Surface water dynamics analysis based on sentinel imagery and Google Earth Engine Platform: a case study of Jayakwadi dam

Abstract

Surface water is important for the urban and agriculture ecosystem, the accurate and very easy to detect and analysis of the surface water based on the remote sensing data and google earth engine platform. It is a very much important for irrigation and water resource management during the dry period and rabi with summer season. In this paper, we have extracted the surface water analysis from Sentinel-2A images using a new technique of Google Earth Engine (GEE) and machine learning coding. Monitoring the dynamics of surface water is a very helpful to study of the irrigation, and drinking water requirement, which climate change factors most damages on the surface water, natural environmental health and understand the impacts of global changes and human actions on the planning and development of the water resources in the semi-arid region. Currently, so much research has focused on the surface water extraction and dynamic monitoring using remote sensing imagery and softwares. But downloading a big number of remote sensing images are covering that area and then process in the remote sensing and GIS software as per the traditional method. The traditional process is much lengthy, also very time-wasting for such kind of time series dataset analysis. In this view, GEE platform has been given easier to access any satellite data within less time. The GEE is a total cloud-based platform enthusiastic to satellite image data processing based on the machine learning coding. So for many excellent remote sensing image processing coding, algorithms have been integrated in the platform of Google Earth Engine (GEE). These data do not require to save images and collection, which are very easy to doing satellite data processing and effective output capability. The results show that 2019 year has observed that there are increased water area, i.e. 202.4871 km² and for the year of 2019, water spread-out area is 410.9113 km² in dam. To compare the results of 2015–2019, there is much increase in the water area due to various reasons like pumping, heavy rain, etc. in the Jayakwadi dam. This water can use for various purposes and irrigation water management and drinking purposes during the drought condition. The developed algorithms have given better information and results of water spread-out by google earth engine. The current techniques are reliable, novel, and quick to get the maximum and minimum extent of the surface water. The results can be very useful for the surface water planning and management in the study area.

Appendix

Code for Accessing Image in GEE

var wetSeason = ee.ImageCollection('COPERNICUS/S2').

print (wetSeason)

Code for Visualizing Image

var visParams = { bands: ['B4', 'B3', 'B2'], min: 0, max:3000};

Map.setCenter (74.89, 19.75, 5);

Map.addLayer(wetSeason.clip(Studyarea), visParams, 'wetSeason');

Code for Mosaicing

var wetSeasonMosaic = wetSeason.mosaic()

var visParams = { bands: ['B4', 'B3', 'B2'], min: 0, max:3000};

Map.setCenter(74.89, 19.75, 5);

Map.addLayer(wetSeasonMosaic.clip(Studyarea), visParams, 'Mosaicwet');

Code for Water Index Calculation

var NDWI = wetSeasonMosaic.expression(

'(GREEN-NIR)/(GREEN + NIR)',

{'GREEN':wetSeasonMosaic.select('B3'),

'NIR':wetSeasonMosaic.select('B8')

});

var visParams2 = { min:-1,

max:1,

palette:['yellow','black'],

};

Map.addLayer(NDWI.clip(Studyarea),visParams2,'NDWI').

Export to drive

Export.image.toDrive({

image: NDWI,

description: 'NDWI_2019_10m',

scale: 10,

region: Studyarea});