Forecasting Future Land Use and Land Cover Changes Using Spatial Modeling: A Case Study of Phitsanulok Province

Abstract

Background and Objectives: Rapid urban expansion and infrastructure development along economic corridor frameworks in Phitsanulok Province, which lies at the intersection of the East-West Economic Corridor (EWEC) and the North-South Economic Corridor (NSEC), have continuously driven land-use and land-cover (LULC) changes in the province. These ongoing changes directly affect habitat quality and ecosystem integrity. This study aimed to (1) analyze the spatiotemporal dynamics of LULC change in Phitsanulok Province across three time points, namely 2007, 2014, and 2021; (2) forecast future LULC patterns to the year 2028 using the CA–Markov model; and (3) assess the impacts of land-use change on habitat quality and habitat degradation using the InVEST Habitat Quality module.

Methodology: Satellite imagery was analyzed using Landsat 5 TM data for 2007 and Landsat 8 OLI/TIRS data for 2014 and 2021, both at 30-meter spatial resolution, obtained from the USGS Earth Explorer archive. Dry-season imagery was selected to minimize interference from cloud cover. Land use was classified into five categories—agricultural land, forest land, urban and built-up area, miscellaneous land, and water body—using the Maximum Likelihood Classification (MLC) technique. The CA-Markov model was then validated by using the 2007 and 2014 LULC data to simulate a land-use map for 2021, which was compared against the actual reference map from the Land Development Department. Once validated, the 2014–2021 transition probability matrix was applied to forecast land use for 2028. In parallel, habitat quality and habitat degradation were assessed using the InVEST Habitat Quality module, incorporating four threat factors—urban and built-up areas, agricultural land, major roads, and railways—with parameters assigned following a literature-based approach drawn from the InVEST User's Guide and relevant tropical-context studies.

Main Results: The CA–Markov model demonstrated satisfactory predictive performance for land-use forecasting, achieving an Overall Accuracy of 0.87 and a Kappa coefficient of 0.79, indicating a high level of agreement between the simulated and actual land-use maps. The 2028 projection showed that urban and built-up areas were expected to increase from 6.08% to 6.42% of the provincial area, while agricultural land was projected to expand by 6.41%, and forest land showed a slight recovery of 0.57%. In contrast, miscellaneous land was projected to decline sharply, by as much as 40.51%, indicating that this land type functioned as the primary reservoir absorbing conversion into both agricultural and urban built-up areas as the province continued to develop. Assessment using the InVEST Habitat Quality model further showed that overall habitat quality declined from 0.544 in 2007 to 0.528 in 2028, while the corresponding habitat degradation index rose from 0.137 to 0.154 over the same period. Spatially, areas experiencing high habitat degradation expanded outward from the urban core toward peri-urban zones and the surrounding agricultural areas, particularly along the major road network and the transitional zones bordering forest edges. Notably, these areas of high degradation coincided closely with the same locations classified as having very-low habitat quality, underscoring a consistent spatial pattern linking urban and infrastructure expansion to ecological deterioration.

Conclusions: The coupled CA-Markov and InVEST Habitat Quality modeling framework effectively demonstrates the land-use dynamics of Phitsanulok Province, showing that these changes are driven primarily by transportation systems and continued urban expansion, both of which exert a direct and inverse effect on ecosystem quality. Although the province's extensive forest cover helps to moderate the overall provincial-level rate of decline in habitat quality, urban fringes, major roads, and the transition zones between urban areas and forest land remain at considerable long-term risk of habitat fragmentation. The knowledge generated through this study can serve as a foundation for supporting urban planning efforts, the delineation of ecological buffer zones, and the management of urban expansion, all of which are necessary to achieve a sustainable balance between continued economic development and the conservation of biodiversity in the province.