Nexus of urbanization and changes in agricultural land in Bangladesh

Agriculture is the backbone of the economy of many countries; however, loss of agricultural land has become widespread due to urbanization-induced land use/cover change (LU/CC) worldwide. Understanding the spatial distribution of agricultural land and its changes is important for policy development. This study is aimed at investigating the spatiotemporal changes in agricultural lands and evaluating their relationship with urbanization. Landsat and Nightime Light (NTL) satellite images were utilized. The Normalized Difference Vegetation Index (NDVI), Modified Normalized Difference Water Index (MNDWI), and Normalized Difference Built-up Index (NDBI) were used to classify agricultural land cover types using the Decision Tree (DT) algorithm. The digital number (DN) of the NTL images was used to segregate urban and rural areas. Moreover, the lighting pixels were used to classify districts into four categories, e.g., high, moderate, low, and very low urbanized areas. Subsequently, we computed basic statistics of the changes in agricultural lands and enumerated yearly loss or gain during 26 years, 1992–2018. This study indicated that Bangladesh experienced a significant loss in agricultural lands; however, the loss rates varied. A high rate of change was observed in rural areas compared to urban areas. The results also showed a strong coefficient of determination (R2 = 0.59) between changes in agricultural land and lighting areas in urban areas of Bangladesh.