Phytolacca americana L. is an invasive perennial plant that has become increasingly widespread in China, but its current climatic suitability and future redistribution under climate change remain insufficiently quantified. This study aimed to identify the major environmental drivers of P. americana distribution and to project its potential habitat suitability under future climate scenarios. We compiled a national occurrence dataset and retained 683 quality-controlled presence records after taxonomic verification, coordinate checking, and 5 km spatial thinning. A BIOMOD2 ensemble modelling framework was used to integrate nine algorithms, and future projections were generated using CMIP6 climate data under SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5 across four time periods from 2021 to 2100. The ensemble model showed strong predictive performance, with TSS = 0.804 and ROC = 0.967. May shortwave radiation, January mean temperature, and annual temperature range were identified as the dominant predictors of habitat suitability. Under current climate conditions, highly suitable habitats were mainly concentrated in warm and humid regions of eastern and southern China. Future projections indicated that suitable habitats may expand toward northern, northwestern, and higher-elevation regions, whereas highly suitable habitats may become redistributed or fragmented under stronger climate forcing. Centroid analyses further suggested non-linear, scenario-dependent shifts rather than a simple poleward expansion. These findings provide a spatial basis for early warning, targeted monitoring, and pathway-focused management of P. americana in China. 5. Conclusions Using a performance-filtered BIOMOD2 ensemble, this study offers a high-accuracy assessment of the climatic drivers and future distribution of P. americana in China. Three coordinated variables—May shortwave radiation, January mean temperature, and annual temperature range—emerged as dominant controls, jointly shaping early-season growth, overwinter survival, and tolerance to thermal variability. These mechanisms explain both the species’ current concentration in warm–humid eastern China and its projected expansion into northern and higher-elevation regions as climates warm. Future scenarios indicate that while geographic opportunity increases, high-quality habitats may fragment under stronger forcing, producing non-linear, pathway-specific centroid shifts rather than a simple poleward trend. This highlights the central role of climatic variability and extreme events in invasion dynamics. Our findings refine climate-driven invasion theory by integrating physiological mechanisms into distribution forecasts and provide actionable guidance for surveillance timing, pathway management, and habitat interventions aimed at limiting spring energy availability. Figure 1. Distribution sites of P american in China after data cleaning and spatial filtering. Red dots represent the occurrence records used in the BIOMOD2 ensemble modelling.The dashed lines in the lower-right inset indicate the maritime boundary lines associated with the South China Sea Islands in the standard base map and are shown for cartographic reference only; they were not used as occurrence records or environmental predictors in the modelling analysis. Figure 1. Distribution sites of P american in China after data cleaning and spatial filtering. Red dots represent the occurrence records used in the BIOMOD2 ensemble modelling.The dashed lines in the lower-right inset indicate the maritime boundary lines associated with the South China Sea Islands in the standard base map and are shown for cartographic reference only; they were not used as occurrence records or environmental predictors in the modelling analysis. Figure 2. Correlation analysis of candidate environmental predictors for P. americana distribution modelling. Numbers in the lower triangle represent Pearson’s correlation coefficients (r), while the upper triangle visualizes the magnitude and direction of correlations. This analysis was used to identify collinearity among variables and to retain ecologically meaningful predictors for subsequent BIOMOD2 ensemble modelling. Figure 2. Correlation analysis of candidate environmental predictors for P. americana distribution modelling. Numbers in the lower triangle represent Pearson’s correlation coefficients (r), while the upper triangle visualizes the magnitude and direction of correlations. This analysis was used to identify collinearity among variables and to retain ecologically meaningful predictors for subsequent BIOMOD2 ensemble modelling. Figure 3. Partial response curves for key environmental predictors of P. americana distribution in the BIOMOD2 ensemble model. The blue curves represent the predicted probability of presence in response to changes in Srad08, Srad10, and Prec10, and the green dashed lines indicate the threshold value used for suitability interpretation. These response relationships suggest that the occurrence probability of P. americana is sensitive to seasonal variation in solar radiation and precipitation. Figure 3. Partial response curves for key environmental predictors of P. americana distribution in the BIOMOD2 ensemble model. The blue curves represent the predicted probability of presence in response to changes in Srad08, Srad10, and Prec10, and the green dashed lines indicate the threshold value used for suitability interpretation. These response relationships suggest that the occurrence probability of P. americana is sensitive to seasonal variation in solar radiation and precipitation. Figure 4. Spatial distribution of P. americana in China. Grey indicates unsuitable area, green indicates low suitability, yellow indicates moderate suitability, and red indicates high suitability. The predicted suitable habitats are mainly distributed in central, eastern, and southern China, with highly suitable areas concentrated in regions with favorable hydrothermal conditions. Figure 4. Spatial distribution of P. americana in China. Grey indicates unsuitable area, green indicates low suitability, yellow indicates moderate suitability, and red indicates high suitability. The predicted suitable habitats are mainly distributed in central, eastern, and southern China, with highly suitable areas concentrated in regions with favorable hydrothermal conditions. Figure 5. Spatial distribution of habitat suitability for P. americana in China under the SSP1-2.6 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable are. Figure 5. Spatial distribution of habitat suitability for P. americana in China under the SSP1-2.6 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable are. Figure 6. Spatial distribution of habitat suitability for P. americana in China under the SSP2-4.5 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 6. Spatial distribution of habitat suitability for P. americana in China under the SSP2-4.5 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 7. Spatial distribution of habitat suitability for P. americana in China under the SSP3-7.0 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 7. Spatial distribution of habitat suitability for P. americana in China under the SSP3-7.0 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 8. Spatial distribution of habitat suitability for P. americana in China under the SSP5-8.5 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 8. Spatial distribution of habitat suitability for P. americana in China under the SSP5-8.5 scenario during four future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). Grey indicates unsuitable area, green indicates low suitability, yellow indicates suitable area, and red indicates highly suitable area. Figure 9. Changes in the area of suitable habitats for P. americana under future climate scenarios. ( a) 2021–2040; ( b) 2041–2060; ( c) 2061–2080; ( d) 2081–2100. Green, light green, yellow, and red bars represent changes in total, low-suitability, moderately suitable, and highly suitable habitat areas, respectively. Figure 9. Changes in the area of suitable habitats for P. americana under future climate scenarios. ( a) 2021–2040; ( b) 2041–2060; ( c) 2061–2080; ( d) 2081–2100. Green, light green, yellow, and red bars represent changes in total, low-suitability, moderately suitable, and highly suitable habitat areas, respectively. Figure 10. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP1-2.6 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 10. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP1-2.6 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 11. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP2-4.5 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 11. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP2-4.5 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 12. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP3-7.0 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 12. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP3-7.0 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 13. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP5-8.5 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Figure 13. Standard deviation ellipses and centroid shifts of suitable habitats for P. americana under the SSP5-8.5 climate scenario. The ellipses represent the spatial distribution ranges of suitable habitats during the current period (1970–2000) and future periods (2021–2040, 2041–2060, 2061–2080, and 2081–2100). The inset shows the migration trajectory of the habitat centroid across different periods. Table 1. Environmental variables considered for P. americana distribution modelling. Table 1. Environmental variables considered for P. americana distribution modelling. Variable Description Unit BIO01 Annual Mean Temperature °C BIO02 Mean Diurnal Range °C BIO03 Isothermality (BIO2/BIO7) (×100) - BIO04 Temperature Seasonality - BIO05 Max Temperature of Warmest Month °C BIO06 Min Temperature of Coldest Month °C BIO07 Temperature Annual Range (BIO5–BIO6) °C BIO08 Mean Temperature of Wettest Quarter °C BIO09 Mean Temperature of Driest Quarter °C BIO10 Mean Temperature of Warmest Quarter °C BIO11 Mean Temperature of Coldest Quarter °C BIO12 Annual Precipitation mm BIO13 Precipitation of Wettest Month mm BIO14 Precipitation of Driest Month mm BIO15 Precipitation Seasonality mm BIO16 Precipitation of the Wettest Quarter mm BIO17 Precipitation of the Driest quarter mm BIO18 Precipitation of Warmest Quarter mm BIO19 Precipitation of Coldest Quarter mm Elev Elevation m Prec01-12 Monthly precipitation from January to December mm Srad01-12 Monthly solar radiation from January to December kJ/(m 2 d) Wind01-12 Monthly wind speed from January to December m/s Vapr01-12 Monthly vapor pressure from January to December kPa Tmin01-12 Monthly minimum temperature from January to December °C Tmax01-12 Monthly maximum temperature from January to December °C Tavg01-12 Monthly mean temperature from January to December °C Table 2. Criteria used to evaluate model performance based on AUC, TSS, and Kappa. Table 2. Criteria used to evaluate model performance based on AUC, TSS, and Kappa. Indicator Excellent Good Fair Poor Fail AUC 1.00–0.90 0.90–0.80 0.80–0.70 0.70–0.60 0.50–0.00 TSS 1–0.85 0.85–0.70 0.70–0.55 0.55–0.4 0.40–0.00 Kappa 1–0.85 0.85–0.70 0.70–0.55 0.55–0.4 0.40–0.00 Table 3. Centroid displacement distance and direction of suitable habitats for P. americana under future climate scenarios. Table 3. Centroid displacement distance and direction of suitable habitats for P. americana under future climate scenarios. Scenario Period Displacement Distance (km) Direction/Azimuth (°) Main Direction SSP1-2.6 2021–2040 514.3 93.2 East-northeast SSP1-2.6 2041–2060 817.2 101.5 East-northeast SSP1-2.6 2061–2080 525.6 352.8 Northwest/north-northwest SSP1-2.6 2081–2100 754.1 28.7 Northeast SSP2-4.5 2021–2040 232.5 145.2 Southeast SSP2-4.5 2041–2060 461.3 84.6 East-northeast SSP2-4.5 2061–2080 383.1 98.3 East SSP2-4.5 2081–2100 253.8 183.5 South SSP3-7.0 2021–2040 732.4 82.1 East-northeast SSP3-7.0 2041–2060 638.3 86.3 East-northeast SSP3-7.0 2061–2080 549.8 24.5 Northeast SSP3-7.0 2081–2100 324.8 116.7 Southeast SSP5-8.5 2021–2040 742.8 104.3 East-southeast SSP5-8.5 2041–2060 784.5 110.7 East-southeast SSP5-8.5 2061–2080 603.2 137.5 Southeast SSP5-8.5 2081–2100 300.6 201.8 Southwest Direction is expressed as degrees clockwise from north.
