Dynamics and determinants of land change in India: integrating satellite data with village socioeconomics

Prasanth Meiyappan, Parth S. Roy, Yeshu Sharma, Reshma M. Ramachandran, Pawan K. Joshi, Ruth S. DeFries, Atul K. Jain

Research output: Contribution to journalArticlepeer-review


We examine the dynamics and spatial determinants of land change in India by integrating decadal land cover maps (1985–1995–2005) from a wall-to-wall analysis of Landsat images with spatiotemporal socioeconomic database for ~630,000 villages in India. We reinforce our results through collective evidence from synthesis of 102 case studies that incorporate field knowledge of the causes of land change in India. We focus on cropland–fallow land conversions, and forest area changes (excludes non-forest tree categories including commercial plantations). We show that cropland to fallow conversions are prominently associated with lack of irrigation and capital, male agricultural labor shortage, and fragmentation of land holdings. We find gross forest loss is substantial and increased from ~23,810 km2 (1985–1995) to ~25,770 km2 (1995–2005). The gross forest gain also increased from ~6000 km2 (1985–1995) to ~7440 km2 (1995–2005). Overall, India experienced a net decline in forest by ~18,000 km2 (gross loss–gross gain) consistently during both decades. We show that the major source of forest loss was cropland expansion in areas of low cropland productivity (due to soil degradation and lack of irrigation), followed by industrial development and mining/quarrying activities, and excessive economic dependence of villages on forest resources.

Original languageEnglish (US)
Pages (from-to)753-766
Number of pages14
JournalRegional Environmental Change
Issue number3
StatePublished - Mar 1 2017


  • Agriculture
  • Causes
  • Deforestation
  • Drivers
  • Food security
  • Land use change

ASJC Scopus subject areas

  • Global and Planetary Change


Dive into the research topics of 'Dynamics and determinants of land change in India: integrating satellite data with village socioeconomics'. Together they form a unique fingerprint.

Cite this