Predicting suitable cultivation regions of medicinal plants with Maxent modeling and fuzzy logics: a case study of Scutellaria baicalensis in China

Abstract

Scientific prediction of suitable cultivation regions is an effective way for the assessment of habitat suitability and resource conservation to protect endangered medicinal plants. In recent years, the natural habitat of Scutellaria baicalensis Georgi has been degenerating and disappearing in China owing to excessive market demand of medicinal plant resource. This paper reports a new approach to predict potential suitable cultivation regions and to explore the key environmental factors affecting the content of active ingredients in S. baicalensis using integrated Maxent (maximum entropy) modeling and fuzzy logics. The modeling procedure used 275 occurrence records and baicalin contents of S. baicalensis collected through 2000–2014, and 16 Worldclim environmental factors as well as HWSD soil data. The result showed that six environmental factors (alt, prec7, prec1, bio4, bio1 and t_ph) were determined as key influential factors that mostly affect both the habitat distribution and baicalin content of S. baicalensis. The highly suitable cultivation regions of S. baicalensis mainly distribute (with probability ≥0.50) in the northeast, the north-central and the northwest of China (total 419,857 km2). The statistically significant AUC (area under the curve) value (0.952) of ROC (receiver operating characteristic) curve indicated that integrated Maxent modeling and fuzzy logics could be used to predict the potential suitable cultivation regions of medicinal plants. These results could pave the road for the habitat conservation and resource utilization of endangered medicinal plants.