Multi criteria decision analysis and artificial neural network for assessing soil quality variation under different land use and land cover in Samsun, Türkiye

Soil quality refers to the ability of soil to function within natural or managed ecosystems to maintain productivity, protect environmental quality, and promote plant and animal health. Evaluating and monitoring soil quality (SQ) across various land use types can provide valuable insights for identifying degraded soils and promoting sustainable land practices. This enables land managers and decision-makers to adopt strategies to prevent soil quality degradation. This study aims to assess soil quality indicators under different land use types, focusing on the effects of cultivated land, forest, and grazing land. A total of 54 soil samples were collected from three land use types: agricultural, forest, and pasture lands. Twenty-six soil quality indicators (including physical, chemical, biological, and nutrient properties) were analyzed. These indicators were categorized based on their contribution to overall soil quality, and a standard scoring function was applied to normalize the results. Using Analytic Hierarchy Process (AHP) techniques, the weight of each parameter was determined. The soil quality index (SQI) was calculated by multiplying the scoring function by the assigned weights. Additionally, an artificial neural network (ANN) was used to model SQI in MATLAB. The results classified the SQI into five categories: very low, low, moderate, high, and very high. Most of the study area (48.39%) fell into the moderate category, while 31% showed high to very high soil quality. Only 20.65% exhibited low or very low SQI. Understanding soil quality and its relationship to land use is essential for sustainable land management and ecosystem preservation. • Analyzed 26 soil quality parameters, including physical, chemical, biological, and nutrient properties. • Applied standard scoring and Analytic Hierarchy Process (AHP) techniques to determine soil quality index (SQI). • Modeled soil quality using Artificial Neural Networks (ANN) in MATLAB. • Majority of the area (48.39%) exhibited moderate soil quality, with 31% showing high to very high SQI. • Study highlights the importance of soil quality monitoring for sustainable land management and ecosystem preservation.