Evaluation of the factors explaining the use of agricultural land: A machine learning and model-agnostic approach

• Machine learning and model-agnostic approaches are relevant in agricultural land research. Biophysical bioclimatic variables explained most global interpretability. Drainage density is important for globally explaining wheat, maize, olive grove plantation. Socioeconomic became equally at the parcel level. Degree of mechanisation influences specific parcels wheat. To effectively plan manage use land, it crucial to identify evaluate multiple human environmental factors that influence it. In this study, we propose a model framework potentially plantations regional By developing machine-learning coupled with approach, provide local interpretations influential factors. We collected nearly 140 related biophysical, bioclimatic, socioeconomic conditions. Overall, results indicated biophysical conditions were more than At interpretation level, proposed identified strong contribution drainage density, slope, soil type. contrast, level such as degree could be As demonstrated, analytical approach has potential serve decision-making tool instrument better control land.