Assessment of Spatial Variations and the Interaction of Climate-Soil-Human Drivers of India’s Terrestrial Carbon Use Efficiency

Abstract:

The conversion of atmospheric CO₂ into biomass takes place through plant communities and soil microorganisms. A part of this efficiency is controlled by vegetation and is termed as vegetation carbon use efficiency (CUE). It is a measure of carbon storage as useful biomass. It is defined as the ratio of net to gross primary productivity. The CUE is important to calculate carbon turnover rates, carbon flux dynamics and regionalized carbon assessments. The CUE varies across ecosystems, climate and nutrient types. Earlier studies have depicted the impact of climate on CUE (He et al, 2018, Agr Forest Meteorol). Soil nutrients were also assessed as additional drivers of CUE for China’s ecosystems (Chen & Yu, 2019, Sci Rep). Anthropogenic stress also plays a vital control on CUE, especially for India where vast scale land cover change and fossil fuel emissions exert an additional perturbation on the carbon cycle. Examination of the interactive behaviour of the climate-soil-human drivers on CUE is seldom addressed.

Assessment of CUE helps in planning future carbon sequestration policies. The MODIS product has been used for Indian ecosystems (Sharma & Goyal, 2017, Glob Chang Biol). This product is used to find mean annual CUE from 2000-2014 in this study. The study examines spatial variation of CUE across major Indian ecosystems (forests, croplands, grasslands, shrublands and savannas). Trends of climatic (temperature, precipitation, water scarcity), soil (pH, soil organic carbon, cation exchange capacity, sand%, clay%) and anthropogenic (carbon emissions, population density) drivers are studied. An unbiased machine learning based approach is used to find important drivers of CUE and it is modelled using the best four predictors for each ecosystem using a mixed model estimate. The results reveal largest CUE for grasslands and forests and lowest for shrublands. Temperature and population density exert a quadratic control while other factors control CUE linearly. Temperature is the most important predictor across all ecosystems. Anthropogenic stress and soil nutrient factors also feature among the best four predictors across all ecosystems. Mixed model with 2-way interactions capture a significant variability of CUE accurately across all ecosystems of India.