Publications

Maurya, AK; Murugan, D; Singh, D (2021). An approach for soil moisture estimation using urban and vegetation fraction cover from coarse resolution Scatsat-1 data. ADVANCES IN SPACE RESEARCH, 68(3), 1329-1340.

Abstract
Soil moisture (SM) estimation using coarse resolution satellite data, like Scatsat-1 scatterometer data (2 km spatial resolution), is still challenging due to various land cover classes present within a single pixel. To estimate SM from coarse resolution data, there is a necessity to study the contribution of different land covers to the total backscatter signal of the single pixel. The estimation of SM is very much dependent on the accuracy of the backscatter signal from the soil component. Therefore, this paper proposes a modified water cloud model to compute the soil-backscattering signal by correcting the backscatter contribution due to urban and vegetation fraction cover. The backscattering signal from the soil surface is then used to retrieve SM. The retrieved SM is compared with the ground truth SM using the error metrics such as root mean square error (RMSE), mean bias (b), correlation coefficient (r), and average relative error (ARE). The proposed model is tested in three conditions: first, when only vegetation correction is applied, RMSE, mean bias, r, and ARE are found to be 0.086 m(3)/m(3), 0.70 m(3)/m(3), 0.534, and 0.24, respectively, and results show an overestimation in SM. Second, when both vegetation and urban correction are applied using the actual fraction, the accuracy is improved with an RMSE, mean bias, and ARE of 0.071 m(3)/m(3), 0.056 m(3)/m(3), and 0.20, respectively. However, in the third condition, when the urban fraction cover is considered twice the actual value, the RMSE and mean bias are found to be 0.22 m(3)/m(3) and -0.10 m(3)/m(3), respectively, and results show an underestimation in SM. (C) 2021 COSPAR. Published by Elsevier B.V. All rights reserved.

DOI:
10.1016/j.asr.2021.03.022

ISSN:
0273-1177