Non-industrial private forest expansion in Andhra Pradesh


Outside forests, Andhra Pradesh is greening, and we are learning by how much— and why. Our objectives are to (1) map smallholder forest plantations in Andhra Pradesh using multitemporal HLS S10 and/or very-high spatial resolution commercial satellite data, and (2) determine the drivers of plantation forest establishment. We have (1) developed a land use model that integrates land quality and spatial aspects of the farm with market variables and farmer production decisions, (2) implemented a household-level socio-economic survey, and (3) completed a Sentinel 2-era classification that separates natural from planted forest with 94% accuracy. Plantation forestry is rapidly expanding in Asia, and understanding the extent, drivers, and ramifications of these new trees outside forests is vital.



land use, land quality, economics, remote sensing, Machine learning, Julia, LCLUC, land cover, Sentinel, forestry