From Farmland to Cityscape: Urban Growth Simulation in Surkhet Valley, Nepal Using Remote Sensing and CA-Markov Modeling

Urbanization is rapidly transforming the spatial and socioeconomic landscape of many emerging cities in Nepal, yet relatively little research has explored these dynamics outside the Kathmandu Valley. This study applies a cellular automata-Markov (CA-Markov) model to simulate and predict land use and land cover (LULC) changes in Surkhet Valley, the core of Birendranagar Municipality, one of Nepal’s fastest-growing urban centers. Using Landsat imagery from 1999, 2009, and 2019, alongside spatial and socioeconomic factors, the model captures historical LULC transitions and projects future changes for the years 2029, 2039, and 2049. Model validation was conducted against the 2019 classified LULC map, yielding an overall agreement of 80.65% and a standard kappa statistic of 70.31%, confirming the model’s predictive reliability. Results indicate a clear trajectory of urban expansion at the expense of agricultural land. Built-up surfaces is projected to more than double – from 12.43 km² in 2019 to 31.38 km² in 2049, while cultivated land is expected to decline by over 20 km² in the same period. Spatial analysis shows urban growth intensifying around existing centers, highways, and transitional ecotones between forest and cultivation zones. Compared to similar studies in Kathmandu and Biratnagar, Surkhet exhibits a higher normalized rate of urban expansion, highlighting its emerging role in regional development. This research underscores the value of remote sensing and spatial modeling in urban planning and land management. The findings provide essential insights for policymakers to guide sustainable development in Surkhet and other rapidly urbanizing areas across Nepal.