Remote Sensing Tells Kinship: Agricultural Land use Based Study in Rural Bangladesh

Abstract

Remote sensing senses the land cover, which is indirectly linked to social structure, economic condition, and environmental issues. For instance, collateral information from remote sensing can be used in certain economic conditions and social aspects such as kinship detection, family structure, economic condition, habitat fragmentation, biodiversity, and structure of the villages. The Manikdi Mouza, Netrokona, Bangladesh was chosen as a case study because of its homogeneous landscape. This study conducted a household questionnaire survey, involving 370 respondents with a 95% confidence level whereas the margin of error is 5%. CS, RS, and Worldview-3 images were used to map the agricultural boundary and interpret the associated family structure changes over the study period. Besides that, three primary methods such as cluster analysis, multiple regression modeling, and Fuzzy Cognitive Mapping (FCM) were employed to examine plot parcelization, identify the driving forces influencing land parcelization, and determine their respective weights. The results indicate that, over the study period from 1950 to 2023, the total number of plots increased from 359 to 1,048 due to land fragmentation, while 404 plots remained unchanged. Unaltered plots are associated with combined families, while fragmented plots correspond to nuclear families. From the Cadastral Survey (CS) to the Revisional Survey (RS), approximately 26.98% of plots remained unchanged, and from the RS to the present, 38.55% of plots have stayed unaltered. The study's investigation into agricultural land fragmentation and its effects on kinship dynamics in rural Bangladesh highlights crucial implications for local governance and national policy frameworks, advocating for informed, community-centric strategies in land management.

J. Asiat. Soc. Bangladesh, Sci. 51(1): 51-72, June 2025