Arabica coffee, as one of the world's three native coffee species, requires rational planning for its growing areas to ensure ecological and sustainable agricultural development. This study aims to establish a decision-making framework using Geographic Information Systems (GIS) and Multi-Criteria Decision-Making (MCDM), with a focus on assessing the habitat suitability of Arabica coffee in Yunnan Province, China. The impacts of climate, topography, soil, and socio-economic factors were considered by selecting 13 criteria through correlation analysis. Indicator weights were determined using the Best-Worst Method (BWM), while weighted processing was conducted using the Base-Criterion Method (BCM). Sensitivity analysis was performed to verify the accuracy and stability of the model. Additionally, several decision models were evaluated to investigate regionalizing Arabica coffee habitats in Yunnan. The results highlighted that minimum temperature during the coldest month is crucial for evaluation purposes. The BWM-GIS model identified suitable areas comprising 13.55% of the total area as most suitable, 27.46% as suitable, and 59.00% as unsuitable, whereas corresponding values for the BCM-GIS model were 9.97%, 30.43%, and 59.59%. Despite employing different decision-making methods, both models yielded similar and consistent results. The suitable areas mainly encompass Dehong, Pu'er, Lincang, Xishuangbanna, Baoshan, southern Chuxiong, eastern Honghe, southern Yuxi, and parts of Wenshan. BWM-GIS achieved an area under curve (AUC) value of 0.891, while BCM-GIS obtained an AUC value of 0.890, indicating the stability and reliability of the models. Among them, the evaluation process of BCM-GIS was simpler and more realistic. Therefore, it has high feasibility and practical value in practical application. The findings from this study provide a significant scientific foundation for optimizing Yunnan Province.
Keywords: Arabica coffee; Base-criterion method; Best–worst method; Habitat suitability; Multi-criteria decision-making; Planting regionalization.
© 2024. The Author(s).