Multidimensional climatic niche conservatism and invasion risk of .

The cotton mealybug , a globally invasive insect, is listed as a national quarantine pest in agriculture and forestry, which seriously threatens biological safety of China. Niche conservatism is a key assumption of species distribution model. An evaluation of the applicability of using ecological niche models to assess the invasion risk of cotton mealybug, and further optimizing model complexity, are of both theoretical and practical significance. Based on 706 occurrence records and key bioclimatic variables, we used -dimensional hypervolume niche analysis method to quantify the climatic niche hypervolumes of this pest in both native and invasive sites, and further tested the niche conservatism hypothesis. MaxEnt model parameters were optimized to predict the invasion risk of the mealybug under current and future climate scenarios in China. The results showed that four climatic variables (annual mean temperature, mean temperature of wettest quarter, mean temperature of warmest quarter, and precipitation of driest quarter) were the key climate factors affecting the distribution of cotton mealybug. Compared with native climatic niche (hypervolume volume, HV=40.43), the niche hypervolume of cotton mealybug in the invasive areas was significantly reduced (HV=6.04). Niche contraction (the net differences between the amount of space enclosed by each hypervolume was 0.84) explained 98.8% of niche differentiation, whereas niche shift (the replacement of space between hypervolumes was 0.01) contributed less than 2%. The direction of climatic niche contraction of the pest in different invasive areas was not exactly consistent. The default parameters of MaxEnt model were unreliable (ΔAICc=14.27), and the optimal parameter combination was obtained as follows: feature combination was linear-quadratic-hinge-product and regularization multiplier was 0.5. The most suitable habitats of cotton mealybug were concentrated in the south of Huaihe River-Qinling Mountains line, and the north-central provinces contained a large area of low suitable habitat. The increase of suitable habitat was not significant at the end of 21 century (SSP1-2.6: 1.7%, SSP5-8.5: 0.7%). The multidimensional climatic niche of was highly conservative. The species distribution model was suitable for analyzing its invasion risk. The northward spread was obvious, and climate change had less impact on the pest.