3.1  Database and Software used

Remotely sensed orthorectified CartoDEM, SRTM and ASTER GDEM datasets were used for detection and delineation of WBA. These DEMs could be used to monitor, detection and delineation of the surface WBA especially in inaccessible areas. For a comparative analysis of detecting and delineated water bodies, CartoDEM, SRTM and ASTER GDEM were re-projected to a common projection (UTM WGS84, Zone 43) in ArcGIS software. The ‘Hydrology’ toolset in ArcGIS, developed by ESRI (2004), has been used for DEM preprocessing and surface stream simulation. The precise co-registration among the DEMs is a pre-requisite because DEMs are sensitive to mis-registration for areas and relief as a shift of a fraction of a pixel may cause significant changes in the elevation difference (Ni et al., 2014). Details of datasets are given in Table 1.

3.1.1 CartoDEM

The Cartosat-1 DEM (CartoDEM version 1) is an Indian DEM developed by the Indian Space Research Organization (ISRO). This DEM has been used in multiple sectors for planning and decision making. It is derived from the Cartosat-1 stereo payload launched in May 2005 by ISRO, India. The primary output is a tile of 7.5ʹ × 7.5ʹ extents with DEM spacing of 1/3 arc-sec, and co-registered ortho-image at resolution 1/12 arc-sec (NRSC, 2015). The CartoDEM is a land surface model of digital elevation and covers land and covers 

land surfaces within India (Das et al., 2016). Freely available 1 arc second spatial resolution NRSC CartoDEM  V1 and CartoDEM V1.1R1 1×1 degree tile was downloaded from the NRSC website http://bhuvan.nrsc.gov.in. The operational procedure of this DEM generation comprised of stereo strip triangulation of 500 × 27 km segment with 30-m posting (Muralikrishnan et al., 2013). The vertical accuracy for CartoDEM (V1) claimed by NRSC is 8 m (Muralikrishnan et al., 2011). Considering the appearance of water bodies as a flatter surface over digital elevation model, this CartoDEM data was selected and used for detection of water bodies.  

3.1.2  SRTM

SRTM and ASTER GDEM dataset is available at http://gdem.ersdac.jspacesystems.or.jp/ at 90 m and 30 m resolution. SRTM was an international project, spearheaded by the US National Geospatial-Intelligence Agency (NGA) and the National Aeronautics and Space Administration (NASA). The SRTM data provided a globally consistent digital elevation data for approximately 80% of the world’s land surface area. It’s produced by NASA and available at two spatial resolutions: 1-arc-second in the USA and a degraded 3-arc-second for the rest of the world. SRTM elevation data are based on the WGS84 ellipsoid as horizontal datum, while its mean sea level is defined by EGM96 geoid as vertical datum. The specified accuracy requirement for SRTM DEM products was 16 m absolute vertical error (at 90% confidence level). This paper deals with the better known and widely available 3-arc-second products, which are publicly available (Jing et al., 2014). Considering the smoothness in digital appearance and the visualization of WBA as a flatter surface over this DEM, this was used to achieve the objectives of the study.

3.1.3  ASTER GDEM

The ASTER GDEM purports to be the highest resolution, 30 m (1 arc second) spacing as compare to CartoDEM and SRTM DEM (Guth, 2010; Arefi and Reinartz, 2011). Moreover, the ASTER sensor on board of the Terra spacecraft has an along-track stereoscopic capability due to a nadir and 27.6º backward-looking telescopes in the Near Infrared (NIR) spectral band. This instrumental setup allows for photogrammetric DEM generation with vertical accuracies of ±15-30m (Toutin, 2008). It is freely available and would make a significant contribution to understand the water bodies. ASTER GDEM is a significant for representing digital topography at 30 m resolution, which was generated by stereo matching techniques, contributed by the Ministry of Economy, Trade, and Industry (METI) and NASA. Remotely sensed orthorectified ASTER GDEM (Version 1) dataset is available at free of cost from Earth Remote Sensing Data Analysis Center (ERSDAC) of Japan and NASA (ASTER GDEM, 2015). Complete GDEM covers land surfaces, which is wider than SRTM (Jing et al., 2014). Topographic parameters such as elevation point, slope, aspect, cross profile can be generated using this DEM and automated procedures for detection of water bodies with cross verifications. This DEM has good spatial resolution with clear appearance of the water bodies through falter digital surface over. 

Global Mapper software (v11.01) together with ArcGIS (v10.0) was used for CartoDEM, SRTM and ASTER GDEM data processing, i.e. generation of EPFs (7194906), slope and contour map preparation. The obtained results from the EPFs, slope map and contour map were cross-validated from Landsat 8 satellite imagery (Dec 2013) and topographical maps (No. 47/E/5 and 47/E/9) obtained from Survey of India (SOI). Moreover, these water bodies were also verified from high-resolution Google Earth images. It is a complementary reference to help distinguish confusing water bodies from background noise like mountain shadows (Jiang et al., 2014).

3.2  Approaches    

3.2.1  EPF Approach

The smooth water bodies act as specular reflector and thus reflect most of the transmitted energy from the sensor that results the low backscatter values from water bodies (Hahmann et. al., 2008). Such water bodies are digitally appearing as flattening surface over inland water bodies on CartoDEM, SRTM and ASTER GDEM. In some cases, the water bodies may be normalized to have a slope value around zero degree.  Therefore, the flattening surface over these DEMs is considered as water bodies that may be reservoir, lake, dam, etc. over inland surface. Scientifically, each and every water bodies are smooth and plain because of its liquid nature. Thus, water bodies are absolutely smooth and flat (Slope: 0°-0.5°) appeared on these DEMs images as compare to other region were considered as water bodies. Flat digital pixels over CartoDEM, SRTM and ASTER GDEM are seemed, as a water body (Frey and Paul, 2011) which can be used to detect the surface water body area (WBA) using slope and contour maps using combination logic.

EPF is the grid points which are generated in unique array and consisted with specific height on the DEMs. It is the grid point which has the unique elevation value at specific intervals. In this study, CartoDEM, SRTM and ASTER GDEM data were processed for generation of EPFs (Figure 3). Conceptually, the unique DEM values (EPF) are picked up along the boundaries of water body as an input for determining the elevation of each water body. The logic behind unique EPFs is the flattening surface. Each of the water body has assigned flat elevation value in the DEMs. The EPF (7194906) was created at Elevation Grid Cell Centers (EGCC) by automated procedures (Figure 3). Spatial query was run in the query builder over the selected DEMs sample EPF for detection of the WBA in the selected study area. Generated EPFs were exported into point vector data shape file and put up in the software for further processing as quarry analysis. In the entire EPFs few outlier (non-water body) points are occurred due to unique elevation with the existed WBA. These identical spatial heights were an error with relevant water bodies in the surrounding area. Thus, the data combination approach was used to remove the non-water bodies’ points.

Figure 3. Elevation Points Feature (EPF) from ASTER GDEM

3.2.2  Combination Approach

Generated EPFs are integrated with prepared contour and slope map using query analysis task by combining the generated 7194906 EPFs. The contour and slope maps were used to detect the WBA (Figure 4). Example: the contour line (576 m) detection at the edge of water bodies of Darna reservoir (Figure 5) was performed for cross verification of obtaining results from advance query and combination of slope map. The slope map with 0^º to 0.5^º was used to compare the WBA in the existed EPFs cluster. This detected EPFs (590m and 576m) area was cross-validated from Landsat 8 satellite imagery and topographical maps of the same region. The omission of unnecessary EPF (non-water bodies) is essential to measure the WBA, precisely. The delineation of contour at the edges of water bodies and its incorporation with slope map was therefore attempted (Figure 5). These were the non-clustered points (non-water bodies) and located in the outlier part of the existed water bodies. For the purpose of clear identification of boundaries of water bodies, nearest neighbor (for discrete surface) to cubic convolution (for continues surface) conversion of DEMs were carried out (Figure 6). Finally, the combination of EPFs, slope map, recognition of contour line at the edge of water bodies and nearest neighbor to cubic convolution of processed DEMs were attempted for final detection of the water bodies in the study area (Figure 7).

Figure 4. (A) Slope map; (B) Superimposing of contour line (Interval: 100 m) over the water bodies

4.1  EPF Result

The water bodies are digitally appearing over the 3 DEMs as a flattening surface. The digital topography seemed with flattering digital pixels (Frey and Paul, 2011) on the selected DEMs can be consider as WBA. The selected DEMs were processed for EPFs generations (7194906). The generated EPFs are unique height vector points, distributed with same interval over these DEMs. The heights of all EPFs were found same, which was considered as the WBA.  Derived slope and contour from DEMs are probably useful for the visual appearance of WBA. Available ASTER GDEM and CartoDEM purports to be the highest resolution, with 30 m (1 arc second) spacing (Guth, 2010) as compare to SRTM data, thus, the smooth water bodies act as precise specular reflectors over these DEMs. Hence, the outcome of the precise resolution from ASTER GDEM and CartoDEM were better for detection and delineation of water bodies in the area. SRTM data overestimate the elevations. For SRTM, overestimation of elevation is predictable since radar signal returns are affected by vegetation cover (Guth, 2006; Shortridge 2006). It might be because of the spatial resolution of the same data. Therefore, large error was noticed in the water body detection from the SRTM data (Figure 7). These errors mean, highlighting the non-water body area over the SRTM DEM. Probably, the EPFs are without the clustering precincts over SRTM data as compare to ASTER and CartoDEM data.

4.2  Data Combination

Identical height of water bodies EPFs are seen in the surrounded non-water body area which could be an error in the present assessment. In view of this, in the current task, from the spatial quires results the shape of EPF was found irregular due to the location of water bodies is consisted with undulating hilly topography in the study area. The result reveled from spatial query on EPFs for Dam 1 and Dam 2 sites were shown the water body area with unnecessary EPFs as an error points (Figure 5). Thus, for appropriate detection of water bodies with the slope and contour map combination increased the preciseness in the WBA. The slope maps generated from CartoDEM, SRTM and ASTER GDEM were also supported for water body detection. In this sense, the slope is absolutely plain between 0° to 0.5°; flattening the surface of DEMs, absence of contours in the WBA and clustering of EPFs in the area logic helped to detect the precise water bodies in the area. The data combination approach improves the result of WBAs appearance. At the same time, the conversions of detecting WBA from the nearest neighbor to cubic convolution were also helped to enhance the detected water bodies after the appearance of proper boundary of water bodies (Figure 6). To avoid the bias delineation the incorporated approach of slope, contour and its cross validation from high resolution satellite imagery, etc. are more supportive to enhanced detection and delineation of the WBA in the present hilly region.

Figure 5. (A) Primarily detected water bodies by query analysis with outlier EPF; (B) Detected water bodies by combining contour and advance spatial query

Figure 6. Nearest neighbor to cubic convolution conversion for detecting the edges of water bodies: (A) water bodies with nearest neighbor; (B) water bodies with cubic convolution

5.3 Detection and Delineation of Water Bodies

Results obtained for the selected water bodies indicating that, both CartoDEM and ASTER GDEM data meet their predefined vertical accuracy specifications as compare to SRTM data. It was realized that ASTER GDEM data and CartoDEM have a higher horizontal accuracy than SRTM for the study area. In this study, these spatial and vertical resolutions of both the dataset are influencing the preciseness of detected WBA. The underestimation of ASTER and CartoDEM is more pronounced on flat and less complex terrain and of a greater magnitude than the overestimation of SRTM (Gerald and Ben, 2012). Water bodies are appeared similar over all DEMs: SRTM 90m, ASTER 30m and CartoDEM 30m. This comparison shows that DEMs data can be used to detect the water bodies precisely. However, WBA could be detected from CartoDEM and ASTER GDEM with precision. The distribution analysis shows bias estimation using DEM using SRTM data with less precision. Assimilation of query analysis, contour line delineation at the edge of water bodies, clustering of EPF, non-contour area and flattening water bodies over ASTER GDEM and CartoDEM with 0° to 0.5° slope area were finally considered as WBA in the present study area (Figure 7). Water bodies in the study area were estimated as: 3.09% of reviewed area from CartoDEM, 2.22% from ASTER GDEM and 4.38% from SRTM DEM (Table 2). CartoDEM and ASTER GDEM show potentials of precise estimation of surface water bodies especially in inaccessible hilly region as it has higher resolution and good digital flattering surface texture. However, smaller water bodies have been detected and delineated more precisely from CartoDEM (v1) than ASTER GDEM.

Figure 7. Detected Water Bodies from CartoDEM, SRTM and ASTER GDEM data

The authors declare no conflict of interest.

Authors are thankful to Department of Geography, S.N. Arts, D.J.M. Commerce and B.N.S. Science College, Sangamner, Maritime, Technology and Environment Hub, Research and Innovation, Southampton Solent University, and Department of Geography, H.P.T. Arts and R.Y.K. Science College, Nashik, for support and providing necessary research facilities. Author gratefully acknowledge the anonymous reviewers for constructive comments and suggestions for improvement in the draft.

ASTER: Advanced Spaceborne Thermal Emission and Reflection; DEM: Digital Elevation Model; EGCC: Elevation Grid Cell Centers; EPF: Elevation Point Feature; GDEM: Global Digital Elevation Model; GIS: Geographical Information System; InSAR: Interferometric Synthetic Aperture Radar; ISRO: Indian Space Research Organisation; NASA: National Aeronautics and Space Administration; NDVI: Normalized Difference Vegetation Index; NDWI: Normalized Difference Water Index; NRSC: National Remote Sensing Centre; SRTM: Shuttle Radar Topography Mission; Water Body Area; WGS: World Geodetic System.