3.1. Current potential distribution
The current total suitable habitat area of E. torus Evans is approximately 28.67 km², accounting for 2.99% of the total land area. Of this, highly suitable habitat spans an area of 1.73 km², which is roughly 0.18% of the total land area. These highly suitable habitats are predominantly found in Taiwan, with sporadic distributions in select regions of Guangxi, Guangdong, and Fujian. The moderately suitable habitats, covering an area of 1.07 km² (0.11% of the total land area), are distributed near and around highly suitable habitats (Fig. 2a, Table. 2).
The current total suitable habitat area of R. ferrugineus is estimated to be around 225.65 km², making up approximately 23.50% of the total land area. The highly suitable habitats for this species cover about 4.25 km², or 0.44% of the total land area. These habitats are primarily located in central and southern Yunnan, Hainan Island, Taiwan, and the southeastern coastal regions. The moderately suitable habitats, spanning an area of 5.45 km² (or 0.57% of the total land area), are found surrounding the highly suitable habitats (Fig. 2b, Table.2).
The current total suitable habitat area for the O. arenosella is approximately 4.64 km², which is about 0.48% of the total land area. The highly suitable habitats cover an area of 0.73 km², accounting for around 0.08% of the total land area. These habitats are mainly located in most parts of Hainan Island, with sporadic areas along the coastal regions of Guangxi. The moderately suitable habitats, covering 2 km² (or 0.21% of the total land area), are distributed similarly to the highly suitable habitats and are primarily on Hainan Island (Fig. 2c, Table.2).
3.2. Future changes in suitable habitats
Under the future climate scenarios of SSP 126 and SSP 585 (2041–2060, 2061–2080, 2081–2100), there are varying degrees of changes in the suitable habitat areas of the three invasive insects (Table. 2, Fig .3). The suitable habitat of E. torus Evans exhibited overall stability with some significant expansion in the SSP126 (2081–2100) scenario and substantial contraction in the SSP585 (2081–2100) scenario. The majority of the suitable habitat for R. ferrugineus showed overall contraction, with substantial contraction in the SSP585 (2081–2100) scenario. The suitable habitat of O. arenosella demonstrated a notable overall expansion trend, with the most pronounced expansion observed in the SSP585 (2061–2080) scenario.
In the context of climate change, the potential suitable habitats for the E. torus Evans in China are primarily concentrated in Hainan, Guangdong, Guangxi, Yunnan, southeastern Guizhou, southeastern Tibet, Fujian, and Jiangsu (Fig. 4). In the SSP126 scenario, the potential suitable habitat area of the E. torus Evans slightly decreased in 2041–2060, followed by an increase in 2061–2080 and 2081–2100, resulting in an overall net increase primarily due to the expansion of low-suitability regions. In the SSP126 (2081–2100) scenario, the maximum increase in low-suitability regions was observed, with an expansion of 9.20 km2. However, in the SSP585 scenario, the potential suitable habitat area slightly increased in 2041–2060 but decreased in 2061–2080 and 2081–2100, leading to an overall net decrease mainly due to the reduction in low-suitability regions. In the SSP585 (2081–2100) scenario, the most significant reduction was observed in the low-suitability regions, amounting to 6.43 km2. Overall, the total area of suitable habitat for E. torus Evans remained relatively stable, with slight changes occurring in the peripheral regions (Table.2).
Under the future climate scenarios, the potential suitable habitats for R. ferrugineus in China are mainly found in Hainan, Taiwan, Guangdong, Guangxi, Yunnan, Guizhou, southeastern Tibet, Sichuan, Chongqing, Hubei, Jiangsu, Anhui, Fujian, Jiangxi, and Zhejiang (Fig. 5). There is an overall decrease in the potential suitable habitat area for R. ferrugineus primarily due to a significant reduction in low-suitability areas and a notable increase in non-suitable areas. Under the SSP 126 (2061–2080) scenario, the most significant reduction in the low-suitability area was − 152.46 km², while the most significant increase in non-suitable area was 154.16 km². Similarly, under the SSP585 scenario, there was an overall decrease in the potential suitable habitat area, characterized by a significant reduction in low-suitability areas and a notable increase in non-suitable areas. The most significant reduction in the low-suitability area under the SSP585 (2081–2100) scenario was − 162.92 km², while the most significant increase in the non-suitable area was 167.94 km². Overall, the total suitable habitat area for R. ferrugineus is decreasing, with a gradual reduction at the borders (Table.2).
In the context of climate change, the potential suitable habitats for O. arenosella in China are primarily located throughout Hainan Island and Taiwan, coastal regions of Guangdong and Guangxi, with sporadic distributions in Yunnan and Fujian (Fig. 6). There is an overall increase in the potential suitable habitat area for O. arenosella in future climate scenarios. This is mainly due to a reduction in the non-suitable areas and an increase in the high-, medium-, and low-suitability areas. Under the SSP 126 (2081–2100) scenario, the areas with high, medium, and low suitability all increased. Similarly, under the SSP 585 scenario, there is an overall increase in the potential suitable habitat area, characterized by increases in the high-, medium-, and low-suitability areas. Overall, the total suitable habitat area for O. arenosella is increasing (Table.2).
3.3. Suitability dynamics of individual species
Compared to the current, the overall suitable area for E. torus Evans remains relatively stable with minor variations within the range (Fig. 4). In the SSP126 scenario, compared to the current, the changes in suitable habitats for E. torus Evans in the 2041–2060 and 2061–2080 exhibit relative stability, with an equilibrium between expansion and contraction areas. From 2081 to 2100, there is a larger expansion area, primarily concentrated in the southeastern part of Yunnan Province, southern Guizhou, the entire Guangxi Province, Guangdong Province, Fujian Province, and some regions in Chongqing and Hubei. Under the SSP585 scenario, significant contractions could occur in 2081–2100, primarily occurring in the northwestern part of Yunnan Province, southeastern Guizhou Province, Guangxi Province, Guangdong Province, and Fujian Province. Overall, under both SSP 126 and SSP 585, the range of suitable areas for E. torus Evans does not change significantly.
Compared to the present, the overall suitable area for R. ferrugineus decreases, with some variations in its range, under climate change scenarios, and in these scenarios it is mainly concentrated in southeastern Tibet, Guangdong, Guangxi, Yunnan, Hunan, and Jiangxi (Fig. 5). Under SSP 126, most of the suitable areas remain stable in 2041–2060, with reductions mainly in the coastal areas of Jiangsu, Shanghai, Zhejiang, Yunnan, Guangxi, and eastern and southern Sichuan. Expansions are predicted in southeastern Tibet, southern Hubei, Chongqing, Jiangxi, Fujian, and Taiwan. In 2061–2080, the suitable areas significantly contract, mainly in Zhejiang, Jiangxi, Fujian, Guizhou, Yunnan, Guangxi, Chongqing, and Sichuan, with minor expansions in southeastern Tibet, Hubei, and Zhejiang. Under SSP 585, the suitable areas for R. ferrugineus contract considerably in 2041–2060, 2061–2080, and 2081–2100, primarily in Yunnan, Guangxi, Guangdong, Guizhou, Sichuan, Hunan, Zhejiang, Shanghai, and Jiangsu. Minimal expansion of suitable areas is predicted.
Compared to the present, the overall suitable area for the O. arenosella will significantly expand. Under climate change scenarios, the potential suitable areas for O. arenosella in China are mainly concentrated in the coastal areas of Guangdong, Guangxi, and Taiwan (Fig. 6). Under the SSP126 and SSP585 scenarios, the expansion of suitable areas mainly occurs along the coastal regions of Guangxi, Guangdong, and Taiwan. There are also some expansions in the southern part of Yunnan and the southeastern coastal areas of Fujian. Overall, there is a trend of northward expansion for O. arenosella.
3.4. Centroid distribution transfer of the three invasive insects
The quantitative description of the centroid movement in potential suitable areas under climate change scenarios provides insights into habitat shifts for the invasive insect species. The current centroid of suitable habitats for E. torus Evans is located in the northeastern part of Guangxi Province (111.032664°E, 24.381571°N)(Fig. 2a). Under future climate scenarios, the centroid experiences comparatively smaller shifts, all within Guangxi Province. Overall, the centroid shifts toward the southwest direction under the SSP585 scenario, while it moves toward the south direction under the ssp126 scenario. Under SSP 126, the centroid will move to the southwest (110.811127°E, 24.323439°N) during 2041–2060, southeast (111.193172°E, 24.272996°N) during 2061–2080, and northwest (111.151359°E, 24.322719°N) during 2081–2100. Under SSP 585, the centroid will move to the northeast (111.393446°E, 24.539894°N) during 2041–2060, southwest (111.151826°E, 24.194223°N) during 2061–2080, and continue southwest (110.891164°E, 24.129056°N) during 2081–2100.
The current centroid of suitable habitats for R. ferrugineus is located in the northern part of Guangxi Province (110.225126°E, 25.447637°N) (Fig. 2b). Under future climate scenarios, the centroid experiences relatively small shifts that are all within Guangxi Province. Overall, the centroid shifts toward the southwest direction under the ssp585 scenario and moves toward the east direction under the SSP126 scenario. Under SSP 126, the centroid will move to the southwest (110.069562°E, 25.357406°N) during 2041–2060, southeast (110.14699°E, 25.184387°N) during 2061–2080, and northeast (110.389835°E, 25.302924°N) during 2081–2100. Under SSP 585, the centroid will move to the southwest (109.903544°E, 25.083778°N) during 2041–2060, continue southwest (109.30784°E, 24.947314°N) during 2061–2080, and southeast (109.396298°E, 24.828778°N) during 2081–2100.
The centroid of suitable habitats for O. arenosella experiences simpler shifts, and is currently located in the northeastern maritime area of Hainan Province (111.133218°E, 20.40704°N) (Fig. 2c). Overall, the centroid shifts toward the northeast direction under both the SSP 585 and SSP 126 scenarios. Under SSP 126, the centroid will move to the northeast (111.985993°E, 21.151916°N) during 2041–2060, continue northeast (112.716081°E, 21.182953°N) during 2061–2080, and northwest (112.337654°E, 21.41479°N) during 2081–2100. Under SSP 585, the centroid will move northeast (112.569632°E, 21.43052°N) during 2041–2060, northwest (111.679366°E, 22.171874°N) during 2061–2080, and southwest (111.352594°E, 21.798452°N) during 2081–2100.
3.5. Key drivers of the invasive insect species distributions
Internal jackknife tests of the MaxEnt model reveal the importance of several factors (Table 3). The precipitation of the coldest quarter (bio19, 31.5%), population density (31.4%), annual mean temperature range (bio7, 27.1%), and Human Impact Index (HII, 4.7%) collectively contribute to 94.7% of the E. torus Evans distribution model. Additionally, population density accounts for 5.3% of the overall contribution. For R. ferrugineus, mean temperature of the driest quarter (bio9, 45.1%), precipitation of the driest month (bio14, 16.6%), temperature seasonality (bio4, 9.5%), and HII (8.2%) account for 79.4% of the distribution model. For O. arenosella, the minimum temperature of the warmest month (bio 6, 88.2%), population density (population density, 4.3%), precipitation of the warmest quarter (bio18, 2.5%), and mean diurnal range (bio2, 1.3%) account for 96.3% of the distribution model.
Table.3 Percentage contributions of the bioclimatic variables included in the Maxent models for E. torus Evans, R. ferrugineus and O. arenosella in China.
Species | Ranking of the importance of variables | First | Second | Third | Fourth | Fifth | Sixth | Seventh | Eighth | Ninth | Tenth | Eleventh | Twelfth | Thirteenth | Fourteenth | Fifteenth |
Erionota torus Evans | Variable | bio19 | Population density | bio7 | HII | Ele | NDVI | bio12 | Slope | bio3 | bio13 | bio2 | bio1 | ChinaCover | Lucc | / |
Percent contribution | 31.5 | 31.4 | 27.1 | 4.7 | 1.2 | 0.8 | 0.8 | 0.6 | 0.5 | 0.4 | 0.4 | 0.4 | 0.2 | 0.1 | / |
Rhynchophorus ferrugineus | Variable | bio9 | bio14 | bio4 | HII | Population density | bio2 | bio15 | Ele | bio8 | lucc | bio16 | NDVI | ChinaCover | bio10 | Slope |
Percent contribution | 45.1 | 16.6 | 9.5 | 8.2 | 5.3 | 4.8 | 1.8 | 1.5 | 1.5 | 1.4 | 1.2 | 0.9 | 0.9 | 0.6 | 0.6 |
Opisina arenosella | Variable | bio6 | Population density | bio18 | bio2 | Ele | Aspect index | NDVI | bio8 | ChinaCover | lucc | bio19 | HII | bio15 | / | / |
Percent contribution | 88.2 | 4.3 | 2.5 | 1.3 | 0.9 | 0.5 | 0.4 | 0.4 | 0.4 | 0.4 | 0.4 | 0.2 | 0.1 | / | / |
Note:bio1 = Annual mean temperature;bio2 = Mean diurnal range; bio3 = Isothermality; bio4 = Temperature seasonality; bio5 = Maximum temperature of warmest month; bio6 = Minimum temperature of warmest month; bio7 = Annual mean temperature range; bio8 = Mean temperature of wettest quarter; bio9 = Mean temperature of driest quarter; bio10 = Mean temperature of warmest quarter; bio11 = Mean temperature of coldest quarter; bio12 = Annual precipitation; bio13 = Precipitation of wettest month; bio14 = Precipitation of driest month; bio15 = Precipitation seasonality; bio16 = Precipitation of wettest quarter; bio17 = Precipitation of driest quarter; bio18 = Precipitation of warmest quarter; bio19 = Precipitation of coldest quarter; HII = Human Influence Index; NDVI = Normalized difference vegetation index; ChinaCover = china land cover data; Ele = digital elevation model data; Lucc = the Land use and Cover Change. |