Climatic niche comparison across a cryptic species complex

According to current molecular evidence, the Chionaspis pinifoliae heterophyllae species complex has been recognized as 10 cryptic species. In this study, we construct potential distribution maps for seven cryptic species based on climatic variables. This was done to assess the main environmental factors that have contributed to the distribution map and test the degree of niche overlap across the seven cryptic species. We used MaxEnt to build the climatic niche models under climatic variables. For these models, the similarities and differences of the niches across the cryptic species were estimated. By comparing the potential distribution model of each cryptic species, our results suggested parapatric, sympatric and allopatry populations for this cryptic species complex. Our results showed high variability in niche overlap, and more often niche conservatism than niche divergence. The current species delimitation of the Chionaspis pinifoliae heterophyllae complex by molecular information and the hypothesis that the niche overlap in the sympatric population is higher than that of the allopatry population were supported based on the findings. This study will provide baseline data and a distribution range to facilitate the further control of these insects and formulate quarantine measures.

138 commission errors (Lobo et al., 2008;Peterson et al., 2008). In addition, this method can not 139 provide information about the spatial distribution of model errors the total extent to which 140 models are carried out highly influences the rate of well-predicted absences, and the AUC scores 141 (Lobo et al., 2008). Thus, an alternative Partial ROC metric approach was employed to model 142 evaluations (Peterson et al., 2008). Partial ROC statistics were implemented using online the 143 Niche Toolbox site (http://shiny.conabio.gob.mx:3838/nichetoolb2/) with 1000 replicates and E 144 = 0.05. To improve predictions made in this study, the predicted continuous suitability maps by 145 MaxEnt were converted into suitable/unsuitable areas (binary habitat) using an applied threshold. 146 Here, the lowest presence threshold (LPT = minimum training presence threshold of MaxENT) 147 was selected for each species (Pearson et al., 2007;Wisz et al., 2008). This threshold can 148 identify the minimum predicted area possible whilst maintaining zero omission error in the 149 training dataset. 154 was used to transform the environmental space of the investigated or selected environmental 155 variables into a two-dimensional space defined by the first and second principal components 156 (Strubbe et al., 2015). The two-dimensional environmental space was then projected onto a 157 100×100 PCA grid of cells bounded by the minimum and maximum PCA values in the 158 background data. This method also corrects the potential sampling bias in occurrence records 160 species complex was measured by the mean of Schoener's D directly from the ecological niche 161 space (Warren et al., 2008). The Schoener's D is an index which varies from 0 (no overlap) to 1 162 (overlap). In addition, there are two common statistical tests considering hypotheses (niche 163 equivalence test and niche similarity test) of niche divergence or conservatism. . For the similarity test, a p value >0.05 was considered to 172 indicate that niches were no more similar than expected by chance. Niche similarity tests wre 173 used in the current study to estimate niche differentiation. The overlap value between two ENMs 174 was either above the 95% confidence interval of the null hypothesis, supporting niche 175 conservatism, or below the 95% confidence interval of the null hypothesis, supporting niche 176 divergence.

177
In addition, the null hypothesis of niche equivalency was also tested using ENMTools 1.4.4 178 (Warren et al., 2010). ENMTools uses MaxEnt to generate an ENM from each species, and then 179 uses the model and predicted suitability scores generated by Maxent for each species to calculate 180 niche equivalency test. This test is based on the metrics of niche overlap (Schoener's D and I) 181 that ranging vary from 0 (no overlap) to 1 (complete overlap) (Warren et al., 2008). The 224 (0.014), S1 and S7 (0.042), S2 and S8 (0.0146), and S2 and S6 (0.097). The low niche overlap 225 suggested that they occupy considerably different environmental niches. All niche overlap values 226 are illustrated in Table 3. 227 According to the identify test, the true calculated niche overlap of all species pairs, are 228 outside of the 95% confidence interval of the null hypothesis ( Figure S7) and confirm the 229 separation between them. Thus, these species pairs showed that the species' ENMs were not 230 equivalent.

231
The above test results indicate that the niches of CPCH species complex are similar, but 232 they are not identical.
233 Geography of speciation 234 The degree of geographic overlap by ENM between the species pairs provides evidence that 235 suggests that some sympatric, parapatric and allopatric species exist in our analysis based on the 236 similarity test. Of the 21 total pairs, two cases of parapatry between species pairs in these species 237 complex were known (S1 and S7; and S1 and S10). Four cases of allopatry between species pairs 238 in these species complex were known (S1 and S2; S1 and S5; S1 and S6; and S1 and S8). In 239 addition, 15 species pairs in the species complex were found to exhibit sympatry. The parapatric 240 species pairs had a small niche overlap (S1 vs S7= 0.041; S1 vs S10=0.231). As predicted, all 241 allopatric (except S1 vs S5 = 0.476) species pairs also suggest a small niche overlap. All 242 speciation types also showed niche divergence in same direction.   Manuscript to be reviewed

Niche comparisons for the cryptic species complex
Niche overlap values are presented for the comparisons of niche similarity and equivalency of species 1 with species 2. All of the comparisons between the species highlight the nonequivalency of their niche.
1 Table 3 Niche comparisons for the cryptic species complex. Niche overlap values are presented for the comparisons of 2 niche similarity and equivalency of species 1 with species 2. All of the comparisons between the species highlight the 3 nonequivalency of their niche.