Dubey et al. (2025) Forest-savanna stability in India under human interventions and changing climate
Identification
- Journal: Communications Earth & Environment
- Year: 2025
- Date: 2025-12-10
- Authors: Nivedita Dubey, Tejasvi Chauhan, Subimal Ghosh
- DOI: 10.1038/s43247-025-03076-5
Research Groups
- Department of Civil Engineering, Indian Institute of Technology Bombay, Mumbai 400076, India
- Biosphere Theory and Modelling Group, Max Planck Institute for Biogeochemistry, 07745 Jena, Germany
- Interdisciplinary Program in Climate Studies, Indian Institute of Technology Bombay, Mumbai 400076, India
Short Summary
This study investigates the stability of forest-savanna ecosystems in India under human interventions and climate change, projecting a future shift towards savanna-like conditions due to increased precipitation variability, while highlighting the mitigating potential of strategic anthropogenic conservation efforts.
Objective
- To investigate the dynamics of India's tree cover (2001-2020) and its relationship with regional hydroclimate.
- To project India's tree cover distribution for the end of the century (2071-2100) under the SSP585 climate scenario.
- To assess the effect of non-climatic drivers, including anthropogenic interventions, on tree cover distribution.
- To understand the resilience and stability of forest-savanna systems in India to climate change and human activities.
Study Configuration
- Spatial Scale: India, with tree cover data at 500 meter resolution and climate data at 0.5 degree latitude by 0.5 degree longitude resolution.
- Temporal Scale: Observed changes from 2001 to 2020; present climate baseline periods 1971-2000 and 1991-2020; future projections for 2071-2100.
Methodology and Data
- Models used:
- Gaussian mixture model (GMM) for tree cover classification.
- Generalized linear models (GLM) with binomial distribution and logit link function for likelihood estimation.
- Dynamical systems approach using the Langevin equation to construct potential landscapes for tree cover stability.
- Coupled Model Intercomparison Project Phase 6 (CMIP6) Earth System Models (ESMs) for future climate projections.
- Data sources:
- MOD44B Version 6.1 Vegetation Continuous Fields (VCF) product for annual percent tree cover (500 meter resolution).
- MCD12Q1 International Geosphere-Biosphere Programme (IGBP) product for land-cover classification (500 meter resolution).
- CRU-TS (Climatic Research Unit gridded Time Series) V4.07 for present climate precipitation (0.5 degree resolution).
- CMIP6 ESM outputs for future climate under the SSP585 scenario (regridded to 0.5 degree resolution).
- Landsat 8 TCC (tree canopy cover) data for validation.
Main Results
- India's total tree cover grids declined from 30% to 28% between 2001 and 2020.
- Non-climate-driven compositional shifts were observed: low tree cover grids decreased (13.49% to 7.26%), while savanna increased (9.21% to 11.27%) and forest grids increased (7.18% to 8.60%).
- A 40% tree cover threshold statistically delineates savanna (10-40% tree cover) from forest (>40% tree cover).
- Nearly half of India's landscape exhibits forest-savanna alternative states (unistable savanna), primarily controlled by low (high) precipitation variability and seasonality.
- The likelihood of forests increased in the Himalayas, parts of northeast India, east-central India, and the Western Ghats from 2001 to 2020, attributed to non-climate drivers such as CO2 fertilization, natural regrowth, and anthropogenic interventions.
- CMIP6 projections for 2071-2100 (SSP585 scenario) indicate an increased likelihood of savanna and a decreased likelihood of forest in parts of the western and eastern Himalayas, central India, and the Western Ghats.
- This projected shift is mainly driven by an increase in precipitation variability and seasonality, which counteracts the effects of increased mean annual precipitation.
- The potential landscape analysis shows two stable tree cover basins (~10-30% and ~60-80% tree cover) separated by an unstable equilibrium of 40-50%.
- Regions with low rainfall variability and seasonality maintain bimodal forest-savanna states, whereas regions with high variability and seasonality tend towards a unistable savanna-dominated state.
- Anthropogenic interventions demonstrate the potential to mitigate the adverse effects of climate change on forest ecosystems.
Contributions
- Provides the first comprehensive assessment of forest-savanna stability in India under the combined influence of human interventions and climate change, addressing a critical research gap in the region's ecological resilience.
- Quantifies observed non-climate-driven compositional shifts in India's tree cover from 2001 to 2020, highlighting the positive impact of anthropogenic conservation efforts.
- Delivers end-of-century projections for India's tree cover distribution under a high-emission climate scenario (SSP585), emphasizing the critical role of precipitation variability and seasonality in future ecosystem shifts.
- Utilizes a multi-model approach combining generalized linear models and non-linear dynamical systems to robustly analyze tree cover-hydroclimate relationships, revealing how hydroclimatic factors modulate forest-savanna bimodality.
- Offers crucial, policy-relevant insights for strategic forest conservation and management in India, advocating for policies that consider future climate impacts and the potential of human interventions to build ecosystem resilience.
Funding
- Department of Science and Technology Swarnajayanti Fellowship Scheme (project no. DST/ SJF/ E&ASA-01/2018-19; SB/SJF/2019-20/11)
- Strategic Programs, Large Initiatives and Coordinated Action Enabler (SPLICE) and Climate Change Program (project no. DST/CCP/CoE/140/2018)
Citation
@article{Dubey2025Forestsavanna,
author = {Dubey, Nivedita and Chauhan, Tejasvi and Ghosh, Subimal},
title = {Forest-savanna stability in India under human interventions and changing climate},
journal = {Communications Earth & Environment},
year = {2025},
doi = {10.1038/s43247-025-03076-5},
url = {https://doi.org/10.1038/s43247-025-03076-5}
}
Original Source: https://doi.org/10.1038/s43247-025-03076-5