1 23

Arabian Journal of Geosciences ISSN 1866-7511 Arab J GeosciDOI 10.1007/s12517-012-0761-9

Assessing the impacts of changing landcover and climate on Hokersar wetland inIndian Himalayas

Shakil Ahmad Romshoo & Irfan Rashid

1 23

Your article is protected by copyright and all

rights are held exclusively by Saudi Society

for Geosciences. This e-offprint is for personal

use only and shall not be self-archived in

electronic repositories. If you wish to self-

archive your work, please use the accepted

author’s version for posting to your own

website or your institution’s repository. You

may further deposit the accepted author’s

version on a funder’s repository at a funder’s

request, provided it is not made publicly

available until 12 months after publication.

ORIGINAL PAPER

Assessing the impacts of changing land cover and climateon Hokersar wetland in Indian Himalayas

Shakil Ahmad Romshoo & Irfan Rashid

Received: 31 March 2012 /Accepted: 31 October 2012# Saudi Society for Geosciences 2012

Abstract Monitoring the spatiotemporal changes in wet-lands and assessing their causal factors is critical for devel-oping robust strategies for the conservation and restorationof these ecologically important ecosystems. In this study, thespatiotemporal changes in the land cover system within aHimalayan wetland and its catchment were assessed andcorrelated using a time series of satellite, historical, and fielddata. Significant changes in the spatial extent, water depth,and the land system of the Hokersar wetland were observedfrom the spatiotemporal analysis of the data from 1969 to2008. The wetland area has shrunk from 18.75 km2 in 1969to 13 km2 in 2008 with drastic reduction in the water depthof the wetland. The marshy lands, habitat of the migratorybirds, have shrunk from 16.3 km2 in 1969 to 5.62 km2 in2008 and have been colonized by various other land covertypes. The land system and water extent changes within thewetland were related to the spatiotemporal changes in theland cover and hydrometeorological variables at the catch-ment scale. Significant changes in the forest cover (88.33–55.78 km2), settlement (4.63–15.35 km2), and water bodies(1.75–0.51 km2) were observed in the catchment. It is con-cluded that the urbanization, deforestation, changes in thehydrologic and climatic conditions, and other land systemchanges observed in the catchment are the main causesresponsible for the depleting wetland extent, water depth,and biodiversity by adversely influencing the hydrologicerosion and other land surface processes in the catchment.All these causes and effects are manifest in the form ofdeterioration of the water quality, water quantity, the biodi-versity changes, and the decreasing migratory bird popula-tion in the wetland.

Keywords Spatiotemporal changes . Catchment . Remotesensing . Biodiversity . Hydrometeorology

Introduction

Wetlands occupy about 7 % of the Earth’s land surface (MEA2005; Mitsch and Gosselink 1986); and in the mountainousregion of Kashmir Himalayas alone, there are 3,813 wetlandsand water bodies (Romshoo et al. 2010). Sustainable manage-ment of these wetland ecosystems is necessary as wetlandsprovide a variety of services and functions and contributetremendously to the livelihoods and human wellbeing in theregion. Most important wetland ecosystem services affectingthe human wellbeing involve fisheries, food products, fresh-water supplies, water purification and detoxification, andglobal climate change regulation (Costanza et al. 1997; Davis1993; Hruby 1995; MEA 2005). Wetlands deliver a widearray of hydrological services, for instance, flood regulation,promote groundwater recharge, and regulate river flows(Bullock and Acreman 2003). Further, wetlands are amongthe most productive ecosystems and a rich repository ofbiodiversity and are known to play significant role in carbonsequestration (Kraiem 2002). The world’s wetlands aredegrading at an alarming rate, more than other ecosystemsseriously affecting their biodiversity (Vorosmarty et al. 2010).Due to accelerated rate of human intervention and human-induced modification of natural processes, natural wetlandlandscapes also are today under acute seasonal water scarcity.Wetland areas have been gradually squeezing, permanentwetland areas have transmuted into semipermanent wetlandswith groundwater table slashing down rapidly (Pal andAkoma 2009). Permanent and seasonal changes within wet-lands occur in response to a range of external factors, such asthe changes in the land systems at the catchment scale(Dooner 2003; Gillies et al. 2003), fluctuations in water table

S. A. Romshoo (*) : I. RashidDepartment of Earth Sciences, University of Kashmir, Hazratbal,Srinagar Kashmir 190006, Indiae-mail: shakilrom@yahoo.com

Arab J GeosciDOI 10.1007/s12517-012-0761-9

Author's personal copy

(Funk et al. 1994), climate change (Kraiem 2002), or otherassociated human activities. Climate change may exacerbateimpacts of threats to wetlands through predicted reductions inrainfall and increased temperature, decreasing flow, and alteringtiming and variability of flow regimes (Kingsford 2011). Thetiming, magnitude, and frequency of rainfall or snowmelt inmany wetland catchments is predicted to change (Klausmeyerand Shaw 2009; Palmer et al. 2009; Viers and Rheinheimer2011), with increasing temperatures predicted to augment flowsearly in spring as snow melt and produce flow reductions insummer (Aldous et al. 2011).

Worldwide, the lack of understanding of the values andfunctions of the wetlands have led to their conversion foragriculture, settlements, plantations, and other developmentactivities (Joshi et al. 2002; Wetlands International 2007).Similar scenario is being witnessed in the mountainousHimalayan region where unplanned urbanization, recklessdeforestation, and the depleting snow and glacier resourcesare the major causes of the wetland depletion.

The need for management, protection, and restoration ofthese valuable systems, as well as the need to understand thewetland hydrology and ecology, have spurred the investigation

Fig. 1 Study area

Arab J Geosci

Author's personal copy

of new technologies for mapping and monitoring of wetlands(Gondwe et al. 2010; Hess et al. 1995; Lyon 2001; Tanis et al.1994). Though the entire wetlands in the Indian Himalayashave been mapped at 1:50,000 scale (Romshoo et al. 2010),there is very little information, if at all, on how they have beenchanging over the time in response to the changes in theclimatic variables and the land system changes occurring atthe catchment level (Anonymous 1990; Bourgeau-Chavez etal. 2001; Garg et al. 1998). It is therefore essential to use a timeseries of satellite data for assessing the spatiotemporal changeswithin the wetlands and the catchment areas to determine thecause–effect relationship so that robust strategy for the man-agement, conservation, and restoration of the wetland is devel-oped. Multitemporal monitoring of wetlands using remotesensing and geographic information systems gives a completeunderstanding of distribution, structure, and functionality ofwetland ecosystem (Munyati 2000; Ramsey 1998; Romshoo2004; Touzi et al. 2007) and the spatiotemporal dynamics ofvarious variables in the catchment areas (Basnyat et al. 2000;Omernik et al. 1981; Saxena et al. 2000; Rashid et al. 2011). Anumber of studies related to limnological variables in lakes andwetlands have been attempted using remote sensing (Birkett1995; Kapetsky 1987; Olmanson et al. 2002; Ozesmi andBauer 2002; Roeck et al. 2008; Romshoo and Sumira 2010;Romshoo and Muslim 2011).

In the present study, time series multisensor satellite datawas used to determine the spatiotemporal changes in Hoker-sar wetland that has tremendous ecohydrological and socio-economic importance. These spatiotemporal changes wererelated with the changes at the catchment scale and hydro-meteorology. Most of the studies conducted on the Hoker-sar, the Queen of wetlands in Kashmir Himalayas are eitherfocused on the hydrobiology or hydrochemistry (Gangooand Makaya 2000; Handoo and Kaul 1982; Handoo 1978;Kak 1990; Khan 2000; Kaul and Zutshi 1967; Kaul 1982;Pandit 1980; Pandit and Kumar 2006; Rather et al. 2001).However, very few studies have used the geoinformaticsapproach to study spatiotemporal dynamics and limnologi-cal variables of the Hokersar wetland (Humayun and Joshi2000; Joshi et al. 2002; Romshoo et al. 2011). The presentstudy assumes significance, in view of the fact that a longertime series of satellite and other spatial data stretching from1969 to 2008 has been used to monitor and assess thespatiotemporal evolution of the wetland for the last fourdecades and the changes that have occurred in the land useand land cover types within the catchment spread over anarea of about 732 km2. Further, the linkages between theobserved changes in the wetland have been correlated withthe hydrometeorological data to investigate if there are anyimpacts of the changing climate on the wetland.

Fig. 2 Scheme of methodology

Arab J Geosci

Author's personal copy

Study area

Doodhganga, located in Kashmir Himalaya, India is one ofthe major left-bank catchments of Jhelum River. It is situat-ed between 33° 15′–34° 15′ latitudes and 74° 45′–74° 83′longitudes covering an area of 732.6 km2. It is bounded bylofty Pir Panjal Mountain Range on south. The catchmenthas a varied topography and exhibits altitudinal extremes of1,548 to 4,634 m (above mean sea level (amsl)). Its relief isdiverse, comprising of steep slopes, plateaus, plains, andalluvial fans. The plains of the catchment are very fertile,hence, ideal for agriculture, whereas the higher reachescomprise dense pine forests and lush green alpine pastures.Geologically, the area consists of Panjal traps, limestone,

Karewa Formation, and Recent Alluvium. The characteristicKarewa Formation in relatively lower elevations is ideal forhorticulture. The study area experiences temperate climatewith the average winter and summer temperatures rangingfrom 5 to 25 °C, respectively. The average annual precipita-tion is about 660 mm in the form of rain and snow. Doodh-ganga stream, one of the important perennial tributaries ofriver Jhelum, is the main drainage and water resource in thecatchment. Doodhganga stream flows for a course of about56 km before emptying into Hokersar wetland.

Hokersar wetland (34° 06′ N latitude, 74° 05′ Elongitude) lying in the Northern most part of Doodh-ganga catchment is a protected wildlife reserve and aRamsar site at an altitude of 1,584 m (amsl). Thewetland harbors about two million migratory waterfowlduring winter that migrate from Siberia and the CentralAsian region. The wetland is fed by two inlet streams—Doodhganga (from east) and Sukhnag Nalla (fromwest). The wetland attains a maximum depth of 2.5 min spring due to appreciation in discharge from thesnow-melt water in the upper reaches of Doodhgangacatchment. The water depth in autumn is minimum at0.7 m. Figure 1 shows the location of the study area.

Fig. 3 Hokarsar boundary extents at different points in time

Table 1 Spatial extentof the Hokersar wetlandat different points intime

Year Area (km2)

1969 18.75

1992 14.94

2001 14.71

2005 14.33

2008 13.00

Arab J Geosci

Author's personal copy

Materials and methods

Datasets used for changes in Hokersar wetland

Multitemporal datasets from various sources from 1969 to2008 were used for analyzing changes in Hokersar wetland.Survey of India (SOI) topographical maps of 1969 at 1:50,000scale were used for generating the base map of the Hokersarwetland. Time series of satellite data from various satelliteswas chosen for monitoring the spatial and temporal changes inthe wetland. In order to minimize the impacts of the changingseason on the mapping, it was ensured, wherever possible, touse the data of the same season with minimum possible gapsbetween them. Landsat TM (15 Oct 1992) with a spatialresolution of 30 m and Path/Row 149/36; Landsat ETM+ (30Sept 2001), with a spatial resolution of 30 m and Path/Row-149/36; IRS LISS-III (19 Oct 2005) with a spatial resolution of23.5 m and Path/Row-92/46 and IKONOS (11 Jan 2008) witha spatial resolution of 1 m were used. Though, all the satellitedata, except IKONOS, pertain to the autumn season, when thedischarge and water depth of the wetland is at the minimum,however, due to unavailability of the cloud-free satellite data inautumn, the January IKONOS data was used for monitoringthe changes in the wetland up to 2008.

Datasets used for changes in the catchment of Hokersar

Time series of satellite data (1972–2005) from various sat-ellites was used to analyze changes at catchment level.Landsat MSS (17 Nov 1972) with a spatial resolution of57 m and Path/Row 160/36; Landsat TM (15 Oct 1992) witha spatial resolution of 30 m and Path/Row 149/36; IRSLISS-III (19 Oct 2005) with a spatial resolution of 23.5 mand Path/Row-92/46, 92/47 were used.

Hydrometeorological data

A time series of the hydrometeorological data comprising ofprecipitation and river discharge data from 1979 to 2009 wasstatistically analyzed to investigate if there is any link betweenthe changing climate and the declining water extent of theHokersar wetland.

For accomplishing the research objectives, the multi-source and multitemporal satellite data was used at twospatial scales; at the catchment scale and the wetland scale.The flowchart of the methodology adopted in this researchis given in Fig. 2. In this research, we adopted twoapproaches for extracting the information from the images;(a) onscreen digitization of the image data to delineate the

Fig. 4 Land use and land cover types delineated from the scanned topographic map of 1969

Arab J Geosci

Author's personal copy

wetland boundaries and for mapping the land use and landcover within the wetland boundary and (b) digital imageclassification for extracting land use and land cover infor-mation at the catchment scale.

For delineating the wetland boundary from the topo-graphic map, onscreen digitization method was employed.Using the digitized boundary of the wetland from SOI mapas the base map, the wetland boundary extents at differentpoints in time were delineated from the satellite images. Theland use and land cover types within the wetland boundarywere also digitized from the scanned map and the satelliteimages in order to determine the spatiotemporal changesthat have occurred during the observation period (1969–2008). For extracting the land use and land cover informa-tion in the catchment of the Hokersar wetland, supervisedimage classification technique based on the maximum

likelihood classifier was used (Fu 1976; Tso and Mather2001). National Natural Resources Management System(NNRMS) standards (ISRO 2005) were used for cate-gorizing land use and land cover in the Doodhgangacatchment that drains into the Hokersar wetland. Whilechoosing various training samples for the maximumlikelihood classifier, homogeneity of the samples wasensured for achieving higher classification accuracy.The land use and land cover map of 2005 at thecatchment scale was validated in the field to determineits accuracy. Seventy-one sample points were chosen forverification of the land use and land cover map in thefield. The accuracy estimation is essential to assessreliability of the classified map (Foody 2002). Kappacoefficient, the robust indicator of the accuracy estima-tion for the final land use and land cover map, wasestimated by the following formula:

k ¼NPr

i¼1Xii �

Pr

i¼1Xiþ:Xþið Þ

N2 �Pr

i¼1Xiþ:Xþið Þ

Fig. 5 Land use and land cover types delineated from 1992 satellite data

Table 2 Land use andland cover types delin-eated from the scannedtopographic map of1969

Class name Area (km2)

Marshy 16.30

Open water 1.74

Plantation 0.64

Road 0.05

Arab J Geosci

Author's personal copy

where r is number of rows in error matrixxii is number of observations in row i and

column i (on the major diagonal)xi+ is total of observations in row i (shown as

marginal total to right of the matrix)x+i total of observations in column i (shown as

marginal total at bottom of the matrix)N is total number of observations included in

the matrix

In addition, the overall accuracy, user’s accuracy, pro-ducer’s accuracy, errors of omission and commission were

also computed to assess the accuracy of the land use andland cover at the catchment scale. The wetland boundaryand the land use and land cover types within the wetlandwere extensively verified on the ground with respect to2008 data only as the ground truth was not available forthe other time periods. In order to determine the changes inthe land use and land cover within the wetland, thathave occurred over the observation period from 1969to 2008, change detection analysis was performed (Bak-er et al. 2007; Schmid et al. 2005). Similarly, thechange detection analysis was also performed at thecatchment scale between 1972 and 2005.

Fig. 6 Spatial distribution of the land use and land cover data for the year 2001

Table 3 Area covered by different land use and land cover types from 1992 to 2008 within Hokersar wetland

Class name Area 1992 (km2) Area 2001 (km2) Area 2005 (km2) Area 2008 (km2) Change from 1992 to 2008 (km2) % Change

Agriculture 4.26 3.69 3.23 4.95 0.69 3.69

Aquatic vegetation 2.5 3.48 4.56 4.46 1.96 10.73

Built up 0.01 0.05 0.12 0.11 0.1 0.55

Fallow 0.88 0.21 0.27 0.48 −0.4 −2.22

Marshy 7.74 8.06 7.27 5.62 −2.12 −11.86

Open water 0.85 0.43 0.31 0.36 −0.49 −2.72

Plantation 1.82 2.18 2.32 2.16 0.34 1.83

Road 0.03 0.03 0.03 0.03 0 0

Arab J Geosci

Author's personal copy

Results and discussion

Wetland depletion

The wetland has shrunk and depleted over a period of time.During the observation period from 1969 to 2008, the spatialextents of wetland have reduced from 18.75 km2 in 1969 to13.00 km2. The extent of the wetland area at different points intime is given in Table 1. As is evident from the data, an area of5.75 km2 has been lost during the last four decades. Figure 3shows the thematic representation of the Hokersar boundaryextents at different points in time. There is progressive deple-tion of the wetland area from 1969 to 2008.

Land use and land cover changes within the wetland

To analyze and map the land use and land cover within thewetland area, onscreen digitization approach was adopted.Eight types of land use and land cover classes were delineatedfrom the satellite data (1992, 2001, 2005, and 2008) and thescanned topographical map (1969) at a 1:25,000 scale. Theland use and land cover types are agriculture, fallow, planta-tion, marshy lands, aquatic vegetation, built up, open water,and road. For delineating the land use and land cover types

from the images, image elements and other contextual infor-mation was used for improved accuracy. Figure 4 shows theland use and land cover types delineated from the scannedtopographic map that has symbols for these types. From theanalysis of Fig. 4 and Table 2, we can see that the area undermarshes was 16.30 km2, plantation was 0.64 km2, and openwater was 1.74 km2 (which includes flood channel 0.07 km2)out of the total area of 18.75 km2. There is no built up,agriculture, fallow, and aquatic vegetation category shownon the topographic map and hence these three categories aremissing in Fig. 4. It could be assumed that there was no built-up, agriculture and fallow within the wetland in 1969. How-ever, the marshy land category shown on the map may con-stitute some aquatic vegetation as well that has been shownunder marshy land.

From analysis of the 1992 data, as shown in Fig. 5 andTable 3, all the eight categories of the land use and landcover are present in the wetland. Marshy lands dominate thewetland area covering an area of 7.74 km2 that constitutes42.7 % of the wetland area. Agriculture, that was non-existent in 1969, is the second major land use type in thewetland covering about 23.51 % of the wetland area. Sim-ilarly, the built up has emerged within the wetland that wasnot present before 1969 and covers an area of 0.01 km2

Fig. 7 Distribution of the land use and land cover types mapped during the 2005

Arab J Geosci

Author's personal copy

(0.09 %). The area under the open waters has also drasticallyreduced in 1992 compared to the baseline data (1969). Theopen water body within the wetland has drastically reducedfrom 1.74 km2 in 1969 to 0.85 km2 in 1992.

Figure 6 shows the spatial distribution of the land use andland cover data within the wetland for the year 2001 mappedfrom the LANDSAT ETM+ data. From the analysis of thedata in Table 3, we observe that marshy lands are predom-inant in the wetland followed by agriculture. The area underbuilt up has increased from 0.01 km2 in 1992 to 0.05 km2 in2001. However, the open water body has shrunk by morethan a half from 0.85 to 0.43 km2.

The distribution of the land use and land cover typesmapped during the 2005 is shown in Fig. 7 and the propor-tionate spatial statistics are given in Table 3. From theanalysis of the data, it is observed that the area under theaquatic vegetation has significantly increased from 3.48 km2

in 2001 to 4.56 km2 in 2005. Similarly, the built-up isshowing an increase. Marshy lands that have tremendousecological importance for the migratory birds as they nestand breed in these areas are showing a decrease from 8.06 to7.27 km2 during the 2001–2005 period.

Figure 8 shows the areal distribution of the land use andland cover types delineated from 2008 IKONOS data. Table 3

gives the area estimates and the proportionate spatial statisticsfor each of the land use and land cover type observed withinthe wetland. From the analysis of the data, it is observed thatthe area estimates of the land use and land cover types derivedfrom 2008 high-resolution IKONOS data are not showingconsistent trend as observed from 1969 to 2005 except formarshy and aquatic vegetation categories. In fact, due to dif-ferent image acquisition date of 2008 data, i.e., January, whenthe water discharge is usually a bit higher than the autumnwhen it is at the minimum, there is increase in the water extentobserved from the 2008 data. However, compared to the areaestimates of the dominant land cover types observed in 1969,there are sharp changes in the open water body, marshy lands,aquatic vegetation, and built-up area observed in 2008.

Land use and land cover change in the catchment

In order to analyze the causes of this deterioration anddepletion of the Hokersar wetland, multitemporal land useand land cover of the Doodhganga catchment of the wet-land, spread over an area of 732.6 km2, was determinedusing the three date satellite data from 1972 to 2005. Thir-teen land use and land cover classes were delineated basedon NNRMS standards; agriculture, exposed rock, fallow,

Fig. 8 Spatial distribution of the land use and land cover types delineated from 2008 IKONOS data

Arab J Geosci

Author's personal copy

forests, orchards, pasture, plantation, river, riverbed, scrub,settlement, snow, and water body from all the satellite data.Figure 9 shows the thematic map of the land use and landcover types of 1972. It is observed from the information that

fallow was the dominant class covering 38.59 % area.Agriculture covered 20.6 % of the area followed by forest(11.37 %), snow (11 %) while as the area under orchardsand settlements was 2.79 and 0.63 %, respectively (Table 4).

Fig. 9 Spatial distribution ofland use and land cover types in1972

Table 4 Area under different land use and land cover classes from 1972 to 2005 of Doodhganga catchment

Class Name Area 1972 (km2) Area 1992 (km2) Area 2005 (km2) Change from 1972 to 2005 (km2) % Change

Agriculture 150.92 140.42 130.67 −20.25 −2.76

Exposed rocks 27.66 49.18 61.58 33.92 4.63

Fallow 282.71 246.51 200.16 −82.55 −11.27

Forest 83.33 78.44 55.78 −27.55 −3.76

Horticulture 20.46 70.88 99.44 78.98 10.78

Pasture 11.3 10.74 7.54 −3.76 −0.51

Plantation 29.71 31.22 55.3 25.59 3.49

River 6.96 5.93 5.93 −1.03 −0.14

River bed 12.42 9.65 9.65 −2.77 −0.38

Settlement 4.63 12.4 15.35 10.72 1.46

Scrub 20.13 40.6 49.99 29.86 4.08

Snow 80.62 36.03 40.7 −39.92 −5.45

Water body 1.75 0.6 0.51 −1.24 −0.17

Arab J Geosci

Author's personal copy

The land use and land cover map of 1992 is shown inFig. 10. From the analysis of the thematic and the tabulardata, it is observed that the area under agriculture andfallow, taken together, has marginally decreased between1972 and 1992 (Table 4). Similarly, the area under thepasture and plantation has almost remained static. The areasunder snow and forest have shown a decline. The built-uparea has significantly increased from 4.63 to 12.4 km2.Similarly, the land under orchards, scrub, and exposed rockhas shown marked increase in area.

Figure 11 shows the spatial distribution of the land useand land cover types in the catchment delineated from the2005 IRS LISS-III data. From the analysis of the data, it isevident that the agriculture and fallow, taken together, showa significant decrease in areal extent. Similar trend is ob-served in case of forest, which has decreased by about22.66 km2 compared to 1992 data. The area under settle-ments has increased to 15.35 km2 in comparison to 12.4 km2

in 1992 (Table 4). Plantation, orchards, and exposed rockare also showing increase in their spatial extents.

An accuracy assessment of the land use land cover typesderived from the supervised classification of the 2005 satel-lite data was also carried out. The accuracy of the land useland cover delineated from 2005 satellite data was 94.09 %(Table 5). The error of omission, i.e., probability of exclud-ing a pixel that should have been included in the class washighest for river bed (0.30) followed by exposed rock (0.11),plantation (0.272), scrub (0.087), horticulture (0.066), andfallow (0.076). Similarly, the error of commission which isthe probability of including a pixel in a class when it shouldhave been excluded was highest for river bed (0.125) fol-lowed by exposed rock (0.11), scrub (0.087), fallow (0.076),and pasture (0.071).

Kappa is lower than overall accuracy and differences inthese two measures are to be expected in that each incorpo-rates different forms of information from the error matrix.While overall accuracy only includes the data along themajor diagonal and excludes the errors of omission andcommission, kappa incorporates the non diagonal elementsof error matrix as a product of row and column marginal.

Fig. 10 Spatial distribution ofland use and land cover types in1992

Arab J Geosci

Author's personal copy

Kappa coefficient for the classified data of 2005 was foundto be 0.935.

Hydrometeorological data analysis

A time series of the hydrometeorological data comprising oftemperature, precipitation, and river discharge data from1979 to 2009 was analyzed to investigate if there is any linkbetween these parameters and the declining water extent ofthe Hokersar wetland. Temperature data shows an increas-ing trend Fig. 12a with r2 of 0.08. Lowest temperatures wasrecorded in 1991 (10.63 °C) while as highest temperaturewas recorded in 2001 (14.73 °C). The last decade (2000–2009) was the hottest decade with an average temperature of13.75 °C as compared to average of 12.74 °C from 1979 to1999. The precipitation data shows a declining trend, eventhough weak, as seen from the Fig. 12b with r2 of 0.14.Highest precipitation of 943 mm has been recorded in 1983while 2000 recorded the lowest precipitation of 423 mm.Relatively low precipitation has been recorded in late 2000sas compared to that in early 1980s. Similarly, the analysis ofthe time series of the discharge data of the Doodhganga

tributary from 1970 to 2009, the main feeder tributary ofthe wetland, at both the head (Branwar) and tail (Barzulla)indicate decreasing tendency of the river discharge with r2

of 0.26 and 0.15 respectively (Fig. 13a, b). The lowering ofwater discharge may be attributed to the depleting snowcover and reduction in annual precipitation in Doodhgangacatchment. The decreasing extent of water spread and depthof the Hokersar could partly be attributed to the changingclimate in the Himalayan region (Akhtar et al. 2008; Dahal2005; ICIMOD 2009). The decreasing trend in precipitationand discharge of Doodhganaga stream has a direct bearingon the changing land use land cover in Doodhganga catch-ment. Particularly, agriculture lands are being converted toapple orchards as the latter require less amount of water andhence are climatologically more viable.

The depletion in the wetland extent are mainly attributedto the encroachment by the farmers, increase in the settle-ments, conversion of wetland area into agriculture, planta-tion and built-up, and climate change (Joshi et al. 2002;Kraiem 2002). From the data, it is evident that the openwater extent in the wetland has receded from 1.74 km2 in1969 to 0.31 km2 in 2005 (Fig. 14a). However, the 2008

Fig. 11 Distribution of the landuse and land cover typesdelineated from 2005 IRSLISS-III data

Arab J Geosci

Author's personal copy

high-resolution IKONOS data shows an increase of0.05 km2 in the water extent with respect to the 2005 datamainly due to its acquisition in winter when the dischargefrom the feeder stream is higher compared to the autumnseason when all other images, used in the spatiotemporalanalysis, were acquired. Similarly, the marshy lands thathave tremendous ecological importance for the migratorybirds, serving as the nesting and breeding grounds, haveshowed consistent decline from 16.3 km2 in 1969 to5.62 km2 in 2008. The marshy land was the predominantland cover in 1969 covering more than 85 % of the wetlandarea. The depletion of the marshy lands within the wetlandhas adversely affected the breeding patterns of the migra-tory birds. The emergence of the built-up areas within thewetland and its immediate surroundings has also responsi-ble for the destruction of the wetland ecology and func-tionality (Fig. 14b). The built up, that was non-existentwithin the wetland in 1969, has emerged and has colonizedalmost 0.11 km2 of the wetland in 2008. Due to theseencroachments and human settlements, the agriculture andolericulture activity has got a boost within the wetlandboundary and large areas of the wetland, spread over anarea of 5.40 km2, have come under agriculture/fallow since1969. All these anthropogenic influences within the wet-land have accelerated the deterioration of the wetland

structure and functions. There are umpteen studies thathave demonstrated the adverse impacts of the human influ-ences on the wetlands all over the world (UNEP 2007)

Further, the land use and land cover dynamics in thecatchment of the wetland have profound impacts on thefunctionality and health of the wetland. From the spatialand temporal analysis of the land use and land cover in thecatchment, it is observed that there have been significantchanges from 1972 to 2005. There is marked increase inthe horticulture, plantation, scrub, settlements, and exposedrock while as the area under agriculture, fallow, forest,pasture, and water resources has decreased. Settlementshave increased about four times from 4.63 km2 in 1972to 15.35 km2 in 2005. Similarly, horticulture and plantationshow a significant increase in area from 1972 to 2005.Permanent snow cover has decreased by about 40 km2

resulting in increase in the area of exposed rock. Areaunder agriculture and fallow has decreased by about100 km2 responsible for increase in spatial extent of horti-culture and plantation. Similarly, forest and pasture areashave been transformed into scrub because of deforestationfrom the past 33 years. These changes observed in thecatchment have adverse impacts on the wetland ecologyand hydrology. The impacts of these changes in the catch-ment and those in the vicinity of the wetland are reflected

Table 5 Accuracy assessment of land use land cover delineated from 2005 satellite data

Reference data

AG ER FA FO HO PA PL RI RB SE SC SN WB Rowtotal

User’saccuracy(%)

Classification data

AG 19a 0 0 0 1 0 0 0 0 0 0 0 0 20 95.00

ER 0 8a 1 0 0 0 0 0 0 0 0 0 0 9 88.89

FA 0 0 12a 0 0 0 0 0 1 0 0 0 0 13 92.31

FO 0 0 0 32a 0 0 2 0 0 0 0 0 0 34 94.12

HO 0 0 0 0 14a 0 1 0 0 0 0 0 0 15 93.33

PA 0 0 0 0 0 13a 0 0 0 0 1 0 0 14 92.86

PL 0 0 0 0 0 0 8a 0 0 0 0 0 0 8 100.00

RI 0 0 0 0 0 0 0 12a 0 0 0 0 0 12 100.00

RB 0 0 0 0 0 0 0 0 7a 0 1 0 0 8 87.50

SE 0 0 0 0 0 0 0 0 1 16a 0 0 0 17 94.12

SC 0 1 0 0 0 0 0 0 1 0 21a 0 0 23 91.30

SN 0 0 0 0 0 0 0 0 0 0 0 6a 0 6 100.00

WB 0 0 0 0 0 0 0 0 0 0 0 0 7a 7 100.00

Column total 19 9 13 32 15 13 11 12 10 16 23 6 7 186

Producer’saccuracy (%)

100.00 88.89 92.31 100.00 93.33 100.00 72.73 100.00 70.00 100.00 91.30 100.00 100.00

a Overall accuracy = [(19 + 8 + 12 + 32 + 14 + 13 + 8 + 12 + 7 + 16 + 21 + 6 + 7)/186] × 100 = 94.09 %

AG agriculture, ER exposed rock, FA fallow, FO forest, HO horticulture, PA pasture, PL plantation, RI river, RB river bed, SE settlement, SC scrub,SN snow, WB waterbody

Arab J Geosci

Author's personal copy

in the form of changes in the marshy and aquatic vegeta-tion within the wetland. Increase in spatial extent of ex-posed surfaces is responsible for enhanced silt load inDoodhganga stream which finds its entry into the Hokersarwetland thereby decreasing the depth and water retentioncapacity. The changes in the composition and distributionof marshy land and aquatic vegetation are showing adverseimpacts on the migratory birds, hydrobiology, and hydro-chemistry of the wetland (DEARS 2001; Khan 2000;Pandit and Kumar 2006).

Further, the symptoms of the wetland deterioration areattributed to the reckless use of fertilizers and pesticidesfor agriculture and horticulture in the catchment, whichultimately find their way into wetland through DudhgangaRiver. This fact has been substantiated by the physico-chemical characteristics of the wetland as reported byPandit and Kumar (2006). The analysis shows an increasefor nitrate and ammonical nitrogen from 1978 to 2002.Due to the increase of these nutrients, the ecology of thewetland is changing and adversely affecting the aquatic

flora and fauna. This nutrient enrichment boosts thegrowth of aquatic vegetation (Fig. 14c) found in thewetland like Nymphoides peltatum, Myriophyllum verticil-latum, Trapa natans, Typha angustata, and Phragmitesaustralis (Dar et al. 2002).

Within the wetland, various changes have been ob-served in the composition and distribution of the aquaticvegetations. Some macrophytes like Nelumbo nucifera,Euryale ferox, and Acorus calamus have disappearedand some new species have been observed like Meny-nanthese trifoliate (Kaul and Zutshi 1967). The mainreason for the disappearance of these macrophytes isattributed to the increase of silt load to the wetlandbrought from the catchment by Doodhganga Nallah. Anincrease in the number of macrophytic species from 24(Pandit 1980) to 46 (Pandit and Kumar 2006) has beenreported. A possible reason for this may be the improve-ment in the flood situation following dry weather con-ditions leading mostly to summer draw-down during therecent years (Pandit and Kumar 2006). As a result of this

Fig. 12 a Graph showingaverage annual temperaturefrom 1979 to 2009. b Graphshowing total annualprecipitation from 1979 to 2009

Arab J Geosci

Author's personal copy

increased aquatic vegetation, a drop in the oxygen con-tent has been observed causing eutrophication which hasdirect effects on aquatic fauna like fishes. Vegetablegardens and paddy fields, that have come up in thevicinity of the wetland since 1969, have also increasedthe nutrient loading to the wetland as the use of fertil-izers and irrigation in large quantities is practiced inthese vegetation gardens and paddy fields.

The massive deforestation in the upper reaches of thecatchment has increased the silt load to the downstreamwater bodies including Hokersar wetland. Due to the in-creased siltation, the predominant land cover type in thewetland, the marshy lands, has fragmented and is replacedby several land use and land cover classes, particularlyaquatic vegetation. Excess load of siltation has also adverse-ly affected the depth of the wetland which was 1.12 m(Pandit 1980) and has reduced to 0.63 m only (Rather andPandit 2002). Currently, the depth of the water has furtherreduced resulting in decrease in the water spread.

The time series analysis of the precipitation and dis-charge data of the Doodhganga catchment from 1979 to

2009 shows a declining trend. This means that the re-duction of the water inflow to the wetland, as a conse-quence of climate change, is responsible for the reductionin the depth and spread of the water level of the wetland.Wetland biodiversity, ecosystem, and services are indeedunder threat from the impacts of the climate change butproper management of the wetlands can reduce theseimpacts (O’Reilly et al. 2003; Verburg et al. 2003). Thedrastic reduction in the area of marshy lands, the waterdepth, and the water spread, have changed the ecologicalconditions within the lake and thus, adversely affectedthe arrival of migratory birds, as less number of waterfowl has been reported since the past few years. Analysisof time series temperature data of Doodhganga catchmentfrom 1979 to 2009 showed an increasing trend. Theincreasing temperatures are possibly cause for bloom ofalien aquatic invasive tropical water fern Azolla sp. inHokersar wetland which causes decrease in light penetra-tion, dissolved oxygen content of Hokersar wetland be-sides competing with the macrophytic species within thewetland (Uheda et al. 1999). Dissolved oxygen shows an

Fig. 13 a Graph showingdischarge of Doodhgangastream at head (Branwar). bGraph showing discharge ofDoodhganga stream at Tail(Barzulla)

Arab J Geosci

Author's personal copy

inverse relationship with temperature as per Henry’s Law(IUPAC 1997, b). Hence, the increasing temperatures arealso responsible for reduction in dissolved oxygen con-tent in the Hokersar wetland. Further, the direct dischargeof the effluents and sewage from the surrounding areasinto the wetland because of increase in spatial extent ofsettlements has increased the nutrient loading of the

wetland. The construction of the network of roads aroundthe wetland and the proliferation of the willow planta-tions within the wetland has hampered the natural flowof the drainage adversely affected the wetland hydrologyand the environmental flows. Willows have been recog-nized as a serious threat to wetland as they cause a rangeof deleterious morphological and ecological changes towetlands and aquatic ecosystems (Poppe et al. 2006).

Conclusion

Various land use and land cover changes within thewetland and its catchment have tremendous ecologicaland socio-economic importance and it aptly depicts theway people are treating the wetland ecosystems in themountainous Himalayan region. The water quality of thewetland has deteriorated and changes in the vegetationcomposition and distribution have been very significantlyaffecting the biodiversity of the wetland. The wetlanddepletion has serious implications not only on our floraand fauna but also on livelihood of the people dependenton the service and goods provided by the wetland. Thedepletion and degradation of this wetland shall haveadverse impacts on the efficacy of the wetland in retain-ing flood waters during peak discharge and flash floodsand thus endanger the lives and property of the Srinagarcity dwellers. The degradation of the marshy habitat ofthe millions of the migratory birds from Siberia andCentral Asia has affected the arrival of these birds asnoticed by their less numbers in the recent years. De-creasing trend in the precipitation has a direct bearing onland use land cover dynamics in Doodhganga catchment.Agriculture lands are getting converted in orchards main-ly because less amount of water is required in the lattercase. Increase in temperature causes interference in thehatching of eggs of birds besides disturbing the speciescomposition of natural vegetation. From the analysis anddiscussion of the results, it is thus concluded that themain reasons for the deterioration of the Hokersar wet-land are increase in the nutrient and silt load from thecatchment due to deforestation and reckless use of pesti-cides and fertilizers, encroachment, unplanned urbaniza-tion in the vicinity of the wetland, and the decreaseddischarge to climate change observed in the region. It issuggested that an appropriate mechanism is establishedfor continuous monitoring of the wetland, its immediatesurrounding and the catchment for land system changes,hydrochemistry, biodiversity, and wetland hydrology sothat a robust strategy and action plan is developed forthe conservation and restoration of this important wet-land, commonly called the Queen of Wetlands inKashmir Himalayas.

Fig. 14 a Loss in water spread due to ingress of silt from Doodhgangacatchment and partly because of the encroachment. b Built-up areascoming-up just around the main wetland body as a consequence offaulty regulatory framework. c Reckless growth of macrophytes withinthe main wetland body as a consequence of urbanization and agricul-tural practices around Hokersar

Arab J Geosci

Author's personal copy

References

Akhtar M, Ahmad N, Booij MJ (2008) The impact of climate change onthe water resources of Hindukush–Karakorum–Himalaya regionunder different glacier coverage scenarios. J Hydrol 355:148–163

Aldous A, Fitzsimons J, Richter B, Bach L (2011) Droughts, floodsand freshwater ecosystems: evaluating climate change impactsand developing adaptation strategies. Mar Freshw Res 62(3):2230–2231. doi:10.1071/MF09285

Anonymous (1990) Directory of wetlands in India. Ministry of Envi-ronment and Forests, Government of India, New Delhi, p 52

Baker C, Lawerence R, Montagne C, Pattern D (2007) Change detec-tion of wetland ecosystem using Landsat imagery and changevector analysis. Wetlands 27(3):610–619

Basnyat P, Teeter LD, Lockaby GG, Flynn KM (2000) The use ofremote sensing and GIs in watershed level analyses of non-pointsource pollution problems. For Ecol Manag 128:65–73

Birkett CM (1995) The global remote sensing of lakes, wetlands andrivers for hydrological and climate research. Proceedings of IEEEIGARSS Conference, Firenze, pp. 1979–1981

Bourgeau-Chavez LL, Kasischke ES, Brunzell SM, Mudd JP, SmithKB, Frick AL (2001) Analysis of space-borne SAR data forwetland mapping in Virginia riparian ecosystems. Int J RemoteSens 22(18):3665–3687

Bullock A, Acreman M (2003) The role of wetlands in the hydrologicalcycle. Hydrol Earth Syst Sci 7(3):358–389

Costanza R, D’Arge R, de Groot R, Farber S, Grasso M, Hannon B,Limburg K, Naeem S, O’Neill RV, Paruelo J, Rsakin RG, SuttonP, Van den Belt M (1997) The value of the world’s ecosystemservices and natural capital. Nature 387:253–260

Dahal N (2005) Perceptions in the Himalayas. Tiempo 56:19–24Dar GH, Bhagat RC, Khan MA (2002) Biodiversity of Kashmir

Himalayas. Valley, Srinagar, p 399Davis TJ (1993) Towards the wise use of wetlands. Ramsar Conven-

tion Bureau, Kuala Lampur, p 45DEARS (2001) Land use/land cover mapping of Jammu and Kashmir

State—a report Directorate of Environment, Ecology and RemoteSensing, Government of Jammu and Kashmir, p 92

Donner S (2003) The impact of cropland cover on river nutrient levels inthe Mississippi River Basin. Glob Ecol Biogeogr 12(4):341–355

Foody GM (2002) Status of land cover classification accuracy assess-ment. Remote Sens Environ 80(1):185–201

Fu KS (1976) Pattern recognition in remote sensing of the Earthresources. IEEE Trans Geosci Electron 14(1):10–18

Funk DE, Pullman E, Peterson K, Crill P, Billings W (1994) Influenceof water table on carbon dioxide, carbon monoxide and methaneflux from taiga bog microcosms. Glob Biogeochem Cycles 8(3):271–278

Gangoo SA, Makaya AS (2000) Changes in vegetation pattern ofHokersar (Wetland reserve), Kashmir. In: Environmental Biodi-versity and Conservation, p. 170–112

Garg JK, Singh TS, Murthy TVR (1998) Wetlands of India. Projectreport: RSAM/SAC/resa/pr/01/98. Space Application Centre, In-dian Space Research Organization (ISRO), Ahmadabad, p 240

Gillies RR, Box JB, Symanzik J, Rodemaker EJ (2003) Effects ofurbanization on the aquatic fauna of the Line Creek Watershed,Atlanta—a satellite perspective. Remote Sens Environ 86(3):411–422

Gondwe BRN, Sang-Hoon H, Wdowinski S, Bauer-Gottwein P (2010)Hydrologic dynamics of the ground-water dependent Sian Ka’anwetlands, Mexico, derived from InSAR and SAR data. Wetlands30:1–13

Handoo JK (1978) Ecological and production studies of some typicalwetlands of Kashmir. Ph.D Thesis, University of Kashmir, Srina-gar, 141 p

Handoo JK, Kaul V (1982) Phytosociological and standing crop stud-ies in wetlands of Kashmir. In: Gopal B, Turner RE, Wetzel RG,Whigam DF (eds) Wetlands: ecology and management, part1.National Institute of Ecology and International Scientific Publi-cation, Jaipur, pp 187–197

Hess LL, Melack JM, Filoso S, Wang Y (1995) Delineation of inun-dated area and vegetation along the Amazon floodplain with SIR-C synthetic aperture radar. IEEE Trans Geosci Remote Sens 33(4):896–904

Hruby T (1995) Estimating relative wetland values for regional plan-ning. Wetlands 15(2):93–107

Humayun R, Joshi PK (2000) Evaluation of waterfowl habitat in Hoker-sar Wetland Reserve, Jammu&Kashmir: a geospatial approach. PGDiploma Dissertation Report, IIRS, Dehradun, India, p 37

ICIMOD (2009) The changing Himalayas: impact of climate changeon water resources and livelihoods in the Greater Himalayas.International Centre for Integrated Mountain Development, Kath-mandu, Nepal, p 25

International Union for Pure and Applied Chemistry (IUPAC) (1997)Henry’s Law: IUPAC Compendium of Chemical Terminology.2nd Edition

International Union for Pure and Applied Chemistry (IUPAC) (1997)Compendium of Chemical Terminology, 2nd ed. Online version:http://goldbook.iupac.org/H02783.html. Accessed on: 24th March

ISRO (2005) NNRMS standards: a national standard for EO images,thematic and cartographic maps, GIS databases and spatial out-puts. ISRO NNRMS Tech Rep No 112:235pp

Joshi PK, Humayun R, Roy PS (2002) Landscape dynamics in Hokersarwetland—an application of geospatial approach. J Indian Soc Re-mote Sens 30(1&2):2002

Kak AM (1990) Aquatic and wetland vegetation of Kashmir Himalaya.J Econ Taxon Bot 14(1):1–14

Kapetsky JM (1987) Satellite remote sensing to locate and inventorysmall water bodies for fishing management and aquaculture devel-opment in Zimbabwe, CIFA Occasional Paper, no. 14. Rome: FAO(Fisheries and Aquaculture Department). http://www.fao.org/docrep/008/ad768e/ad768e00.htm. Accessed: 1 March 2012

Kaul S (1982) Community architecture, biomass and production insome typical wetlands of Kashmir. Indian J Ecol 9:320–329

Kaul V, Zutshi DP (1967) A study of aquatic and marshland vegetationof Srinagar lakes. Proc Natl Inst Sci India 33B:111–128

Khan MA (2000) Wetland biodiversity in the Kashmir Himalaya:assessment and conservation strategies. In: Khan MA (ed) Envi-ronmental biodiversity and conservation. APH Publishing, NewDelhi, pp 69–93

Kingsford RT (2011) Conservation management of rivers and wetlandsunder climate change—a synthesis. Mar Freshw Res 62(3):217–222

Klausmeyer KR, Shaw MR (2009) Climate change, habitat loss, pro-tected areas and the climate adaptation potential of species inMediterranean ecosystems worldwide. PLoS One 4(7):e6392.doi:10.1371/journal.pone.0006392

Kraiem H (2002) Biophysical and socio-economic impacts of climatechange on wetlands in Mediterranean. Proceedings of the Medi-terranean Regional Workshop on Water, Wetlands and ClimateChange: Building Linkages for their Integrated Management,Athens, Greece, 10–11December, 2002

Lyon J (2001) Wetland landscape characterization: GIS, remote sens-ing and image analysis. Ann Arbor Press, Chelsea, p 160

Millennium Ecosystem Assessment (MEA) (2005) Ecosystems andhuman wellbeing: wetlands and water synthesis. World ResourcesInstitute, Washington, p 80

Mitsch WI, Gosselink IG (1986) Wetlands. Van Nostrand Reinhold,New York

Munyati C (2000) Wetland change detection on the Kafue Flats,Zambia, by classification of a multi-temporal remote sensingimage dataset. Int J Remote Sens 21(9):1787–1806

Arab J Geosci

Author's personal copy

O’Reilly CM, Alin SR, Pilsnier PD, Cohen AS, McKee BA (2003)Climate change decreases aquatic ecosystem productivity of LakeTanganyika, Africa. Nature 424:766–768

Olmanson LG, Bauer ME, Brezonik PL (2002) Aquatic vegetationsurveys using high resolution IKONOS imagery. Proceedings ofthe ISPRS Commission-IV Symposium Integrated Remote Sens-ing at the Global, Regional and Local Scales, 10–15 Nov 2002,Denver, CO USA

Omernik JM, Abernathy AR, Male LM (1981) Stream nutrient levelsand proximity of agricultural and forest lands to streams: somerelationships. J Soil Water Conserv 36:227–231

Ozesmi SL, Bauer ME (2002) Satellite remote sensing of wetlands.Wetl Ecol Manag 10(5):381–402

Pal S, Akoma OC (2009) Water scarcity in wetland area within KandiBlock of West Bengal: a hydro-ecological assessment. Ethiop JEnviron Stud Manag 2(3):1–12

Palmer MA, Lettenmaier DP, Poff NL, Postel SL, Richter B, Warner R(2009) Climate change and river ecosystems: protection and ad-aptation options. Environ Manag 44(6):1053–1068. doi:10.1007/S00267-009-9329-1

Pandit AK (1980) Biotic factor and food chain structure in sometypical wetlands of Kashmir. Ph.D. Thesis. University of Kash-mir, Srinagar, J&K

Pandit AK, Kumar R (2006) Comparative studies on ecology ofHokersar wetland, Kashmir: present and past. J Himal Ecol Sus-tain Dev 1:73–81

Poppe L, Rutherford I, Price P, Lovett S (2006) River and riparian landmanagement: controlling willows along Australian rivers. Tech-nical guideline No. 6, Land and Water Australia. Australian Govt,Canberra, p 17

Ramsey EW (1998) Radar remote sensing of wetlands. In: Lunetta RS,Elvidge CD (eds) Remote sensing change detection: environmen-tal monitoring methods and applications. Ann Arbor Press, Chel-sea, p 318

Rashid M, Lone MA, Romshoo SA (2011) Geospatial tools for assess-ing land degradation in Budgam district, Kashmir Himalaya. JEarth Syst Sci 120(3):423–434

Rather SA, Pandit AK (2002) Phytoplankton dynamics in Hokersarwetland, Kashmir. J Res Dev 2:25–46

Rather SA, Bhat SA, Pandit AK (2001) Water quality of Hokersar, atypical wetland of Kashmir. J Res Dev 1:38–43

Roeck ER, Verhoest NEC, Miya MH, Lievens H, Batelaan O, ThomasA, Brendonck L (2008) Remote sensing and wetland ecology: aSouth African case study. Sensors 8(5):3542–3556

Romshoo SA (2004) Radar remote sensing for monitoring of dynamicecosystem processes related to biogeochemical exchanges in trop-ical peatlands. Vis Geosci 9(1):9–28

Romshoo SA, Muslim M (2011) Geospatial modelling for assessingthe nutrient load of a Himalayan lake. Environ Earth Sci.doi:10.1007/s12665-011-0944-9

Romshoo SA, Sumira J (2010) Geospatial tools for watershedcharacterization of the Dudhganga catchment, Jhelum basin.PG Diploma Dissertation (RS &GIS), Department of Geolo-gy and Geophysics, University of Kashmir, Srinagar, India,123p

Romshoo SA, Ali N, Rashid I (2011) Geoinformatics for characterizing andunderstanding the spatio-temporal dynamics (1969–2008) of Hokarserwetland in Kashmir Himalayas. Int J Phys Sci 6(5):1026–1038

Romshoo SA, Qadri T, Rashid I, Muslim M, Panigrahy S, Singh TS,Patel JG (2010) National Wetland Atlas: Jammu and Kashmir.SAC/RESA/AFEG/NWIA/ATLAS/16/2010. Space ApplicationsCentre, ISRO, Ahmadabad, p 176

Saxena RK, Verma KS, Chary GR, Srivastava R, Barthwal AK (2000)IRC-IC data application in watershed characterization and man-agement. Int J Remote Sens 21(17):3197–3208

Schmid T, Koch M, Gumuzzio J (2005) Multisensor approach to deter-mine changes of wetland characteristics in semiarid environments(Central Spain). IEEE Trans Geosci Remote Sens 43(11):2516–2525

Tanis FJ, Bourgeau-Chavez LL, Dobson MC (1994) Applications ofERS-1 SAR for coastal inundation. Proceedings of the InternationalGeoscience and Remote Sensing Symposium (IGARSS’94), Pasa-dena, California, 8–12 August 1994, IEEE Catalog no. 94CH3378-7, pp. 1481–1483

Touzi R, Deschamps A, Rother G (2007) Wetland characterizationusing polarimetric RADARSAT-2 capability. Can J Remote Sens33(1):56–67

Tso B, Mather PM (2001) Classification methods for remotely senseddata. Taylor and Francis, UK, pp 186–229

Uheda E, Kitoh S, Shiomi N (1999) Response of six Azolla species totransient high-temperature stress. Aquat Bot 64:87–92

UNEP (2007) Global environmental outlook 2007, Geo-4. ProgressPress Ltd, Valletta, p 540

Verburg P, Hecky RE, Kling H (2003) Ecological consequences of acentury of warming in Lake Tanganyika. Science 301:505–507

Viers JH, Rheinheimer DE (2011) Freshwater conservation options fora changing climate in California’s Sierra Nevada. Mar Freshw Res62(3):266–278. doi:10.1071/MF09286

Vorosmarty CJ, McIntyre PB, Gessner MO, Dudgeon D, Prusevich A,Green P, Glidden S, Bunn SE, Sullivan CA, Liermann CR, DaviesPM (2010) Global threats to human water security and riverbiodiversity. Nature 467:555–561. doi:10.1038/nature09440

Wetlands International (2007) The comprehensive management actionplan on Wular lake, Kashmir. Wetlands International, South Asiafinal report, New Delhi, India, pp. 221

Arab J Geosci

Author's personal copy