan Suomessa (2017). Vuotuinen vaikutus virtaamaolosuhteisiin Taivalkoskenpadolla vuosina 2013, 2014 ja 2014 oli 74 %, 84 % ja 61 %, kun taas kuukausittainen vaikutusvaihteli 27 %:sta 100 %:iin.
Asiasanat: energiamarkkinat, lyhytaikaissäännöstely, padot, säännöstely, tuntivirtaama,virtaamat
Rivers
8
9
Acknowledgements
I thank Nessling Foundation (Maj ja Tor Nesslingin Säätiö), Maa-ja vesitekniikan tuki ry (MVTT), Sven Hallin Research Foundation, University of Oulu Graduate school, and Olvi Foundation for their continued financial support during my PhD studies. This work would not have been possible without their support.
I express sincere gratitude to my principal supervisor, Professor Bjørn Kløve, for his continuous support, patience, and motivation during my PhD studies. I also thank my co-supervisors, Dr. Hannu Marttila and Dr. Ali Torabi Haghighi, for their insightful comments, encouragement, and constructive criticism which impelled me to broaden my research from various perspectives. Their involvement with the project was immensely helpful at every step, from conception of the research project until the writing of this thesis. Hannu, your maddening attention to detail finally drove me to get the color and font combination in my graphs correct; Ali, your sense of humor came in handy at times when I had lost mine (your ‘Two negatives make a positive’ motto is legendary). I could not have imagined having a better group of supervisors and mentors for my PhD studies, their selfless time and care kept me going. I am also thankful to Dr. Anna-Kaisa Ronkanen and Dr. Pekka Rossi for being my follow-up group members and giving valuable inputs during my studies.
I am indebted to Uzair Akbar Khan for being a remarkable office buddy. On writing this, I am reflecting about countless conversations and moments we shared, and I want to tell how grateful I am for our time spent together as colleagues and, more importantly, as friends. I am also thankful to Markus Saari, you have been my window into Finnish culture and Finland in general, thank you for answering all the questions that have baffled me about Finland over the years. I feel lucky to also have friends like Mohsin Jahan Qazi, Zubair Akhone, and Zakaria Laskar, who have always been very supportive and understanding.
I thank my fellow lab mates for exciting discussions, for the long working days together before deadlines, and for all the fun we have had in the past four years. Finally, I would like to thank my family: my parents, brother, and sister, for supporting me throughout this journey.
Oulu, January 2020 Faisal B. Ashraf
10
11
Abbreviations
AF Annual flow (m3) AGWPS Annual average global wavelet power spectrum value AQ Allocated flow CA Catchment area cm3 Cubic centimeters CSI Connectivity status index CWT Continuous wavelet transform DCC Daily contribution coefficient DFT Discrete Fourier transform Dfc Subarctic climate DHA Degree of hydrologic alteration DOF Degree of fragmentation DOR Degree of regulation EF Environmental flow EFC Environmental flow component ELVMax Maximum water level in reservoir ELVMin Minimum water level in reservoir ELVNor Normal water level in reservoir EU European Union EU WFD European Union Water Framework Directive FC Flood control FFRs Free-flowing rivers GWh Gigawatt hours GWPS Global wavelet power spectrum HA Hydrologic alteration HAV Hydrologic alteration values HHH High-head hydropower HP Hydropeaking HP1 Hydropeaking indicator 1 (intra-day day change) HP2 Hydropeaking indicator 2 (intra-day ramping rate) HPcap Hydropower capacity HPS Hydropower scale IHA Indicators of hydrologic alteration IPCC Intergovernmental Panel on Climate Control IQR Interquartile range
12
IR Irrigation kW Kilowatts LTR Lower threshold release Max Maximum MCC Monthly contribution coefficient mdt Month, day, time (hour) MIF Magnitude impact factor Min Minimum MW Megawatts MWh Megawatt hour(s) NAH Natural annual hydrograph ndym Number of days in month m PMI Power market impact PPM Power per meter of available dam head Qhp Outflow from hydropower (m3 s-1) Qin Inflow to reservoir (m3 s-1) QSp Outflow from spillway (m3 s-1) RDD Road density RE Renewable energy RI River impact ROR Run of the river RRs Regulated rivers RRI River regime index RVA Range variability approach SED Sediment trapping index SER System efficiency ratio TCC Hourly contribution coefficient TRHP1 Hydropeaking threshold for HP1 TRHP2 Hydropeaking threshold for HP2 URB Urban areas USE Consumptive water use UTR Upper threshold release VIF Variation impact factor VMax Maximum volume in reservoir VMin Minimum volume in reservoir VNor Normal volume in reservoir VRES Variable renewable energy resources
13
WS Water supply WPS Wavelet power spectrum μ Mean annual flow (m3 s-1) τ Kendall's Tau 𝜂 Efficiency
14
15
List of original publications
This thesis is based on the following publications, which are referred to throughout the text by their Roman numerals:
I Ashraf, F. B., Haghighi, A. T., Marttila, H., & Kløve, B. (2016). Assessing impacts of climate change and river regulation on flow regimes in cold climate: a study of a pristine and a regulated river in the sub-arctic setting of Northern Europe. Journal of Hydrology, 542, 410-422.
II Ashraf, F. B., Haghighi, A. T., Riml, J., Alfredsen, K., Koskela, J. J., Kløve, B., & Marttila, H. (2018). Changes in short term river flow regulation and hydropeaking in Nordic rivers. Scientific reports, 8(1), 17232.
III Ashraf, F. B., Haghighi, A. T., Riml, J., Kløve, B., & Marttila, H. (2020). Assessment of sub-daily flow alterations using globally mapped free-flowing rivers and wavelet analysis. Manuscript.
IV Haghighi, A. T., Ashraf, F. B., Riml, J., Koskela, J., Kløve, B., & Marttila, H. (2019). A power market-based operation support model for sub-daily hydropower regulation practices. Applied Energy, 255, 113905.
The author’s contribution to Papers I- IV was as follows:
I Designed the study with guidance from the supervisor and co-authors. Collected the necessary data required for the study, performed the data analysis, and wrote the paper. The co-authors provided critical and helpful information during writing the paper.
II Designed the study with guidance from the supervisor and co-authors. Collected the necessary data required for the study, performed the data analysis and wrote the paper. The co-authors provided critical and helpful information during writing the paper.
III Designed the study with guidance from the supervisor and co-authors. Formulated the methodology, performed the data analysis and wrote the paper. The co-authors provided critical and helpful information during writing the paper.
IV Designed the study with guidance from the supervisor and co-authors. Helped in formulation of the methodology, performed the data analysis. The co-authors provided critical and helpful information during writing the paper.
16
17
Contents Abstract Tiivistelmä Acknowledgements 9 Abbreviations 11 List of original publications 15 Contents 17 1 Introduction 19
1.1 Background on river regulation for hydropower generation ................... 19 1.2 Environmental concerns regarding excessive river regulation ................ 22 1.3 Motivation for this study ......................................................................... 25
1.3.1 Needs arising due to climate change (Paper I) ............................. 25 1.3.2 Needs arising due to energy production and demand
change (Paper II) .......................................................................... 25 1.3.3 Need for new indices (Paper III) .................................................. 26 1.3.4 Need to connect river flow alteration to the power markets
(Paper IV) ..................................................................................... 26 2 Research objectives and thesis structure 29 3 Brief overview of the study area, data sources, and the Nordic
hydropower system 33 4 Methodology 37
4.1 Indicators of hydrologic alteration and monthly flow regime index ........................................................................................................ 37
4.2 Hydropeaking indicators, thresholds, and hydropeaking pressure classes ..................................................................................................... 38 4.2.1 Hydropeaking indicators .............................................................. 38 4.2.2 Thresholds and hydropower pressure classes ............................... 39
4.3 Hydrologic time series analysis by continuous wavelet transform ......... 40 4.4 Formulation of an index to categorize sub-daily flow variation
levels ....................................................................................................... 41 4.5 Setting flow variability boundaries ......................................................... 44 4.6 Designing different regulation practices ................................................. 45 4.7 Release simulations for different scenarios ............................................. 47 4.8 Power market impact (PMI) and system efficiency ratio (SER) ............. 49
5 Results and discussion 51 5.1 Effects of long-term river regulation ....................................................... 51
18
5.1.1 Impacts of river regulation on river regime in a sub-Arctic setting ........................................................................................... 51
5.1.2 Changes in extreme flow events ................................................... 51 5.1.3 Flow alteration at different river points ........................................ 52
5.2 Effect of short-term regulation on sub-daily flow dynamics of Nordic rivers ........................................................................................... 52 5.2.1 Measure of hydropeaking in the region ........................................ 52 5.2.2 Relationship between hydropeaking, catchment and
hydropower plant properties ......................................................... 53 5.2.3 Current hydropeaking trends in the region. .................................. 54
5.3 Evaluation of a new classification method to assess hydropeaking ........................................................................................... 55 5.3.1 Assessing and characterizing annual hydropeaking in
terms of sub-daily variations in free-flowing rivers ..................... 55 5.3.2 Seasonal hydropeaking in terms of sub-daily variations .............. 55
5.4 Interaction between power market and short-term river flow regime ...................................................................................................... 58 5.4.1 Effect of energy price variations on flow regime
thresholds ...................................................................................... 58 5.4.2 Combined effect of fluctuating power prices and
regulation practices on hydropeaking regime ............................... 58 5.4.3 Power market impact (PMI) and system efficiency ratio
(SER) ............................................................................................ 60 5.4.4 Advantages of the new methodology ........................................... 61 5.4.5 Limitations of the framework ....................................................... 62
6 Conclusions 63 7 Future studies 67 List of references 69 Appendices 77 Original publications 79
19
1 Introduction
1.1 Background on river regulation for hydropower generation
Globally, river regulation, water abstraction, and climate change are major causes of river regime alteration. River regulation is mainly attributable to construction of dams for hydropower, flood control, and irrigation. Around 47,000 large dams and 800,000 smaller dams have been constructed during the past 100 years (Richter & Thomas, 2007; Rosenberg et al., 2000), causing extensive morphological and flow changes in the world’s major rivers. Change is natural river flow dynamics is recognized as a key threat to endangered species in many river systems (Bunn & Arthington, 2002). The European Water Framework Directive (European Union [EU], 2009) and Blueprint (European Commission [EC], 2012) recognize the importance of water quantity for biotic and recreation needs in defining sustainable discharge flow conditions.
There is global consensus on the United Nations Sustainable Development Goals (SDGs), which include “protect and restore water-related ecosystems” (target 6.6). The rising demand for energy has resulted in an increase in the number of dams being constructed worldwide for a seemingly flexible and non-carbon emitting power source. More than 3700 hydropower dams (>1 MW) are currently being planned or are under construction worldwide (Zarfl et al., 2015). According to the 2018 hydropower status report by the International Hydropower Association, growth was fastest in East Asia and the Pacific, with 9.8 GW of capacity added in 2017, while China, Brazil and India together added around 16 GW of installed capacity in 2017 (International Hydropower Association, 2019). Ecologically sensitive and regions like the Amazon and the Himalayas are also observing an increasing trend in major hydropower construction (Winemiller et al., 2016). Thus, at present, only one-third of large global rivers as classified as free-flowing (Grill et al., 2019).
Hydropower has an important role in Europe’s power market and is surging in many developing counties, because of its flexibility (Baron, 2013). High sub-daily flow variance is typically driven by management practices governed by intra-day differences in electricity prices to achieve maximum economic benefit. Hydropeaking is the process of releasing more water through turbines during peak hours of electricity/power demand and less or no water during low demand hours. This increases short-term stream flow variability and causes unnaturally fluctuating flow conditions in downstream river reaches. Climate change is also an important
20
factor that is altering hydropower operation practices and river flows, especially in the Nordic countries. According to some climate change and precipitation projection studies, in the 21st century the Nordic region may experience a significant rise in both annual mean temperature and precipitation (Irannezhad, Chen, & Kløve, 2014). As a result of temperature increases, snow accumulation has declined in the area (Irannezhad, Ronkanen, & Kløve, 2016; Park et al. 2013; Park et al., 2013). Similar variations in boreal regions worldwide have been documented in various climate change studies and most climate models for the next century (Beldring et al., 2008; Vehviläinen et al., 2012; Veijalainen et al., 2010). These meteorological changes have altered river discharge and hydropower reservoir management practices, with e.g., premature spring peak flooding and increased spring and winter runoff (Korhonen & Kuusisto, 2010; Wilson et al., 2010).
Power markets are changing rapidly due to an increasing share of renewable energy production (wind, solar, etc.). The energy-water nexus and stricter environmental restrictions have complex influences on riverine flow patterns, and changes in both may adversely affect hydropower plant operation and, thus, operation (Gaudard & Romerio, 2014; Vehviläinen et al., 2012). The increasing share of renewables demands a shift in short-term regulation of hydropower, to compensate for deficits between supply and demand. Future hydropower regulation practices may also change under a changing climate (Patro et al., 2018) and due to the economic benefits of reducing greenhouse gas emissions (Boehlert et al., 2016), thus further altering hydrological regimes. In order to evaluate the influence of hydropower on natural stream flow variability, investigations capturing flow alterations across a wide spectrum of temporal scales are required. In the Nordic countries, hydropower generation is partly controlled by a common electricity market. Wholesale electricity is traded through the Nord Pool Spot power exchange, which uses Elspot (day-ahead) and Elbas (intra-day) to set the market price. In the deregulated power market, hydropower production is bid into the day-ahead market (Aasgård et al., 2019) and is also used to provide load balancing at intra-day and intra-hour time steps (Vardanyan & Hesamzadeh, 2017). To maximize total generation of revenue by providing load balancing at intra-day and intra-hour time steps, operators prefer to release more water through turbines when the price is high (often during peak demand hours) and store water or release less water when the price is low (often during low demand hours). The release rate can change at intra-hour time steps, causing hydropeaking, which is an important issue for major Nordic rivers (Figure 1). As the share of variable renewable energy sources increases, the role of hydropower as a load balancing source of power also increases
21
(Gaudard & Romerio, 2014), since energy production from emerging renewable sources (e.g., wind, sun) varies widely on short scales (Mileva et al., 2016; Raynaud et al., 2018). This growing demand for balancing energy from hydropower puts pressure on operators to increase hydropeaking in regulated river systems, exerting additional pressure on riverine ecosystems.
During the past two decades, river management policies/practices have employed more holistic approaches, which has led to wide acceptance of the natural flow regime as the reference condition of river flow dynamics (Poff et al., 1997). Hydropower operations thus use natural flow regime as a benchmark for ecologically sustainable flow dynamics. Conventionally, natural flow regime corresponds to pre-dam or unregulated river flow conditions and is now considered when granting new dam licenses (Jager & Bevelhimer, 2007). Unregulated, free-flowing rivers were previously simply described as “unimpounded” or “unregulated” rivers (Nilsson et al., 2005; Palmer et al., 2008). Although this definition may be acceptable to assess the changes from a purely hydrological perspective, it does not take into account all aspects of truly free-flowing river flow dynamics. In a recent study, Grill et al. (2019, p. 216) defined free-flowing rivers as “rivers where ecosystem functions and services are largely unaffected by changes to the fluvial connectivity, allowing unobstructed movement and exchange of water, energy, material and species within the river system and with surrounding landscapes”. Their methodology thus offers improved benchmark status for evaluating regulated rivers.
To characterize natural and human-impacted flow regimes, numerous hydrologic indices have been developed e.g. (Haghighi & Kløve, 2013; Poff & Zimmerman, 2010; Vanzo et al., 2016). These hydrologic indices are mainly used to assess flow regime alterations or evaluate environmental flow designs. The most commonly used approaches at present for quantifying flow variability are based on statistical analyses of monthly- or daily-averaged flow records (Haghighi et al., 2014; Richter et al., 1996). These include the indicators of hydrologic alteration (IHA) method (Richter et al., 1996), which is the most widely used methodology (Richter et al., 1997; Yang et al., 2008), and the associated range variability approach (RVA), which can be used to evaluate flow variations (Poff et al., 1997; Stanford et al., 1996). Others have developed hydrologic parameters and indices to quantify hydropeaking and sub-daily flow variations (Meile et al., 2011). However, these existing methodologies can mask hourly variation effects, and hence do not provide a comprehensive impact assessment for sub-daily flow variations. In recent years, wavelet analysis methods have improved flow regime assessment by
22
enabling simultaneous detection and quantification of flow variability, especially at sub-daily temporal scales (Percival & Walden, 2000; Steel & Lange, 2007; Torrence & Compo, 1998; Wu et al., 2015). Wavelet analysis allows localization in time, i.e., non-stationary processes can be analyzed, so it may be useful when analyzing market-driven hydropower production. It can also be utilized to formulate a framework that categorizes sub-daily flow variations due to hydropeaking, a method for which is currently lacking.
Considering the major changes currently affecting the energy industry and the growing use of hydropower for balancing supply and demand, more comprehensive indices for evaluation of high-frequency, non-stationary temporal flow patterns are required. This is particularly the case for rivers subjected to hydropeaking reservoir operations, where rapid fluctuations are typically present at sub-daily to daily scales (Bevelhimer et al., 2015; Meile et al., 2011; Zimmerman et al., 2010).
1.2 Environmental concerns regarding excessive river regulation
Alteration in natural river flow regime is recognized as a key threat to many riverine species and ecosystem services which rivers provide (Arthington et al., 2010; Dynesius & Nilsson, 1994; Tonkin et al., 2018). Hydropower impacts river systems by altering the patterns of water flow (Zhang et al., 2018) and affecting natural habitats (Ansar et al., 2014; Hennig et al., 2013; Rosenberg et al., 2000; Sternberg, 2006). Damming and regulation of river systems result in homogenization of long-term river dynamics (Dynesius & Nilsson, 1994; Haghighi & Kløve, 2015; Poff et al., 2007) and, hydropeaking (Arheimer et al., 2017; Bejarano et al., 2017). Hydropeaking poses challenges in maintaining current river ecological status, and thus the potential effects should be evaluated in more details.
Flow regulation and damming of river systems cause change in natural river flow dynamics, therefore adversely affecting aquatic biodiversity (Dynesius & Nilsson, 1994; Mustonen et al., 2016; Poff et al., 2007). Biotic communities are influenced by variations in both temporal and spatial flow conditions (Poff & Ward, 1989), and hence studying alterations in natural flow regimes is a first step in determining the extent of total influence. In the Nordic region in particular, climate warming will rapidly change the hydrological cycle, and therefore information on potential changes in water systems is urgently required (Prowse et al., 2015). In Fennoscandinavia, many major rivers and lakes are regulated for hydropower purposes (Hellsten & Riihimäki, 1996; Svensson, 2000) which has caused drastic changes in river biotic structures, especially in salmonid species populations.
23
Boreal rivers react differently to regulation than temperate rivers because of the importance of ice disturbance during winter (river freeze-up) and spring (river ice break-up), which dominates annual flooding (Nilsson et al., 1993).
Streamflow regulation, intended for hydropower generation purposes, causes river flow alteration from the natural flow state by changing the monthly, seasonal, and sub-daily flow patterns of regulated rivers, and adversely affects the ecological state of rivers (Dynesius & Nilsson, 1994; Mustonen et al., 2016; Poff et al., 2007). Negative impacts of sub-daily flow alteration may include direct impacts on aquatic organisms (Casas-Mulet et al., 2016; Haas et al., 2016; Harnish et al., 2014) a decline in fish habitat quality (García et al., 2011), thermal regime change in regulated river water (Krause, 2011), restricted recreational use of the river corridor, and alteration of the hyporheic habitat (Bruno et al., 2010) river thermal regime (Choi & Choi, 2018; Maheu et al., 2016) and recreational value (Brown et al., 1991; Brunson & Shelby, 1993; Rossel & de la Fuente, 2015; Teigland, 1999).
There is a knowledge gap regarding the impacts of major changes now affecting the energy industry on hydropower reservoir operation practices and causing environmental concerns. Sustainable river management requires the development of methodologies to quantify hydropeaking and its relationship with the power market. This thesis sought to address this knowledge gap by: 1) distinguishing between river regime alteration due to climate change and due to river regulation; 2) documenting current levels of hydropeaking in the Nordic countries; 3) developing new indices to measure hydropeaking at multiple temporal scales; and 4) modelling interactions between the power market and sub-daily river flows.
24
Fig. 1. Map of the Nordic region, showing location of unregulated and regulated river sites at which hydropeaking was measured (under CC-BY license from Paper II © 2018 Authors).
25
1.3 Motivation for this study
Long-term river regime change under river regulation is a well document topic (Paper I; Rheinheimer & Viers, 2015). But hydropeaking studies requires underlying data to have short time steps, which is not always readily available, hence fewer studies have quantified river regime alteration at sub-daily time period or examined seasonal variations in hydropeaking (Bejarano et al., 2017; Bevelhimer et al., 2015; Sauterleute et al., 2015). Some studies have also measured the impacts of climate and operational changes on hydropeaking in regulated rivers (Bevelhimer et al., 2015; Carolli et al., 2015; Marcarelli et al., 2010). However, hydropeaking in a Nordic context still needs to be investigated, as studies done in the region are limited to small study areas or have examined the effect of catchment properties, while the relationship with hydropower production capacity and other catchment properties has not been considered (Arheimer et al., 2017). Hence, the studies comprising this thesis investigates the issue of hydropeaking in Nordic region. Following is the further breakdown of reasons justifying the need for these studies.
1.3.1 Needs arising due to climate change (Paper I)
Ecosystem services provided by rivers and lakes are affected by alteration in natural river flow regime. Typically, long term flow variability plays an important role in maintaining natural hydraulic complexity, sediment transport, hyporheic exchanges, floodplain connections, and habitat structure etc. There is a reported increase in mean annual temperature (Omstedt et al., 2004; Tietäväinen et al., 2010; Tuomenvirta & Heino, 1996) and total precipitation (Irannezhad, Marttila, & Klöve, 2014) in the region, which makes estimation or separation of the effects of human actions and climate change on the flow regimes important. In Paper I in this thesis, a method to separate the extent of flow alteration due to damming from that due to climate-induced changes is presented.
1.3.2 Needs arising due to energy production and demand change (Paper II)
The share of low-carbon energy can increase by over three-fold in the stringent RCP 2.6 mitigation scenarios of the Intergovernmental Panel on Climate Change (IPCC) (Bruckner et al., 2014). Moreover, the European Union (EU) is aiming to
26
increase its use of renewable energy to 20% of total energy consumption by 2020 (European Union, 2009). This is causing a change in Nordic countries power market (introduction of renewable energy sources such as wind, solar, etc.), as a result the sub-daily flow patterns also being altered. Abrupt variations in power demand and production will most likely be balanced by hydropower, causing hydropeaking (Hauer et al., 2014; Schmutz et al., 2015). In combination with the effects of climate change on river flows (Berg et al., 2013; Huntington et al., 2009), unnatural sub-daily flow variations (hydropeaking) calls for a more sustainable reservoir regulation and water release patterns. Thus, water resources management for more efficient hydropower production (and as a back-up source for wind and solar energy) has become an important issue in the Nordic countries (Sokka, 2017).
1.3.3 Need for new indices (Paper III)
The indices used to date to quantify stream flow variations cover a wide range of hydrologic regime parameters, but they do not capture the full spectrum of temporal variation in regulated stream/river flow. Use of wavelet analysis was tested in Paper III to improve flow regime assessment by enabling simultaneous detection and quantification of flow variability, especially at sub-daily temporal scales. The key feature of wavelet analysis (compared with other spectral methods, e.g., power spectrum) is that it allows localization in time, i.e., that non-stationary processes can be analyzed. This is useful when analyzing market-driven hydropower production. Several studies have used wavelet transform for assessing flow regime alterations in different river systems (Shiau & Wu, 2013; Wu et al., 2015; Zolezzi et al., 2009). However, prior to this thesis work, a common framework that categorizes flow variations in general, and sub-daily flow variations due to hydropeaking, was still lacking.
1.3.4 Need to connect river flow alteration to the power markets (Paper IV)
The short-term regulation causes a drastic increase in downstream river flow fluctuations and ecosystem changes (Anderson et al., 2015). To fully assess the impacts of the power market (demand) on regulation practices (supply) the sub-daily flow variations needs to be studied. Rising global hydropower production is making it important to assess the effects of hydropeaking on the natural flow regime at finer time scales, in order to help devise policies for more sustainable
27
hydropower production. Understanding the interaction between power markets and river flow can serve up as a bottom-up approach to help maintain environmentally sustainable flow regime. Thus, sustainable river management calls for methodologies to quantify hydropeaking and its relationship with power market demands. Paper IV sought to address a knowledge gap by linking the impacts of major changes underway in the energy industry with hydropower reservoir operation practices.
28
29
2 Research objectives and thesis structure The work presented in Papers I-IV in this thesis addressed the following research questions:
– How has river regulation changed flow regimes in Nordic rivers? – What are the characteristics of sub-daily flow conditions (hydropeaking) due
to river regulation operations in Nordic countries? – How can hydropeaking on multiple sub-daily temporal scales be quantified? – How can the relationship between price fluctuations and hydropeaking be
modelled?
Specific objectives in Papers I-IV were to quantify sub-daily fluctuations in the major regulated rivers in the Nordic countries, develop methodology to quantify flow regime alterations related to hydropeaking, apply wavelet analysis to assess sub-daily flow alterations, and model the interactions between power markets and hydropeaking. In order to meet these objectives, the research was broken down into three main areas (Figure 2):
Determine the current situation of hydropeaking in Nordic rivers, analyze drivers behind observed changes in river regimes and identify the overall driving forces for hydropeaking.
Observing hydropeaking in rivers at sub-daily resolution can help reveal flow characteristics that might otherwise be overlooked when analyzing discharge data with daily or monthly resolution. Hydropeaking is widespread in the Nordic countries and several studies have sought to quantify it, but these studies have been limited to local scale or small catchments. Therefore, in this thesis the levels of hydropeaking were quantified in a large study region, using high-resolution data (hourly) covering multiple years and large spatial areas for Nordic rivers (Finland, Sweden, Norway). High-frequency variations at a given time were analyzed and seasonal changes in hydropeaking were investigated. Simple statistical analysis was used to quantify and compare hydropeaking measured with respect to hydropower plant type and catchment properties across multiple rivers. The main aim was to investigate the scale of hydropeaking and trends in many rivers reaches.
Develop an easy-to-use procedure that can quantify hydrological alterations at multiple sub-daily time scales caused by hydropeaking, using commonly available flow data, and compare alterations in regulated rivers and free-flowing rivers.
30
To address this issue, an approach was developed to characterize sub-daily flow variability, specifically identifying the time and quantifying the magnitude of high-frequency flow periodicities induced by hydropeaking. Wavelet analysis was used to quantify sub-daily flow variability, which was then used to quantify alterations compared with reference conditions for natural riverine ecosystem functions, based on free-flowing rivers mapped by Grill et al. (2019). Hence the framework utilizes unregulated discharge data for free-flowing rivers to categorize sub-daily variations in regulated rivers.
Devise a framework and scenario-based approach for modeling the interaction between sub-daily river flows and the power market.
To connect power markets with river flow regime, and thus bridge the gap between hydrology and the power market, a novel framework and methodology for estimating interactions between power market demand and actual regulation practices were developed. Two new indices (power market impact (PMI) and system efficiency ratio (SER)) were developed for sustainable hydropower reservoir management under a changing energy supply and demand situation. Methodology was devised to quantify the impact of capacity constraints (including reservoir volume and hydropower capacity) on sub-daily flow regime alteration of rivers, to comply with the instant energy demand change. The methodology was tested on the Kemijoki, one of the most regulated rivers in Finland.
The remainder of this thesis is organized as follows. Chapter 3 describes in detail the areas in which the research was concentrated. Chapter 4 begins by describing methodology that was used to enable separation of climate change and river regulation impacts on flow regimes (sections 4.1). It then describes the methodology behind analysis of hydropeaking, which was applied to rivers in Finland, Sweden and Norway (section 4.2). Next, it presents a simple framework for characterizing hydropeaking at multiple sub-daily temporal scales, with reference flow conditions set by choosing a comparable free-flowing river (sections 4.3 and 4.4). The final section of Chapter 4 describes a power market-based operation model for sub-daily hydropower regulation practices (section 4.5 - 4.8). Chapter 5 presents main results and provides a discussion of these results, which are described and tackled for the main research areas, broken into sub-areas (sections 5.1 to 5.10). Conclusions and suggestions for future research are given in Chapter 6.
31
Fig. 2. Schematic diagram summarizing the work performed in this thesis.
Indi
ces u
sed
Changing river regimes
Types of alteration
High intraday flow variation, high hydropeaking levels in summer, increasing hydropeaking trend in the last decade, ROR dams showing higher levels of hydropeaking than HHH dams
Indices like IHA and monthly regime index used
Using wavelet analysis to unmask the high power periodicities at short temporal scales
Short-term
Increased winter and autumn flow, decreased summer flow, increase in minimum flow condition and decrease in maximum flow condition, 50% alteration attributed to climate
Long term
Wavelet analysis based index proposed
Need for new indices to characterize subdaily flow variations at multiple sub-daily scales
Price variation in power market, snowmelt, highclimate variabilityM
ajor
caus
es
Climate change + Dam operation
No need for new indices
Modelling the interaction between power markets and short-term river flow regimes to fill in the existing knowledge gap
32
33
3 Brief overview of the study area, data sources, and the Nordic hydropower system
This work began by investigating river regime alterations generally in Finnish, and Swedish catchments (Paper I). First, river regime alterations were investigated for the rivers Kemijoki and Tornionjoki in Finland, two major rivers in the Northern Baltic Sea region (Figure 3). These two rivers provided a unique opportunity to compare river regime alteration in a totally pristine river (Tornionjoki) with that in a highly regulated river (Kemijoki). These rivers flow close to each other and are in same climate zone (Dfc according to the Köppen climate classification).
The Kemijoki (length 550 km, catchment size 50,683 km2, mean annual discharge 515 m3 s-1) is a highly regulated rivers in Finland. Damming started on the Kemijoki in 1949 for hydropower purposes, and since then 18 new hydropower plants have been constructed on it. Porttipahta (1353 km3) and Lokka (1460 km3) are two major reservoirs that were constructed at the headwaters of the Kemijoki in 1967 and 1981to regulate water downstream for hydropower plants. Water is transferred between these reservoirs by the Vuotso Canal, which links them together. Corine Land Cover data (CLS) was used to estimate land use and area of drained peatland in the catchment from data produced by Finnish Environment Institute (SYKE) (see Figure 1 and Table 1 in Paper I).
Small-scale regulation has been carried out in a few lakes on small tributaries of the Tornionjoki, with Lakes Raanujärvi, Vietonen, and Portimojärvi on the Finnish side and Lake Puostijärvi on Swedish side regulated for hydroelectric power generation. These regulated lakes only contribute 9% of Tornionjoki’s annual discharge, with little to no impact on Tornionjoki’s flow regime. In contrast to the Kemijoki, the Tornionjoki (length 522 km, catchment size 40,157 km2, mean annual discharge 370 m3 s-1) has very low levels of regulation (Paper I). Tornionjoki provides one of the last natural habitats for Baltic salmonids. Both rivers have similar seasonal flow regime typical of a snow fed river in the region (Romakkaniemi et al., 2000). Further details on the catchments can be found in Appendices of this thesis (Tables A1 and A2).
34
Fig. 3. Location of the study rivers Kemijoki and Tornionjoki, gauging stations and hydropower plants (reprinted with permission from Paper I © 2016 Elsevier B.V.).
In Paper II, discharge data from 157 (80 regulated and 77 unregulated) gauging stations or hydropower dams on major pristine and regulated rivers spread across Norway, Sweden, and Finland (see Figure 1) were collected and analyzed. Discharge data at hourly resolution (m3 s−1) were acquired from the Finnish Environmental Institute (SYKE), Swedish Meteorological and Hydrological Institute (SMHI), and Norwegian Water Resources and Energy Directorate (NVE).
35
Time span of flow rate varied from one year to 10 years for regulated stations and to 20 years for unregulated gauging stations (up to the end of 2017). Hourly discharge data from the SYKE database were unchecked, and hence needed correction. This was done using the following equations
Qdiff = Qdaily unchecked − Qdaily checked (1)
and
Qhourly corrected = Qhourly unchecked − Qdiff, (2)
where Qdaily unchecked is daily aggregated mean discharge calculated from unchecked hourly discharge data, Qdaily checked is the checked daily discharge value available from the SYKE database, Qdiff is the difference between checked and unchecked daily discharge data, and Qhourly corrected is the corrected hourly discharge value.
In the analysis in Paper III, four different types of regulated river systems in Finland were selected, with four free-flowing rivers in the region used as their unregulated counterparts (Figure 4). The selected river systems are all located inside the boreal geographical and climatic conditions (Dfc according to Köppen climate classification). Hourly discharge data from eight gauging stations, one from each river system, were analyzed. Taivalkoski, Skatila, Lieksanjoki, and Billnäs were selected as regulated river stations, as these represent river systems of varying sizes ranging from mean annual flow of 515 m3 s−1 (Taivalkoski) to 38 m3 s−1 (Skatila) (Table 1 in Paper III). The selected stations also represented high to low levels of regulation, e.g., Taivalkoski is on the Kemijoki, the largest river in Finland and one of the most regulated; Skatila represents a medium-size river system in Finland, Lieksanjoki represents a river system between two large lakes providing stable inflow to the river system; and Billnäs represents a small regulated river system. Marraskoski, Haukipudas, Nokisenkoski, and Kaukolankoski were selected as stations on unregulated rivers providing reference free-flow conditions. The Taivalkoski station on the Kemijoki river was also used in Paper IV.
36
Fig. 4. Location in northern Finland of four pairs of gauging stations (1-4) on (a) regulated rivers (red) and (b) their unregulated counterparts (green): Marraskoski (1a), Taivalkoski (1b), Haukipudas (2a), Skatila (2b), Kaukolankoski (3a), Billnäs (3b), Nokisenkoski (4a), and Lieksanjoki (4b) (reprinted with permission from Paper III © 2020 Authors).
37
4 Methodology
4.1 Indicators of hydrologic alteration and monthly flow regime index
To measure river regime alteration, two set of indices with daily and monthly temporal scales were used. The indicators of hydrologic alteration (IHA) approach developed by Richter et al. (1996) for the US Nature Conservancy was used for daily scale, and the monthly flow regime alteration developed by Torabi Haghighi et al. (2014), was used for monthly scale. The IHA method measures hydrological alteration of 33 hydrological parameters (Nature Conservancy, 2009). The IHA program analyzes the extent of hydrologic alteration in terms of i) frequency, ii) rate of change, iii) timing, iv) magnitude, and v) duration. Hydrologic alteration is quantified using range of variability approach (RVA) described by Richter et al. (1997). Degree of hydrologic alteration (DHA), which was used as a measure of hydrologic alteration in this thesis, is calculated as
DHA = – × 100, (3)
where Expected frequency is the frequency of occurrence of a flow parameter in the pre-impact period, and Observed frequency is the actual observed frequency of the same parameter in the post-impact period. Hence DHA (unitless) is the difference in values of a flow parameter between two periods.
To quantify seasonal flow alterations, the river impact (RI) approach developed by Torabi Haghighi and Kløve (2013) was used. This method evaluates effects of dam construction for three main characteristics of monthly hydrographs: i) magnitude, ii) variability, and iii) timing of flow. The alteration is measured alteration as river impact (RI)
RI = MIF × (TIF + VIF), (4)
where MIF is flow magnitude impact factor, TIF is timing impact factor, and VIF is variability impact factor. RI values between 0 and 1 are divided into five categories, where values between 0–0.2, 0.2–0.4, 0.4–0.6, 0.6–0.8, and 0.8–1
38
indicate drastic, severe, moderate, incipient, and low impact, respectively (Haghighi et al., 2014).
Both the IHA and RI approaches are purely statistical, and hence relevant for use in any climate. However, these indices cannot be used for direct measurement of the ecological status of rivers. Rather, they quantify changes over time in the flow dynamics of a river, which can have ecological consequences.
4.2 Hydropeaking indicators, thresholds, and hydropeaking pressure classes
4.2.1 Hydropeaking indicators
Hydropeaking was measured in terms of highest intraday flow difference (HP1) and ramping rates (HP2) (Carolli et al., 2015). These parameters were selected as they are simple and have been demonstrated to quantify hydropeaking in snow-fed rivers. For the large-scale data analysis across Fenno-Scandinavia in the present thesis, it was not feasible to use several indices and thus a decision was made to focus on one well-defined index to illustrate regional differences. HP1 is dimensionless measure of hydropeaking and is calculated as the median of annual daily HP1 values (HP1i) as follows
HP1= Median (HP1i), (5)
where
HP1 = , ,, . (6)
In Eq. (6) subscript i is day of the year (1–365), Qmax,i and Qmin,i are maximum and minimum discharge, respectively, and Qmean,i is mean daily discharge. HP1monthly
(which will be used further in the study) is aggregated monthly mean of value of HP1i.
Ramping rates were measured as HP2, it is computed as the 90th percentile (P90) of annual median of daily values of HP2 (HP2i), (Carolli et al., 2015) and mathematically expressed as
HP2 = Median (HP2i), (7)
where
39
HP2i = |P (HP2 ) | (8)
and
(HP2 ) = ∆∆ = . (9)
Subscript k refers to each available discharge datum (e.g., [1≤ k ≤ 24] for data sampled hourly), subscript i is day of the year (1–365), and |…| denotes absolute value in Eqs. (7) – (9). HP2 is expressed in m3 s-1 h-1 and HP2monthly is aggregated monthly mean of value of HP2i.
4.2.2 Thresholds and hydropower pressure classes
Hydropeaking threshold values (TR) for both indicators was set i.e., TRHP1 and TRHP2 (Carolli et al., 2015). These were estimated based on discharge data of 70 unregulated stations from Finland, Sweden and Norway. Thresholds were defined as
TR = 𝑃 HP1 + 1.5(P − P ) HP1 (10)
and
TR = P HP2 + 1.5(𝑃 − 𝑃 ) HP2 , (11)
where HP1 and HP2 are the daily values of the two indicators for unpeaked stream gauges, and P75 and P25 are the 75th and 25th percentile of the distribution, respectively.
Threshold values of HP1 and HP2 were classified according to the following four levels (Carolli et al., 2015):
1. Class 1: Low level of hydropeaking. HP1 < TRHP1 and HP2 < TRHP2. 2. Class 2a: Moderate level of hydropeaking. HP1 > TRHP1 and HP2 < TRHP2. 3. Class 2b: Moderate level of hydropeaking. HP2 >TRHP2 and HP1 < TRHP1. 4. Class 3: High level of hydropeaking. HP1 > TRHP1 and HP2>TRHP2.
For identifying and estimating trends in hydropeaking, the Mann-Kendall trend test was used along with Sen’s slope estimator. To assess the significance of shifts in the median of HP1 and HP2 pre- and post- 2013 (which was the year when wind power production increased significantly in Finland) the Mann-Whitney U test was used.
40
4.3 Hydrologic time series analysis by continuous wavelet transform
Continuous wavelet transform (CWT) is an effective tool to analyze non-stationary time series. Unlike Fourier transform (FT), which assumes stationarity when decomposing a signal into its inherent frequencies, CWT unfolds a time series not only in frequency, but also in time (Percival & Walden, 2000). CWT provides information on frequencies present in a signal, while still maintaining the localization in time, and thus offers a means to display intermittent temporal patterns in the time series. Wavelet analysis has been widely applied across many areas of research since its introduction in the early 1980s (see e.g., Percival and Walden, 2000). More recently, it has also been used to analyze variability in hydrologic time series data at multiple temporal scales (Shiau & Wu, 2013; Steel & Lange, 2007; Wu et al., 2015; Zolezzi et al., 2009).
As river flow data are typically noisy, irregular, and non-stationary, wavelet analysis can be used for characterizing periodicities at varying scales. The wavelet function is stretched by varying its scale to reveal periodic components of the time series change over time (Daubechies, 1992). A brief outline of the CWT procedure is provided below (for a more in-depth description of the approach, see (Torrence & Compo, 1998). The wavelet function is mathematically expressed as
𝛹 (𝜂) = 𝜋 / 𝑒 𝑒 / , (12)
where 𝛹 (𝜂) is a wavelet function and ω0 is the non-dimensional frequency, taken to be 6 (Farge, 1992).
The continuous wavelet transform 𝑊 (𝑠) of a discrete time series 𝑥 for (𝑛 = 0 ... N – 1), at a temporal scale 𝑠 with equal time spacing 𝛿𝑡, is then the convolution of 𝑥 with a scaled and translated version of 𝛹(𝜂):
𝑊 (𝑠) = 𝑥 𝛹∗ , (13)
where N is length of time series, 𝛹∗ is a complex conjugate of Ψ, and δt is data sampling interval.
To approximate the continuous wavelet transform, convolutions are performed in the Fourier space using discrete Fourier transform (DFT), more details of which
41
can be found in (Torrence & Compo, 1998). The wavelet function Ψ is obtained by normalizing 𝛹 (𝜂) , where 𝜂 = is non-dimensional time parameter. The Morlet wavelet was used as the wavelet base function in this thesis, since it is the preferred choice for streamflow data (Shiau & Wu, 2013; Steel & Lange, 2007). Finally, wavelet power can be measured in terms of the amplitude of CWT, called wavelet power spectrum (WPS) and defined as
WPS (𝑠) = |𝑊 (𝑠)| , (14)
where WPS (𝑠) is the wavelet power spectrum at scale 𝑠 and |𝑊 (𝑠)| is the amplitude of the continuous wavelet transform 𝑊 (𝑠).
4.4 Formulation of an index to categorize sub-daily flow variation levels
Wavelet power in a given period averaged across time was used here to categorize the spectral differences at sub-daily scales between regulated and free-flowing rivers. The sub-daily flow variations were characterized based on a temporal averaged WPS (𝑠) of flow, referred to as the global wavelet power spectrum (GWPS), defined as:
GWPS(𝑠) = ∑ WPS (𝑠). (15)
The averaged WPS for a given period can be used to quantify variations in a hydrograph (Figure 5). In this thesis, GWPS of sub-daily time scales, calculated from observed discharge data of free-flowing rivers, was used to estimate the deviation in flow variability between regulated rivers and unregulated rivers.
42
Fig. 5. Example (a) Hydrograph from regulated river system and (b) its wavelet transform and (c) corresponding GWPS values at multiple temporal scales (reprinted with permission from Paper III © 2020 Authors).
Free-flowing rivers for the study were selected based on an integrated connectivity status index (CSI) developed by Grill et al. (2019) that quantifies connectivity ranging from 0% to 100%. Grill et al. (2019) calculated CSI by applying a set of weights within a multi-criteria model. Table 2 in Paper III shows the free flow status of study sites and their CSI values. These CSI values were calculated by first choosing five major pressure factors that impact river connectivity according to an extensive literature review: (1) river fragmentation, (2) flow regulation, (3) sediment trapping, (4) water consumption (surface or groundwater abstractions), and (5) infrastructure development in riparian and floodplain areas. These five pressure factors were quantified using the following six proxy indicators: (1) degree of fragmentation (DOF), (2) degree of regulation (DOR), (3) sediment trapping index (SED), (4) consumptive water use (USE), (5) road density (RDD), and (6) urban areas (URB). Grill et al. (2019) calculated CSI for each river reach
43
by producing a weighted average of the six individual pressure indicators, each defined within a range of 0%–100%, subtracted from the maximum of 100%:
CSI = 100− , , (16)
where CSI is the CSI for river reach 𝑥 ,, is the value of pressure indicator i at reach j , 𝑤 is the weight applied to the pressure indicator 𝑖 , and 𝑛 is the number of pressure indicators (six in the present case). Grill et al. (2019) prescribe the sum of 𝑤 to be 100%, and hence the resulting CSI values can range from 0% (not connected) to 100% (fully connected). Rivers with CSI at or above 95% over their entire length from source to river outlet were defined here as free-flowing rivers (FFRs). See Grill et al. (2019) for more detailed explanations of CSI, FFRs, and the size of globally mapped FFRs.
In the first part of the analysis in this thesis (Paper III), sub-daily variations were measured and categorized on an annual level. This type of analysis should be done in cases where an ideal free-flowing counterpart to a regulated river is not available. Here, GWPS was caculated for all selected FFRs, and the highest value among those was chosen as a threshold to represent natural variability. An index was then formed, which measures the sub-daily variation levels in multiples of the levels found in FFRs. Hence, the ‘Natural variation levels’ section of the index comprised all GWPS values that were equal to or less than the highest value calculated from all the unregulated gauging stations in Paper III (Haukipudas, Kaukolankoski, Nokisenkoski, and Marraskoski). The highest annual average GWPS value (AGWPS) was 0.04 (at Kaukolankoski) and lowest was 0.0005 (at Haukipudas) (see Figure 4 in Paper III), and hence all values ≤0.04 were categorized as natural flow variation levels. In the second step in Paper III, a more detailed seasonal analysis was performed. This type of analysis can be done when an ideal free-flowing counterpart is available. Here, GWPS in each season for each of the FFRs was compared with seasonal GWPS from its regulated counterpart. December, January, February, and March were taken to represent winter, April and May represented spring, June, July, and August represented summer, and October and November represented autumn.
Note that the values represented free-flowing conditions for four Finnish rivers that were chosen to represent different types of rivers, so other regions may need value adjustment according to local free-flowing conditions. Free-flowing
44
counterpart rivers should only be chosen from catchments with comparable climate and geomorphology to the regulated river catchments. In ideal cases where a single regulated river system under analysis has a free-flowing counterpart flowing in the same region, with similar climate and geomorphology, GWPS from that FFR could be used to set that natural flow variability threshold. In cases where finding a perfect free-flowing counter part of a regulated river is not possible, multiple FFRs in the region can be analyzed to set the natural variability threshold.
4.5 Setting flow variability boundaries
Sub-daily flow fluctuations river affected by hydropeaking, can be assumed to be higher than a free-flowing river. Based on this assumption in a novel approach developed in paper IV of this thesis, two thresholds to encompass hydropower release induced flow fluctuations were set. Free flowing natural flow regime was set as the lower threshold release (LTR). For our study case Ounasjoki river, a pristine tributary of the Kemijoki river, was chosen. The upper threshold is set based on turbine released flow and hourly energy prices, which include a large storage reservoir and hydropower capacity allowing very flexible operation possibilities. Upper threshold (UTR) was set as an hourly allocation plan where annual flow release is proportional to its hourly price for each single hour of the year. This is described by the equation
𝐴𝑄 = 𝐶𝑇 × AF, (17)
where AQmdt is hourly allocated flow (m3) for hydropower generation, AF is volume of annual flow (m3), and CTmdt (combination of three other coefficients) is hourly contribution coefficient for a specific month (m), day (d), and hour (t) according to
𝐶𝑇 = 𝑀𝐶𝐶 × 𝐷𝐶𝐶 × 𝑇𝐶𝐶 . (18)
MCCm in Eq. (18) is monthly (January-December = 1-12) contribution coefficient calculated as
𝑀𝐶𝐶 = ∑ ∑∑ ∑ ∑ (19)
45
where
∑ 𝑀𝐶𝐶 = 1,
Pmdt is the price of one MGWh of energy during month m, day d, and time t, and ndym is the number of days in month m.
DCCmd in Eq. (18) is daily contribution coefficient, calculated (for each day) as
𝐷𝐶𝐶 = ∑∑ ∑ , (20)
where
𝐷𝐶𝐶 = 1. 𝑇𝐶𝐶 in Eq. (20) is the hourly contribution factor, calculated as:
𝑇𝐶𝐶 = ∑ , (21)
where
∑ 𝑇𝐶𝐶 = 1.
If reservoir water allocated for hydropower generation is strictly based on hourly energy prices, then outflow according to Eq. (17) will theoretically have highest sub-daily flow fluctuation, hereafter referred to as upper threshold release (UTR). Both LTR and UTR are expressed in hydrological terms; their range as discharge (m3 s-1) is shown in Figure 4 in Paper IV.
4.6 Designing different regulation practices
Achieving UTR in real life situation will almost never happen, and the hourly flow release fluctuations will fall between the upper and lower thresholds. Hence, we
46
simulated 96 different flow releases patterns based on varying constraint values of reservoir size (8 options) and hydropower generation capacity (12 options). These simulations resulted in intermediate flow regime patterns with sub-daily fluctuation levels (due to hydropeaking) between natural flow regime and UTR. For these simulations, a constant 15 m dam height was assumed with spillway placed at an elevation of 10.0 m (normal water level) and the minimum water level for operating hydropower was assumed to be 5.0 m. Dam height assumption was made based on the similar river morphology which flow through landscapes characterized by a flat topography and are not suited for construction of high head dams. It is important to note that constant head was not considered in this thesis, but rather constant variation in head (5–15 m) for different reservoir volumes (from 1% to 320% of annual flow), equipped with 12 varying capacities for hydropower production (see Paper IV). These assumptions can be changed to be used on other sites according to the river morphology and landscape characteristics area. To make the comparison of river flow variation of varying river sizes comparable, the hourly time series of flow data were normalized by dividing each hourly flow value by annual mean flow of the time series as follows
𝐸𝑄 = , (22)
where EQt is the normalized value of Qt (m3 s-1) at time t and μ is mean annual flow (m3 s-1). This scaled time series will still contain the flow variations of its parent time series data.
For a unit flow of 1 m3 s-1 and 10 m head over the hydropower, the mean energy production (potential for energy generation), henceforth denoted as hydropower scale (HPS), would be 353 MWh calculated as
HP = 3600 × 𝜂𝛾𝑄ℎ, (23)
where HP (W) is the amount of energy generated, Q (m3 s-1) is flow release from hydropower, h (m) is the water head over the hydropower, 𝜂 is efficiency of hydropower (here = 0.9), and 𝛾 is the unit weight of water.
Based on energy production from a unit discharge of 1 m3 s-1 (HPS = 353 MWh), 12 different hydropower options were considered, that covered electricity generation within the range 0–2648 MWh. This covered river regulation practices
47
representing no hydropower producing capacity, very small hydropower capacity (10% of HPS) and large hydropower capacity (750% of HPS). These hydropower producing capacities were combined with eight different reservoir storage volume values at normal water level (range 0.03 to 100 million m3). This reservoir size range covers reservoirs from volume <1% of annual flow (AF) to volume 310% of AF). 96 simulated flow release patterns were formulated by combining 12 options of hydropower production capacities with the eight reservoir volume options.
4.7 Release simulations for different scenarios
Following hourly water balance equation was used to calculate the outflow from dam (see Figure 1 in Paper IV)
𝑉 = 𝑉 + 𝑄 − 𝑄 −𝑄 , (24)
where Vt+1 and Vt (m3) are volume of the reservoir at time t+1 and t (hour), Qin (m3 s-1) is inflow to the reservoir (between t and t+1), Qhp (m3 s-1) is outflow from the hydropower unit (between t and t+1), which is released to produce electricity (controlled outflow), and Qsp (m3 s-1) is uncontrolled outflow from the spillway (between t and t+1), occuring when the water level is higher than the normal water level in the reservoir (Figure 1a in Paper IV). At each computational interval, the simulation loop strives for achieving UTR, but values of reservoir size, head of water, and the hydropower generation capacity constrain the UTR.
Here we assumed that volume and water level have a linear relationship. Water level according to our assumption varies between 5 m (ElvMin) and 15 m (ElvMax). Hourly flow release from the reservoir (Qhp and Qsp) is decided by the following conditions.
– When water level is between ElvMin and ElvNor, energy production between the time steps t and t+1 can be calculated as
𝐴𝐻𝑃 = 3600 × 𝐴𝑄 𝜂𝛾ℎ, (25)
where AHPmdt (W) is the amount of energy produced according to the allocated flow (𝐴𝑄 ) for the specific time (1 ≤ t ≤24) on a specific day (1 ≤ d ≤ 31 (or 30, 29,
48
28)) during a specific month (1≤ m≤ 12), 𝜂 is efficiency of hydropower, h is head over the hydropower, and 𝛾 is the unit weight of water.
– When water level falls below ElvMin, outflow is reduced by reducing Qhp. Reducing Qhp is placed in a loop by dropping ε (here equal to 0.01 m3s-1) in each stage and recalculating Eq. (24) until the water level surpasses or equates to ElvMin (Loop 1 in Figure 1 in Paper IV). In this condition, the system is unable of producing energy according to UTR, due to inadequate inflow or stored water in the reservoir.
– When water level gets higher than ElvNor, additional water will be released through the spillway. To calculate Qsp, the outflow equation for ogee spillways is used (USBR, 1987):
𝑄 = 𝑐𝑙ℎ ⁄ , (26)
where h (m) is the difference between water level in the reservoir and ElvNor, c is a spillway coefficient (with a suggested value of 2.1 according to USBR, 1987), and l (m) is the length of the spillway.
For water level higher than ElvNor, the algorithm of the model first checks the amount of energy produced (Eq. (25)), because as a default condition the outflow will be considered according to UTR. If the energy generated is lower than the hydropower capacity, the full hydropower production capacity will be utilized, and then extra water will outflow through the spillway (see Loop 2 in Figure 1a in Paper IV). Therefore, the system will generate extra energy than in the allocated plan due to limitation of storage capacity, exceeding UTR (New Qhp=𝐴𝑄 + QEhp) at time t. This is described by equations
HP = 3600 × 𝑄 𝜂𝛾ℎ (27)
and
𝑃𝑄 = × , (28)
where HPall is the amount of energy produced by the allocated flow (AQmdtp), HPcap is the hydropower capacity MWh, and PQEhp is the extra potential discharge for
49
hydropower at time t. If PQEhp ≥ Qsp, then QEhp = Qsp = 0. If PQEhp < Qsp, then QEhp = PQEhp and Qsp = Qsp − PQEhp.
4.8 Power market impact (PMI) and system efficiency ratio (SER)
Following power market impact (PMI) can estimate the impact of each flow allocation plan on flow regime, the impact value is quantified between 0 and 1:
PMI = ∑∑ , (29)
where 𝐼 is the current regulated discharge downstream of the hydropower plant, Ift is the river analyzed in each scenario, 𝑁 is scaled value of natural flow upstream in the river or lower threshold, 𝐴 is the allocated flow, and Sh and Eh are the start and end hours in the period for which PMI is being calculated. For example, to calculate annual PMI, Sh = 1 and Eh = 8760 (365x24), while to calculate PMI for January, Sh = 1 and Eh = 744 (31 x 24).
System efficiency ratio (SER) calculates the efficiency of different flow allocation plans to use water for energy production. It is a function of spillway outflow and total outflow over a certain period
SER=1- ∑∑ , (30)
where Vtt and Vspt are volume of total outflow and spillway outflow at time t, respectively, and Sh and Eh are start and end hours in the period for which SER is being calculated. SER varies between 0 and 1, where SER = 1 means all flow passes through hydropower and 0 means the reservoir does not have any hydropower.
50
51
5 Results and discussion
5.1 Effects of long-term river regulation
5.1.1 Impacts of river regulation on river regime in a sub-Arctic setting
Moderate to high monthly flow hydrologic alteration (HA) from December to April was observed across all three stations on Kemijoki river (Paper I). Post dam period saw an increase in the frequency of observed values in the high RVA category both the Ounaskoski and Taivalkoski stations. Only the last station (Karunki gauging station) at the mouth of catchment on unregulated Tornionjoki river recorded moderate to high levels of HA for monthly flow at the. At Karunki, February to April flows showed an increase and June a decrease in the frequency of values in the high RVA category flows (Paper I). River regulation has caused an increase in winter flows increased while summer and autumn flows have decreased (see Figure 4 in Paper I), which is in line with recent findings for Finnish rivers (Olsson et al., 2015) and for flooding in northern Swedish catchments (Matti et al., 2016). These results show the extent of flow homogenization on Kemijoki river. The decrease in seasonal flow in the second quarter was more prominent for the Kemijoki, because seasonal changes become coupled with reservoir operation (Figure 9 in Paper I). The small increase in discharge and low HA levels at Tornionjoki can be attributed to climate change and variability (McClelland et al., 2004), including increased precipitation (Irannezhad, Marttila, & Kløve, 2014), which has increased river discharge e.g., in the Eurasian region (Peterson et al., 2002).
5.1.2 Changes in extreme flow events
There was a statistically significant decrease in high flow events and an increase in low flow events on the Kemijoki river. On Tornionjoki there was increase in both low and high flow events (Paper I). These results show the result of loss in natural flow dynamics on Kemijoki flow. There also is an indication that climate change has decreased the 90-day maximum flow events in the unregulated Tornionjoki river and increase in low flow events possibly an effect of climate change was also evident for Tornionjoki. Apart from the annual minimum flow timing for the
52
Kemijoki, which moved from the 95th day in the Julian calendar year to the 215th day, annual maximum and minimum flows timing did not change in either river.
5.1.3 Flow alteration at different river points
Data analysis from stream gauges at different locations on the Kemijoki river showed that on moving from upstream (Kemijärvi) to downstream (Taivalkoski) stations, the measure of flow alteration on mean monthly flow was dampened to some degree. Comparison of flow alteration assessments between the stations on Kemijoki revealed that the alteration reduced from moderate (RI = 0.65) at Kemijärvi to incipient (RI = 0.82) at Taivalkoski (Paper I). This type of retrieval of flow dynamics is attributed to the addition of the unregulated catchment. The most regulated section of the Kemijoki river is located upstream of Kemijärvi, with Porttipahta and Lokka reservoirs, forms about 53% (27,285 km2) of the total catchment area (50,600 km2). There is no big reservoir in the remaining 47% of the catchment (e.g., the pristine tributary river Ounasjoki), this dampens the levels of alteration (Haghighi & Kløve, 2015). Still, the Kemijoki does not fully recuperate to its pre-impact state as the low flow months (November, December, January, February, March, and April) evidently show an increase in monthly mean flow, due to water release from reservoirs. In the Tornionjoki river, mean monthly flow alteration was low (RI = 0.96–0.99). Even though the monthly regime alteration due to climate change was not substantial, the regime has clearly altered more at the stations near the Baltic Sea coast. This implies a higher climate change impact in coastal areas, a key issue mentioned in other studies (Prowse et al., 2015).
5.2 Effect of short-term regulation on sub-daily flow dynamics of Nordic rivers
5.2.1 Measure of hydropeaking in the region
Finnish rivers on average had the highest levels hydropeaking impact, followed by Norwegian and Swedish rivers. In all, 64% of studied Finnish and 77% of Norwegian regulated rivers exhibited moderate to high levels of hydropeaking. Only one regulated river gauging station in Sweden was part of the study with hourl flow data, which showed low levels of hydropeaking (Paper II). Hydropeaking levels had high seasonal variation (Paper II). Hydropeaking indicators had higher
53
values during months with low power production and lower values in months with high power production.
As predictability and seasonality are both vital for riverine ecology (Mustonen et al., 2018; Vollset et al., 2016) and for recreational needs, we analyzed the seasonal variation in hydropeaking in details. Hydropeaking varied noticeably between seasons, with higher values in summer and autumn months, when power production and consumption are lowest in the Nordic region. During winter months hydropeaking levels were low, which could be due to consistently higher energy demand demanding consistently high energy production all day and thereby consistent dam out flow. Higher hydropeaking during low flow months (summer and autumn) might also be the result of higher demand for adjustment energy in markets or of smaller available water (low flow) in rivers causing more sensitive conditions for hydropeaking. Low HP2 and high HP1 during March and April could mean a continuous change in sub-daily flow discharge at a considerably small rate of change. Flow variations during this time of the year follow a more natural seasonal flow pattern, because spillways of dams are opened to release large amounts of snowmelt water which cannot be contained or channeled for electricity production.
5.2.2 Relationship between hydropeaking, catchment and hydropower plant properties
Annual ramping rates were higher for Finnish ROR-type plants than for HHH-type plants, without any significant seasonal changes. This could be because the small head ROR plants requires a higher discharge to produce the same amount of power as produced from a comparable discharge on HHH plant. Moreover, in most of the analyzed HHH only a part of the main channel flow is regulated, thus causing less hydropeaking. There seemed no effect of catchment size and power generated per meter (PPM) on HP1, but HP2 was higher for larger catchment areas and higher PPM (Paper II). This may indicate a stronger relationship between hydropower plant size and ramping rate than between plant size and net intra-day discharge difference. Hydropeaking levels were low with an increase in the catchment’s lake percentage.
Overall, Norwegian high-head hydropower (HHH) plants showed lower hydropeaking levels than Finnish small head and/or run-of-the-river (ROR) plants. HP1 had higher values for low flow years indicating that hydropeaking is more intense at regulated sites during dry years, when there may already be challenges
54
in sustaining ecologically critical environmental flows. Due to lack of data for highly regulated sites in Sweden, hydropeaking values for Swedish rivers were low. However, there is a high probability that other rivers in Sweden are more affected by hydropeaking, as shown in previous studies (Bejarano et al., 2018; Bejarano et al., 2017). Stored energy in larger reservoirs is used to cushion Nordic power market demand for flexible energy, which results in higher ramping rates. Therefore, large rivers are more useful for balancing the demand, causing hydropeaking and unnatural flow settings in the river systems. Analysis in Paper II also showed high levels of HP1 values in smaller river systems with catchment area of less than 500 m2.
5.2.3 Current hydropeaking trends in the region
Out of the 92 regulated river gauging stations that were analyzed for hydropeaking in terms of sub-daily variations in Paper II, only 54 had more than five years of hourly discharge data available. Trend analysis (Mann-Kendall trend test) of hydropeaking data at those stations showed a statistically significant trend for one or both hydropeaking parameters at 40 stations. Hydropeaking levels at 26 stations showed a statistically significant increasing and at 14 stations a decreasing trend. In the region, particularly in Finland, there has been a surge in wind power production since 2013. For all the regulated Finnish rivers analyzed, there was a statistically significant difference (Mann-Whitney U-test) in values of both the hydropeaking parameters between the periods 2014-2017 and 2011-2013 (HP1 p-value <0.001) and HP2 p-value <0.001). The sites Pamilo and Vatajankoski (Finland) and Sokna (Norway) showed the highest values for Sen’s slope (see Figure 10 in Paper II).
Thus, it can be said that hydropeaking is already at elevated levels in many Nordic rivers, with an increasing trend on many rivers (Paper II). Also, in Finnish rivers the average annual hydropeaking levels have increased significantly since 2013. Thus, hydropeaking has increased in recent years, possibly due to increased production of renewable energy and the need for a load balancing source. This shows a need for optimization of hydropower operations in the Nordic countries, as hydropower will continue to play an increasingly substantial part in future power markets. Different scenarios for carbon-neutral power systems in the Nordic countries show a greater role of Norwegian hydropower as a load balancing source (Graabak & Warland, 2014; Sauterleute et al., 2015). Future Norwegian energy export plans are mainly focused on peaking between reservoirs, which would
55
elevate the current levels of hydropeaking. Therefore, the effects of load balancing by means of hydropower to propel renewable energy penetration, needs to be studied further, to help policy makers set limits for load balancing by hydropower.
5.3 Evaluation of a new classification method to assess hydropeaking
5.3.1 Assessing and characterizing annual hydropeaking in terms of sub-daily variations in free-flowing rivers
Continuous wavelet transform and GWPS values for the selected regulated gauging stations displayed annual sub-daily variation levels that varied between 0.44 and 0.003, with high variation in power levels in different periods (Paper III). Taivalkoski is a highly regulated station, but it has a head of only 15 m and reservoir capacity of only 50 million m3, restricting the levels of attainable hydropeaking, but high levels of overall annual sub-daily flow variations were observed even there. Other stations studied have smaller reservoirs with low levels of sub-daily variations.
For the Taivalkoski station, two high power variation patterns were visible at a period of 0.5 and 1 (see Figure 5a in Paper III), which appeared as peaks in the GWPS graph. Similar peaks, but of smaller magnitude, were observed for Billnäs and Skatila (also regulated river stations). For Skatila, overall a much less variable sub-daily flow pattern was observed. The Lieksanjoki station showed a similar single high power variation pattern, but in a much earlier time period than other stations. Using the index developed to categorize the variation levels, at Taivalkoski sub-daily flow variation levels were found to be 20-fold times higher than those observed in unregulated rivers. Billnäs and Lieksanjoki had twice the natural sub-daily flow variation levels and Skatila had natural levels of sub-daily flow variation (see Figure 6 in Paper III).
5.3.2 Seasonal hydropeaking in terms of sub-daily variations
Seasonal hydropeaking at each of the regulated stations was measured against the variation levels of its reference free-flowing counterpart. Hence a hydropeaking index was provided for each river for each season. Among the chosen FFRs, the highest AGWPS in winter was 0.05 at Nokisenkoski station, the highest in spring
56
was 0.12 at Kaukolankoski station, and the highest in summer was 0.07 and the highest in autumn was 0.08, both at Nokisenkoski (see Table 3 in Paper III). Seasonal analysis showed that sub-daily variations at Taivalkoski were 20-fold greater than natural levels for the winter, summer, and autumn periods measured at its free-flowing counterpart gauging station at Marraskoski. Compared with the natural sub-daily variation levels of summer, Lieksanjoki exhibited around 20-fold higher levels, and Billnäs around 10-fold higher levels, than their respective free-flowing counterpart gauging station. Due to the snow-fed nature of the rivers studied and small reservoir size, no regulation during spring flood period was observed in discharge time series, and hence variation levels were close to natural levels during spring for all sites studied. Hydropeaking levels at the station on the most regulated and largest of the rivers studied (Kemijoki) were highest for winter, followed by autumn, summer, and spring. For other stations on smaller rivers with lower levels of regulation, summer season with the highest variable flow, followed by autumn, winter, and spring.
The methodology developed in this thesis can characterize sub-daily river regimes at hourly to daily scales, which is an important benefit for impact assessment of many aspects of ecology, water resource management, and energy-water-food nexus systems. Its main advantage is that it is a relatively simple method, requiring only sub-daily river discharge data, but is still able to quantify hydropeaking at sub-daily time scales. Although gauging stations were selected to cover a wide range of rivers in Finland and the proposed methodology gives a framework that is suitable for use in Finland, it can be applied in various river systems globally. By increasing the number of gauging stations on free-flowing natural rivers, a higher range of natural variation could be captured for a region, thus enabling more accurate categorization of hydropeaking. Another convenient characteristic of the proposed index is that it can also be calibrated to assess variation levels at multiple time scales, e.g., annual and seasonal, as demonstrated in Paper III, allowing analysis of critical time periods for ecosystems and services.
Three of the four regulated stations studied displayed variation levels exceeding eight-fold the highest variation levels found in their free-flowing counterparts for ecologically crucial time periods or seasons. This finding is important, because it has been shown that high sub-daily flow variability leads to sediment and macroinvertebrate scouring, unfavorable juvenile fish-rearing habitats, and fish stranding (Fette et al., 2007; Halleraker et al., 2003; Korman & Campana, 2009; Scruton et al., 2008). Assessing seasonal hydropeaking is especially important because, during ecologically critical time periods, e.g., many
57
salmon species reach river estuaries during early summer (May-June) and migrate for spawning during autumn (September-November), the results in this thesis clearly show elevated levels of sub-daily flow variability. At Taivalkoski and Skatilla stations, the levels were up to 20-fold greater than those in natural free-flowing rivers in winter months. Fish eggs that remain in gravel on the riverbed throughout the winter in regulated rivers are also vulnerable to sudden unnatural flow variations due to hydropower reservoir operation and hydropeaking.
Kaukolankoski, a station on an unregulated river, showed considerably higher sub-daily variation levels than stations on other unregulated rivers in the region. This could be because Kaukolankoski’s catchment comprises 41% agricultural area with fast runoff response rate, compared with <5% agricultural area in other unregulated catchments. To account for the fact that the sub-daily period value of AGWPS in unregulated catchments can also vary due to other differences in land use and land type, the highest sub-daily period value of AGWPS was chosen as a threshold below which all values were classified as natural levels of sub-daily variation. This was done to evaluate variation levels that were higher than natural levels and could thus be attributed to reservoir operation or hydropeaking.
Assessing hydropeaking is also important because it can affect the recreational value by worsening the water quality (Rossel & de la Fuente, 2015) and biophysical environment (Brown et al., 1991; Brunson & Shelby, 1993; Teigland, 1999). River recreation has economic, psychological, physiological, social, cultural, and ecological benefits. Many existing dams were built when environmental and recreational issues were not regarded as highly important, and when generation of electricity was seen as an overriding national interest (Robbins & Lewis, 2008). However, demand for recreational use of watercourses has been growing due to increased leisure time and urbanization, along with increased environmental awareness. This trend poses new challenges for management of regulated river systems. For example, there are currently policy initiatives such as the EU Water Framework Directive (EU, 2009) to restore the multi-functionality of riverine ecosystems.
58
5.4 Interaction between power market and short-term river flow regime
5.4.1 Effect of energy price variations on flow regime thresholds
In Paper IV of the thesis we estimated an upper threshold release (UTR) for hydropower reservoir release taking in to account the hourly energy price. In 2017, energy prices in the Nordic energy market were highest for September and lowest for June (see Table 1 in Paper IV). Based on the aggregated mean weekly price, Wednesday had the highest and Sunday the lowest energy prices (see Figure 2 in Paper IV). Contribution factor for different timesteps ((𝑀𝐶𝐶 ,𝐷𝐶𝐶 ,𝑇𝐶𝐶 ; see Eqs. (19)-(21)) were combined using Eq. (18) to calculate the hourly contribution coefficient for an exact hour (t), which was a function of hourly price of energy (Paper IV). Taking a unit mean annual discharge of 1 m3 s-1 and hourly coefficient, the UTR flow regime was defined. Lower threshold (LTR) was set based on the scaled (time series divided by mean flow) hourly flow regime in the unregulated Ounasjoki river at Marraskoski station.
5.4.2 Combined effect of fluctuating power prices and regulation practices on hydropeaking regime
Total flow release, including four types of outflow (Figure 6, 1-4) was determined in Paper II and is shown for five selected scenarios (Figure 6, a-e). The priority for outflow plan is releasing flow approaching to UTR. If the water level is higher than ElvNor and the hydropower capacity is not fully utilized additional water is released. And in case the full hydropower production capacity is reached, and water level is still higher than ElvNor water is released through the spillway. Summation of these three release patterns constitutes the total outflow from the reservoir, which is total regulated flow downstream. Flow release from hydropower reservoir is reliant on inflow, reservoir capacity, hydropower generation capacity, and energy demand (see Figure 5 in Paper IV). To comply with the UTR will require large reservoir capacity, This interaction is displayed for the scenario using reservoir size ≥320% of annual flow and hydropower capacity ≥750% of HPS, which causes a very regulated condition for the system and complies with the UTR (see Figure 6e1-4).
59
Fig. 6. Hourly flow release patterns as a result of varying hydropower capacity and reservoir size. (1) Hydropower flow release according to the upper threshold release (UTR) plan, (2) hydropower flow release outside the UTR plan, (3) uncontrolled flow from the spillway, and (4) total flow release into the river. (a) hydropower capacity at hydropower scale (HPS) and storage at 55% of annual flow (AF), (b) small reservoir (4% AF) and small hydropower capacity (20% HPS), (c) large reservoir (320% AF) and small hydropower capacity (20% HPS), (d) large hydropower capacity (750% HPS) and small reservoir (4% AF) and (e) large hydropower capacity (750% HPS) and large reservoir (320% AF) (reprinted with permission from Paper IV © 2019 Elsevier B.V.).
Full compliance with UTR is impossible, in absence of large enough reservoir capacity, as is displayed in Figure 6d where effect of grouping a small reservoir with large hydropower capacity on outflow is modelled. In this scenario higher power generation was possible only during the high flow period when there is an abundance of available water.
High hydropower capacity led to more energy production only during the high-flow season (outflow to the second priority of release increased), and the amount of extra flow release was not fully served for UTR. Thus, having abundant inflow
60
without enough hydropower generation capacity is inadequate to attain UTR, since in this scenario flood flow was released through the spillways. Similarly, we have further modelled in Paper IV the relationship between different quantities of reservoir capacity, inflow and hydropower producing capacity. According to the results, to fulfil the UTR release plan, a minimum reservoir capacity of 1.4 times the annual inflow (options 6, 7, and 8 in Figure 6 in Paper IV) and minimum hydropower generation capacity of 2.6 times HPS (j6–j8, k6–k8, and l6–l8 in Figure 6) would be required. The effect of small reservoirs on flow was inconsequential and flow release from the system was like the natural flow regime (Paper IV).
5.4.3 Power market impact (PMI) and system efficiency ratio (SER)
For situation with lower than HPS hydropower producing capacity, reservoir size was the main variable affecting annual PMI. The highest PMI was 0.65 for the scenario with reservoir volume ≥30% of annual flow (Paper IV). In scenarios with hydropower capacity exceeding HPS, PMI was found to be independent of hydropower generation capacity and, for a given reservoir volume, PMI was constant. Moreover, in all scenarios with reservoir capacity >30% of annual flow and HP/HPS>1, PMI was found to be 1.
Consequently, the maximum impact on hourly flow regime occurs due to any combination of: i) hydropower capacity >1 HPS and ii) reservoir capacity ≥30% of annual flow. Other combinations of hydropower and reservoir size in the absence of one of these two conditions result in an impact ranging from zero to 99%. Smaller reservoirs, even when associated with considerable hydropower plants, have lower PMI. The highest PMI obtained for the smallest reservoir size analyzed in Paper IV (storage: 0.01 of AF) was less than 10%.
Greater hydropower producing capacity meant less spillage, and thus an increase in the annual efficiency (SER) (see Figure 7b in Paper IV). For small reservoirs to have high SER, 100%, 30%, 20%, and 10% of AF storage capacity, equipped with 100%, 140%, 300%, and 750% of HPS, respectively, would be needed. Impact of power price fluctuation (i.e., the PMI) at our test site on Kemijoki river during 2013, 2014, and 2015 was assessed to be 0.74, 0.84, and 0.61, respectively (see Figures 8a2-c2 in Paper IV). Range of variation of PMI for different months was high ranging from 0.27 to 1.00. PMI also varied interannually, showing clearly the influence of natural flow regime on PMI fluctuations (Figure 8b in Paper IV).
61
5.4.4 Advantages of the new methodology
With greater variation in hourly energy demand and prices, the need to use flexible hydropower to balance energy production is increasing. These new regulation strategies exert pressures on riverine ecosystems, particularly impacting their ecological and hydrological functioning. Understanding drivers and causes of sub-daily alterations in hydrological conditions would improve management and aid impact assessment of flow regime changes. Proposed methodology offers following advantages in fulfilment of these goals.
1. It is a complete framework that not only can assist in formulating economically sound hydropower release options, but also quantify regime alteration due to these options.
2. It helps to differentiate between the efficiency of hydropower systems of different capacities to comply with high energy demand fluctuations and their impacts on natural flow regime alteration.
3. It is a flexible method and can use any regulation plan (in theory) as UTR to estimate the impact of any desired reservoir operation plan on any desired time scale. The framework can thus be adjusted to any situation and conditions and used to evaluate pressures from power market demand or to calculate the impact of hourly flow regulation practices.
4. This method can quantify seasonal power market impact on flow regime alteration. can help in. Thus, monthly PMI can be calculated to get an impact scale to measure intra-annual flow alteration, especially during the ecologically critical river flow months.
5. Proposed model can evaluate the existing hydropeaking status in regulated river systems and its relation to the power market. The method allows for direct comparison of power market demand and flow regime variations, and thus can be an efficient tool for sustainable river management balancing different demands and drivers. This is highly important, since in future more flexible energy sources such as hydropower will be needed to balance sudden production changes in distributed energy production (Graabak & Warland, 2014; Sauterleute et al., 2015). Simultaneously, societal demand for other ecosystem services provided by regulated rivers is increasing, creating a need for impact assessment methodologies. Against this background, the framework could be used to evaluate the consequences of intra-day power market
62
balancing for flow conditions in regulated river systems and for ecosystem services.
It should be noted that the main contribution of the method lies in modeling the relationship between energy price variation and sub-daily river flow variation based on historical energy and flow data, and not optimization of hydropower operation. The approach can be used in a coupled energy-economic and hydraulic model that forecasts energy prices and river runoff, which can then be used for optimization of hydropower operation.
5.4.5 Limitations of the framework
The framework is flexible and uses open access data from power markets, but it also has following limitations.
1. One is that estimation of LTR was done using the flow from a comparable unregulated river (Ounasjoki). The best option for setting LTR would have been to base it on measured hourly reservoir inflow. However, such time series data are not available for most rivers, because flow upstream is affected by other dams and by water consumption by other users, hence this adds uncertainty to the methodology.
2. Another limitation is lack of availability high resolution time series discharge data on regulated river systems, as such information is not often available from open access data sources or even at authority level. When applying the framework to other regions, local datasets on the power market and river system should always be used. In this thesis, Nordic power market data sources were used, but the energy supply and demand conditions for UTR can vary annually and regionally. The LTR (natural river flow) also varies for different river systems and may be affected especially by a changing climate (Palmer et al., 2008).
3. Full compliance with the UTR plan is highly improbable, due to economic and site topography limitations. Installing very large hydropower capacity (e.g., 750% of HPS) could provide flexible conditions for regulation according to UTR, but it would not work at full capacity throughout the year, so in practice would not be an economically feasible option.
63
6 Conclusions Comparing flow regimes in two large adjacent rivers in Finland revealed major differences in regime alteration pattern of regulated and unregulated rivers experiencing similar climate conditions. Total flow alteration (average DHA 25% for all 33 IHA parameters) in the pristine Tornionjoki river was almost wholly attributable to climate change and climate variability. Due to additional regulation impacts, DHA was 100% higher in the regulated Kemijoki river (average DHA 51% for all 33 IHA parameters). Hence, regulation and climate change had similar degrees of effect on flow alteration in the Kemijoki (50% each). Daily variation in natural stream flow regime was altered in both rivers, but DHA for daily flow values in winter months was significantly higher for the Kemijoki. Hence, river regulation modified daily low flow patterns more in winter than in other seasons, but similar modification was not observed in monthly mean values. Degree of alteration in both rivers was dependent on gauging station location in the river layout. The impact of flow regulation in downstream parts of the regulated river was moderated by an increase in unregulated catchment area. For the unregulated Tornionjoki, there were increases in mean monthly flow for August and December and in annual maximum flow, a decrease in annual minimum flows, and shifts in peak flow timing of around a week. These changes could be climate related. The changes were even greater in the regulated Kemijoki river, where they were amplified by regulation. Overall, IHA and monthly analysis methods allowed the impacts of climate change on overall flow regime to be distinguished from those of river regulation. Averaging the parameters at all stations masked important information but enabled a simple and quick approach to distinguish between degree of alteration due to climate change and river regulation.
Analysis of hydropeaking in Nordic rivers showed frequent incidence, with an increase especially in recent years, and an ongoing upward trend in hydropeaking levels on many regulated rivers. A significant increase in wind power may require an increase in hydropower as a balancing source, which could further increase hydropeaking. Run-of-river hydropower plants constructed on rivers with small elevation head were found to have higher hydropeaking than high-head hydropower plants. Sub-daily flow variation and ramping rate displayed a positive correlation with power per meter available head. Hydropower operations in the Nordic region thus need to be optimized using an integrated economic and ecological framework, in order to address the high levels of hydropeaking in regulated river systems. According to the results presented above it can be
64
concluded that hydropeaking: i) occurs at considerably high levels in the region, ii) varies with lake percentage (based on significant variation in HP1 with lake percentage), hydropower dam type, and catchment size, and iii) shows substantial seasonal variation. Many HHH-type plants in Nordic river systems still operate on base load, and hence seasonal analysis is important for their bypass reaches. However, Nordic river systems rarely suffer from severe droughts, and thus evaluations of environmental flow conditions, especially at ROR-type plants, should focus on hydropeaking, and not on analyzing daily or monthly flow conditions. While hydropower is highly important for power markets and in providing a flexible energy source, hydropeaking should be further examined and discussed in a Nordic perspective.
Methodology to quantify and characterize hydropeaking in terms of sub-daily variations using time averaging of GWPS was developed. The methodology does not consider the distribution of variability at specific periodicities and low power of non-peaked sub-daily flow is overridden by high power sub-daily hydropeaking events, thus masking important information. However, the methodology can be useful for classification of hydropeaking levels at river stations using variation levels in unregulated rivers as reference. Setting hydropeaking thresholds and measuring sub-daily variations can be useful in river restoration, hydrodynamic modeling, and fish habitat restoration projects. The methodology could also be useful in water management, to minimize the difference in flow variations between regulated and free-flowing rivers, which could indirectly help mitigate the ecological impacts of hydropeaking.
An approach developed for evaluating the coupled effects of the power market and sub-daily regulation practices on river regime offers a novel tool for sustainable river management. Future changes in energy markets can result from climate change, increasing share of renewable energy, and technological advances in battery powered appliances and cars, etc. This will change future hydropower planning and operation. Modeling the interactions between various energy generation scenarios and river flow helps provide an approximate estimate of future sub-daily river flow conditions. The approach uses a theoretical regulation plan together with a scenario-based approach incorporating key constraints to maximize the flexible nature of hydropower generation and to meet erratic energy supply-demand situations. Water release plans are designed for varying reservoir size, hydropower capacity, natural inflow, and energy prices, and coefficients based on hourly energy prices are used to allocate upper threshold release from the hydropower plant based on the power market. Two simple metrics, power market
65
impact (PMI) and system efficiency ratio (SER), were developed to quantify altered flow regime under proposed or actual regulation practices, which can help policymakers decide on optimum regulation practices. The major advantage of the approach is that each river is compared with its natural dimensioned flow as lower threshold release (no changes) and the regional power market demand is considered as higher threshold release (demand). This makes each river system unique, with a quantified impact varying between 0 and 1, which enables comparison of different river systems. The approach can be applied to estimate the consequences of future power market development and associated demands on regulated river systems.
66
67
7 Future studies Future studies should examine the effect of utilizing hydropower to balance integration of increasing share of variable renewable energy sources (VRES), the impacts of electricity demand flexibility on hydropeaking, and the effects of tighter environmental hydropower flow constraints on flexibility of electricity supply. There is an assumption that profit-maximizing hydropower operations are conditional on the share and production pattern of VRES, and on price elasticity of demand. The profit-maximizing hydropeaking level will be higher with higher share of VRES, but lower with more flexible demand. Tighter environmental flow constraints can be set for hydropower if consumption resources are active in the spot and balancing markets.
Future studies should also explore complex interactions between societal benefits and management options of hydropower-regulated rivers. An ecosystem services methodology, providing monetary estimates of final services, could be applied to evaluate changes in societal benefits due to alternative river management activities and energy scenarios. An integrated approach for assessing the benefits and costs of potential river management activities could be developed. More precisely, hydrodynamic modelling and assessment, hydropower and energy markets, and valuation of ecosystem services could be integrated in a transparent manner for comprehensive impact assessment and communication in decision-making processes. Additionally, future studies should investigate separation of sub-daily variation due to natural causes and hydropeaking, which was not addressed in the thesis.
68
69
List of references Aasgård, E. K., Fleten, S. E., Kaut, M., Midthun, K., & Perez-Valdes, G. A. (2019).
Hydropower bidding in a multi-market setting. Energy Systems, 10(3), 543-565. Ansar, A., Flyvbjerg, B., Budzier, A., & Lunn, D. (2014). Should we build more large dams?
The actual costs of hydropower megaproject development. Energy Policy, 69, 43–56. Arheimer, B., Donnelly, C., & Lindström, G. (2017). Regulation of snow-fed rivers affects
flow regimes more than climate change. Nature Communications, 8(1), 62. Arthington, Á. H., Naiman, R. J., Mcclain, M. E., & Nilsson, C. (2010). Preserving the
biodiversity and ecological services of rivers: new challenges and research opportunities. Freshwater Biology, 55(1), 1–16.
Baron, R. (2013). Renewable Energy: a Route to Decarbonisation in Peril? Document Préparé Pour La 29e Table Ronde Sur Le Développement Durable Organisée Au Siège de l’OCDE, à Paris, 4–5.
Bejarano, M.D, Jansson, R., & Nilsson, C. (2018). The effects of hydropeaking on riverine plants: a review. Biological Reviews, 93(1), 658–673.
Bejarano, M.D, Sordo-Ward, Á., Alonso, C., & Nilsson, C. (2017). Characterizing effects of hydropower plants on sub-daily flow regimes. Journal of Hydrology, 550, 186–200.
Beldring, S., Engen-skaugen, T., Førland, E. J., & Roald, L. A. (2008). Climate change impacts on hydrological processes in Norway based on two methods for transferring regional climate model results to meteorological station sites. Tellus, Series A: Dynamic Meteorology and Oceanography, 60 A(3), 439–450. https://doi.org/10.1111/j.1600-0870.2008.00306.x
Berg, P., Moseley, C., & Haerter, J. O. (2013). Strong increase in convective precipitation in response to higher temperatures. Nature Geoscience, 6(3), 181.
Bevelhimer, M. S., McManamay, R. A., & O'connor, B. (2015). Characterizing sub-daily flow regimes: implications of hydrologic resolution on ecohydrology studies. River Research and Applications, 31(7), 867–879. https://doi.org/10.1002/rra.2781
Boehlert, B., Strzepek, K. M., Gebretsadik, Y., Swanson, R., McCluskey, A., Neumann, J. E., McFarland, J., & Martinich, J. (2016). Climate change impacts and greenhouse gas mitigation effects on US hydropower generation. Applied Energy, 183, 1511–1519.
Brown, T. C., Taylor, J. G., & Shelby, B. (1991). Assessing the direct effects of streamflow on recreation: a literature review. Journal of the American Water Resources Association, 27(6), 979–989. https://doi.org/10.1111/j.1752-1688.1991.tb03147.
Bruckner, T., Bashmakov, I. A., Mulugeta, Y., Chum, H., De la Vega Navarro, A., Edmonds, J., … Zhang, X. (2014). Energy systems. In O. Edenhofer, R. Pichs-Madruga, Y. Sokona, E. Farahani, S. Kadner, K. Seyboth, & J. C. Minx (Eds.), Climate change 2014: Mitigation of climate change. Contribution of working group III to the fifth assessment report of the intergovernmental panel on climate change (pp. 511–597). Cambridge: Cambridge University Press.
Bruno, M. C., Maiolini, B., Carolli, M., & Silveri, L. (2010). Short time-scale impacts of hydropeaking on benthic invertebrates in an Alpine stream (Trentino, Italy). Limnologica, 40(4), 281-290.
70
Brunson, M. W., & Shelby, B. (1993). Recreation substitutability: A research agenda. Leisure Sciences, 15(1), 67–74. https://doi.org/10.1080/01490409309513187.
Bunn, S. E., & Arthington, A. H. (2002). Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management, 30(4), 492–507. https://doi.org/10.1007/s00267-002-2737-0
Carolli, M., Vanzo, D., Siviglia, A., Zolezzi, G., Bruno, M. C., & Alfredsen, K. (2015). A simple procedure for the assessment of hydropeaking flow alterations applied to several European streams. Aquatic Sciences, 77(4), 639–653.
Casas-Mulet, R., Saltveit, S. J., & Alfredsen, K. T. (2016). Hydrological and thermal effects of hydropeaking on early life stages of salmonids: A modelling approach for implementing mitigation strategies. Science of the Total Environment, 573, 1660–1672.
Choi, B., & Choi, S. U. (2018). Impacts of hydropeaking and thermopeaking on the downstream habitat in the Dal River, Korea. Ecological Informatics, 43, 1-11.
Daubechies, I. (1992). Ten Lectures on Wavelets. Society for Industrial and Applied Mathematics. https://doi.org/doi:10.1137/1.9781611970104.
Dynesius, M., & Nilsson, C. (1994). Fragmentation and flow regulation of river systems in the northern third of the world. Science, 266(5186), 753-762.
European Union [EU]. (2009). Directive 2009/28/EC of the European Parliament and of the Council of 23 April 2009 on the promotion of the use of energy from renewable sources and amending and subsequently repealing Directives 2001/77/EC and 2003/30/EC. Official Journal of the European Union, 5, 2009.
European Commission. (2012). Report from the Commission to the European parliament and the council on the implementation of the Water Framework Directive (2000/60/EC) River Basin management plans. Brussels: Author
Fette, M., Weber, C., Peter, A., & Wehrli, B. (2007). Hydropower production and river rehabilitation: A case study on an alpine river. Environmental Modeling & Assessment, 12(4), 257–267. https://doi.org/10.1007/s10666-006-9061-7
García, A., Jorde, K., Habit, E., Caamaño, D., & Parra, O. (2011). Downstream environmental effects of dam operations: changes in habitat quality for native fish species. River Research and Applications, 27(3), 312–327.
Gaudard, L., & Romerio, F. (2014). The future of hydropower in Europe: Interconnecting climate, markets and policies. Environmental Science & Policy, 37, 172–181.
Graabak, I., & Warland, L. (2014). A carbon neutral power system in the Nordic region in 2050. SINTEF Energy Research, Trondheim, Norway, TR A, 7365.
Grill, G., Lehner, B., Thieme, M., Geenen, B., Tickner, D., Antonelli, F., Babu, S., Borrelli, P., Cheng, L., Crochetiere, H., Macedo, H. E., Filgueiras, R., Goichot, M., Higgins, J., Hogan, Z., Lip, B., McClain, M. E., Meng, J., Mulligan, M., … Zarfl, C. (2019). Mapping the world’s free-flowing rivers. Nature, 569(7755), 215–221. https://doi.org/10.1038/s41586-019-1111-9
Haas, C., Zinke, P., Vollset, K. W., Sauterleute, J., & Skoglund, H. (2016). Behaviour of spawning Atlantic salmon and brown trout during ramping events. Journal of Applied Water Engineering and Research, 4(1), 25–30.
71
Haghighi, A. T., & Kløve, B. (2013). Development of a general river regime index (RRI) for intra-annual flow variation based on the unit river concept and flow variation end-points. Journal of hydrology, 503, 169-177.
Haghighi, A. T., & Kløve, B. (2015). Development of monthly optimal flow regimes for allocated environmental flow considering natural flow regimes and several surface water protection targets. Ecological Engineering, 82, 390–399.
Haghighi, A. T., Marttila, H., & Kløve, B. (2014). Development of a new index to assess river regime impacts after dam construction. Global and Planetary Change, 122, 186–196.
Halleraker, J. H., Saltveit, S. J., Harby, A., Arnekleiv, J. v, Fjeldstad, H.-P., & Kohler, B. (2003). Factors influencing stranding of wild juvenile brown trout (Salmo trutta) during rapid and frequent flow decreases in an artificial stream. River Research and Applications, 19(5), 589–603. https://doi.org/10.1002/rra.752
Harnish, R. A., Sharma, R., McMichael, G. A., Langshaw, R. B., & Pearsons, T. N. (2014). Effect of hydroelectric dam operations on the freshwater productivity of a Columbia River fall Chinook salmon population. Canadian Journal of Fisheries and Aquatic Sciences, 71(4), 602–615.
Hauer, C., Unfer, G., Holzapfel, P., Haimann, M., & Habersack, H. (2014). Impact of channel bar form and grain size variability on estimated stranding risk of juvenile brown trout during hydropeaking. Earth Surface Processes and Landforms, 39(12), 1622–1641.
Hellsten, S., & Riihimäki, J. (1996). Effects of lake water level regulation on the dynamics of littoral vegetation in northern Finland. Hydrobiologia, 340(1-3), 85-92.
Hennig, T., Wang, W., Feng, Y., Ou, X., & He, D. (2013). Review of Yunnan’s hydropower development. Comparing small and large hydropower projects regarding their environmental implications and socio-economic consequences. Renewable and Sustainable Energy Reviews, 27, 585–595.
Huntington, T. G., Richardson, A. D., McGuire, K. J., & Hayhoe, K. (2009). Climate and hydrological changes in the northeastern United States: recent trends and implications for forested and aquatic ecosystems. Canadian Journal of Forest Research, 39(2), 199–212.
International Hydropower Association. (2018). Hydropower status report. International Hydropower Association: London, UK. https://www.hydropower.org/publications/2018-hydropower-status-report.
Irannezhad, M., Chen, D., & Kløve, B. (2015). Interannual variations and trends in surface air temperature in Finland in relation to atmospheric circulation patterns, 1961–2011. International Journal of Climatology, 35(10), 3078-3092.
Irannezhad, M., Marttila, H., & Kløve, B. (2014). Long‐term variations and trends in precipitation in Finland. International Journal of Climatology, 34(10), 3139–3153.
Irannezhad, M., Ronkanen, A., & Kløve, B. (2016). Wintertime climate factors controlling snow resource decline in Finland. International Journal of Climatology, 36(1), 110–131.
72
Jager, H. I., & Bevelhimer, M. S. (2007). How run-of-river operation affects hydropower generation and value. Environmental Management, 40(6), 1004–1015. https://doi.org/10.1007/s00267-007-9008-z
Korhonen, J., & Kuusisto, E. (2010). Long-term changes in the discharge regime in Finland. Hydrology Research, 41(3–4), 253–268.
Korman, J., & Campana, S. E. (2009). Effects of hydropeaking on nearshore habitat use and growth of age-0 rainbow trout in a large regulated river. Transactions of the American Fisheries Society, 138(1), 76–87. https://doi.org/10.1577/T08-026.1
Krause, F (2011). River management. Technological challenge or conceptual illusion? Salmon weirs and hydroelectric dams on the Kemi River in Northern Finland. In: M. Schmidt, V. Onyango, & D. Palekhov (Eds.), Implementing Environmental and Resource Management pp. (229–248). Berlin: Springer.
Maheu, A., St-Hilaire, A., Caissie, D., El-Jabi, N., Bourque, G., & Boisclair, D. (2016). A regional analysis of the impact of dams on water temperature in medium-size rivers in eastern Canada. Canadian Journal of Fisheries and Aquatic Sciences, 73(12), 1885–1897. https://doi.org/10.1139/cjfas-2015-0486.
Marcarelli, A. M., Kirk, R. W. V., & Baxter, C. V. (2010). Predicting effects of hydrologic alteration and climate change on ecosystem metabolism in a western US river. Ecological Applications, 20(8), 2081-2088.
Matti, B., Dahlke, H. E., & Lyon, S. W. (2016). On the variability of cold region flooding. Journal of Hydrology, 534, 669–679. https://doi.org/10.1016/j.jhydrol.2016.01.055
McClelland, J. W., Holmes, R. M., Peterson, B. J., & Stieglitz, M. (2004). Increasing river discharge in the Eurasian Arctic: Consideration of dams, permafrost thaw, and fires as potential agents of change. Journal of Geophysical Research-Atmospheres, 109(D18), D18102. https://doi.org/10.1029/2004JD004583
Meile, T., Boillat, J.-L., & Schleiss, A. J. (2011). Hydropeaking indicators for characterization of the Upper-Rhone River in Switzerland. Aquatic Sciences, 73(1), 171–182. https://doi.org/10.1007/s00027-010-0154-7
Mileva, A., Johnston, J., Nelson, J. H., & Kammen, D. M. (2016). Power system balancing for deep decarbonization of the electricity sector. Applied Energy, 162, 1001–1009.
Mustonen, K. R., Mykrä, H., Marttila, H., Haghighi, A. T., Kløve, B., Aroviita, J., ... & Muotka, T. (2016). Defining the natural flow regimes of boreal rivers: relationship with benthic macroinvertebrate communities. Freshwater Science, 35(2), 559-572.
Mustonen, K. R., Mykrä, H., Marttila, H., Sarremejane, R., Veijalainen, N., Sippel, K., Muotka, T., & Hawkins, C. P. (2018). Thermal and hydrologic responses to climate change predict marked alterations in boreal stream invertebrate assemblages. Global change biology, 24(6), 2434-2446.
Nature Conservancy, (2009). IHA software version 7.1 user’s manual. The Nature Conservancy, https://www.conservationgateway.org/Documents/IHAV7.pdf.
Nilsson, C., Nilsson, E., Johansson, M. E., Dynesius, M., Grelsson, G., Xiong, S., ... & Danvind, M. (1993). Processes structuring riparian vegetation. Current topics in botanical research, 1, 419-431.
73
Olsson, T., Jakkila, J., Veijalainen, N., Backman, L., Kaurola, J., & Vehviläinen, B. (2015). Impacts of climate change on temperature, precipitation and hydrology in Finland-studies using bias corrected Regional Climate Model data. Hydrology and Earth System Sciences Discussions, 12, 2657–2706.
Omstedt, A., Pettersen, C., Rodhe, J., & Winsor, P. (2004). Baltic Sea climate: 200 yr of data on air temperature, sea level variation, ice cover, and atmospheric circulation. Climate Research, 25(3), 205–216. https://doi.org/10.3354/cr025205
Palmer, M. A., Liermann, C. A. R., Nilsson, C., Flörke, M., Alcamo, J., Lake, P. S., & Bond, N. (2008). Climate change and the world’s river basins: anticipating management options. Frontiers in Ecology and the Environment, 6(2), 81–89.
Park, H., Walsh, J. E., Kim, Y., Nakai, T., & Ohata, T. (2013). The role of declining Arctic sea ice in recent decreasing terrestrial Arctic snow depths. Polar Science, 7(2), 174–187.
Patro, E. R., Michele, C. de, & Avanzi, F. (2018). Future perspectives of run-of-the-river hydropower and the impact of glaciers’ shrinkage: The case of Italian Alps. Applied Energy, 231, 699–713.
Percival, D. B., & Walden, A. T. (2000). Wavelet Methods for Time Series Analysis. Cambridge University Press. https://doi.org/10.1017/CBO9780511841040
Peterson, B. J., Holmes, R. M., McClelland, J. W., Vorosmarty, C. J., Lammers, R. B., Shiklomanov, A. I., Shiklomanov, I. A., & Rahmstorf, S. (2002). Increasing river discharge to the Arctic Ocean. Science, 298(5601), 2171–2173. https://doi.org/10.1126/science.1077445
Poff, N.L., Allan, J. D., Bain, M. B., Karr, J. R., Prestegaard, K. L., Richter, B. D., Sparks, R. E., & Stromberg, J. C. (1997). The natural flow regime. Bioscience, 47(11), 769–784. https://doi.org/10.2307/1313099
Poff, N.L., & Zimmerman, J. K. H. (2010). Ecological responses to altered flow regimes: a literature review to inform the science and management of environmental flows. Freshwater Biology, 55(1), 194–205.
Poff, N.L., Olden, J. D., Merritt, D. M., & Pepin, D. M. (2007). Homogenization of regional river dynamics by dams and global biodiversity implications. Proceedings of the National Academy of Sciences of the United States of America, 104(14), 5732–5737. https://doi.org/0609812104.
Poff, N. L., & Ward, J. V. (1989). Implications of streamflow variability and predictability for lotic community structure - a regional-analysis of streamflow patterns. Canadian Journal of Fisheries and Aquatic Sciences, 46(10), 1805–1818. https://doi.org/10.1139/f89-228
Prowse, T., Bring, A., Mård, J., Carmack, E., Holland, M., Instanes, A., Vihma, T., & Wrona, F. J. (2015). Arctic freshwater synthesis: Summary of key emerging issues. Journal of Geophysical Research: Biogeosciences, 120(10), 1887–1893.
Raynaud, D., Hingray, B., François, B., & Creutin, J. D. (2018). Energy droughts from variable renewable energy sources in European climates. Renewable Energy, 125, 578–589.
74
Rheinheimer, D. E., & Viers, J. H. (2015). Combined effects of reservoir operations and climate warming on the flow regime of hydropower bypass reaches of California’s Sierra Nevada. River Research and Applications, 31(3), 269–279.
Richter, B., Baumgartner, J., Wigington, R., & Braun, D. (1997). How much water does a river need? Freshwater Biology, 37(1), 231–249. https://doi.org/10.1046/j.1365-2427.1997.00153.x
Richter, B. D., Baumgartner, J. V., Powell, J., & Braun, D. P. (1996). A method for assessing hydrologic alteration within ecosystems. Conservation Biology, 1163–1174.
Richter, B. D., & Thomas, G. A. (2007). Restoring environmental flows by modifying dam operations. Ecology and society, 12(1).
Robbins, J. L., & Lewis, L. Y. (2008). Demolish it and they will come: Estimating the economic impacts of restoring a recreational fishery1. Journal of the American Water Resources Association, 44(6), 1488–1499. https://doi.org/10.1111/j.1752-1688.2008.00253.x
Romakkaniemi, A., Lilja, J., Nykanen, M., Marjomaki, T. J., & Jurvelius, J. (2000). Spawning run of Atlantic Salmon (Salmo salar) in the River Tornionjoki monitored by horizontal split-beam echosounding. Aquatic Living Resources, 13(5), 349–354. https://doi.org/10.1016/S0990-7440(00)01083-4
Rosenberg, D M, McCully, P., & Pringle, C. M. (2000). Global-scale environmental effects of hydrological alterations: Introduction. Bioscience, 50(9), 746–751. https://doi.org/10.1641/0006-3568(2000)050[0746:GSEEOH]2.0.CO;2
Rossel, V., & de la Fuente, A. (2015). Assessing the link between environmental flow, hydropeaking operation and water quality of reservoirs. Ecological Engineering, 85, 26-38.
Sauterleute, J., Wolfgang, O., & Graabak, I. (2015). Scenarios for large-scale balancing and storage from Norwegian hydropower. Water Resources, 2015, 2–26.
Schmutz, S., Bakken, T. H., Friedrich, T., Greimel, F., Harby, A., Jungwirth, M., Melcher, A., Unfer, G., & Zeiringer, B. (2015). Response of fish communities to hydrological and morphological alterations in hydropeaking rivers of Austria. River Research and Applications, 31(8), 919–930.
Scruton, D. A., Pennell, C., Ollerhead, L. M. N., Alfredsen, K., Stickler, M., Harby, A., Robertson, M., Clarke, K. D., & LeDrew, L. J. (2008). A synopsis of “hydropeaking” studies on the response of juvenile Atlantic salmon to experimental flow alteration. Hydrobiologia, 609, 263–275. https://doi.org/10.1007/s10750-008-9409-x
Shiau, J.-T., & Wu, F.-C. (2013). Optimizing environmental flows for multiple reaches affected by a multipurpose reservoir system in Taiwan: Restoring natural flow regimes at multiple temporal scales. Water Resources Research, 49(1), 565–584. https://doi.org/10.1029/2012WR012638
Sokka, L. (2017). Wood-based energy as a strategy for climate change mitigation in the arctic-perspectives on assessment of climate impacts and resource efficiency with life cycle assessment. In: K. Latola & H. Savela, (Eds.), Springer Polar Sciences pp. (61–66). Saint Petersburg: Springer.
75
Stanford, J. a, Ward, J. v, Liss, W. J., Frissell, C. a, Williams, R. N., Lichatowich, J. A, & Coutant, C. C. (1996). A general protocol for restoration of regulated rivers. Regulated Rivers: Research & Management, 12(4–5), 391–413. https://doi.org/10.1002/(SICI)1099-1646(199607)12:4/5<391::AID-RRR436>3.0.CO;2-4
Steel, E. A., & Lange, I. A. (2007). Using wavelet analysis to detect changes in water temperature regimes at multiple scales: effects of multi-purpose dams in the Willamette River basin. River Research and Applications, 23(4), 351–359. https://doi.org/10.1002/rra.985
Sternberg, R. (2006). Damming the river: a changing perspective on altering nature. Renewable and Sustainable Energy Reviews, 10(3), 165–197.
Svensson, B. S. (2000). Hydropower and instream flow requirements for fish in Sweden. Fisheries Management and Ecology, 7(1‐2), 145–155.
Teigland, J. (1999). Predictions and realities: Impacts on tourism and recreation from hydropower and major road developments. Impact Assessment and Project Appraisal, 17(1), 67–76. https://doi.org/10.3152/147154699781767972
Tietäväinen, H., Tuomenvirta, H., & Venäläinen, A. (2010). Annual and seasonal mean temperatures in Finland during the last 160 years based on gridded temperature data. International Journal of Climatology, 30(15), 2247–2256.
Tonkin, J. D., Merritt, D. M., Olden, J. D., Reynolds, L. v, & Lytle, D. A. (2018). Flow regime alteration degrades ecological networks in riparian ecosystems. Nature Ecology & Evolution, 2(1), 86.
Torrence, C., & Compo, G. P. (1998). A practical guide to wavelet analysis. Bulletin of the American Meteorological Society, 79(1), 61–78. https://doi.org/10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2
Tuomenvirta, H., & Heino, R. (1996). Climatic changes in Finland—recent findings. Geophysica, 32(1-2), 61-75.
United States. Bureau of Reclamation. (1987). Design of small dams (3rd edn.). Washington, D.C: US Department of the Interior, Bureau of Reclamation.
USBR. 1977. Design of Small Dams. Department of the Interior, US Bureau of Reclamation. US Government Printing Office: Washington,DC; 816.
Vanzo, D., Siviglia, A., Carolli, M., & Zolezzi, G. (2016). Characterization of sub‐daily thermal regime in alpine rivers: quantification of alterations induced by hydropeaking. Hydrological Processes, 30(7), 1052–1070.
Vardanyan, Y., & Hesamzadeh, M. R. (2017). The coordinated bidding of a hydropower producer in three-settlement markets with time-dependent risk measure. Electric Power Systems Research, 151, 40–58.
Vehviläinen, B., Huttunen, M., Alfredsen, K., Harby, A., Linnansaari, T., & Ugedal, O. (2012). Climate change and water resources in Finland. River Research and Applications, 28(6), 731–739. https://doi.org/10.1002/rra.1550
Veijalainen, N., Lotsari, E., Alho, P., Vehviläinen, B., & Käyhkö, J. (2010). National scale assessment of climate change impacts on flooding in Finland. Journal of Hydrology, 391(3), 333–350.
76
Vollset, K. W., Skoglund, H., Wiers, T., & Barlaup, B. T. (2016). Effects of hydropeaking on the spawning behaviour of Atlantic salmon Salmo salar and brown trout Salmo trutta. Journal of Fish Biology, 88(6), 2236–2250.
Wilson, D., Hisdal, H., & Lawrence, D. (2010). Has streamflow changed in the Nordic countries?–Recent trends and comparisons to hydrological projections. Journal of Hydrology, 394(3), 334–346.
Winemiller, K. O., McIntyre, P. B., Castello, L., Fluet-Chouinard, E., Giarrizzo, T., Nam, S., Baird, I. G., Darwall, W., Lujan, N. K., Harrison, I., Stiassny, M. L. J., Silvano, R. A. M., Fitzgerald, D. B., Pelicice, F. M., Agostinho, A. A., Gomes, L. C., Albert, J. S., Baran, E., Petrere, M., …Sáenz, L. (2016). Balancing hydropower and biodiversity in the Amazon, Congo, and Mekong. Science, 351(6269), 128. https://doi.org/10.1126/science.aac7082
Wu, F.-C., Chang, C.-F., & Shiau, J.-T. (2015). Assessment of flow regime alterations over a spectrum of temporal scales using wavelet-based approaches. Water Resources Research, 51(5), 3317–3338. https://doi.org/10.1002/2014WR016595
Yang, T., Zhang, Q., Chen, Y. D., Tao, X., Xu, C., & Chen, X. (2008). A spatial assessment of hydrologic alteration caused by dam construction in the middle and lower Yellow River, China. Hydrological Processes, 22(18), 3829–3843. https://doi.org/10.1002/hyp.6993
Zarfl, C., Lumsdon, A. E., Berlekamp, J., Tydecks, L., & Tockner, K. (2015). A global boom in hydropower dam construction. Aquatic Sciences, 77(1), 161–170. https://doi.org/10.1007/s00027-014-0377-0
Zhang, Y., Zhai, X., & Zhao, T. (2018). Annual shifts of flow regime alteration: new insights from the Chaishitan Reservoir in China. Scientific Reports, 8(1), 1414.
Zimmerman, J. K. H., Letcher, B. H., Nislow, K. H., Lutz, K. A., & Magilligan, F. J. (2010). Determining the effects of dams on subdaily variation in river flows at a whole-basin scale. River Research and Applications, 26(10), 1246–1260. https://doi.org/10.1002/rra.1324
Zolezzi, G., Bellin, A., Bruno, M. C., Maiolini, B., & Siviglia, A. (2009). Assessing hydrological alterations at multiple temporal scales: Adige River, Italy. Water Resources Research, 45(12). https://doi.org/10.1029/2008WR007266
77
Appendices Table A1. Land use classification in the Tornionjoki and Kemijoki catchments (reprinted with permission from Paper I © 2016 Elsevier B.V.).
Land use Tornionjoki Kemijoki
(ha) % (ha) %
Urban area 23,585 0.6 38,082 0.77
Agricultural area 33,688 0.85 37,187 0.76
Forest area 3,020,642 76.2 3,838,244 77.56
Total peatland area 662,732 16.7 804,899 16.27
Drained peatland areaa 169,644 4.28 476,451 12.02
Lake area 225,364 5.65 228,280 4.64 aPart of total peatland area.
Table A2. Pre- and post-impact periods used in spatial and temporal analysis (reprinted with permission from Paper I © 2016 Elsevier B.V.).
Gauging station
River Location For spatial analysis For temporal analysis
Pre-impact period
Post-impact period
Pre-impact period
Post-impact period
Seitakorva Kemijoki Upstream 1971–1981 1981–2015 1952–1966 2001–2015
Ounaskoski Kemijoki Midstream 1921–1949 2001–2015 1952–1966 2001–2015
Taivalkoski Kemijoki Downstream 1921–1949 2001–2015 1952–1966 2001–2015
Enontekio Tornionjoki Upstream 1972–1986 2001–2015 a
a
Pello Tornionjoki Midstream 1956–1978 2001–2015 a
a
Kilpisjärvi Tornionjoki Upstream a
a 1952–1966 2001–2015
Muonio Tornionjoki Midstream a
a 1952–1966 2001–2015
Karunki Tornionjoki Downstream 1921–1949 2001–2015 1952–1966 2001–2015
aStation not included in that particular analysis.
78
79
Original publications I Ashraf, F. B., Haghighi, A. T., Marttila, H., & Kløve, B. (2016). Assessing impacts of
climate change and river regulation on flow regimes in cold climate: a study of a pristine and a regulated river in the sub-arctic setting of Northern Europe. Journal of Hydrology, 542, 410-422.
II Ashraf, F. B., Haghighi, A. T., Riml, J., Alfredsen, K., Koskela, J. J., Kløve, B., & Marttila, H. (2018). Changes in short term river flow regulation and hydropeaking in Nordic rivers. Scientific reports, 8(1), 17232.
III Ashraf, F. B., Haghighi, A. T., Riml, J., Kløve, B., & Marttila, H. (2020). Assessment of sub-daily flow alterations using globally mapped free-flowing rivers and wavelet analysis. Manuscript.
IV Haghighi, A. T., Ashraf, F. B., Riml, J., Koskela, J., Kløve, B., & Marttila, H. (2019). A power market-based operation support model for sub-daily hydropower regulation practices. Applied Energy, 255, 113905.
Reprinted with permission from Elsevier (Papers I and IV), Authors (Paper III), and under Creative Commons CC-BY license from Springer Nature (Paper II).
Original publications are not included in the electronic version of the dissertation.
80
A C T A U N I V E R S I T A T I S O U L U E N S I S
Book orders:Virtual book store
http://verkkokauppa.juvenesprint.fi
S E R I E S C T E C H N I C A
730. Ashraf, Muhammad Ikram (2019) Radio resource management in device-to-deviceand vehicle-to-vehicle communication in 5G networks and beyond
731. Elbamby, Mohammed S. (2019) Flexible duplexing and resource optimization insmall cell networks
732. Ruokamo, Henna (2019) Time-gating technique for a single-photon detection-based solid-state time-of-flight 3D range imager
733. Piñero, Pablo (2019) The metabolism of socio-economic systems : combination ofinput-output analysis and material flow accounting for footprint-type indicators
734. Mahmood, Shahriare (2020) The substantive approach for managing demand andsupply sustainably in fashion industry
735. Laukka, Aleksi (2020) The effects of microalloying on the scale formation of AISI304 stainless steel in walking beam furnace conditions
736. Ali, Samad (2020) Learning-based predictive resource allocation for machine typecommunications
737. Vatjus-Anttila, Jarkko (2020) Energy efficient rendering of extensible 3D virtualenvironments on mobile devices
738. Wijewardhana, Uditha (2020) Sparse recovery algorithms for streaming andmultidimensional signals
739. Li, Panpan (2020) Deep eutectic solvents in the production of wood-basednanomaterials
740. Nuortimo, Kalle (2020) Public acceptance in energy technology development anddeployment : opinion mining case study
741. Juuti, Timo (2020) The precipitation of Laves phase and its contribution tomechanical properties in novel high-Cr ferritic stainless steels
742. Myllymäki, Pekka (2020) Chemical precipitation in the simultaneous removal ofNH4?N and PO4?P from wastewaters using industrial waste materials
743. Alcaraz López, Onel Luis (2020) Resource allocation for machine-typecommunication : from massive connectivity to ultra-reliable low-latency
744. Frondelius, Tero (2020) Development of methods in engine design process
746. Särestöniemi, Mariella (2020) UWB in-body propagation and radio channelcharacteristics for wireless body area network applications
C747etukansi2.fm Page 2 Tuesday, April 7, 2020 10:40 AM
UNIVERSITY OF OULU P .O. Box 8000 F I -90014 UNIVERSITY OF OULU FINLAND
A C T A U N I V E R S I T A T I S O U L U E N S I S
University Lecturer Tuomo Glumoff
University Lecturer Santeri Palviainen
Postdoctoral researcher Jani Peräntie
University Lecturer Anne Tuomisto
University Lecturer Veli-Matti Ulvinen
Planning Director Pertti Tikkanen
Professor Jari Juga
University Lecturer Anu Soikkeli
University Lecturer Santeri Palviainen
Publications Editor Kirsti Nurkkala
ISBN 978-952-62-2604-0 (Paperback)ISBN 978-952-62-2605-7 (PDF)ISSN 0355-3213 (Print)ISSN 1796-2226 (Online)
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
U N I V E R S I TAT I S O U L U E N S I SACTAC
TECHNICA
OULU 2020
C 747
Faisal Bin Ashraf
RIVER REGIMES AND ENERGY DEMAND INTERACTIONS IN NORDIC RIVERS
UNIVERSITY OF OULU GRADUATE SCHOOL;UNIVERSITY OF OULU,FACULTY OF TECHNOLOGY
C 747
AC
TAFaisal B
in Ashraf
C747etukansi2.fm Page 1 Tuesday, April 7, 2020 10:40 AM
