Prediction of Potentially Suitable Distributions of Codonopsis pilosula in China Based on an Optimized MaxEnt Model

Species distribution models are widely used in conservation biology and invasive biology. MaxEnt the most among existing modeling tools. In process, default parameters often to build model. However, these tend be overfit. Aiming at this problem, study uses an optimized model analyze impact of past, present future climate on distributions Codonopsis pilosula , economic species, provide a theoretical basis for its introduction cultivation. Based 264 records eight environmental variables, potential areas C. last interglacial, middle Holocene current periods 2050 2070 were simulated. Combined with percentage contribution, permutation importance, jackknife test, factors affecting suitable area species discussed. The results show that optimal are: regularization multiplier is 1.5, feature combination LQHP (linear, quadratic, hinge, product). main temperature annual mean temperature, diurnal range, isothermality. precipitation seasonality, wettest quarter, driest which average contributes species. With warming, exhibits northward expansion trend. It estimated 2070, will expand maximum, reaching 2.5108 million square kilometers. highly mainly Sichuan, Gansu, Shaanxi, Shanxi, Henan Provinces. Our findings can support related avoiding blind .