Assessing the impacts of landscape change and habitat degradation on mammal diversity and distribution in the northern Eastern Ghats, Andhra Pradesh, using ecological, geographic and social information

Abstract

Anthropogenic activities are causing landscape modification and biodiversity loss. Mammals in particular are among the worst affected by landscape change. Mammal extinction risks are highest in tropical forest landscapes beyond hotspots and other high-priority landscapes with low protection. The Northern Eastern Ghats (NEG) is one such ‘neglected landscape’. Only 3.53% of area is protected. Dams, plantations, mining and agriculture have led to landscape changes. This thesis attempts to understand how landscape change and protection levels impacted mammals and local communities in the NEG. Specifically, I aim to 1) understand how land cover has changed between Papikonda National Park (PNP), and its relatively unprotected buffer (BUP); 2) identify drivers of these changes; 3) measure responses of mammals to change and differential levels of protection; and 4) study local communities’ perception regarding these changes. I address these objectives through quantitative and qualitative approaches. 

I reviewed land-use in NEG and show how this has changed between 1991 and 2014 using NDVI change analysis. I analysed forest change at landscape and village levels to understand proxies at different scales. My analysis shows that 12% of forest in PNP became degraded between 1991 and 2014, whereas 32.68% became degraded in BUP. Impact of villages on forests was greater inside PNP due to higher dependence of communities on forests. Communities mainly attributed changes in forest cover to shifting cultivation, plantations and over-extraction of forest resources. Most habitat changes occurred in low elevation dry deciduous forests, along roads, and close to villages. These habitats are further threatened by the upcoming Polavaram dam, which will result in their submergence.

I estimated mammal community parameters (presence, diversity and abundance) across different habitats in PNP and BUP using 271 camera traps (CTs) distributed across 58 (2x2 km) grids following an occupancy approach. I quantified habitat structure and disturbance from covariates in 10x10m plots around CTs. I enumerated 55 mammal species from surveys, secondary data and interviews with local communities. Moist deciduous forests, particularly at higher elevations, supported higher mammal presence, diversity and abundance. PNP supported significantly higher mammal presence and abundance than the BUP. The mammal community was highly skewed and dominated by barking deers and wild pigs. Land cover changes most negatively affected large herbivores like gaur and sambar than other mammals. Dry deciduous forests supported high mammal richness, but are most threatened by landscape change, particularly dams. 

I analysed community observations of landscape change. Community dependence on forests varied with tribal groups and their location. Correlation between community observations of land cover change and satellite analysis was strongest near villages, regardless of their location in the conservation matrix, showing that in neglected landscapes, degradation is independent of protection. I discuss these results in the context of forest management, forest rights and tribal development schemes. This thesis presents the first account of mammals from NEG, and how they’ve been impacted by landscape change. It reveals the role of less known protected areas in conserving biodiversity even in neglected landscapes.