Feedback From Vegetation to Interannual Variations of Indian Summer Monsoon Rainfall

Abstract

Interannual variations of indian summer monsoon rainfall (ismr) are modulated by external forcings such as el niño southern oscillation, indian ocean dipole, and the atlantic niño. Vegetation over land responds to variations in ismr, but the feedback from vegetation to ismr variability has not been fully explored yet. To address this gap, we perform two simulations with the regional weather research and forecasting model coupled to the community land surface model (wrf-clm) for the period of 2004–2018. We use the same boundary forcing from era-interim reanalysis for the two experiments, but with two different vegetation prescriptions, (1) observed, interannually varying leaf area index (lai), obtained from satellite images/data (var-lai); and (2) climatological leaf area index from the same product, to suppress interannual lai variations (clim-lai). We find that the correlation coefficient of simulated total seasonal rainfall with the observed data is higher for var-lai simulation as compared to clim-lai. To elicit causality among eco-hydro-climatological variables, we develop a network based on information theory, i.e., A process network. We find that lai plays a major role in influencing precipitation in the network through evapotranspiration. The number of links originating from lai and evapotranspiration increases during drought years, making the eco-hydro-climatological network denser. Our findings indicate that the ismr predictions and projections need to represent the time-varying lai to fully capture the varying feedbacks from evolving vegetation to the atmosphere especially during drought years. © 2021. American geophysical union. All rights reserved.