Prediction of potential suitable habitats of Aphis sp. in Ningxia under future climate scenarios based on ensemble model

Prediction of potential suitable habitats of Aphis sp. in Ningxia under future climate scenarios based on ensemble model | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Prediction of potential suitable habitats of Aphis sp. in Ningxia under future climate scenarios based on ensemble model Zhou Chenhong, Liu Yao, Liu xiaoli, Jing Bo, Li Fusheng, He Qi, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7258894/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Aphis sp. is a major pest of Goji berry in Ningxia, significantly affecting its yield and quality. To explore the potential distribution changes of Aphis sp. under climate change, this study used data from 104 valid occurrence points collected from 2019 to 2024 in the main Goji-producing areas of Ningxia, combined with 19 climatic variables. The Biomod2 ensemble modeling framework, incorporating 10 individual models, was employed to predict the current and future (2030s–2090s) suitable habitats for Aphis sp. under four climate scenarios (SSP126, SSP245, SSP370, and SSP585).The results indicate that under current climatic conditions, highly suitable habitats for Aphis sp. are primarily concentrated in the central and northern regions of Ningxia, with a total suitable area of 3,902.67 km², accounting for 5.88% of the region’s total land area. Environmental factor analysis revealed that the mean temperature of the coldest quarter (Bio11), temperature seasonality (Bio4), and annual mean temperature (Bio1) are the key variables influencing the distribution of Aphis sp. , with a combined contribution rate of 41.2%. The ensemble models (EMca and EMwmean) demonstrated significantly higher predictive accuracy (AUC > 0.95, TSS > 0.89) compared to individual models. In particular, the EMca model more effectively captured fluctuations in the extent of suitable habitats. Under four climate scenarios, the suitable habitat area for Aphis sp . is projected to expand significantly, with the greatest increase observed under the SSP370 scenario, reaching 40,723 km² by the 2090s. Moreover, the suitable range is expected to shift from the central-northern region toward the northwest and southwest. This study provides a theoretical foundation for the targeted management of Aphis sp . in Ningxia and highlights the need to closely monitor the impact of climate warming on the expansion of their suitable habitat. Earth and environmental sciences/Climate sciences Biological sciences/Ecology Earth and environmental sciences/Ecology Earth and environmental sciences/Environmental sciences Aphis sp. Ensemble model Climate change Prediction of suitable area Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction As an important economic crop in the north, Goji berry is highly susceptible to insect pests due to its luxuriant branches and leaves. Among them, Aphis sp. is the largest pest population, which harms Goji berry by stinging and sucking, seriously affects the yield and quality of Goji [ 1 ] . Aphis sp. occurs more frequently and has a close relationship with the meteorological environment [ 2 ] , and occurs over a long span of time in a year, resulting in a high number of prevention and treatment times and high application costs [ 3 ] . In recent years, Goji berry is the leading industry with the most local characteristics and brand advantages in Ningxia Hui Autonomous Region, with the abnormal changes in the climate and the expansion of the planting area year by year, it has made the occurrence of Aphis sp. more and more serious [ 4 ] . It is of great significance to understand the spatial distribution of Aphis sp. and the environmental factors affecting its propagation and spread, to analyze the suitable area of Aphis sp. in Ningxia and to conduct risk assessment for scientific and effective prediction of Aphis sp. and active prevention and control. Species distribution models (SDMs), as a method used to calculate the potential suitable distribution area of organisms, has been widely used in the prediction of the potential distribution area of species under the condition of climate change by combining the location information of the known distribution points of organisms with the data of the habitat factors in the study area, and simulating the potential distribution of species through a specific model [ 5 – 6 ] . In recent years, with the wide application of 3S and other technologies in the study of the relationship between species distribution and environmental factors, ecological niche model simulation of species distribution has become a research hotspot. Liu Y et al. used the Maximum Entropy (Maxent) habitat modeling analysis method to evaluate the habitat suitability of the purple marten population for the first time in the context of multiple resolution scales [ 7 ] . Peng et al. used the Maxent model to predict the habitat suitability of Solidago decurrens in China [ 8 ] . At present, most of the studies on species' fitness zones use single models (Ji Yelin et al. 2019; Ye Jiangxia et al. 2021; Guo et al. 2024) [ 9 – 11 ] , like generalized linear model [ 12 ] and so on. However, their principles, assumptions and algorithms are diverse, with differences in the scope of application and predictive performance, different results may be obtained by using different modeling techniques for the same species, and different methods often have specific differences among species [ 13 ] . More and more researchers are working to establish the integration of multiple models to mitigate instability caused by different principle algorithms or differences in input data [ 14 ] . Biomod2, an integrated species distribution prediction platform developed based on the R language, is capable of comprehensively evaluating current species and integrating the model with better accuracy in the model to predict species, thus maximizing the accuracy of the model and predicting the accuracy of future species distribution [ 15 ] . In order to solve the instability of the prediction performance of a single model when facing a large amount of data, this paper uses the Biomod2 program package based on the free and open-source R language, and uses 10 models: generalized linear model (GLM), generalized boosted regression models (GBM), generalized boosted regression models (GBM), and generalized boosted regression models (GBM). regression models (GBM), generalized additive model (GAM), classification and regression tree analysis (CTA), artificial neural network (ANN), and the generalized linear model (GLM). network (ANN), Flexible discriminant analysis (FDA), Surface range envelope (SRE), MaxEnt, Multivariate adaptive regression model (MARM), and the model for the classification and regression tree analysis (CTA). (SRE), Maximum Entropy Model (MaxEnt), Multivariate adaptive regression splines (MARS), and Random forest (RF) [ 16 ] to construct a combined model to predict the fitness zone of Aphis sp . in Ningxia, China, under the current and future climatic conditions, and to explore and analyze the fitness pattern of Aphis sp. in Ningxia Hui Autonomous Region to provide certain theoretical support for its prevention and control work. Data and methods Meteorological and geographic information data acquisition Climate is the main environmental factor that determines the distribution of species, and this study used 19 (Biol-19) climatic variables with a spatial resolution of (0.0083°x0.0083°) in the Worldclim Global Climate Database ( http://www.worldclim.org/curent ) to study the climatically suitable distribution area of the Goji berry aphid. The 19 environmental variables including annual mean temperature (Bio1), mean diurnal range (Bio2), isothermality (Bio3), temperature seasonality (Bio4), max temperature of warmest month (Bio5), min temperature of coldest month (Bio6), temperature annual range (Bio7) and so on are shown in Table 1 [ 17 ] . Future climate scenarios select the sixth international coupled model (CMIP6) shared socio-economic pathway (SSP) [ 18 ] , compared with the typical concentration emission pathway (RCP) proposed in the fifth international coupled model (CMIP5), the CMIP6 model integrates the consideration of the direction of socio-economic development, and integrates the consideration of the SSP and the RCP, and puts forward four socio-economic development scenarios [ 19 ] :1) SSP126, which represents zero carbon emission, low economic development and low radiative forcing; 2) SSP245 represents current carbon emission, economic development level and medium radiative forcing; 3) SSP370 represents low economic growth, relatively high carbon emission and radiative forcing; 4) SSP585 represents high carbon emission, high economic growth and high radiative forcing. Since other possible correlation factors, such as elevation, slope direction, slope, solar radiation, vegetation index, etc., were not considered, the distribution range of Aphis sp. predicted in this study is its potential distribution area and does not represent its actual distribution area. Table 1 Description of environmental variables Variables Description of variables Unit Variables Description of variables Units bio1 Annual mean temperature ℃ bio11 Mean temperature of coldest quarter ℃ bio2 Mean diurnal range ℃ bio12 Annual precipitation mm bio3 Isothermality - bio13 Precipitation in the wettest month mm bio4 Temperature seasonality - bio14 Precipitation in the driest month mm bio5 Max temperature of warmest month ℃ bio15 Seasonal coefficient of variation of precipitation - bio6 Min temperature of coldest month ℃ bio16 Wettest quarter precipitation mm bio7 Temperature annual range ℃ Bio17 Precipitation in the driest quarter mm bio8 Mean temperature of wettest quarter ℃ bio18 Hottest quarterly precipitation mm bio9 Mean temperature of driest quarter ℃ bio19 Coldest quarterly precipitation mm bio10 Mean temperature of warmest quarter ℃ In addition to model selection, how to screen for good explanatory variables is also a difficulty in conducting species spatial distribution simulation studies [ 20 ] . To achieve the best arithmetic results, in most cases, researchers screen the environmental factors input into the model through their own experience [ 21 ] . This method, although simple and easy to follow, is highly subjective. In order to reduce the uncertainty caused by the combination of environmental factors, this study adopts the method of repeated random sampling of environmental factors, i.e., randomly selecting n out of N climate factors as explanatory variables and repeating it m times, i.e., randomly generating m groups of climate factor combinations. For each combination of climate factors, simulations are conducted separately with 10 spatial distribution models to find the combination of environmental factors with universal significance for different models. In this paper, for 19 climate factors, 5 are randomly selected each time and repeated 30 times, and each environmental factor combination is numbered in order. The basic geographic information data were mainly provided by Ningxia Meteorological Research Institute, including administrative boundaries (provincial, municipal and county boundaries), which were used for the spatial characterization of the meteorological conditions of Aphis sp. disaster and the mapping of the results of disaster risk assessment. Aphis sp. data acquisition and processing The Aphis sp. data were mainly obtained from the Institute of Agricultural Economics and Information Technology of Ningxia Academy of Agriculture and Forestry, as well as the manual survey organized by this project. The distribution data of Aphis sp. collected from 21 large-scale Goji berry planting bases in the main Goji berry producing areas of Ningxia during the 6-year period of 2019–2024 were selected, and the point data totaled 1998. In order to avoid spatial auto correlation, ENMTools [ 22 ] was used to remove redundant points and automatically match the raster size of environmental factors used in the analysis, instead of deleting the data based on the distance method, and finally 104 valid points were retained. Five-point sampling was used for each monitoring point, and two Goji berry trees were randomly investigated at each point, and one branch was randomly selected from each tree in five directions: east, south, west, north, and center, and the number of different insect states of Aphis sp. within 30 cm of the branch was recorded, including the number of alate aphids and nymphs. Construction of the habitat suitability model and evaluation Biomod2 was run using R-Studio sourced from the website https://www.rstudio.com/products/rstudio/ and R speech version (4.2.1) sourced from https://mirrors.tuna.tsinghua.edu.cn/CRAN/ . When building the model, Biomod2 needs two sets of data, the presence points and the missing points of the species. Since the missing points are difficult to obtain, Biomod2 uses its own "random" command to randomly generate 10 sets of 1,000 pseudo-missing points, and randomly sets 80% of the constructed dataset as the training set, and 20% of the remaining data as the test set. 80% of the constructed data set is used as the training set, and the remaining 20% is used as the test set. We set the same weights for the existence points and pseudo-missing points, and set the parameters of a single model to the default values, and run each model 10 times to produce a total of 1,000 sets of training results (10 models x 10 sets of data x 10 operations), and normalize the model results. We used three of the most widely used current model evaluation metrics, Kappa, True Skill Statistic(TSS) and Area Under Curve(AUC) [ 23 – 24 ] .Kappa was used to assess the agreement between the sample data and the simulation results. AUC was used as a criterion to evaluate the simulation results of the model using the Receiver operating characteristic (ROC) curve. TSS is a modified method based on Kappa, which retains the advantages of Kappa and corrects the disadvantage that Kappa is affected by the wide range of species distributions, and the AUC value usually ranges from 0.5-1, with the closer the AUC value is to 1, the better the prediction effect is. Kappa and TSS values can be modeled by sensitivity and specificity of the difference between true positive and false positive rates, and the closer the difference is to 1, the better the prediction is. When TSS > 0.7, AUC > 0.8 and Kappa > 0.6,it indicates that the model has good predictive performance [ 25 – 26 ] . In the construction of the combined model, all three indicators were taken into account, and from the model results, a single model with excellent results (TSS > 0.8, Kappa > 0.8, AUC > 0.9) was screened out, and these models were outputted a comprehensive result by weighted average method, and Ensemble Model based on weighted mean(EMwmean). The other is the cluster analysis method, the similar prediction results of each model are grouped into one category, and representative models are selected in each cluster category for integration, which can construct the combined model- Ensemble Model committee averaging (EMca) [ 27 ] . Results and analysis Model optimization and accuracy evaluation Comparing the 10 models in Biomod2, the Kappa, AUC and TSS values predicted by a single model are shown in Fig. 1 and Table 2 . It can be seen that for predicting the aptitude zone of Goji berry in Ningxia by single model, the simulation effect of GBM is the best, and the Kappa, TSS and AUC reached 0.97, 0.99 and 0.99, respectively. Followed by RF and ANN, which both can reach Kappa > 0.8, TSS > 0.9 and AUC > 0.9. The worst performance is SRE with Kappa of 0.17, TSS of 0.42 and AUC of 0.71.The GBM, GAM, ANN, RF and MAXENT models are the models with high accuracy and good stability under different evaluation indexes. The worst performing model, SRE, failed the KAPPA and TSS score accuracy tests. The other four models performed between the above models. Table 2 Kappa, TSS and AUC values of different models GLM GBM GAM CTA ANN SRE FDA MARS RF MAXENT KAPPA 0.48 0.97 0.74 0.65 0.85 0.17 0.70 0.76 0.80 0.86 TSS 0.62 0.99 0.90 0.89 0.96 0.42 0.84 0.87 0.93 0.85 AUC 0.81 0.99 0.98 0.96 0.99 0.71 0.95 0.97 0.94 0.97 From the model results, the excellent models with TSS values greater than 0.8 were screened out, and the final results were constructed by constructing numerous simple models and then averaging or weighted averaging them, which can effectively avoid the overfitting risk of the algorithmic parameter estimation brought about by too many predictors and a small number of observations. The final prediction results obtained by the two combined models are Kappa > 0.90, AUC > 0.95, and TSS > 0.89, indicating that the combined models have excellent prediction effects and are better than a single model. Importance of environmental factors From the results of the importance of environmental variables (Fig. 2 ), the six environmental variables with the highest contribution to the Ningxia Aphis sp. in descending order were the mean temperature of coldest quarter (Bio11), the temperature seasonality (Bio4), the annual mean temperature (Bio1), the hottest quarterly precipitation (Bio18), the mean temperature of warmest quarter (Bio10), and the precipitation in the wettest month (Bio13), respectively. Among them, the mean temperature of the coldest season directly affected the Aphis sp. 's overwintering survival rate and population base, and became the core limiting factor for the prediction of the aphid's aptitude zone in Ningxia. Among all the environmental factors, the total contribution rate of temperature factor was 41.2%, the contribution rate of precipitation factor was 34.3%, and the contribution rate of terrain factor was 12.2%. It shows that the effect of precipitation factor on the distribution of Aphis sp. was higher than that of temperature factor and terrain factor. Prediction of current Aphis sp. habitat distribution Through 10 single models and 2 ensemble models, the results of predicting the suitable area of Aphis sp. in China with environmental variables in the current period are shown in Fig. 3 . With the same precision prediction results, the simulation effect of GBM is the best among the models, and it is the closest to the 2 ensemble models both in terms of the scale of the suitable area and the division of the different levels of the suitable area. On the contrary, SRE had the worst effect, with a large area of highly suitable area for Aphis sp. , which was inconsistent with the reality. As can be seen from Fig. 3 , the range of the predicted fitness zones of Aphis sp. under the current climatic conditions is mainly located in the north-central region of Ningxia. Among them, the high aptitude area is located in the central part of Zhongning County, the northeastern part of Xixia District of Yinchuan City and scattered areas in the southern part of Helan County; the medium aptitude area is located in the central part of Pingluo County of Shizuishan City, the northern part of Xixia District of Yinchuan City and central part of Helan County, the central part of Litong District of Wuzhong City and scattered areas of Qingsongxia City, the northeastern part of Shapotou District of Zhongwei City and central part of Zhongning County; and the low aptitude area is concentrated in the northern part of Litong District of Wuzhong City and central part of Qingsongxia City, Zhongwei City Zhongning County, and the eastern part of Haiyuan County. The output result of the model is the existence probability of Aphis sp. , using ArcGIS software to convert ASCII data into Raster data, to get the existence probability distribution of Aphis sp. . The prediction results of the Aphis sp. in Ningxia were classified into risk levels using the "natural break point classification method", and quantitatively analyzed by the probability of suitability(P,P∈[0,1]). The classification criteria were: non-suitability (P < 0.4), low suitability (0.4 ≤ P < 0.6), medium suitability (0.6 ≤ P < 0.8) and high suitability (P ≥ 0.80) [ 28 ] . The reclassification tool was used to count the proportion of regional pixels at each level and to calculate the area of different levels of distribution areas by combining the land use data [ 29 ] . The area of Aphis sp . fitness zone with different degrees in the current period of the two combined models was obtained in Table 4 . Table 4 Current suitable area of Aphis sp. (km 2 ) city EMca EMwmean High Moderate High Moderate Moderate Moderate Low Wuzhong 1.00 333.91 737.43 79.11 375.11 547.82 Guyuan 8.13 31.21 10.88 57.37 19.25 80.25 Zhongwei 1231.21 543.75 1062.33 1428.74 597.22 501.11 Shizuishan 0.00 8.74 39.21 5.16 13.17 87.37 Yinchuan 23.53 26.12 45.56 294.32 25.75 96.16 Total 1263.87 943.73 1695.41 1864.70 1030.50 1312.71 From Table 4 , it can be seen that the area of suitable area for Aphis sp. in EMca of the integrated model is 3902.67 km 2 , accounting for 5.88% of the total land area (66,400 km 2 )in Ningxia [ 30 ] , and the areas of high, medium and low suitable areas are 1263.87 km 2 、830.50 km 2 and 902.71 km 2 , accounting for 36.08%、26.94% and 26.94%; of the total area of suitable areas 36.08%、26.94% and 48.38%, respectively; the suitable area of EMwmean model was 4207.91 km 2 , accounting for 6.34% of the total land in the Ningxia region, and the areas of high, medium, and low suitable areas were 1864.7km 2 , 1030.5km 2 , and 1312.71km 2 , accounting for 44.31%, 24.49%, 31.49%, and 31.71%, respectively, of the total area of the suitable area. The total area of suitable area is 1864.7km 2 , 1030.5km 2 and 1312.71km 2 respectively, accounting for 44.31%, 24.49% and 31.20% of the total area of suitable area. The suitable areas for Ningxia Aphis sp. in the current period obtained by the two models are basically the same and more concentrated, which are mainly distributed in Helan County in the north and Lingwu City in the south of Yinchuan City, Pingluo County and Huinong District in the southeast of Shizuishan City, Zhongning County in the north of Zhongwei City, and Hongsibao District in Wuzhong City. Ningxia Aphis sp. high suitable area mainly in Zhongning County, Tongxin County, Huinong District, Helan County and other places, the climate in these areas is relatively warm and humid, which is highly conducive to the reproduction of Aphis sp. species, particularly during the spring and summer when high temperatures promote rapid population expansion of aphids. [ 31 ] . Among these, the cultivation of Goji berry in Shizuishan City is more dispersed, leading to a more scattered distribution of Aphis sp. . As a result, these areas exhibit a moderate to low risk of aphid infestation, consistent with the findings of the two combination models. When compared to the current distribution of Aphis sp. in Ningxia, the results of the two models were more accurate in EMca than in EMwmean. Prediction of suitable growing area of Aphis sp. under future climate scenarios Based on the two species distribution ensemble models developed in this study, simulations were conducted to predict the suitability distribution range of Aphis sp. in Ningxia under the influence of future climate change. Using the spatial distribution pattern of Aphis sp. under current climatic conditions, derived from the more accurate ensemble model EMca as a reference, four future climate scenarios were integrated to predict changes in the distribution of potential habitats at different time intervals (2030s, 2050s, 2070s, and 2090s). Following the current suitability distribution area classification standard [ 32 ] , the spatial distribution of Aphis sp. under future climate scenarios was also categorized into four levels, and the area of each level was calculated to determine the distribution of potentially suitable habitats in future periods. Figure 4 illustrates the changes in the extent of potentially suitable areas under various future scenarios predicted by the two models. Based on the combined analysis of changes in potentially high-risk areas for Aphis sp. and the spatial variations shown in Table 5 and Fig. 5 , it is evident that under the SSP126 scenario, the most rapid expansion occurs in the 2030s and 2090s, with average increases of 501% and 557%, respectively, compared to the current suitable area. Under the SSP245 scenario, the greatest growth is projected for the 2070s, with an average increase of 540%. Under the SSP370 scenario, the highest expansion is observed in the 2090s, with an average increase of 726%. For the SSP585 scenario, while the suitable area begins expanding in the 2040s, the fastest growth is expected in the 2070s and 2090s, with average increases of 550% and 515%, respectively, relative to the current suitable area. Comparing the two models in terms of total suitable area, the EMwmean model indicated a consistent and significant increase in the suitable habitat of Aphis sp. across different periods and future climate scenarios. In contrast, the EMca model, except under the SSP370 scenario, also showed an overall increasing trend; however, in the other three scenarios, a decline in suitable area was observed during certain periods. These reductions were mainly concentrated in the northern part of Zhongwei and along the border of Wuzhong City, and the magnitude of increase was notably smaller than that predicted by the EMwmean model. The simulation results from EMca exhibited noticeable fluctuations but corresponded more closely with actual observations, demonstrating a better overall simulation performance. Therefore, the EMca model was selected for predicting future suitable habitats of Aphis sp. , and the potential risk areas in the Ningxia Autonomous Region under future climate change scenarios were identified (Fig. 6 ). Table 5 Comparison of suitable areas of Aphis sps . in Ningxia under current and future climatic conditions (× 10 2 km 2 ) Decade Decade EMwmean SSP125 SSP245 SSP370 SSP585 SSP125 SSP245 SSP370 SSP585 2030s 158.65 -15.54 14.62 16.05 305.89 146.5 119.57 143.81 2050s -13.74 150.9 142.46 86.10 28.31 364.04 232.10 343.59 2070s 148.97 113.29 128.6 220.11 286.34 393.77 412.62 391.43 2090s 187.25 95.66 282.92 234.35 317.12 380.76 468.53 348.53 "-": indicates area reduction compared to current climate conditions; SSP: shared socioeonomie pahways As can be seen from Fig. 6 , compared with the predicted results under the current climatic conditions, the total suitable area of Aphis sp. was increasing, among which the high and medium suitable areas in Yinchuan and Shizuishan City increased significantly; the boundary of the low suitable area showed a clear tendency to shift southeast, especially in the cities of Zhongwei and Wuzhong. The predicted total suitable area of Aphis sp. under current climatic conditions is 39.03×10 2 km 2 , accounting for 5.88% of Ningxia’s total area. Under the SSP370 scenario in the 2090s, the total suitable area increased the most, reaching 407.23×10 2 km 2 , or 61.33% of Ningxia’s total area. The predicted highly suitable area of Aphis sp. under current climatic conditions is 15.75×10 2 km 2 . Under future climatic conditions, the area of highly suitable habitat decreases, even the area will be reduced to zero under the scenarios of SSP245, SSP370, and SSP585 in the 2030s, and under the SSP126 and SSP370 scenarios in the 2050s. The area of medium-suitable habitat under current conditions is 8.31×10 2 km 2 . In the future, the area of medium-suitable habitat increases under all scenarios, with the largest area under the SSP370 scenario in the 2090s, reaching 221.62×10 2 km 2 . The area of low-suitable habitat under current climatic conditions is 9.01×10 2 km 2 . In the future, the area of low-suitable habitat increases under all scenarios, except for the SSP126 and SSP585 scenarios in the 2050s. Under projected climate change scenarios, the potential suitable habitats for Aphis sp. in Ningxia were primarily concentrated in the cities of Zhongwei, Wuzhong, Yinchuan, and most parts of Shizuishan, with the overall distribution pattern remaining largely stable. Compared to the baseline period (current), the total area suitable for Aphis sp. increased significantly, particularly in Yinchuan and Shizuishan. The suitable areas expanded from the north-central region toward the northwestern and southwestern parts of Ningxia. The model predicted a similar overall trend in habitat suitability for Aphis sp. across all four climate scenarios; however, the predicted suitability levels for the same time periods varied between scenarios. For example, during the 2030s, the scenarios produced varying predictions, with the most notable difference being that areas identified as highly or moderately suitable in the SSP126 scenario were classified as having low suitability in the other three scenarios. In the 2050s, Yinchuan and most of Shizuishan were projected to be moderately suitable under the SSP245 and SSP370 scenarios, but were predicted to be low or unsuitable under SSP126 and SSP585. During the 2070s and 2090s, the extent of high and medium suitability areas under the SSP126 scenario was significantly larger than in the other scenarios, with the most suitable zones concentrated in Yinchuan and northwestern Zhongwei. These areas may represent high-risk zones for potential Aphis sp. outbreaks. Discussion In this study, we used an integrated model combined with different climate scenarios to predict the potential distribution of Aphis sp. in Ningxia. By analyzing the climate data under the current climate and four future climate scenarios (SSP126, SSP245, SSP370, SSP585), we found that the area of the suitable zone for the Aphis sp. showed significant changes under different climate scenarios in different periods in the future. Overall, future climate warming will likely expand the suitable habitat zone for Aphis sp. , with the most pronounced increase in suitable habitat area occurring under the high-emission scenario (SSP585) [ 33 ] . Among the two combined models used in this study, the EMca model provided more realistic predictions compared to the EMwmean model [ 34 ] . The EMca model better captured fluctuations in Aphis sp. fitness area size, particularly in spatial distribution changes over time and across different climate scenarios. Unlike the relatively smooth predictions of the EMwmean model, the EMca model accounted for more complex climatic influences, such as fluctuations in temperature extremes and precipitation patterns, which provided more accurate simulations of Aphis sp. Distribution [ 35 – 36 ] . The environmental factor analysis indicated that temperature and precipitation were key climatic factors influencing the distribution of Aphis sp. In particular, the mean temperature of coldest quarter (Bio11) and the temperature seasonality (Bio4) were the most important factors influencing the prediction of suitable areas for the aphid. The temperature factor, especially the temperature change in the coldest season, directly affects the aphid's overwintering survivability and population base [ 37 – 38 ] , and is the core limiting factor for the distribution of Aphis sp. in the suitable habitat range in Ningxia. In this study, Ensemble Models were introduced to predict the potential habitat of Aphis sp. in Ningxia to improve the accuracy and stability of prediction. Compared with the single modeling method, especially the widely used MaxEnt, the Ensemble Models make up for the limitations of the single model prediction results by integrating multiple modeling algorithms (e.g., GBM, RF, GLM, etc.) in various aspects [ 39 – 40 ] . The results show that the integrated model can better simulate the potential distribution area of Aphis sp. under different climate scenarios, with higher AUC and TSS evaluation indexes, reflecting stronger adaptability and generalization ability. The study provides a theoretical reference for the prevention and control strategy of characteristic economic crop pests in the northern region of China, and also provides a methodological demonstration for similar studies in the future. The study still has some shortcomings. Regarding the selection of environmental variables, only climate factors were selected as model inputs, failing to cover other ecological factors such as topography, vegetation, soil, etc. [ 41 ] , which led to the simulation results focusing more on reflecting the influence of climatic conditions on the fitness distribution of Aphis sp. , and making it difficult to comprehensively reveal their spatial distribution characteristics in different ecological contexts [ 42 ] . Secondly, the model was constructed based on the basic ecological niche theory, which only considered the influence of abiotic factors on the species distribution, and ignored the regulating effect of biotic factors [ 43 – 44 ] , such as the migratory ability of Aphis sp. , the interference of natural enemies, and the competitive relationship. The distribution of species in real ecosystems is the result of the joint action of abiotic factors and biotic interactions, so the ecological niche range predicted by the model only represents an idealized state, which is often wider than the real distribution range [ 45 ] . Conclusion Based on the integrated model and climate data under different climate scenarios, this study predicted the potential habitat of Aphis sp. in Ningxia. The results showed that future climate change, especially the increase in temperature, will likely lead to the expansion of the fitness zone of Aphis sp. , especially under the high emission scenario (SSP585), where the increase in the area of the fitness zone is particularly obvious. Nevertheless, the changes in the suitable area under different scenarios varied greatly and the fluctuation of the area of the suitable area in different time periods was more significant, which provides an important reference for future Aphis sp. prevention and control work. This study provides theoretical support for the control of Aphis sp. and suggests that local governments and agricultural administrations should strengthen the monitoring of Aphis sp. potential risk areas and take appropriate preventive and control measures, especially under climate change scenarios, to adjust prevention and control strategies in time to cope with the possible impacts of Aphis sp. However, this study was limited to predicting the effects of climate on Aphis sp . and analyzing its distribution and habitat suitability within the Ningxia region. Several avenues for enhancing future research are apparent. First, the integration of multi-source data and the inclusion of additional ecological variables, such as soil type, vegetation cover, and elevation combined with an expansion of the study to a national scale, would significantly improve the model’s ecological adaptability. Second, the incorporation of biological interactions, particularly the relationships between Aphis sp . and its host plants or natural enemies, would facilitate the development of a dynamic model that better reflects real-world ecological processes. Third, model validation should be further strengthened through comparative analysis with field monitoring data, thereby improving the predictive accuracy and overall reliability of the model. Declarations Competing interests The authors declare no competing interests. Funding This research was funded by the National Natural Science Foundation of China (NNSFC) under Project Number 32160639, 2024 Ningxia Hui Autonomous Region Young Top Homo Sapiens Talent Cultivation Project, 2025 New Era Meteorological High-level Scientific and Technological Innovation Homo Sapiens Talent Program. Author Contribution Z.H.: Conceptualization, methodology, writing - original draft,writing-review and editing. L.Y.: Conceptualization, methodology, software, formal analysis and writing - original draft. S.L.: Writing - review and editing, supervision. L.L. and H.Q.: Resources, writing and formal analysis. L.S.:Investigation, resources and funding acquisition. J.B., M.J. and L.Y.: Investigation. Data Availability The authors confirm that the data supporting the findings of this study are available within the article, and further requests can be directed to the corresponding author. References Jiahui et al. Lycium barbarum studies: A system review on molecular biology, cultivation, and quality characteristics of goji berries. Biochem. Syst. Ecol. 121 , 105020 (2025). Panwar, T. S. & Singh, S. B. Vinod Kumar Garg. Influence of meteorological parameters on population dynamics of thrips (Thrips tabaci Lindeman) and aphid (Aphis gossypii Glover) in Bt and non Bt cotton at Malwa region of Madhya Pradesh. J. Agrometeorology . 17.1 , 136–138 (2015). Chaieb et al. Chemical composition and aphicidal potential of Citrus aurantium peel essential oils. Entomol. Generalis . 37 , 63–75 (2018). Wu, Y. et al. Investigating the impact of climate warming on phenology of aphid pests in China using long-term historical data. Insects 11 (3), 167 (2020). Wang, F., Yuan, X., Sun, Y. & Liu, Y. Species distribution modeling based on MaxEnt to inform biodiversity conservation in the Central Urban Area of Chongqing Municipality.Ecological Indicators. 158 , 111491. (2024). Hosseini, N. & Hossein Mostafavi. Habitat potential modelling and the effect of climate change on the current and future distribution of three Thymus species in Iran using MaxEnt. Sci. Rep. 14 (1), 3641 (2024). Liu, Y. H. et al. Habitat suitability evaluation of sable (Martes zaibellina) based on MAXENT model in Beijicun National Nature Reserve Heilongiiang Province, China. Acta Ecol. Sin. 44 (2), 559–569 (2024). Peihao, P. Feng Yu.Habitat suitability evaluation for Paeonia decomposita based on a MaxEnt model. Eco.mont. J. Protected Mountain Areas Res. 14 (1), 25–32 (2022). Ji Yelin, S. & Xiyou, Y. Potential habitat prediction of Hyphantria cunea based on a random forest model in China. J. Nanjing Forestry Univ. 43 (6), 121–128 (2019). Jiangxia, Y. & Jingwen, W. Zhang Mingsha,Zhou Ruliang & Shi Lei. Risk Pattern Analysis of Hyphantria cunea Based on Spatial Matrix Model and0-1 Measure.Scientia silvae sinicae. 57(1),140–152 (2021). Guo, Y. et al. Prediction of Chinese suitable habitats of Panax notoginseng under climate change based on MaxEnt and chemometric methods. Sci. Rep. 14 , 16434 (2024). Li, S. et al. Estimation and inference for high-dimensional generalized linear models with knowledge transfer. J. Am. Stat. Assoc. 119 , 1274–1285 (2024). Takolander, A. et al. Cross-realm transferability of species distribution models–Species characteristics and prevalence matter more than modelling methods applied. Ecol. Model. 499 , 110950 (2025). Yang, L. Qing Hua. Predicting suitable habitats of parasitic desert species based on Biomod2 ensemble model: Cynomorium songaricum rupr and its host plants as an example. BMC Plant Biol. 25 (1), 351 (2025). Gu, R. et al. Predicting the impacts of climate change on the geographic distribution of moso bamboo in China based on biomod2 model. Eur. J. For. Res. 143 (5), 1499–1512 (2024). Zhaio et al. Analyzing potential distribution and disturbance intensity of plateau pika in the source region of the Yellow River via BIOMOD2 integrated model. Chin. J. Ecol. 43 (4), 1192 (2024). Sosa-Guillén, P. et al. Bioclimatic indicators dataset for the orographically complex Canary Islands archipelago. Sci. Data . 11 (1), 1323 (2024). Deepa, R. & Kumar, V. Suchithra Sundaram. A systematic review of regional and global climate extremes in CMIP6 models under shared socio-economic pathways. Theoret. Appl. Climatol. 155 (4), 2523–2543 (2024). Xu, C. Yu-Hao Mao, & Hong Liao. Future ozone changes and their impacts on vegetation and human health in China under the Shared Socio‐economic Pathways. J. Geophys. Research: Atmos. 129 (4), (2024). Austin, M. P., Kimberly, P. & Van Niel Improving species distribution models for climate change studies: Variable selection and scale. J. Biogeogr. 38 (1), 1–8 (2011). Wideen, M. Jolie Mayer-Smith,& Barbara Moon. A critical analysis of the research on learning to teach: Making the case for an ecological perspective on inquiry. Rev. Educ. Res. 68 (2), 130–178 (1998). Warren, D. L. et al. ENMTools 1.0: an R package for comparative ecological biogeography. Ecography 44 (4), 504–511 (2021). Allouche, O. Ronen Kadmon. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). J. Appl. Ecol. 43 (6), 1223–1232 (2006). Fielding, A. H., John, F. & Bell A review of methods for the assessment of prediction errors in conservation presence/absence models. Environ. Conserv. 24 (1), 38–49 (1997). Leroy, B. et al. Without quality presence–absence data, discrimination metrics such as TSS can be misleading measures of model performance. J. Biogeogr. 45 (9), 1994–2002 (2018). Abrego, N. & Otso Ovaskainen. &. Evaluating the predictive performance of presence–absence models: Why can the same model appear excellent or poor? Ecol. Evol. 13 (12), (2023). Wang, Z. et al. Predicting the Future Geographic Distribution of the Traditional Chinese Medicinal Plant Epimedium acuminatum Franch. in China Using Ensemble Models Based on Biomod2. Plants 14 (7), 1065 (2025). Song, Y. et al. Synergizing multiple machine learning techniques and remote sensing for advanced landslide susceptibility assessment: a case study in the Three Gorges Reservoir Area. Environ. Earth Sci. 83 (8), 227 (2024). Toure, S. I. et al. Land cover and land use change analysis using multi-spatial resolution data and object-based image analysis. Remote Sens. Environ. 210 , 259–268 (2018). Wang, Y. et al. A dynamic evaluation of land use and habitat quality in Ningxia from 1980 to 2020, china. Appl. Ecol. Environ. Res. 23 (1), (2025). Ebert, T. A. Cartwright. Biology and ecology of Aphis gossypii Glover (Homoptera: aphididae). Southwest. Entomol. 22 (1), 116–153 (1997). Schroth, G., Läderach, P., Martinez-Valle, A. I., Bunn, C. & Jassogne, L. Vulnerability to climate change of cocoa in West Africa: patterns, opportunities and limits to adaptation. Sci. Total Environ. 556 , 231–241 (2016). Liao, F. et al. Future change in the Vietnam upwelling under a high-emission scenario. Geophys. Res. Lett. 51 (11), (2024). Shafizadeh-Moghadam, H. et al. Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping. J. Environ. managemen . 217 , 1–11 (2018). Easterling, D. R. et al. Climate extremes: observations, modeling, and impacts. Science 289 (5487), 2068–2074 (2000). Kharin, V. V. et al. Changes in temperature and precipitation extremes in the CMIP5 ensemble. Clim. Change . 119 (2), 345–357 (2013). Bale, J. S. Hayward. Insect overwintering in a changing climate. J. Exp. Biol. 213. (6), 980–994 (2010). Liu, J. et al. Impact of temperature on survival rate, fecundity, and feeding behavior of two aphids, Aphis gossypii and Acyrthosiphon gossypii, when reared on cotton. Insects 12 (6), 565 (2021). Ahmadi, M. et al. MaxEnt brings comparable results when the input data are being completed; Model parameterization of four species distribution models. Ecol. Evol. 13 (2), e9827 (2023). Valavi, R. et al. Predictive performance of presence-only species distribution models: a benchmark study with reproducible code. Ecol. Monogr. 92 (1), e01486 (2022). Mod, H. K. et al. What we use is not what we know: environmental predictors in plant distribution models. J. Veg. Sci. 27 (6), 1308–1322 (2016). Elith, J. & John, R. Leathwick. Species distribution models: ecological explanation and prediction across space and time. Annual review of ecology, evolution, and systematics. 40(1), 677–697 (2009). Hirzel, A. H. Gwenaëlle Le Lay. Habitat suitability modelling and niche theory. J. Appl. Ecol. 45 (5), 1372–1381 (2008). Anderson, R. P. When and how should biotic interactions be considered in models of species niches and distributions? J. Biogeogr. 44 (1), 8–17 (2017). Wisz, M. et al. The role of biotic interactions in shaping distributions and realised assemblages of species: implications for species distribution modelling. Biol. Rev. 88 (1), 15–30 (2013). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7258894","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":501719687,"identity":"f6f7f61f-d206-4e56-87a1-de5e95bafb60","order_by":0,"name":"Zhou Chenhong","email":"","orcid":"","institution":"Key Laboratory of Ecosystem Carbon Source and Sink,CMA","correspondingAuthor":false,"prefix":"","firstName":"Zhou","middleName":"","lastName":"Chenhong","suffix":""},{"id":501719688,"identity":"4d4dc062-d946-4a4a-b06c-92dc6f8dc4bb","order_by":1,"name":"Liu Yao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAx0lEQVRIiWNgGAWjYBACPmYwVcPD2N7Y+PADMVrYIFqOyTD3HG42liBKC4RitmGfkd4mwEOUFnYeM4kfFWw8vDMftjFIMNjJ6TYQdBiPmWTPGRkeydmJbQ8KGJKNzQ4QoUWCt42Nx3B2YruBBMOBxG3EaJH828bMY3/zYJsED7FapHmBWhhnMBKtha3YWubMMR7GnkRgIBsQ4Rd+/sMbb76pqLFnbD/+8OGHCjs5glqAgAUpAg0IKwcBZqKSySgYBaNgFIxgAADvtDdw5m/HXAAAAABJRU5ErkJggg==","orcid":"","institution":"Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculture in Arid Regions,CMA","correspondingAuthor":true,"prefix":"","firstName":"Liu","middleName":"","lastName":"Yao","suffix":""},{"id":501719689,"identity":"a425d7f3-bb38-4916-b667-9971a0d11734","order_by":2,"name":"Liu xiaoli","email":"","orcid":"","institution":"NingXia Academy of Agriculture and Forestry Sciences，Institute of Plant Protection","correspondingAuthor":false,"prefix":"","firstName":"Liu","middleName":"","lastName":"xiaoli","suffix":""},{"id":501719692,"identity":"1e1df1a8-dfdb-4c03-b082-a1d4ae9cd392","order_by":3,"name":"Jing Bo","email":"","orcid":"","institution":"Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculture in Arid Regions,CMA","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Bo","suffix":""},{"id":501719693,"identity":"736519da-d9f7-4f5a-8679-61b382b7c062","order_by":4,"name":"Li Fusheng","email":"","orcid":"","institution":"Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculture in Arid Regions,CMA","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Fusheng","suffix":""},{"id":501719694,"identity":"7499547f-54fa-44a8-af72-d49029e7a6f2","order_by":5,"name":"He Qi","email":"","orcid":"","institution":"NingXia Academy of Agriculture and Forestry Sciences，Institute of Crop Sciences","correspondingAuthor":false,"prefix":"","firstName":"He","middleName":"","lastName":"Qi","suffix":""},{"id":501719695,"identity":"9039bf69-bc2d-4501-88f6-14c703c8011d","order_by":6,"name":"Ma Jin","email":"","orcid":"","institution":"NingXia Academy of Agriculture and Forestry Sciences","correspondingAuthor":false,"prefix":"","firstName":"Ma","middleName":"","lastName":"Jin","suffix":""},{"id":501719696,"identity":"707aa9a0-45b9-4b94-9027-0c3ac4651ae8","order_by":7,"name":"Li Yuan","email":"","orcid":"","institution":"Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculture in Arid Regions,CMA","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Yuan","suffix":""},{"id":501719697,"identity":"1eafc2a7-cf8f-401f-98f3-fb8defa56669","order_by":8,"name":"Sun Yanling","email":"","orcid":"","institution":"Key Laboratory for Meteorological Disaster Monitoring and Early Warning and Risk Management of Characteristic Agriculture in Arid Regions,CMA","correspondingAuthor":false,"prefix":"","firstName":"Sun","middleName":"","lastName":"Yanling","suffix":""}],"badges":[],"createdAt":"2025-07-31 06:53:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7258894/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7258894/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89853392,"identity":"64fd3b02-0adb-4cf7-a276-b06f063a16c3","added_by":"auto","created_at":"2025-08-25 18:10:10","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":257994,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of Kappa and TSS and AUC evaluations of 10 models\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/64cdb46c2c2444ad1fa616e0.png"},{"id":89852791,"identity":"bad161ac-8cd7-4b27-ab6c-e9a70019e0b6","added_by":"auto","created_at":"2025-08-25 18:02:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":208991,"visible":true,"origin":"","legend":"\u003cp\u003eImportance of environmental variables based on ensemble model\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/89c425a90ea1327708da9f67.png"},{"id":89852413,"identity":"f2bec3ad-fe57-499d-ae4a-cc008322a802","added_by":"auto","created_at":"2025-08-25 17:54:10","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":851957,"visible":true,"origin":"","legend":"\u003cp\u003ePrediction of the current habitability zone of 10+2 combinations of models\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/ca7cfc25cc41d556d17911c2.png"},{"id":89852416,"identity":"874aa3e8-eb38-4892-9a89-28ad8f07024f","added_by":"auto","created_at":"2025-08-25 17:54:11","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":240683,"visible":true,"origin":"","legend":"\u003cp\u003ePotential habitat area change of \u003cem\u003eAphis sps\u003c/em\u003e. under 4 climate scenarios (a.SSP126, b.SSP245, c.SSP370, d.SSP585)\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/c2867cedb7f70d558e576a1d.png"},{"id":89852414,"identity":"da61dd6d-57a3-4ed5-97a2-65dad4adcd69","added_by":"auto","created_at":"2025-08-25 17:54:11","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1666805,"visible":true,"origin":"","legend":"\u003cp\u003ePotential habitat range change of \u003cem\u003eAphis sps\u003c/em\u003e. under climate scenarios\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/2fada9c8564819a4d3fbac3e.png"},{"id":89852800,"identity":"0e1f97a2-5027-4780-99c3-15a546ef951f","added_by":"auto","created_at":"2025-08-25 18:02:11","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1008718,"visible":true,"origin":"","legend":"\u003cp\u003ePotential habitat distribution of \u003cem\u003eAphis sp. \u003c/em\u003eunder 4 climate scenarios based on EMca\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/dfaa81fa551622f6aad48ba9.png"},{"id":99794258,"identity":"3c49aa17-22d6-4c98-84eb-c55adf813cb7","added_by":"auto","created_at":"2026-01-08 13:34:22","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5188975,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7258894/v1/99e4b453-064b-4dcb-af40-1e9a52dbb7fd.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Prediction of potential suitable habitats of Aphis sp. in Ningxia under future climate scenarios based on ensemble model","fulltext":[{"header":"Introduction","content":"\u003cp\u003eAs an important economic crop in the north, Goji berry is highly susceptible to insect pests due to its luxuriant branches and leaves. Among them, \u003cem\u003eAphis sp.\u003c/em\u003e is the largest pest population, which harms Goji berry by stinging and sucking, seriously affects the yield and quality of Goji\u003csup\u003e[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]\u003c/sup\u003e. \u003cem\u003eAphis sp.\u003c/em\u003eoccurs more frequently and has a close relationship with the meteorological environment \u003csup\u003e[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]\u003c/sup\u003e, and occurs over a long span of time in a year, resulting in a high number of prevention and treatment times and high application costs\u003csup\u003e[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]\u003c/sup\u003e. In recent years, Goji berry is the leading industry with the most local characteristics and brand advantages in Ningxia Hui Autonomous Region, with the abnormal changes in the climate and the expansion of the planting area year by year, it has made the occurrence of \u003cem\u003eAphis sp.\u003c/em\u003e more and more serious\u003csup\u003e[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]\u003c/sup\u003e. It is of great significance to understand the spatial distribution of \u003cem\u003eAphis sp.\u003c/em\u003e and the environmental factors affecting its propagation and spread, to analyze the suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia and to conduct risk assessment for scientific and effective prediction of \u003cem\u003eAphis sp.\u003c/em\u003e and active prevention and control.\u003c/p\u003e\u003cp\u003eSpecies distribution models (SDMs), as a method used to calculate the potential suitable distribution area of organisms, has been widely used in the prediction of the potential distribution area of species under the condition of climate change by combining the location information of the known distribution points of organisms with the data of the habitat factors in the study area, and simulating the potential distribution of species through a specific model\u003csup\u003e[\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]\u003c/sup\u003e. In recent years, with the wide application of 3S and other technologies in the study of the relationship between species distribution and environmental factors, ecological niche model simulation of species distribution has become a research hotspot. Liu Y et al. used the Maximum Entropy (Maxent) habitat modeling analysis method to evaluate the habitat suitability of the purple marten population for the first time in the context of multiple resolution scales\u003csup\u003e[\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]\u003c/sup\u003e. Peng et al. used the Maxent model to predict the habitat suitability of \u003cem\u003eSolidago decurrens\u003c/em\u003e in China\u003csup\u003e[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]\u003c/sup\u003e. At present, most of the studies on species' fitness zones use single models (Ji Yelin et al. 2019; Ye Jiangxia et al. 2021; Guo et al. 2024)\u003csup\u003e[\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]\u003c/sup\u003e, like generalized linear model\u003csup\u003e[\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003c/sup\u003e and so on. However, their principles, assumptions and algorithms are diverse, with differences in the scope of application and predictive performance, different results may be obtained by using different modeling techniques for the same species, and different methods often have specific differences among species\u003csup\u003e[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]\u003c/sup\u003e. More and more researchers are working to establish the integration of multiple models to mitigate instability caused by different principle algorithms or differences in input data\u003csup\u003e[\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]\u003c/sup\u003e. Biomod2, an integrated species distribution prediction platform developed based on the R language, is capable of comprehensively evaluating current species and integrating the model with better accuracy in the model to predict species, thus maximizing the accuracy of the model and predicting the accuracy of future species distribution\u003csup\u003e[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn order to solve the instability of the prediction performance of a single model when facing a large amount of data, this paper uses the Biomod2 program package based on the free and open-source R language, and uses 10 models: generalized linear model (GLM), generalized boosted regression models (GBM), generalized boosted regression models (GBM), and generalized boosted regression models (GBM). regression models (GBM), generalized additive model (GAM), classification and regression tree analysis (CTA), artificial neural network (ANN), and the generalized linear model (GLM). network (ANN), Flexible discriminant analysis (FDA), Surface range envelope (SRE), MaxEnt, Multivariate adaptive regression model (MARM), and the model for the classification and regression tree analysis (CTA). (SRE), Maximum Entropy Model (MaxEnt), Multivariate adaptive regression splines (MARS), and Random forest (RF)\u003csup\u003e[\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]\u003c/sup\u003e to construct a combined model to predict the fitness zone of \u003cem\u003eAphis sp\u003c/em\u003e. in Ningxia, China, under the current and future climatic conditions, and to explore and analyze the fitness pattern of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia Hui Autonomous Region to provide certain theoretical support for its prevention and control work.\u003c/p\u003e"},{"header":"Data and methods","content":"\u003cp\u003e\u003cb\u003eMeteorological and geographic information data acquisition\u003c/b\u003e\u003c/p\u003e\u003cp\u003eClimate is the main environmental factor that determines the distribution of species, and this study used 19 (Biol-19) climatic variables with a spatial resolution of (0.0083\u0026deg;x0.0083\u0026deg;) in the Worldclim Global Climate Database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.worldclim.org/curent\u003c/span\u003e\u003cspan address=\"http://www.worldclim.org/curent\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) to study the climatically suitable distribution area of the Goji berry aphid. The 19 environmental variables including annual mean temperature (Bio1), mean diurnal range (Bio2), isothermality (Bio3), temperature seasonality (Bio4), max temperature of warmest month (Bio5), min temperature of coldest month (Bio6), temperature annual range (Bio7) and so on are shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e\u003csup\u003e[\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eFuture climate scenarios select the sixth international coupled model (CMIP6) shared socio-economic pathway (SSP)\u003csup\u003e[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]\u003c/sup\u003e, compared with the typical concentration emission pathway (RCP) proposed in the fifth international coupled model (CMIP5), the CMIP6 model integrates the consideration of the direction of socio-economic development, and integrates the consideration of the SSP and the RCP, and puts forward four socio-economic development scenarios\u003csup\u003e[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]\u003c/sup\u003e:1) SSP126, which represents zero carbon emission, low economic development and low radiative forcing; 2) SSP245 represents current carbon emission, economic development level and medium radiative forcing; 3) SSP370 represents low economic growth, relatively high carbon emission and radiative forcing; 4) SSP585 represents high carbon emission, high economic growth and high radiative forcing. Since other possible correlation factors, such as elevation, slope direction, slope, solar radiation, vegetation index, etc., were not considered, the distribution range of \u003cem\u003eAphis sp.\u003c/em\u003e predicted in this study is its potential distribution area and does not represent its actual distribution area.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDescription of environmental variables\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDescription of variables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eUnit\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eVariables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eDescription of variables\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eUnits\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio1\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eAnnual mean temperature\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eMean temperature of coldest quarter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio2\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMean diurnal range\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eAnnual precipitation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio3\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eIsothermality\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePrecipitation in the wettest month\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio4\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTemperature seasonality\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio14\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePrecipitation in the driest month\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMax temperature of warmest month\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio15\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSeasonal coefficient of variation of precipitation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e-\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMin temperature of coldest month\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eWettest quarter precipitation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio7\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eTemperature annual range\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBio17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003ePrecipitation in the driest quarter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio8\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMean temperature of wettest quarter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio18\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eHottest quarterly precipitation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMean temperature of driest quarter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003ebio19\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003eColdest quarterly precipitation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003emm\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ebio10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eMean temperature of warmest quarter\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e℃\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn addition to model selection, how to screen for good explanatory variables is also a difficulty in conducting species spatial distribution simulation studies\u003csup\u003e[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]\u003c/sup\u003e. To achieve the best arithmetic results, in most cases, researchers screen the environmental factors input into the model through their own experience\u003csup\u003e[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]\u003c/sup\u003e. This method, although simple and easy to follow, is highly subjective. In order to reduce the uncertainty caused by the combination of environmental factors, this study adopts the method of repeated random sampling of environmental factors, i.e., randomly selecting n out of N climate factors as explanatory variables and repeating it m times, i.e., randomly generating m groups of climate factor combinations. For each combination of climate factors, simulations are conducted separately with 10 spatial distribution models to find the combination of environmental factors with universal significance for different models. In this paper, for 19 climate factors, 5 are randomly selected each time and repeated 30 times, and each environmental factor combination is numbered in order.\u003c/p\u003e\u003cp\u003eThe basic geographic information data were mainly provided by Ningxia Meteorological Research Institute, including administrative boundaries (provincial, municipal and county boundaries), which were used for the spatial characterization of the meteorological conditions of \u003cem\u003eAphis sp.\u003c/em\u003e disaster and the mapping of the results of disaster risk assessment.\u003c/p\u003e\u003cp\u003e\u003cb\u003eAphis sp.\u003c/b\u003e \u003cb\u003edata acquisition and processing\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThe \u003cem\u003eAphis sp.\u003c/em\u003e data were mainly obtained from the Institute of Agricultural Economics and Information Technology of Ningxia Academy of Agriculture and Forestry, as well as the manual survey organized by this project. The distribution data of \u003cem\u003eAphis sp.\u003c/em\u003e collected from 21 large-scale Goji berry planting bases in the main Goji berry producing areas of Ningxia during the 6-year period of 2019\u0026ndash;2024 were selected, and the point data totaled 1998. In order to avoid spatial auto correlation, ENMTools\u003csup\u003e[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]\u003c/sup\u003e was used to remove redundant points and automatically match the raster size of environmental factors used in the analysis, instead of deleting the data based on the distance method, and finally 104 valid points were retained. Five-point sampling was used for each monitoring point, and two Goji berry trees were randomly investigated at each point, and one branch was randomly selected from each tree in five directions: east, south, west, north, and center, and the number of different insect states of \u003cem\u003eAphis sp.\u003c/em\u003e within 30 cm of the branch was recorded, including the number of \u003cem\u003ealate\u003c/em\u003e aphids and nymphs.\u003c/p\u003e\u003cp\u003e\u003cb\u003eConstruction of the habitat suitability model and evaluation\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBiomod2 was run using R-Studio sourced from the website \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.rstudio.com/products/rstudio/\u003c/span\u003e\u003cspan address=\"https://www.rstudio.com/products/rstudio/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e and R speech version (4.2.1) sourced from \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://mirrors.tuna.tsinghua.edu.cn/CRAN/\u003c/span\u003e\u003cspan address=\"https://mirrors.tuna.tsinghua.edu.cn/CRAN/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. When building the model, Biomod2 needs two sets of data, the presence points and the missing points of the species. Since the missing points are difficult to obtain, Biomod2 uses its own \"random\" command to randomly generate 10 sets of 1,000 pseudo-missing points, and randomly sets 80% of the constructed dataset as the training set, and 20% of the remaining data as the test set. 80% of the constructed data set is used as the training set, and the remaining 20% is used as the test set. We set the same weights for the existence points and pseudo-missing points, and set the parameters of a single model to the default values, and run each model 10 times to produce a total of 1,000 sets of training results (10 models x 10 sets of data x 10 operations), and normalize the model results. We used three of the most widely used current model evaluation metrics, Kappa, True Skill Statistic(TSS) and Area Under Curve(AUC)\u003csup\u003e[\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]\u003c/sup\u003e.Kappa was used to assess the agreement between the sample data and the simulation results. AUC was used as a criterion to evaluate the simulation results of the model using the Receiver operating characteristic (ROC) curve. TSS is a modified method based on Kappa, which retains the advantages of Kappa and corrects the disadvantage that Kappa is affected by the wide range of species distributions, and the AUC value usually ranges from 0.5-1, with the closer the AUC value is to 1, the better the prediction effect is. Kappa and TSS values can be modeled by sensitivity and specificity of the difference between true positive and false positive rates, and the closer the difference is to 1, the better the prediction is. When TSS\u0026thinsp;\u0026gt;\u0026thinsp;0.7, AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.8 and Kappa\u0026thinsp;\u0026gt;\u0026thinsp;0.6,it indicates that the model has good predictive performance\u003csup\u003e[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eIn the construction of the combined model, all three indicators were taken into account, and from the model results, a single model with excellent results (TSS\u0026thinsp;\u0026gt;\u0026thinsp;0.8, Kappa\u0026thinsp;\u0026gt;\u0026thinsp;0.8, AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.9) was screened out, and these models were outputted a comprehensive result by weighted average method, and Ensemble Model based on weighted mean(EMwmean). The other is the cluster analysis method, the similar prediction results of each model are grouped into one category, and representative models are selected in each cluster category for integration, which can construct the combined model- Ensemble Model committee averaging (EMca)\u003csup\u003e[\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Results and analysis","content":"\u003cp\u003e\u003cb\u003eModel optimization and accuracy evaluation\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eComparing the 10 models in Biomod2, the Kappa, AUC and TSS values predicted by a single model are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e. It can be seen that for predicting the aptitude zone of Goji berry in Ningxia by single model, the simulation effect of GBM is the best, and the Kappa, TSS and AUC reached 0.97, 0.99 and 0.99, respectively. Followed by RF and ANN, which both can reach Kappa\u0026thinsp;\u0026gt;\u0026thinsp;0.8, TSS\u0026thinsp;\u0026gt;\u0026thinsp;0.9 and AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.9. The worst performance is SRE with Kappa of 0.17, TSS of 0.42 and AUC of 0.71.The GBM, GAM, ANN, RF and MAXENT models are the models with high accuracy and good stability under different evaluation indexes. The worst performing model, SRE, failed the KAPPA and TSS score accuracy tests. The other four models performed between the above models.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eKappa, TSS and AUC values of different models\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"11\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c10\" colnum=\"10\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c11\" colnum=\"11\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eGLM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eGBM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eGAM\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eCTA\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eANN\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSRE\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eFDA\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eMARS\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c10\"\u003e\u003cp\u003eRF\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c11\"\u003e\u003cp\u003eMAXENT\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eKAPPA\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.48\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.85\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e0.80\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.86\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTSS\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.90\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.42\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.84\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e0.93\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.85\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAUC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.81\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e0.98\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e0.96\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e0.99\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e0.71\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e0.95\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c10\"\u003e\u003cp\u003e0.94\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c11\"\u003e\u003cp\u003e0.97\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFrom the model results, the excellent models with TSS values greater than 0.8 were screened out, and the final results were constructed by constructing numerous simple models and then averaging or weighted averaging them, which can effectively avoid the overfitting risk of the algorithmic parameter estimation brought about by too many predictors and a small number of observations. The final prediction results obtained by the two combined models are Kappa\u0026thinsp;\u0026gt;\u0026thinsp;0.90, AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.95, and TSS\u0026thinsp;\u0026gt;\u0026thinsp;0.89, indicating that the combined models have excellent prediction effects and are better than a single model.\u003c/p\u003e\u003cp\u003e\u003cb\u003eImportance of environmental factors\u003c/b\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eFrom the results of the importance of environmental variables (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), the six environmental variables with the highest contribution to the Ningxia \u003cem\u003eAphis sp.\u003c/em\u003e in descending order were the mean temperature of coldest quarter (Bio11), the temperature seasonality (Bio4), the annual mean temperature (Bio1), the hottest quarterly precipitation (Bio18), the mean temperature of warmest quarter (Bio10), and the precipitation in the wettest month (Bio13), respectively. Among them, the mean temperature of the coldest season directly affected the \u003cem\u003eAphis sp.\u003c/em\u003e's overwintering survival rate and population base, and became the core limiting factor for the prediction of the aphid's aptitude zone in Ningxia. Among all the environmental factors, the total contribution rate of temperature factor was 41.2%, the contribution rate of precipitation factor was 34.3%, and the contribution rate of terrain factor was 12.2%. It shows that the effect of precipitation factor on the distribution of \u003cem\u003eAphis sp.\u003c/em\u003e was higher than that of temperature factor and terrain factor.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePrediction of current\u003c/b\u003e \u003cb\u003eAphis sp.\u003c/b\u003e \u003cb\u003ehabitat distribution\u003c/b\u003e\u003c/p\u003e\u003cp\u003eThrough 10 single models and 2 ensemble models, the results of predicting the suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e in China with environmental variables in the current period are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e. With the same precision prediction results, the simulation effect of GBM is the best among the models, and it is the closest to the 2 ensemble models both in terms of the scale of the suitable area and the division of the different levels of the suitable area. On the contrary, SRE had the worst effect, with a large area of highly suitable area for \u003cem\u003eAphis sp.\u003c/em\u003e, which was inconsistent with the reality. As can be seen from Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, the range of the predicted fitness zones of \u003cem\u003eAphis sp.\u003c/em\u003e under the current climatic conditions is mainly located in the north-central region of Ningxia. Among them, the high aptitude area is located in the central part of Zhongning County, the northeastern part of Xixia District of Yinchuan City and scattered areas in the southern part of Helan County; the medium aptitude area is located in the central part of Pingluo County of Shizuishan City, the northern part of Xixia District of Yinchuan City and central part of Helan County, the central part of Litong District of Wuzhong City and scattered areas of Qingsongxia City, the northeastern part of Shapotou District of Zhongwei City and central part of Zhongning County; and the low aptitude area is concentrated in the northern part of Litong District of Wuzhong City and central part of Qingsongxia City, Zhongwei City Zhongning County, and the eastern part of Haiyuan County.\u003c/p\u003e\u003cp\u003eThe output result of the model is the existence probability of \u003cem\u003eAphis sp.\u003c/em\u003e, using ArcGIS software to convert ASCII data into Raster data, to get the existence probability distribution of \u003cem\u003eAphis sp.\u003c/em\u003e. The prediction results of the \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia were classified into risk levels using the \"natural break point classification method\", and quantitatively analyzed by the probability of suitability(P,P\u0026isin;[0,1]). The classification criteria were: non-suitability (P\u0026thinsp;\u0026lt;\u0026thinsp;0.4), low suitability (0.4\u0026thinsp;\u0026le;\u0026thinsp;P\u0026thinsp;\u0026lt;\u0026thinsp;0.6), medium suitability (0.6\u0026thinsp;\u0026le;\u0026thinsp;P\u0026thinsp;\u0026lt;\u0026thinsp;0.8) and high suitability (P\u0026thinsp;\u0026ge;\u0026thinsp;0.80) \u003csup\u003e[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]\u003c/sup\u003e. The reclassification tool was used to count the proportion of regional pixels at each level and to calculate the area of different levels of distribution areas by combining the land use data \u003csup\u003e[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]\u003c/sup\u003e. The area of \u003cem\u003eAphis sp\u003c/em\u003e. fitness zone with different degrees in the current period of the two combined models was obtained in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e4\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eCurrent suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e (km\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"7\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003ecity\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c4\" namest=\"c2\"\u003e\u003cp\u003eEMca\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"3\" nameend=\"c7\" namest=\"c5\"\u003e\u003cp\u003eEMwmean\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eHigh\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eHigh Moderate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eModerate\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eLow\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWuzhong\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e333.91\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e737.43\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e79.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e375.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e547.82\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGuyuan\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e8.13\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e31.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e10.88\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e57.37\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e19.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e80.25\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eZhongwei\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1231.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e543.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1062.33\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1428.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e597.22\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e501.11\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eShizuishan\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e0.00\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e8.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e39.21\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e5.16\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e13.17\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e87.37\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eYinchuan\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e23.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e26.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e45.56\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e294.32\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e25.75\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e96.16\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTotal\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e1263.87\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e943.73\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e1695.41\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e1864.70\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e1030.50\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e1312.71\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eFrom Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e4\u003c/span\u003e, it can be seen that the area of suitable area for \u003cem\u003eAphis sp.\u003c/em\u003e in EMca of the integrated model is 3902.67 km\u003csup\u003e2\u003c/sup\u003e, accounting for 5.88% of the total land area (66,400 km\u003csup\u003e2\u003c/sup\u003e)in Ningxia\u003csup\u003e[\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]\u003c/sup\u003e, and the areas of high, medium and low suitable areas are 1263.87 km\u003csup\u003e2\u003c/sup\u003e、830.50 km\u003csup\u003e2\u003c/sup\u003e and 902.71 km\u003csup\u003e2\u003c/sup\u003e, accounting for 36.08%、26.94% and 26.94%; of the total area of suitable areas 36.08%、26.94% and 48.38%, respectively; the suitable area of EMwmean model was 4207.91 km\u003csup\u003e2\u003c/sup\u003e, accounting for 6.34% of the total land in the Ningxia region, and the areas of high, medium, and low suitable areas were 1864.7km\u003csup\u003e2\u003c/sup\u003e, 1030.5km\u003csup\u003e2\u003c/sup\u003e, and 1312.71km\u003csup\u003e2\u003c/sup\u003e, accounting for 44.31%, 24.49%, 31.49%, and 31.71%, respectively, of the total area of the suitable area. The total area of suitable area is 1864.7km\u003csup\u003e2\u003c/sup\u003e, 1030.5km\u003csup\u003e2\u003c/sup\u003e and 1312.71km\u003csup\u003e2\u003c/sup\u003e respectively, accounting for 44.31%, 24.49% and 31.20% of the total area of suitable area.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eThe suitable areas for Ningxia \u003cem\u003eAphis sp.\u003c/em\u003e in the current period obtained by the two models are basically the same and more concentrated, which are mainly distributed in Helan County in the north and Lingwu City in the south of Yinchuan City, Pingluo County and Huinong District in the southeast of Shizuishan City, Zhongning County in the north of Zhongwei City, and Hongsibao District in Wuzhong City. Ningxia \u003cem\u003eAphis sp.\u003c/em\u003e high suitable area mainly in Zhongning County, Tongxin County, Huinong District, Helan County and other places, the climate in these areas is relatively warm and humid, which is highly conducive to the reproduction of \u003cem\u003eAphis sp.\u003c/em\u003e species, particularly during the spring and summer when high temperatures promote rapid population expansion of aphids.\u003csup\u003e[\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]\u003c/sup\u003e. Among these, the cultivation of Goji berry in Shizuishan City is more dispersed, leading to a more scattered distribution of \u003cem\u003eAphis sp.\u003c/em\u003e. As a result, these areas exhibit a moderate to low risk of aphid infestation, consistent with the findings of the two combination models. When compared to the current distribution of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia, the results of the two models were more accurate in EMca than in EMwmean.\u003c/p\u003e\u003cp\u003e\u003cb\u003ePrediction of suitable growing area of\u003c/b\u003e \u003cb\u003eAphis sp.\u003c/b\u003e \u003cb\u003eunder future climate scenarios\u003c/b\u003e\u003c/p\u003e\u003cp\u003eBased on the two species distribution ensemble models developed in this study, simulations were conducted to predict the suitability distribution range of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia under the influence of future climate change. Using the spatial distribution pattern of \u003cem\u003eAphis sp.\u003c/em\u003e under current climatic conditions, derived from the more accurate ensemble model EMca as a reference, four future climate scenarios were integrated to predict changes in the distribution of potential habitats at different time intervals (2030s, 2050s, 2070s, and 2090s). Following the current suitability distribution area classification standard\u003csup\u003e[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]\u003c/sup\u003e, the spatial distribution of \u003cem\u003eAphis sp.\u003c/em\u003e under future climate scenarios was also categorized into four levels, and the area of each level was calculated to determine the distribution of potentially suitable habitats in future periods.\u003c/p\u003e\u003cp\u003eFigure \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e illustrates the changes in the extent of potentially suitable areas under various future scenarios predicted by the two models. Based on the combined analysis of changes in potentially high-risk areas for \u003cem\u003eAphis sp.\u003c/em\u003e and the spatial variations shown in Table\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, it is evident that under the SSP126 scenario, the most rapid expansion occurs in the 2030s and 2090s, with average increases of 501% and 557%, respectively, compared to the current suitable area. Under the SSP245 scenario, the greatest growth is projected for the 2070s, with an average increase of 540%. Under the SSP370 scenario, the highest expansion is observed in the 2090s, with an average increase of 726%. For the SSP585 scenario, while the suitable area begins expanding in the 2040s, the fastest growth is expected in the 2070s and 2090s, with average increases of 550% and 515%, respectively, relative to the current suitable area.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eComparing the two models in terms of total suitable area, the EMwmean model indicated a consistent and significant increase in the suitable habitat of \u003cem\u003eAphis sp.\u003c/em\u003e across different periods and future climate scenarios. In contrast, the EMca model, except under the SSP370 scenario, also showed an overall increasing trend; however, in the other three scenarios, a decline in suitable area was observed during certain periods. These reductions were mainly concentrated in the northern part of Zhongwei and along the border of Wuzhong City, and the magnitude of increase was notably smaller than that predicted by the EMwmean model. The simulation results from EMca exhibited noticeable fluctuations but corresponded more closely with actual observations, demonstrating a better overall simulation performance. Therefore, the EMca model was selected for predicting future suitable habitats of \u003cem\u003eAphis sp.\u003c/em\u003e, and the potential risk areas in the Ningxia Autonomous Region under future climate change scenarios were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eComparison of suitable areas of \u003cem\u003eAphis sps\u003c/em\u003e. in Ningxia under current and future climatic conditions (\u0026times; 10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"9\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c9\" colnum=\"9\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\" morerows=\"1\" rowspan=\"2\"\u003e\u003cp\u003eDecade\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e\u003cp\u003eDecade\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colspan=\"4\" nameend=\"c9\" namest=\"c6\"\u003e\u003cp\u003eEMwmean\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eSSP125\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eSSP245\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eSSP370\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003eSSP585\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003eSSP125\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003eSSP245\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003eSSP370\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c9\"\u003e\u003cp\u003eSSP585\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2030s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e158.65\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e-15.54\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e14.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e16.05\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e305.89\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e146.5\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e119.57\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e143.81\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2050s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e-13.74\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e150.9\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e142.46\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e86.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e28.31\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e364.04\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e232.10\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e343.59\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2070s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e148.97\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e113.29\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e128.6\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e220.11\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e286.34\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e393.77\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e412.62\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e391.43\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e2090s\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e\u003cp\u003e187.25\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e\u003cp\u003e95.66\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e\u003cp\u003e282.92\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e\u003cp\u003e234.35\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c6\"\u003e\u003cp\u003e317.12\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c7\"\u003e\u003cp\u003e380.76\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c8\"\u003e\u003cp\u003e468.53\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"char\" char=\".\" colname=\"c9\"\u003e\u003cp\u003e348.53\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003ctfoot\u003e\u003ctr\u003e\u003ctd colspan=\"9\"\u003e\"-\": indicates area reduction compared to current climate conditions; SSP: shared socioeonomie pahways\u003c/td\u003e\u003c/tr\u003e\u003c/tfoot\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eAs can be seen from Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, compared with the predicted results under the current climatic conditions, the total suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e was increasing, among which the high and medium suitable areas in Yinchuan and Shizuishan City increased significantly; the boundary of the low suitable area showed a clear tendency to shift southeast, especially in the cities of Zhongwei and Wuzhong. The predicted total suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e under current climatic conditions is 39.03\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e, accounting for 5.88% of Ningxia\u0026rsquo;s total area. Under the SSP370 scenario in the 2090s, the total suitable area increased the most, reaching 407.23\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e, or 61.33% of Ningxia\u0026rsquo;s total area. The predicted highly suitable area of \u003cem\u003eAphis sp.\u003c/em\u003e under current climatic conditions is 15.75\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e. Under future climatic conditions, the area of highly suitable habitat decreases, even the area will be reduced to zero under the scenarios of SSP245, SSP370, and SSP585 in the 2030s, and under the SSP126 and SSP370 scenarios in the 2050s. The area of medium-suitable habitat under current conditions is 8.31\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e. In the future, the area of medium-suitable habitat increases under all scenarios, with the largest area under the SSP370 scenario in the 2090s, reaching 221.62\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e. The area of low-suitable habitat under current climatic conditions is 9.01\u0026times;10\u003csup\u003e2\u003c/sup\u003ekm\u003csup\u003e2\u003c/sup\u003e. In the future, the area of low-suitable habitat increases under all scenarios, except for the SSP126 and SSP585 scenarios in the 2050s.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eUnder projected climate change scenarios, the potential suitable habitats for \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia were primarily concentrated in the cities of Zhongwei, Wuzhong, Yinchuan, and most parts of Shizuishan, with the overall distribution pattern remaining largely stable. Compared to the baseline period (current), the total area suitable for \u003cem\u003eAphis sp.\u003c/em\u003e increased significantly, particularly in Yinchuan and Shizuishan. The suitable areas expanded from the north-central region toward the northwestern and southwestern parts of Ningxia. The model predicted a similar overall trend in habitat suitability for \u003cem\u003eAphis sp.\u003c/em\u003e across all four climate scenarios; however, the predicted suitability levels for the same time periods varied between scenarios. For example, during the 2030s, the scenarios produced varying predictions, with the most notable difference being that areas identified as highly or moderately suitable in the SSP126 scenario were classified as having low suitability in the other three scenarios. In the 2050s, Yinchuan and most of Shizuishan were projected to be moderately suitable under the SSP245 and SSP370 scenarios, but were predicted to be low or unsuitable under SSP126 and SSP585. During the 2070s and 2090s, the extent of high and medium suitability areas under the SSP126 scenario was significantly larger than in the other scenarios, with the most suitable zones concentrated in Yinchuan and northwestern Zhongwei. These areas may represent high-risk zones for potential \u003cem\u003eAphis sp.\u003c/em\u003e outbreaks.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this study, we used an integrated model combined with different climate scenarios to predict the potential distribution of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia. By analyzing the climate data under the current climate and four future climate scenarios (SSP126, SSP245, SSP370, SSP585), we found that the area of the suitable zone for the \u003cem\u003eAphis sp.\u003c/em\u003e showed significant changes under different climate scenarios in different periods in the future. Overall, future climate warming will likely expand the suitable habitat zone for \u003cem\u003eAphis sp.\u003c/em\u003e, with the most pronounced increase in suitable habitat area occurring under the high-emission scenario (SSP585)\u003csup\u003e[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eAmong the two combined models used in this study, the EMca model provided more realistic predictions compared to the EMwmean model\u003csup\u003e[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]\u003c/sup\u003e. The EMca model better captured fluctuations in \u003cem\u003eAphis sp.\u003c/em\u003e fitness area size, particularly in spatial distribution changes over time and across different climate scenarios. Unlike the relatively smooth predictions of the EMwmean model, the EMca model accounted for more complex climatic influences, such as fluctuations in temperature extremes and precipitation patterns, which provided more accurate simulations of \u003cem\u003eAphis sp.\u003c/em\u003e Distribution\u003csup\u003e[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe environmental factor analysis indicated that temperature and precipitation were key climatic factors influencing the distribution of \u003cem\u003eAphis sp.\u003c/em\u003e In particular, the mean temperature of coldest quarter (Bio11) and the temperature seasonality (Bio4) were the most important factors influencing the prediction of suitable areas for the aphid. The temperature factor, especially the temperature change in the coldest season, directly affects the aphid's overwintering survivability and population base\u003csup\u003e[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]\u003c/sup\u003e, and is the core limiting factor for the distribution of \u003cem\u003eAphis sp.\u003c/em\u003e in the suitable habitat range in Ningxia.\u003c/p\u003e\u003cp\u003eIn this study, Ensemble Models were introduced to predict the potential habitat of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia to improve the accuracy and stability of prediction. Compared with the single modeling method, especially the widely used MaxEnt, the Ensemble Models make up for the limitations of the single model prediction results by integrating multiple modeling algorithms (e.g., GBM, RF, GLM, etc.) in various aspects\u003csup\u003e[\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]\u003c/sup\u003e. The results show that the integrated model can better simulate the potential distribution area of \u003cem\u003eAphis sp.\u003c/em\u003eunder different climate scenarios, with higher AUC and TSS evaluation indexes, reflecting stronger adaptability and generalization ability. The study provides a theoretical reference for the prevention and control strategy of characteristic economic crop pests in the northern region of China, and also provides a methodological demonstration for similar studies in the future. The study still has some shortcomings. Regarding the selection of environmental variables, only climate factors were selected as model inputs, failing to cover other ecological factors such as topography, vegetation, soil, etc.\u003csup\u003e[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]\u003c/sup\u003e, which led to the simulation results focusing more on reflecting the influence of climatic conditions on the fitness distribution of \u003cem\u003eAphis sp.\u003c/em\u003e, and making it difficult to comprehensively reveal their spatial distribution characteristics in different ecological contexts\u003csup\u003e[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]\u003c/sup\u003e. Secondly, the model was constructed based on the basic ecological niche theory, which only considered the influence of abiotic factors on the species distribution, and ignored the regulating effect of biotic factors\u003csup\u003e[\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]\u003c/sup\u003e, such as the migratory ability of \u003cem\u003eAphis sp.\u003c/em\u003e, the interference of natural enemies, and the competitive relationship. The distribution of species in real ecosystems is the result of the joint action of abiotic factors and biotic interactions, so the ecological niche range predicted by the model only represents an idealized state, which is often wider than the real distribution range\u003csup\u003e[\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eBased on the integrated model and climate data under different climate scenarios, this study predicted the potential habitat of \u003cem\u003eAphis sp.\u003c/em\u003e in Ningxia. The results showed that future climate change, especially the increase in temperature, will likely lead to the expansion of the fitness zone of \u003cem\u003eAphis sp.\u003c/em\u003e, especially under the high emission scenario (SSP585), where the increase in the area of the fitness zone is particularly obvious. Nevertheless, the changes in the suitable area under different scenarios varied greatly and the fluctuation of the area of the suitable area in different time periods was more significant, which provides an important reference for future \u003cem\u003eAphis sp.\u003c/em\u003e prevention and control work. This study provides theoretical support for the control of \u003cem\u003eAphis sp.\u003c/em\u003e and suggests that local governments and agricultural administrations should strengthen the monitoring of \u003cem\u003eAphis sp.\u003c/em\u003e potential risk areas and take appropriate preventive and control measures, especially under climate change scenarios, to adjust prevention and control strategies in time to cope with the possible impacts of \u003cem\u003eAphis sp.\u003c/em\u003e\u003c/p\u003e\u003cp\u003eHowever, this study was limited to predicting the effects of climate on \u003cem\u003eAphis sp\u003c/em\u003e. and analyzing its distribution and habitat suitability within the Ningxia region. Several avenues for enhancing future research are apparent. First, the integration of multi-source data and the inclusion of additional ecological variables, such as soil type, vegetation cover, and elevation combined with an expansion of the study to a national scale, would significantly improve the model\u0026rsquo;s ecological adaptability. Second, the incorporation of biological interactions, particularly the relationships between \u003cem\u003eAphis sp\u003c/em\u003e. and its host plants or natural enemies, would facilitate the development of a dynamic model that better reflects real-world ecological processes. Third, model validation should be further strengthened through comparative analysis with field monitoring data, thereby improving the predictive accuracy and overall reliability of the model.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e\u003cp\u003eThis research was funded by the National Natural Science Foundation of China (NNSFC) under Project Number 32160639, 2024 Ningxia Hui Autonomous Region Young Top Homo Sapiens Talent Cultivation Project, 2025 New Era Meteorological High-level Scientific and Technological Innovation Homo Sapiens Talent Program.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eZ.H.: Conceptualization, methodology, writing - original draft,writing-review and editing. L.Y.: Conceptualization, methodology, software, formal analysis and writing - original draft. S.L.: Writing - review and editing, supervision. L.L. and H.Q.: Resources, writing and formal analysis. L.S.:Investigation, resources and funding acquisition. J.B., M.J. and L.Y.: Investigation.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eThe authors confirm that the data supporting the findings of this study are available within the article, and further requests can be directed to the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eJiahui et al. Lycium barbarum studies: A system review on molecular biology, cultivation, and quality characteristics of goji berries. \u003cem\u003eBiochem. Syst. Ecol.\u003c/em\u003e \u003cb\u003e121\u003c/b\u003e, 105020 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePanwar, T. S. \u0026amp; Singh, S. B. Vinod Kumar Garg. Influence of meteorological parameters on population dynamics of thrips (Thrips tabaci Lindeman) and aphid (Aphis gossypii Glover) in Bt and non Bt cotton at Malwa region of Madhya Pradesh. \u003cem\u003eJ. Agrometeorology\u003c/em\u003e. \u003cb\u003e17.1\u003c/b\u003e, 136\u0026ndash;138 (2015).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eChaieb et al. Chemical composition and aphicidal potential of Citrus aurantium peel essential oils. \u003cem\u003eEntomol. Generalis\u003c/em\u003e. \u003cb\u003e37\u003c/b\u003e, 63\u0026ndash;75 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWu, Y. et al. Investigating the impact of climate warming on phenology of aphid pests in China using long-term historical data. \u003cem\u003eInsects\u003c/em\u003e \u003cb\u003e11\u003c/b\u003e (3), 167 (2020).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, F., Yuan, X., Sun, Y. \u0026amp; Liu, Y. Species distribution modeling based on MaxEnt to inform biodiversity conservation in the Central Urban Area of Chongqing Municipality.Ecological Indicators. \u003cb\u003e158\u003c/b\u003e, 111491. (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHosseini, N. \u0026amp; Hossein Mostafavi. Habitat potential modelling and the effect of climate change on the current and future distribution of three Thymus species in Iran using MaxEnt. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (1), 3641 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu, Y. H. et al. Habitat suitability evaluation of sable (Martes zaibellina) based on MAXENT model in Beijicun National Nature Reserve Heilongiiang Province, China. \u003cem\u003eActa Ecol. Sin.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e (2), 559\u0026ndash;569 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003ePeihao, P. Feng Yu.Habitat suitability evaluation for Paeonia decomposita based on a MaxEnt model. Eco.mont. \u003cem\u003eJ. Protected Mountain Areas Res.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (1), 25\u0026ndash;32 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJi Yelin, S. \u0026amp; Xiyou, Y. Potential habitat prediction of Hyphantria cunea based on a random forest model in China. \u003cem\u003eJ. Nanjing Forestry Univ.\u003c/em\u003e \u003cb\u003e43\u003c/b\u003e (6), 121\u0026ndash;128 (2019).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eJiangxia, Y. \u0026amp; Jingwen, W. Zhang Mingsha,Zhou Ruliang \u0026amp; Shi Lei. Risk Pattern Analysis of Hyphantria cunea Based on Spatial Matrix Model and0-1 Measure.Scientia silvae sinicae. 57(1),140\u0026ndash;152 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGuo, Y. et al. Prediction of Chinese suitable habitats of Panax notoginseng under climate change based on MaxEnt and chemometric methods. \u003cem\u003eSci. Rep.\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e, 16434 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLi, S. et al. Estimation and inference for high-dimensional generalized linear models with knowledge transfer. \u003cem\u003eJ. Am. Stat. Assoc.\u003c/em\u003e \u003cb\u003e119\u003c/b\u003e, 1274\u0026ndash;1285 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eTakolander, A. et al. Cross-realm transferability of species distribution models\u0026ndash;Species characteristics and prevalence matter more than modelling methods applied. \u003cem\u003eEcol. Model.\u003c/em\u003e \u003cb\u003e499\u003c/b\u003e, 110950 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eYang, L. Qing Hua. Predicting suitable habitats of parasitic desert species based on Biomod2 ensemble model: Cynomorium songaricum rupr and its host plants as an example. \u003cem\u003eBMC Plant Biol.\u003c/em\u003e \u003cb\u003e25\u003c/b\u003e (1), 351 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eGu, R. et al. Predicting the impacts of climate change on the geographic distribution of moso bamboo in China based on biomod2 model. \u003cem\u003eEur. J. For. Res.\u003c/em\u003e \u003cb\u003e143\u003c/b\u003e (5), 1499\u0026ndash;1512 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eZhaio et al. Analyzing potential distribution and disturbance intensity of plateau pika in the source region of the Yellow River via BIOMOD2 integrated model. \u003cem\u003eChin. J. Ecol.\u003c/em\u003e \u003cb\u003e43\u003c/b\u003e (4), 1192 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSosa-Guill\u0026eacute;n, P. et al. Bioclimatic indicators dataset for the orographically complex Canary Islands archipelago. \u003cem\u003eSci. Data\u003c/em\u003e. \u003cb\u003e11\u003c/b\u003e (1), 1323 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eDeepa, R. \u0026amp; Kumar, V. Suchithra Sundaram. A systematic review of regional and global climate extremes in CMIP6 models under shared socio-economic pathways. \u003cem\u003eTheoret. Appl. Climatol.\u003c/em\u003e \u003cb\u003e155\u003c/b\u003e (4), 2523\u0026ndash;2543 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eXu, C. Yu-Hao Mao, \u0026amp; Hong Liao. Future ozone changes and their impacts on vegetation and human health in China under the Shared Socio‐economic Pathways. \u003cem\u003eJ. Geophys. Research: Atmos.\u003c/em\u003e \u003cb\u003e129\u003c/b\u003e(4), (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAustin, M. P., Kimberly, P. \u0026amp; Van Niel Improving species distribution models for climate change studies: Variable selection and scale. \u003cem\u003eJ. Biogeogr.\u003c/em\u003e \u003cb\u003e38\u003c/b\u003e (1), 1\u0026ndash;8 (2011).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWideen, M. Jolie Mayer-Smith,\u0026amp; Barbara Moon. A critical analysis of the research on learning to teach: Making the case for an ecological perspective on inquiry. \u003cem\u003eRev. Educ. Res.\u003c/em\u003e \u003cb\u003e68\u003c/b\u003e (2), 130\u0026ndash;178 (1998).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWarren, D. L. et al. ENMTools 1.0: an R package for comparative ecological biogeography. \u003cem\u003eEcography\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e (4), 504\u0026ndash;511 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAllouche, O. Ronen Kadmon. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). \u003cem\u003eJ. Appl. Ecol.\u003c/em\u003e \u003cb\u003e43\u003c/b\u003e (6), 1223\u0026ndash;1232 (2006).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eFielding, A. H., John, F. \u0026amp; Bell A review of methods for the assessment of prediction errors in conservation presence/absence models. \u003cem\u003eEnviron. Conserv.\u003c/em\u003e \u003cb\u003e24\u003c/b\u003e (1), 38\u0026ndash;49 (1997).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLeroy, B. et al. Without quality presence\u0026ndash;absence data, discrimination metrics such as TSS can be misleading measures of model performance. \u003cem\u003eJ. Biogeogr.\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e (9), 1994\u0026ndash;2002 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAbrego, N. \u0026amp; Otso Ovaskainen. \u0026amp;. Evaluating the predictive performance of presence\u0026ndash;absence models: Why can the same model appear excellent or poor? \u003cem\u003eEcol. Evol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e(12), (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Z. et al. Predicting the Future Geographic Distribution of the Traditional Chinese Medicinal Plant Epimedium acuminatum Franch. in China Using Ensemble Models Based on Biomod2. \u003cem\u003ePlants\u003c/em\u003e \u003cb\u003e14\u003c/b\u003e (7), 1065 (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSong, Y. et al. Synergizing multiple machine learning techniques and remote sensing for advanced landslide susceptibility assessment: a case study in the Three Gorges Reservoir Area. \u003cem\u003eEnviron. Earth Sci.\u003c/em\u003e \u003cb\u003e83\u003c/b\u003e (8), 227 (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eToure, S. I. et al. Land cover and land use change analysis using multi-spatial resolution data and object-based image analysis. \u003cem\u003eRemote Sens. Environ.\u003c/em\u003e \u003cb\u003e210\u003c/b\u003e, 259\u0026ndash;268 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWang, Y. et al. A dynamic evaluation of land use and habitat quality in Ningxia from 1980 to 2020, china. \u003cem\u003eAppl. Ecol. Environ. Res.\u003c/em\u003e \u003cb\u003e23\u003c/b\u003e(1), (2025).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEbert, T. A. Cartwright. Biology and ecology of Aphis gossypii Glover (Homoptera: aphididae). \u003cem\u003eSouthwest. Entomol.\u003c/em\u003e \u003cb\u003e22\u003c/b\u003e (1), 116\u0026ndash;153 (1997).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eSchroth, G., L\u0026auml;derach, P., Martinez-Valle, A. I., Bunn, C. \u0026amp; Jassogne, L. Vulnerability to climate change of cocoa in West Africa: patterns, opportunities and limits to adaptation. \u003cem\u003eSci. Total Environ.\u003c/em\u003e \u003cb\u003e556\u003c/b\u003e, 231\u0026ndash;241 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiao, F. et al. Future change in the Vietnam upwelling under a high-emission scenario. \u003cem\u003eGeophys. Res. Lett.\u003c/em\u003e \u003cb\u003e51\u003c/b\u003e(11), (2024).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eShafizadeh-Moghadam, H. et al. Novel forecasting approaches using combination of machine learning and statistical models for flood susceptibility mapping. \u003cem\u003eJ. Environ. managemen\u003c/em\u003e. \u003cb\u003e217\u003c/b\u003e, 1\u0026ndash;11 (2018).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eEasterling, D. R. et al. Climate extremes: observations, modeling, and impacts. \u003cem\u003eScience\u003c/em\u003e \u003cb\u003e289\u003c/b\u003e (5487), 2068\u0026ndash;2074 (2000).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eKharin, V. V. et al. Changes in temperature and precipitation extremes in the CMIP5 ensemble. \u003cem\u003eClim. Change\u003c/em\u003e. \u003cb\u003e119\u003c/b\u003e (2), 345\u0026ndash;357 (2013).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eBale, J. S. Hayward. Insect overwintering in a changing climate. \u003cem\u003eJ. Exp. Biol.\u003c/em\u003e \u003cb\u003e213.\u003c/b\u003e (6), 980\u0026ndash;994 (2010).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eLiu, J. et al. Impact of temperature on survival rate, fecundity, and feeding behavior of two aphids, Aphis gossypii and Acyrthosiphon gossypii, when reared on cotton. \u003cem\u003eInsects\u003c/em\u003e \u003cb\u003e12\u003c/b\u003e (6), 565 (2021).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAhmadi, M. et al. MaxEnt brings comparable results when the input data are being completed; Model parameterization of four species distribution models. \u003cem\u003eEcol. Evol.\u003c/em\u003e \u003cb\u003e13\u003c/b\u003e (2), e9827 (2023).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eValavi, R. et al. Predictive performance of presence-only species distribution models: a benchmark study with reproducible code. \u003cem\u003eEcol. Monogr.\u003c/em\u003e \u003cb\u003e92\u003c/b\u003e (1), e01486 (2022).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eMod, H. K. et al. What we use is not what we know: environmental predictors in plant distribution models. \u003cem\u003eJ. Veg. Sci.\u003c/em\u003e \u003cb\u003e27\u003c/b\u003e (6), 1308\u0026ndash;1322 (2016).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eElith, J. \u0026amp; John, R. Leathwick. Species distribution models: ecological explanation and prediction across space and time. Annual review of ecology, evolution, and systematics. 40(1), 677\u0026ndash;697 (2009).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eHirzel, A. H. Gwena\u0026euml;lle Le Lay. Habitat suitability modelling and niche theory. \u003cem\u003eJ. Appl. Ecol.\u003c/em\u003e \u003cb\u003e45\u003c/b\u003e (5), 1372\u0026ndash;1381 (2008).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eAnderson, R. P. When and how should biotic interactions be considered in models of species niches and distributions? \u003cem\u003eJ. Biogeogr.\u003c/em\u003e \u003cb\u003e44\u003c/b\u003e (1), 8\u0026ndash;17 (2017).\u003c/span\u003e\u003c/li\u003e\u003cli\u003e\u003cspan\u003eWisz, M. et al. The role of biotic interactions in shaping distributions and realised assemblages of species: implications for species distribution modelling. \u003cem\u003eBiol. Rev.\u003c/em\u003e \u003cb\u003e88\u003c/b\u003e (1), 15\u0026ndash;30 (2013).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Aphis sp., Ensemble model, Climate change, Prediction of suitable area","lastPublishedDoi":"10.21203/rs.3.rs-7258894/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7258894/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cem\u003eAphis sp.\u003c/em\u003e is a major pest of Goji berry in Ningxia, significantly affecting its yield and quality. To explore the potential distribution changes of \u003cem\u003eAphis sp.\u003c/em\u003e under climate change, this study used data from 104 valid occurrence points collected from 2019 to 2024 in the main Goji-producing areas of Ningxia, combined with 19 climatic variables. The Biomod2 ensemble modeling framework, incorporating 10 individual models, was employed to predict the current and future (2030s\u0026ndash;2090s) suitable habitats for \u003cem\u003eAphis sp.\u003c/em\u003e under four climate scenarios (SSP126, SSP245, SSP370, and SSP585).The results indicate that under current climatic conditions, highly suitable habitats for \u003cem\u003eAphis sp.\u003c/em\u003e are primarily concentrated in the central and northern regions of Ningxia, with a total suitable area of 3,902.67 km\u0026sup2;, accounting for 5.88% of the region\u0026rsquo;s total land area. Environmental factor analysis revealed that the mean temperature of the coldest quarter (Bio11), temperature seasonality (Bio4), and annual mean temperature (Bio1) are the key variables influencing the distribution of \u003cem\u003eAphis sp.\u003c/em\u003e, with a combined contribution rate of 41.2%. The ensemble models (EMca and EMwmean) demonstrated significantly higher predictive accuracy (AUC\u0026thinsp;\u0026gt;\u0026thinsp;0.95, TSS\u0026thinsp;\u0026gt;\u0026thinsp;0.89) compared to individual models. In particular, the EMca model more effectively captured fluctuations in the extent of suitable habitats. Under four climate scenarios, the suitable habitat area for \u003cem\u003eAphis sp\u003c/em\u003e. is projected to expand significantly, with the greatest increase observed under the SSP370 scenario, reaching 40,723 km\u0026sup2; by the 2090s. Moreover, the suitable range is expected to shift from the central-northern region toward the northwest and southwest. This study provides a theoretical foundation for the targeted management of \u003cem\u003eAphis sp\u003c/em\u003e. in Ningxia and highlights the need to closely monitor the impact of climate warming on the expansion of their suitable habitat.\u003c/p\u003e","manuscriptTitle":"Prediction of potential suitable habitats of Aphis sp. in Ningxia under future climate scenarios based on ensemble model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-25 17:54:06","doi":"10.21203/rs.3.rs-7258894/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b7647a7d-f786-4d61-9625-99d8f4d23c2f","owner":[],"postedDate":"August 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53284975,"name":"Earth and environmental sciences/Climate sciences"},{"id":53284976,"name":"Biological sciences/Ecology"},{"id":53284977,"name":"Earth and environmental sciences/Ecology"},{"id":53284978,"name":"Earth and environmental sciences/Environmental sciences"}],"tags":[],"updatedAt":"2026-01-06T18:09:11+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-25 17:54:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7258894","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7258894","identity":"rs-7258894","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}