Spatio-Temporal Variation of Soil Moisture and Surface Fluxes over Indian Subcontinent in Relation to Indian Summer Monsoon: Validation and Inter-Comparison of Data Products

Soil moisture is one of the important drivers of land-atmosphere coupling and feedback systems at the land surface. It's variability directly influences the partitioning of energy between latent and sensible heat fluxes at the surface and eventually the planetary boundary layer, weather and climate processes. The Indian subcontinent comprising of different climatic zones and diverse terrain features, together with unique monsoon rainfall characteristics, lead to highly dynamic soil moisture distributions at varying spatio-temporal scales. The Indian summer monsoon rainfall (ISMR), in addition to introducing large seasonal variations in soil moisture over most of the subcontinent, is also the cause for large intra-seasonal and inter-annual variability in soil moisture patterns. Soil moisture and its variability has influence on the dynamics of the monsoon, especially during the onset and break to active transition phases. Detailed understanding on the mechanisms involved is yet to evolve, due to lack of consistent and reliable spatially distributed soil moisture data sets. Understanding the spatio-temporal evolution of soil moisture over the sub-continent would provide insight into their influences on the modulation of large-scale processes.

The present study is on the validation and inter-comparison of soil moisture data using three soil moisture data products, and investigation on the dynamics of soil moisture over the subcontinent in relation to ISMR. The datasets used are: i) In situ continuous observations carried out by India Meteorological Department (IMD) at nearly 100 locations across the Indian subcontinent ii) Essential Climate Variable (ECV) global soil moisture dataset and iii) MERRA reanalysis soil moisture data. The IMD soil moisture used for the validation is measured at 20 cm depth. The ECV soil moisture data is a satellite data product derived by merging active and passive microwave observations. Although ECV and MERRA data correspond to the surface soil, the 20-cm depth IMD data are used for validation considering the fact that surface soil moisture exhibits strong relationship with deeper layers.

Time series correlation analyses performed using data from 8-month (April – November) pre- to post- monsoon period show good agreement between the three datasets. Correlation coefficients between in situ IMD and MERRA, and IMD and ECV at most of the locations chosen are more than 0.6, suggesting the suitability of derived global soil moisture products for Indian summer monsoon related studies. The correlation between MERRA and ECV soil moisture are more than 0.8 over most of grid points. The soil moisture patterns are found to be in good agreement with the observed rainfall. Time stability analyses performed using the three datasets exhibit identical characteristics on the space-time dynamics of soil moisture over the subcontinent related to monsoon activity and terrain characteristics. Composite analyses performed using all active and break monsoon periods during 2001 – 2010 show distinct patterns with significant spatial variations during active and break periods. The study is further extended to the analyses on the response of surface fluxes to soil moisture variations to understand how soil moisture memory influences latent heat flux variations and modulate boundary layer processes, during different phases of monsoon.

Although the exact quality of different datasets is difficult to ascertain from the comparisons due to uncertainties in all, the similarity in patterns and statistics in terms of correlation, seasonal and intra-seasonal characteristics noted in the results gives the confidence that the data products analyzed in the study do represent realistic soil moisture characteristics over the region. Our study is the first detailed analysis of different soil moisture datasets over the Indian subcontinent in relation to the ISMR.