Theses Doctoral

Forest Degradation and Governance in Central India: Evidence from Ecology, Remote Sensing and Political Ecology

Agarwala, Meghna

There is no clear consensus on the impact of local communities on the resources they manage, primarily due to a shortage of studies with large sample sizes that incorporate multiple causal factors. As governments decentralize resource management to local communities, it is important to identify factors that prevent resource degradation, to inform more effective decentralization, and help the development of institutional characteristics that prevent resource degradation.
This study used remote sensing techniques to quantify forest biomass in tropical deciduous forests in Kanha Pench landscape of Central India, and used these metrics to identify factors associated with changes in forest biomass. Kanha Pench landscape was chosen because of its variation in forest use, and because forests were transferred over a period where satellite imagery was available to track changes. To verify that remote- sensing measured changes indeed constitute degradation, I conducted ecological studies in six villages, to understand changes in biomass, understory, canopy, species diversity and long-term forest composition in intensively used forests. To understand the impact of institutional variables on changes in forest, I interviewed members of forest management committees in fifty villages in the landscape, and tested which institutional variables were associated with changes in forest canopy since 2002, when the forests were decentralized to local communities. The empirical results are of particular conservation significance in India, where further decentralization of forests to local communities in scheduled under the Forest (Dwellers) Rights Act, 2006.
Results indicate that local forest use is associated with decreases in forest biomass, understory, canopy cover, and changes in vegetation structure, species richness and diversity. Most importantly, I found that human use has the potential to alter long- term forest composition as transition of some species to higher size classes is altered where humans use forest more intensively. Particularly, species that are fire and trampling resistant are more likely to become mature trees in intensely used forests. Thus, local forest use is associated with forest degradation as the long-term trajectory of the forest is altered, and forests may not be able to provide ecosystem services including livelihood needs such as fuelwood, construction, and non-timber forest products in the future.
At a broader scale, remote sensing techniques (optical imagery Landsat and RADAR imagery ALOS-PALSAR FBD) were able to quantify forest biomass at an acceptable accuracy (67 percent), while more easily operatable MODIS based EVI was not. Landscape analysis showed that changes in forest biomass from 2007 to 2010 were associated with high population density, high fire radiative power and greater distance to towns. Since people only travel approximately 2 kilometers for subsistence forest use, the significance of greater changes further from towns suggests that, at a broader landscape scale, forest degradation is not primarily due to local use, but may be a result of other factors.
Action taken to exclude outsiders and lower meeting frequency of committees (never) were identified as institutional variables associated with remotely-sensed positive change in canopy over the period when forest management was transferred (2002 to 2010). Villages with no meetings were also associated with higher incumbency of committee Chairpersons and lower incumbency of other committee members. Simultaneously, while economic payments increased awareness and participation in forest management committees, economic payments were not associated with any action to exclude outsiders from forest use. This suggests that managers need to focus on factors besides economic payments to incentivize committees to exclude outsiders, especially as it is associated with positive changes in the forest. Further, while elite capture of resources (as indicated by incumbency and lack of inclusiveness in decision-making) is not helpful for social equity, it does not appear to be detrimental for forests.
Overall, this study suggests a number of management strategies to reduce forest degradation. Managers could focus on forests at a distance from towns and roads, as this is where most negative change in forests appears to occur. They could also work with local communities so that their use of forests does not prevent regeneration of species important for ecosystem services. Managers could also work with committees to find strategies other than economic payments for incentivizing community protection of forests.

Geographic Areas


  • thumnail for Agarwala_columbia_0054D_12432.pdf Agarwala_columbia_0054D_12432.pdf application/pdf 19.1 MB Download File

More About This Work

Academic Units
Ecology, Evolution, and Environmental Biology
Thesis Advisors
DeFries, Ruth S.
Ph.D., Columbia University
Published Here
December 22, 2014