The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Publications Copernicus
Download
Citation
Articles | Volume XLIII-B3-2022
https://doi.org/10.5194/isprs-archives-XLIII-B3-2022-1287-2022
https://doi.org/10.5194/isprs-archives-XLIII-B3-2022-1287-2022
31 May 2022
 | 31 May 2022

DEVELOPMENT OF A GOOGLE EARTH ENGINE BASED APPLICATION TO MONITOR THE SEASONAL WATER SPREAD AREA OF RIVER GANGA

R. N. Tripathi, A. Ramachandran, S. A. Hussain, V. Tripathi, and R. Badola

Keywords: Google Earth Engine, Ganga, seasonal variation, waterbody extraction, river course

Abstract. The National River Ganga, home to the National aquatic animal, Gangetic dolphin along with species of critically endangered otters, gharial crocodile, turtles, waterbirds and numerous aquatic and semi aquatic life forms is of conservation importance to safeguard the biodiversity it harbors. Meanwhile, River Ganga is also the source of highly productive agricultural yield that satiates the nation’s food demands. The last few years have echoed demands for ‘e-flow’ regulations to be formalized and regularized to maintain healthy river conditions with adequate amount of water retained in the rivers while meeting the populations demand for water supply. Though e flow estimations are a complex science in itself, the simplest way is visualization. In this context, there needs to be a transparent mechanism to picturise and depict the ground conditions of the river stretches. Remote sensing provides ‘eyes in the sky’. The river receives water in the form of precipitation, snow melt and ground water recharge and the hydrological flux is evident in the surface water. Thus delineating the water spread area can provide a crude estimate of the water in the river. The Google Earth Engine platform provides a collection of satellite imageries, server side data processing and algorithms. The Landsat and Sentinel data catalogue offers good coverage of the river and its basin. Though sentinel data provides high ground resolution of 10m, Landsat provides multi temporal historic coverage at a good 30 m resolution, that works well for the main stem river. From the various available data processing methods, spectral decomposition technique of Modified Normalised Differential Water Index MNDWI was found to be most effective in delineating surface water. Data was filtered for the three major seasons of Pre monsoon, Monsoon and Post monsoon periods and MNDWI for the same was visualized. While Middle Ganga region through Uttar Pradesh shows seasonal flooding during the monsoon period, the Lower Ganga after the confluence of major tributaries of Son, Ghaghra, Gandak and Kosi flows in full bank width in the Post monsoon period. During the pre monsoon period many stretches of Ganga particularly the Upper middle stretch showed thin shallow channel and intermittent deep pools disconnected by dry sand beds. These regions result in disruption in the river flow and are priority sites to implement e flow regulations. Another key factor is the regulation of flow in the river from barrages. Sudden release of water can cause flooding resulting in disturbance to the life cycle of species that are tuned to the natural annual hydrological pulse of the river. This is because different aquatic species have different river width and depth requirements. Thus monitoring the seasonal variations in the water spread area helps to note the longitudinal and lateral connectivity of the river which is a prime characteristic for the health of a river. This study intended to create an application that offers to visualize the seasonal water spread in the Ganga river for the past years that can assist decision makers to monitor the river and make informed regulatory measures.