Print this article - NepJOL

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc ASSESSMENT AND MONITORING OF SPATIO-TEMPORAL CHANGES IN KEOLADEO GHANA NATIONAL PARK, RAJASTHAN, INDIA USING GEOINFORMATICS C Sudhakar Reddy1, P Hari Krishna1, V Swetha1, T Shivakala1 and Prabhu D Charan2 Address: 1Forestry & Ecology Division, National Remote Sensing Centre, Hyderabad - 500 625, India 2Department of Environmental Sciences, M.D.S. University, Ajmer - 305 001, India *Corresponding Address: [email protected]

Abstract Knowledge of changes in wetlands is becoming a very important issue for sustainable management of natural resources. The present work has been carried out: in Keoladeo Ghana National Park (KNP), a RAMSAR site to study the dynamics of wetlands with reference to longterm and seasonal changes in extent of water spread, vegetation types and to build a spatial database on wetlands. Multi-temporal satellite data of Landsat MSS (1977), Landsat TM data (1989), IRS P6 AWIFS (2005) and IRS P6 LISS III data (2005-2006) was used for mapping and spatial analysis. Comparisons between 1977 and 2005 indicated that changes in overall wetland areas were significant. The total area of wetland in March 1977 was 186 hectares and in March 2005 it was found to be 7.5 ha. There is a drastic reduction in spatial extent of water spread from October to May and results indicated that the total area ranged from 528 ha (October 2005) to 7.5 ha (March-May 2005). The study clearly points that Geoinformatics have become very effective tool for wetlands mapping and monitoring. Key words: wetlands, geoinformatics, satellite data, Keoladeo, Rajasthan INTRODUCTION Geoinformatics is the science and technology for collection, management, analysis and presentation of geographic and other spatially defined data. Remote sensing (RS) and Geographic Information System (GIS) are important tools of Geoinformatics. Knowledge of land cover and land use is important for many planning and management activities (Roy et al 1985). Satellite remote sensing has emerged as one of the powerful technologies for generation of spatial information. Knowledge of changes in land cover is becoming far more important from both ecological and economical point of view. The studies on land cover changes would provide information to understand conservation issues (Reddy et al 2009).

33

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc Wetlands are among the highly productive ecosystems and therefore need to be conserved for quality of environment and existence of biodiversity. The RAMSAR Convention (1971) has defined wetlands as areas of marsh, fern, peat land or water, whether natural or artificial, permanent or temporary, with water that is static or flowing, fresh, brackish or salt, including areas of marine water the depth of which at low tide does not exceed six meters". The Ramsar Convention itself stresses the need for regular monitoring to detect actual or potential change in the ecological character of listed sites. Long-term monitoring of wetlands is an essential element of management for ‘wise use’ (Bennun 2001). The extent of water spread and wetland area is dependent on many parameters like topography, soil, climate, rainfall, etc. In different seasons, the extent of water spread changes and accordingly its usage and functions. The techniques of geoinformatics are very useful for monitoring such land cover changes. The five different established methods are manual overlay, GIS overlay, manual change interpretation, digital change interpretation and multi-temporal image interpretation. The study has been carried out to identify and delineate spatial extent of water spread through visual analysis of post and pre-monsoon season’s satellite data, to identify and delineate aquatic and other vegetation in and around wetland, to build a database of wetlands in the Keoladeo Ghana National Park (KNP). The wetland of Keoladeo Ghana National Park has been studied by considering the changes in the ecosystem. STUDY AREA Keoladeo Ghana National Park (270 07' to 270 12 N to 770 29' to 770 33' E) popularly known as Bharatpur Bird Sanctuary is a unique wetland listed under both the RAMSAR Convention (since 1981) and World Heritage Convention (since 1984). Keoladeo Ghana was established as a National Park (KNP) in 1982, having been designated as a Bird Sanctuary since March 1956. It is not only a bird paradise but is one of the best waterfowl reserves in the world for both resident as well as migratory birds (All & Vijayan 1986). This KNP is one of the three places in the world, known to harbor wintering Siberian cranes (Grus leucogeranus). The population of the Siberian Cranes visiting the KNP declined from around 200 to 2 during the last four decades. They prefer an area with water depth ranging from 0 to 60 cm while the preferred depth of the feeding sites being 10 to 40 cm. Siberian Cranes are omnivorous, although they feed mostly on vegetable matter. The Sarus Crane (Grus antigone) is a resident species (Bhupathy et al. 1998). A unique feature of the wetland ecosystem of KNP is its origin from a natural shallow depression, which was a temporary rainfed wetland situated at the western edge of the Ganges plain. Subsequently, the construction of Ajan Bandh and several sluice gates in the periphery of the KNP facilitated to contain and regulate the water level. Regular flooding and flushing of the wetlands is the only way to manage them. Today KNP is facing a huge shortage of water. With a growing 34

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc shortage of water and feed, the birds no longer find the Park suitable. The KNP requires about 550 million cubic feet water for maintenance of wetland ecosystem. In recent years, only 300 million cubic feet of water flow is possible due to poor rain, which has resulted in the drying of the woodland and wetland flora. The Rajasthan Government has proposed a dedicated pipeline scheme for bringing water from the Chambal river (Pattanaik et al. 2008). KNP has dry tropical monsoon climate. In recent years, the annual rainfall has fluctuated between 270 mm and 662 mm. The mean minimum temperature (January) is 6°C, and the mean maximum (May) 48°C. Research studies on KNP was reviewed by Mathur et al. (2009) The vegetation is scrub and grassland, creating a mosaic of woodland types and open grassland surrounding the lagoons. The majority of the woodland is composed of medium-sized trees (less than 20m) and shrubs. Fruiting trees such as mangos (Mangifera indica) and the figs (Ficus bengalensis and F. religiosa) have been allowed to form mature shady groves in certain locations, or occur as isolates among the more typical thorn scrub. Some of the characteristic woody plant species include Acacia nilotica, Prosopis juliflora, Zizyphus mauritiana, Syzygium cuminii, Phoenix sylvestrix, Salvadora persica, Salvadora oleoides. Grass species include Vetiveria zizanoides, Desmostachia bipinnata, Paspalum distichum, Cynodon dactylon, Chloris dolichostachia, Cenchrus spp. and Aristida spp. (Saxena 1975). The uncontrolled expansion of the Paspalum distichum, Eleocharis plantagenia, Typha angustata and Vetiveria zizanoides communities has caused a considerable change in the whole character of the park. METHODOLOGY The satellite data of IRS-P6 LISS III, Path/Row 96/52, and date of pass 15th Oct, 2005, 25th April 2006 and IRS P6 AWIFS data Path/Row 96/52, January, 2005; February, 2005; March, 2005; April 2005 and May 2005 of the study area were used. IRS Satellite data products were procured from National Remote Sensing Centre, Hyderabad. The orthorectified Landsat MSS (path/row: 157/47 of 8th March 1977), Landsat TM data (path/row: 146/41 of 18th October 1989) (http://glcf.umiacs.umd.edu/) was used as reference for georectification and analysis. Satellite data was chosen based on the cloud free availability and seasonality. The Survey of India (SOI) top sheets (54E/8, 54E/12) at 1:50,000 scale were used as ancillary data. The study area was extracted from satellite images using the image extraction program available in ERDAS IMAGINE. Landsat-MSS, Landsat-TM, IRS P6 LISS III, IRS P6 AWiFS images were used to locate and quantify the changes (Fig.1.). The false color composite images were prepared using near infrared as red, red as green, green as blue band combinations. Wetland maps for the years 1977, 1989 and 2005 were prepared using temporal satellite data. The on-screen visual image interpretation method was followed to map land cover features on satellite data. Comparisons were made between classified data using matrix. Wetlands at two dates were compared by creation of an n x n matrix, where n was equal to the number of classes in the study area. In this case three classes were used for each category, (water body, vegetation, non vegetation). These digital 35

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc analyses allowed exact location of wetland increases and decreases. RESULTS AND DISCUSSION Spatial extent of Vegetation and Wetlands The study revealed that 80.9% of the park area was covered by vegetation in October 2005 (Fig.2). The moist savannah type predominate among the vegetation classes and occupied 1107.1 ha (37.4%) while the scrub-woodland, grasslands and aquatic vegetation covered by 23.1%, 12.3% and 8% respectively. The extent of water bodies in October, 1989 was estimated as 101.5 ha clearly indicate very deficit area as compared to October, 2005 (Table 1). The areal extent of total wetlands in the KNP area has significantly differed from October, 2005 to May, 2005 (Table 2). The tabular data from the interpretation of IRS P6 data from the period October 2005 to May 2005 indicated that the total area of wetlands ranged from 528 ha to 7.5 ha. The maximum difference was 17.9% to 0.3% of total park area. The digital analyses allowed mapping of the specific sites of changes in wetland type. The sites of change were prepared based on the digital analysis. There was a parallel relationship between water level and areas of the other land cover classes. Assessment of Changes Wetlands were accounted to 3.43% of park area during October 1989 and gained in 17.86% percent of the area (426.6 ha) during October 1989 to October 2005 (Fig.3). The net change was also observed in vegetated areas with increase of 97.7 ha which is attributed to growth and regeneration of vegetation. Both spatial and temporal changes were considered for the multitemporal data of 2005 to determine short term changes (Table 2, 3). The change of spatial extent in the month of March was compared to March of another period to understand long term changes. During the period of March 1977 and March 2005 water spread varied by 178.5 ha. The total wetland in March 1977 was 186 hectares and in March 2005 it was found to be 7.5 ha. Comparisons between dry months of 1977 and 2005 and wet months of 1989 and 2005 indicated that changes in overall wetland areas were significant over the 28 years. This analysis indicates that geoinformatics provide a convenient method for documentation and analysis of wetland data. The information developed during this study has provided an opportunity to determine whether the wetlands in the KNP exhibit historical changes. The present study has indicated that the lake is moving towards the degradation stage with low water level and scattered growth of aquatic vegetation. A study on the dynamics of wetlands investigated long-term and seasonal changes that have occurred in the study area for the periods between 1977 and 2005 (Table 4).

36

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc CONCLUSION AND RECOMMENDATIONS The digital data of satellite remote sensing and digital image processing techniques are quite useful for land cover mapping. The techniques in GIS like matrix model have contributed a great deal to reduce the complexity and data uncertainty in change detection. The major conclusions of the wetlands analysis are that there have been significant changes in the total extent of wetland found in the KNP area; that the changes that have occurred, especially in the wetland class, appear to be primarily related to changes in water levels and release of deficit water to the park. This study is an effort to quantify seasonality and extent of vegetation and wetlands with a suite of multi-temporal satellite datasets. The persistent decline in wetland cover has the possibility to threaten both flora and fauna in these wetlands. The decade-long study on bird communities showed that the ecosystem function of KNP depends mainly on the quantum and time of release of water and fish fry every year. Considerable changes have taken place in the Park on account of irregularity in water released and invasion of alien species. The studies conducted in the park suggested to maintain the periodicity of water to sustain the bird populations (Mathur et al. 2009). Among the disturbance factors grazing pressure and expansion of Prosopis juliflora are worsening the KNP. There is an urgent need to conserve the wetlands and biodiversity of Keoladeo National Park. REFERENCES All, S, V Vijayan 1986. Keoladeo National Park Ecology Study: Summary Report (1980-85). Bombay Natural History Society, Bombay. Bennun, LA 2001. Long-term monitoring and the conservation of tropical wetlands: high ideals and harsh realities. Hydrobiologia 458: 9-19. Bhupathy, S., V Vijayan, R. Mathur 1998. Population ecology of migratory waterfowl in Keoladeo National park, Bharatpur. Jour. Bombay Natl. Hist. Soc. 95(2): 287-294. http://glcf.umiacs.umd.edu/ http://www.ramsar.org/ Mathur, VB., K. Sivakumar, B. Singh 2009. A Bibliographical Review for Identifying Research Gap Areas: Keoladeo Ghana National park, Bharatpur. Wildlife Institute of India. Dehradun. Pattanaik, C, SN Prasad, CS Reddy, PM Reddy 2008. Bharatpur wetland: Future desert? Curr. Sci. 95(10): 1384-1385. Reddy, C.S., PRM Rao, C. Pattanaik, C. & PK Joshi 2009. Assessment of large scale deforestation in Nawarangpur district, Orissa, India using remote sensing and GIS Environ. Monit. Assess. 154: 325-335. Roy, PS., RN Kaul, MR Sharma, SS Garbyal 1985. Forest type stratification and delineation of shifting cultivation area in eastern part of Arunachal Pradesh using Landsat MSS data. Int. J. Remote Sens 6: 411-418.

37

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc Saxena, VS 1975. A study of the flora and fauna of Bharatpur Bird Sanctuary. Department of Tourism, Jaipur, Rajasthan. ACKNOWLEDGEMENTS Authors are thankful to Dr. V. Jayaraman, Director, National Remote Sensing Centre and Dr. P.S. Roy, Deputy Director, NRSC and Prof. K.C. Sharma, Head, Department of Environmental Sciences, Ajmer for facilities and encouragement. The present study has been carried out under part of national project on “biodiversity characterization at landscape level using remote sensing and GIS” sponsored by Department of Space and Department of Biotechnology, Govt. of India. ANNEXES Table 1. Areal extent of land cover types in Keoladeo Ghana National Park: 1989 and 2005 October, 1989 October, 2005 Sl.no. Land cover Area-Ha % of Area Area-Ha % of Area Vegetation 1 Scrub-Woodland 685.2 23.2 684.0 23.1 2 Moist Savannah 1093.0 37.0 1107.1 37.4 3 Grassland 368.0 12.4 364.3 12.3 4 Aquatic Vegetation 147.8 5.0 236.2 8.0 Sub total 2294.0 77.6 2391.7 80.9 5 Non Vegetated Area 560.9 19.0 36.6 1.2 6 Water bodies 101.5 3.4 528.1 17.9 Sub total 662.5 22.4 564.7 19.1 Grand Total 2956 100 2956 100

Table 2. Areal extent of water and other land cover in Keoladeo Ghana National Park: 2005 (monthly estimate: area in Ha.) Sl.no. Land cover October January February March April 1 Water 528.1 54.6 48.6 7.5 7.5 2 Vegetated Area 2391.7 1283.0 1258.5 1366.1 1335.7 3 Non Vegetated Area 36.6 1618.4 1649.0 1581.9 1612.8 Total 2956 2956 2956 2956 2956

38

May 7.5 1245.3 1703.7 2956

Journal of Wetlands Ecology, (2010) Vol. 4, pp 33-42 ISSN: 2091-0363, Open access at www.nepjol.info/index.php/jowe Wetland Friends of Nepal at www.wetlandfriends.co.cc

Table 3. Change Area matrix for KNP (October, 05 to May, 05) Vegetated Non Vegetated Sl.no. October/May Water Area Area 1 Water 3.4 113.4 411.4 2 Vegetated Area 4.1 1132.0 1255.7 3 Non Vegetated Area 0.0 0.0 36.6 Total 7.5 1245.3 1703.7

Total 528.1 2391.7 36.6 2956

Table 4. Areal extent of water and other land cover in Keoladeo Ghana National Park: March 1977 to March 2005 (area in Ha.) Sl.no. Land cover March, 1977 March, 2005 Change Area (-,+) 1 Water 186.0 7.5 -178.5 2 Vegetated Area 2416.2 1366.1 -1050.1 3 Non Vegetated Area 353.3 1581.9 1228.6 Total 2956 2956 0.0

39


Figure 1. Landsat MSS, Landsat TM, IRS P6 LISS III and IRS P6 AWIFS images of Keoladeo National Park.

40


Figure 2 .Vegetation and land cover map of Keoladeo National Park (October, 2005)

41


Figure 3. Spatial classified maps of Keoladeo National Park: Mar1977 -May 2005

42