Global temporal distribution of Maca (Lepidium meyenii Walp.) under climate change

Global temporal distribution of Maca (Lepidium meyenii Walp.) under climate change | 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 Research Article Global temporal distribution of Maca (Lepidium meyenii Walp.) under climate change Zeyu Qin, Huasheng Huang, Xuanqi Liu, Xia Meng, Minqiao Li This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6472833/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 Global warming has influenced phenological shifts and community degradation in alpine herbaceous plants. However, the current distribution and future shifts of these plants and their environmental driving factors under climate change are not well known. Here, we focus on Lepidium meyenii , a medicinal herb native to the high-altitude regions of the Andes Mountains in South America, which may face habitat contraction and potential population decline in the future due to changing environmental conditions. We integrate species distribution data with a random forest model to simulate the potential habitats of L. meyenii across time and identify their key environmental drivers. We find that the most significant environmental variables that impact the distribution of L. meyenii include elevation, temperature annual range (bio7), and mean diurnal range (bio2). Since the Last Glacial Maximum, the western Andes in South America has consistently offered suitable habitat for L. meyenii . In contrast, habitat suitability in the Tibetan Plateau (TP) has varied over time, and in the futurer the TP may become a potential area for biodiversity conservation. This study will enhance our understanding of the distribution patterns of alpine herbaceous plants in response to climate change, and contributes to future biodiversity conservation efforts, the establishment of protected areas, and the sustainable management of medicinal plants as biological resources. Climate change Brassicaceae Lepidium meyenii Species distribution Ecological niche model Biogeography Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 1. Introduction Since the Industrial Revolution, the surge in greenhouse gas emissions resulting from industrialization has led to a rise in global average temperatures by 0.5 ℃. It is projected that by 2100, under moderate and high emissions scenarios (SSP2-4.5 and SSP5-8.5), global temperatures will increase by 2.3–4.6 ℃ and 6.6–14.1 ℃, respectively, compared to pre-industrial levels [ 1 , 2 ]. Climate change not only alters weather patterns but also exacerbates the frequency and severity of extreme climatic events, such as heavy rainstorms, droughts, and storms [ 3 ]. As a major consequence of global warming, climate change is increasingly recognized as a key driver of species distribution shifts [ 4 , 5 , 6 ]. Many species are moving to higher altitudes and latitudes as temperatures rise [ 4 , 7 ]. These shifts not only change distribution patterns but also have significant implications for biodiversity and ecosystems [ 8 , 9 ]. Furthermore, species niche shifts occur at a much slower rate than the pace of climate change [ 10 ]. This highlights potential risks of habitat mismatch and biodiversity loss. Therefore, it is essential to understand the relationship between climate change and species distribution for developing effective conservation strategies. Species distribution models (SDMs) are developed to investigate the relationship between climate change and species distribution. They analyze the interactions between species and environmental variables to predict potentially suitable distributions. SDMs also incorporate the physiological responses and adaptation mechanisms of species [ 11 ]. Common algorithms used in species distribution modelling include the maximum entropy model (MaxEnt) [ 12 ], random forest (RF) [ 13 ], generalized additive model (GAM) [ 14 ] generalized linear model (GLM) [ 15 ], among others. The biomod2 package in R is widely adopted for species distribution modelling [ 16 , 17 ]. It integrates various algorithms and allows users to customize parameters for different applications [ 18 ]. Selecting the optimal model is important for accurately predicting potentially suitable areas for species, and this can be achieved through the quantitative evaluation of different SDMs. Among these algorithms, the MaxEnt model, when parameter-tuned using the ENMeval R package, is particularly popular. It effectively reduces overfitting compared to default settings and significantly improves predictive performance [ 19 ]. Lepidium L. is a genus within the Brassicaceae family, and comprises 265 species ( https://www.powo.science.kew.org ). It is widely distributed across all continents except Antarctica and can be found in high-altitude regions [ 20 , 21 ]. The genus originated in Eurasia, evolved, and thrived in the Mediterranean region before its long-distance dispersal to the Americas and Australia during the late Neogene or Quaternary periods [ 22 , 23 ]. L. meyenii is an alpine herbaceous species native to the Andean region of western South America. It diverged ca. 12 million years ago [ 24 ], and has adapted to the extreme conditions of high altitudes. It is commonly referred to as "Peruvian ginseng" and possesses significant economic value due to its rich nutrient profile and numerous health benefit, including enhancing fertility, reducing fatigue, balancing hormone levels, and improving athletic performance [ 25 , 26 , 27 , 28 , 29 , 30 ]. Consequently, products derived from L. meyenii have gained considerable market popularity [ 31 ]. Since the 1990s, cultivation attempts in the United States, Japan, and Germany have met with limited success, while China successfully domesticated L. meyenii in 2002 and has established a substantial market. For example, in Yunnan Province, the cultivation area of L. meyenii has reached 22,000 hm 2 [ 32 , 33 ]. Climate change can lead to alpine gravelization and thus result in the degradation of meadow communities [ 34 ], and significantly affect the phenology of alpine herbaceous species [ 35 ]. The ecological adaptation mechanism of L. meyenii may change in response to global warming which could impact its distribution. Therefore, it is crucial to investigate the changes in the distribution patterns of L. meyenii under future climate change scenarios to develop effective cultivation strategies that ensure its sustainable development and conservation. In this study, we address the following questions: 1) What are the critical environmental variables that influence the distribution of L. meyenii ? 2) How has the suitable habitat for L. meyenii changed from the past to the present? 3) How will L. meyenii respond to future climate change? To answer these questions, we quantitatively compare the MaxEnt model, optimized with the ENMeval package, against 10 other models from the biomod2 package to simulate the potential suitable areas for L. meyenii under various climatic conditions across different time periods. We aim to reconstruct the past, present, and future potential suitable areas for L. meyenii , understand the key environmental factors that drive its distribution, and provide new insights for the conservation and sustainable development of herbaceous plants in the context of global climate change. 2. Data and Methods 2.1 Natural distribution area—western North America L. meyenii is primarily distributed along the western coast of South America within the Andes, encompassing parts of northwestern Argentina, Bolivia, Peru, and the western region of Brazil (Fig. 1 ). The Andes Mountain Range stretches from north to south, with its average elevation that decreasing and then increasing along its length. Within the latitudinal range of 10–30° S, the average altitude is ca. 3000 m, which coincides with the area where L. meyenii is most densely populated [ 36 ]. The climate in western South America is highly complex, shaped by the perennial effects of the Peru Current, the barrier created by the Andes, and the South American monsoon system. These factors contribute to a diverse range of climates from north to south, including tropical arid, tropical humid, savanna, semi-arid, and temperate oceanic climates [ 37 , 38 , 39 ]. 2.2 Species distribution and climate data, and environmental variables The present-day distribution data for L. meyenii were obtained from the Global Biodiversity Information Facility ( http://www.gbif.org ; GBIF.org, 2025) and subsequently cleaned [ 40 ]. We first used ENMTools to eliminate duplicate coordinates and performed random gridding, ensuring that only one occurrence per 5′ × 5′ grid cell was retained. We then applied the “clean_coordinates” function from the “CoordinateCleaner” R package v3.0.1 [ 41 ] to exclude records with potentially erroneous geographical coordinates. Based on previous studies and information from Plants of the World Online ( https://www.powo.science.kew.org ), we confirmed that L. meyenii is native only to northwestern Argentina, Bolivia, northern Chile, and Peru, thus excluded all records outside these regions [ 42 , 43 ]. We eventually obtained 64 valid natural distribution records for L. meyenii . We selected 22 environmental variables from the WorldClim database ( http://worldclim.org ), including 19 bioclimatic and 3 topographic variables (Table S1 ). The bioclimatic variables for these periods were chosen at a spatial resolution of 5′ (ca. 85 km²). Present-day climate data represent average conditions from 1970 to 2020, while past and future climate data were derived from various global climate models (GCMs). We simulated the potential distributions of L. meyenii for the past (Last Glacial Maximum—LGM, ca. 22,000 years ago; Mid-Holocene—MH, ca. 6000 years ago), present, and future (for 2050 and 2090, representing averages from 2040–2060 and 2080–2100, respectively). Averages of climate data for the LGM were obtained from three available GCMs (CCSM4, MIROC-ESM, and MPI-ESM-P), while averages for the MH were obtained from nine GCMs (BCC-CSM1-1, CCSM4, CNRM-CM5, HadGEM2-CC, HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM, MPI-ESM-P, and MRI-CGCM3) in WorldClim v1.4. Future climate data were averaged from three GCMs (FIO-ESM-2-0, MPI-ESM1-2-HR, and MRI-ESM2-0), which performed better than other models across different regions [ 44 ]. The two future periods were chosen to illustrate both short-term and long-term trends in habitat changes. For each period, we incorporated two climate scenarios from CMIP6 GCM: SSP245 (a moderate-emission scenario) and SSP585 (a high-emission and pessimistic scenario) [ 45 ]. To minimize the impact of multicollinearity among environmental variables on model accuracy and to prevent overfitting, we conducted a Pearson correlation analysis using ENMTools software. Subsequently, we excluded highly correlated variables (|r| < 0.8), and selected ten environmental variables for simulation: annual mean temperature (bio1), mean diurnal range (bio2), temperature annual range (bio7), annual precipitation (bio12), precipitation of driest month (bio14), precipitation seasonality (bio15), precipitation of coldest quarter (bio19), aspect, slope and elevation (Fig. S1 ). 2.3 Model selection We employed both ENMeval and biomod2 packages for model analysis [ 18 , 19 ]. All environmental variable data were converted to ASCII format using ArcGIS 10.8.1. The ENMeval package was utilized to optimize MaxEnt model parameters [ 19 ], enabling us to test various feature combination (FC) and regularization multiplier (RM) to identify the most suitable settings. The MaxEnt software offers five types of FC: L (linear), Q (quadratic), H (hinge), P (product), and T (threshold). To refine the model, we tested RM values from 0.5 to 6 in increments of 0.5, resulting in a total of 12 RM values. Additionally, we evaluated six different FC combinations: L, LQ, H, LQH, LQHP, and LQHPT. In total, we assessed 72 parameter combinations using the corrected Akaike Information Criterion (AICc), selecting the configuration with the lowest AICc value to ensure an optimal balance between model fit and complexity. A model is considered optimal when delta.AICc = 0. The best-performing configuration is FC = LQHP and RM = 3, with delta.AICc = 0 (Table S2). We then used the biomod2 package to construct models with 11 different algorithms for simulations: artificial neural network (ANN), classification tree analysis (CTA), flexible discriminant analysis (FDA), generalized additive model (GAM), generalized boosting model (GBM), generalized linear model (GLM), multivariate adaptive regression splines (MARS), maximum entropy model (MaxEnt), random forest (RF), surface range envelope (SRE), and extreme gradient boosting (XGBoost). The optimized MaxEnt model parameters from the previous step were implemented in biomod2, while all other models were executed with their default settings. To minimize model uncertainty, the dataset was randomly partitioned into a training set (75%) and a testing set (25%), with each model repeated 20 times. Following the recommendations of the biomod2 team, we generated 200 pseudo-absence (PA) points, which is three times the number of occurrence points [ 46 ]. Finally, the selection of the best-performing model was based on the receiver operating characteristic (ROC) and true skill statistics (TSS) [ 47 , 48 , 49 ]. We analyzed the area under the curve (AUC) and TSS values to compare the performance of 11 models in simulating the potential suitable habitats for L. meyenii . Model performance was evaluated with the following criteria: (1) poor (AUC < 0.6), (2) fair (0.6 ≤ AUC < 0.8), (3) good (0.8 ≤ AUC < 0.9), and (4) excellent (0.9 ≤ AUC < 1) [ 50 , 51 ]. Additionally, a TSS value greater than 0.8 is generally considered indicative of excellent model performance [ 52 ]. 2.4 Classification of habitat suitability and centroid analysis of suitable areas We applied the natural breaks method in ArcGIS 10.8.1 to classify habitat suitability into four categories: 0–0.2 (suitable), 0.2–0.4 (lowly suitable), 0.4–0.6 (moderately suitable), and 0.6–1 (highly suitable) [ 5 ]. These categories were subsequently reclassified as follows: 0–0.2 = 0, 0.2–0.4 = 1, 0.4–0.6 = 2, and 0.6–1 = 4. Raster calculations were performed by subtracting suitability maps of earlier periods from those of later periods, where values 0 indicate areas of contraction, unchange, and expansion, respectively. This analysis provides insights into the temporal dynamics of potentially suitable distribution areas for L. meyenii . Using the centroid calculation function in ArcGIS 10.8.1, we also calculated the areas of habitat suitability classified as moderate or higher and identified the centroids of potentially suitable habitats for L. meyenii across three time periods: the past, the present and the future in both South America and East Asia. This analysis enables us to track shifts in the centroids of suitable habitats for L. meyenii over time in response to climate change. By examining these centroid movements, we can understand how favorable growing conditions for L. meyenii have geographically relocated as a result of changing climatic factors. 3. Results 3.1 The random forest (RF) has the optimal performance Based on the AUC and TSS values, and the actual simulation outcomes, the random forest (RF) model shows the best performance, with an AUC of 0.9888 and a TSS of 0.8726 (Fig. 2 ). These metrics demonstrate its superior accuracy in predicting the distribution of suitable habitats for L. meyenii . 3.2 Elevation and temperature annual range (bio7) are the most critical factors The present-day distribution of L. meyenii is shaped by both climatic and topographic factors (Fig. 3 ). The most significant environmental variables influencing this distribution are elevation (47.6%), temperature annual range (bio7) (24.1%), mean diurnal range (bio2) (10.4%), slope (8.6%), annual mean temperature (bio1) (4.4%), precipitation seasonality (bio15) (2.2%), precipitation of coldest quarter (bio19) (1%), annual precipitation (bio12) (0.8%), precipitation of driest month (bio14) (0.7%), and aspect (0.2%) (Fig. 3 ). Consequently, elevation and the temperature annual range (bio7) emerge as the most critical factors that influence the current distribution of L. meyenii . The probability of occurrence for L. meyenii shows a positive correlation with elevation, bio2, and slope. Specifically, L. meyenii thrives at elevations above 2000 m, where the mean diurnal range exceeds 13 ℃, and slopes are greater than 1°. Conversely, the probability of occurrence is negatively correlated with temperature annual range (bio7). This indicates that L. meyenii grows better when temperature annual range is less than 20 ℃. 3.3 Potential distribution under past, current, and future climate conditions The simulation results suggest that highly suitable areas are predominantly concentrated in the Andes Mountains of western South America, consistent with current sampling locations, while moderately suitable areas are mainly found in the Tibetan Plateau, and lowly suitable areas are more widely dispersed (Fig. 5 ). Specifically, suitable habitats in the Tibetan Plateau have undergone significant changes through time: 1) During the LGM and MH, the habitat was predominantly of moderate suitability; 2) At present and in 2050 (under the SSP245 scenario), moderately and lowly suitable areas are interspersed; and 3) By 2050 (SSP585) and 2090 (for both SSP245 and SSP585), a larger areaclassified as highly suitable appears along the southern edge of the Tibetan Plateau. This reaches its maximum extent in 2090 under the SSP585 scenario. During the LGM, the total area of potential suitable habitat was 1190.5×10 4 km 2 , the largest among all time periods, although it had the smallest highly suitable area, covering only 68.3×10 4 km 2 . From the past to present, the total suitable habitat has decreased by ca. 12%. From the present to 2090 under the SSP245 scenario, the total suitable habitat area will decrease from 1047.5×10 4 km 2 to 976.1×10 4 km 2 , with a reduction of 6.81%. However, the highly suitable area is projected to decline from 70.0×10 4 km 2 to 74.3×10 4 km 2 , with an increase of 6.1%, while the moderately suitable area will rise from 170.0×10 4 km 2 to 171.2×10 4 km 2 , with an increase of 0.72%. In comparison, from the present to 2090 under the SSP585 scenario, the total suitable habitat area will decrease significantly to 957.4×10 4 km 2 , with a decline of 8.6%. In contrast, both the highly suitable and moderately suitable areas are projected to increase substantially, which reach 78.3×10 4 km 2 with an increase of 11.75% and 174.7×10 4 km 2 with an increase of 2.01%, respectively. Table 1 Temporal changes in the area (10 4 km 2 ) of suitable habitats for Lepidium meyenii . Abbreviations: LGM = Last Glacial Maximum, MH = Mid-Holocene. Period Lowly suitable Moderately suitable Highly suitable Total LGM 895.7 226.5 68.3 1190.5 MH 768.0 231.7 77.9 1078.6 Present 807.4 170.0 70.0 1047.5 SSP245 (2050) 742.5 168.2 72.7 983.3 SSP585 (2050) 726.1 163.8 75.5 965.3 SSP245 (2090) 730.6 171.2 74.3 976.1 SSP585 (2090) 704.4 174.7 78.3 957.4 3.4 Changes in potential habitat suitability since the LGM Generally, both historical data and future projections indicate that the area of habitat contraction for L. meyenii globally exceeds the area of habitat expansion. Additionally, the extent of both habitat expansion and contraction was more pronounced in the past (Fig. 6 , Table 2 ). Between the LGM and the MH, as well as from the MH to the present, an opposing trend emerged in the areas with expansion and contraction across South America, Africa, and East Asia. Specifically, from the LGM to the MH, the global potential suitable habitat contracted by 346×10 4 km 2 mainly in southern and eastern Africa, and expanded by 260.6×10 4 km 2 primarily along the edges of the Tibetan Plateau, Southeast Asia, and northern South America. From the MH to the present, the global potential suitable habitat contracted by 312.5×10 4 km 2 mainly in the Tibetan Plateau and northern South America, with an expansion of 204×10 4 km 2 distributed without significant concentration. In the future, habitat contraction is expected primarily along the southern edge of the Tibetan Plateau, with additional scattered regions of contraction in southern Africa and northwestern China. Under the SSP245 scenario, changes in suitable habitats for L. meyenii are moderate. By 2050, the area of contraction is projected to be 126.7×10 4 km 2 , while the area undergoing expansion is estimated at 71.8×10 4 km 2 ; in 2090, these areas are expected to increase to 149.2×10 4 km 2 and 93.5×10 4 km 2 , respectively. In contrast, under the SSP585 scenario, changes in suitable habitats for L. meyenii are expected to be more pronounced. By 2050, the areas with contraction and expansion will be 152.6×10 4 km 2 and 80.7×10 4 km 2 ; By 2090, these areas are projected to expand further to 210.9×10 4 km 2 for contraction and 147.7×10 4 km 2 for expansion. Table 2 Temporal changes in the distribution area (10 4 km 2 ) for Lepidium meyenii . LGM = Last Glacial Maximum, MH = Mid-Holocene. Period Unchanged Contracted Expanded LGM–MH 12887.4 346.0 260.6 MH–Present 12977.5 312.5 204.0 Present–SSP245 (2050) 13295.5 126.7 71.8 SSP245 (2050)–SSP245 (2090) 13251.3 149.2 93.5 Present–SSP585 (2050) 13260.8 152.6 80.7 SSP585 (2050)–SSP585 (2090) 13135.3 210.9 147.7 3.5 Temporal centroid migration of potential suitable areas for L. meyenii The migration trend is more pronounced in the past compare to the future. In South America, the centroid of suitable habitats for L. meyenii generally shifts northward over time (Fig. 7 a). From the LGM to the MH, the centroid moves north by 135 km, from coordinates 9.71° S, 67.44° W to 8.49° S, 67.41° W. By the present day, it has further shifted 50.8 km to 8.03° S, 67.35° W. Under the SSP245 scenario, the centroid migrates northwest by 28.3 km to 7.87° S, 67.55° W by 2050, and then further to 7.35° S, 67.47° W, with a total movement of 58.1 km by 2090. In the SSP585 scenario, the centroid is expected to move to 7.73° S, 67.48° W by 2050, with a shift of 35.8 km, and then slightly to 7.66° S, 67.44° W by 2090, with a distance of 9.6 km. In East Asia, the centroid migration generally follows an initial southeastward to westward trend (Fig. 7 b). From the LGM to the MH, the centroid shifts southeast by 252.8 km from 34.03° N, 93.93° E to 31.96° N, 95.12° E. By the present day, it has moved west to 32.19° N, 92.68° E with 231.9 km. Under the SSP245 scenario, the centroid shifts with 61.3 km further west to 32.11° N, 92.03° E by 2050, and then to 32.33° N, 92.34° E by 2090, with a further movement of 38.1 km. In the SSP585 scenario, the centroid moves to 32° N, 92.25° E with 45 km, then to 32.45° N, 92.57° E with 56 km. 4. Discussion 4.1 The evaluation of model performance We optimized the MaxEnt model with the ENMeval package and compared it with other models available in the biomod2 package. The results indicate that the random forest (RF) model outperforms others in simulating the potential suitable habitats of L. meyenii through time. The optimized MaxEnt model shows a significant improvement over the default parameters, with an increase of 0.08 in AUC and 0.05 in TSS, though its accuracy remained lower than that of the RF model. As an ensemble learning method based on the CART model, RF is widely used in ecology and biogeography due to its stability, insensitivity to parameter tuning, and ability to maintain high accuracy even with small sample sizes [ 53 , 54 ]. While the RF model showed superior performance in this study, the best model may differ among various species. Many studies have opted for ensemble models in biomod2 to overcome the limitations of single models [ 55 , 56 ]. However, ensemble models do not always outperform an individual model, as an optimized single model can sometimes provide better predictive performance [ 57 ]. Therefore, we emphasize the importance of an integrative evaluation of different models with quantitative methods to identify the optimal model and ensure the accuracy and robustness of the simulation results. 4.2 Key environmental variables and potential distribution in the present day The most critical factors the influence the potential suitable habitat distribution of L. meyenii are elevation, temperature annual range (bio7), mean diurnal range (bio2), and slope (Figs. 3 , 4 ). Elevation and slope are topographic variables, while bio7 and bio2 relate to temperature. Precipitation has a minor role in determining species distribution. L. meyenii is primarily found in areas with elevations exceeding 2,000 m, a mean diurnal temperature range over 15°C, and an annual temperature range below 20°C (Fig. 4 ). This suggests that the species is better adapted to high-altitude environments characterized by significant diurnal temperature variations and relatively low annual temperature fluctuations. The Andes Mountains in western South America represent the largest area of natural distribution for plants, while Yunnan in China is known for the largest area of cultivated introduction [ 58 ]. These regions share similar climatic characteristics influenced by monsoonal climates, which promote the growth of L. meyenii [ 59 , 60 , 61 ]. Genomic sequencing and analysis of L. meyenii reveal its molecular mechanisms that enable its adaptation to extreme high-altitude environments characterized by low temperatures and intense ultraviolet radiation [ 24 ]. This study further confirms earlier results. A low annual temperature range provides a relatively stable environment for growth, while a high diurnal temperature range enables plants to optimize daytime photosynthesis and reduce nighttime respiration, and this thereby promotes organic matter accumulation [ 62 ]. However, species distribution is influenced not only by climate and topography but also by factors such as sunlight, soil, and water availability [ 63 ]. Future studies on L. meyenii distribution should incorporate a wider array of environmental variables. The primary potential suitable habitats for L. meyenii at present are located in western South America and the southern edge of the Tibetan Plateau. Both regions feature alpine and plateau climates, are situated in subtropical zones, and have elevations exceeding 2,000 m, along with an annual temperature range below 20°C and a mean diurnal temperature range above 15°C, and this makes them highly conductive to L. meyenii growth. This climate type is characterized by low annual temperature variation, significant diurnal temperature fluctuations, strong ultraviolet radiation, low atmospheric pressure, and high wind speeds. The current potential suitable habitats for L. meyenii align well with its natural distribution in western South America and its cultivated areas in Yunnan, China (Fig. 3 ). 4.3 Changes in potential suitable habitat from past to present During the LGM–MH, the contraction of potential suitable habitat for L. meyenii predominantly occurred in central South America and eastern and southern Africa (Fig. 6 a), while expansion was primarily observed along the edges of the Tibetan Plateau, Southeast Asia, and northern South America. This pattern may be associated with global warming and regional monsoon dynamics during the LGM–MH, as both factors contributed to a decrease in the annual temperature range (bio7). The intensification of the South Asian and East Asian monsoons during this period created a more favorable climatic condition for the survival of L. meyenii along the edges of the Tibetan Plateau and in Southeast Asia, and thereby this facilitated the expansion of its suitable habitat range [ 64 , 65 ]. In the LGM, cooling in the Northern Hemisphere caused a southward shift of the Intertropical Convergence Zone (ITCZ). In contrast, during the MH, the ITCZ shifted northward, and this led to intensified monsoons in northern South America while weakening them in central and southern regions [ 66 , 67 ]. Consequently, as northern South America experienced significant habitat expansion, central regions saw a greater contraction (Fig. 6 a). By the MH, the centroid of the potential suitable habitat shifted northward (Fig. 7 a). This suggests that, L. meyenii was inclined to expand toward lower latitudes in South America. During the MH, Africa experienced significantly warmer climatic conditions, which were unfavorable for L. meyenii , a cold-adapted alpine herb. This resulted in substantial contractions of suitable habitats in eastern and southern Africa. Similar distribution shifts have been observed in Xerophyta species across Africa [ 68 ]. From the MH to the present, the contraction of suitable habitat for L. meyenii has primarily occurred in the Tibetan Plateau and northern South America, with only limited expansion observed (Fig. 6 b). This contraction may be closely related to elevation-dependent warming (EDW)—a phenomenon where high-altitude regions experience more rapid temperature increases compared to low-altitude areas due to greenhouse gas emissions [ 68 ]. Both the Tibetan Plateau and the Andes Mountains are affected by EDW, which accelerates climate change in alpine ecosystems and modifies vegetation distribution [ 69 ]. This process has led to habitat contraction or shifts, with negative consequences for biodiversity [ 39 ]. When comparing the LGM–MH and MH–Present periods, the primary areas with contraction and expansion for L. meyenii show opposite patterns, particularly in the Tibetan Plateau and northern-central South America. Similarly, many high-altitude, cold-adapted plant species have undergone comparable distribution shifts during these two periods [ 70 ]. We conclude that natural environmental factors primarily drove changes in plant distribution during the LGM–MH, while human activities have increasingly influenced distribution during the MH–present period, particularly under the EDW effect in high-altitude regions. Therefore, future studies should integrate human activities as an environmental variable in distribution models to better understand their impact on shifts in species habitat. 4.4 Changes in potential suitable habitat from the present to the future In the future, the most significant expansion of suitable habitat for L. meyenii is projected to occur along the southern margin of the Tibetan Plateau, while the most substantial contraction is observed in the Tarim Basin of northwestern China and southern Africa. Although the suitable habitat in South America shows minimal overall change, its centroid continues to shift northward. Specifically, in the SSP245 scenario, global average temperatures are anticipated to rise by ca. 2.3–4.6°C above pre-industrial levels, while in the SSP585 scenario, the increase could be 6.6–14.1°C [ 3 ]. The Tarim Basin is thought to become even hotter under these future climate conditions [ 71 ]. Given its significant diurnal and annual temperature variations and an average elevation below 2000 m, this region is likely to become unsuitable for L. meyenii growth. Although southern Africa includes the South African Plateau and the Drakensberg Mountains, its average elevation remains below 2000 m, and it is characterized by a tropical savanna climate, and this makes it a unsuitable habitat for L. meyenii . In contrast, the Tibetan Plateau, with an average elevation exceeding 4000 m, is expected to experience overall warming in the future. This warming trend is particularly pronounced in the northern region, with significant increases in winter temperatures anticipated, and this leads to a further reduction in annual temperature variation [ 72 ]. Yunnan Province in China has a subtropical monsoon climate, marked by large diurnal temperature variations, minimal annual temperature fluctuations, and mild weather throughout the year [ 73 ], and this makes it highly suitable for L. meyenii growth. However, under the SSP585 scenario in 2090, a contraction of suitable habitat in northwestern Yunnan was observed. likely attributed to future climate warming and decreased precipitation in the area [ 74 ]. Given the topography and projected future climatic conditions of both the southern margin of the Tibetan Plateau and Yunnan Province, we propose that the southern Tibetan Plateau may become a more favorable habitat for L. meyenii in the future. Additionally, the distribution of certain plant species and the area of alpine humid forests in the southern margin of the Tibetan Plateau were predicted to increase under future climate conditions [ 75 ]. We therefore suggest that this region could become one of the most suitable habitats for L. meyenii and recommend establishing conservation areas to support its future sustainability. As the climate warms, shifts in plant niches may alter interspecific competition. In planning these conservation areas, it is essential to consider the interactions between herbaceous plants like L. meyenii and woody plants such as those found in alpine humid forests. Understanding their competitive dynamics will be crucial for ensuring harmonious coexistence among plant species, ultimately promoting the stability and development of the ecosystem. 5. Conclusions Using the species distribution data of L. meyenii alongside multiple environmental variables, we evaluated 11 species distribution models and ultimately selected the Random Forest model as the optimal approach. This model allowed us to pinpoint the key environmental factors that influence the distribution of L. meyenii and to simulate its potential suitable habitats through time. We find that the distribution of L. meyenii is primarily influenced by elevation and temperature, while precipitation has a relatively minor impact. Specifically, L. meyenii favors high elevations, a low temperature annual range (bio7), and a high mean diurnal range (bio2). Driven by climate change, the total area of suitable habitat for L. meyenii has decreased from the past to the future, while the area classified as highly suitable has expanded. Historically, highly suitable habitats for L. meyenii were mainly concentrated in western South America, with moderate suitability observed in the Tibetan Plateau. During the LGM–the MH, climate-driven changes led to the expansion of suitable habitats along the margins of the Tibetan Plateau and in Southeast Asia, while significant contractions occurred in southern and eastern Africa. However, from the MH to the present, this distribution pattern has reversed, with a contraction in suitable habitats on the Tibetan Plateau. In the future, western South America is projected to continue serving as a stable and highly suitable region for L. meyenii . Meanwhile, as time progresses and greenhouse gas emissions intensify, highly suitable areas along the southern margin of the Tibetan Plateau are expected to expand further. This suggests that the region may emerge as an important potential area for the cultivation and conservation of L. meyenii , alongside the Andes Mountains. Centroid analysis indicates that in South America, the centroid of suitable habitats for L. meyenii will continue to shift northward. In East Asia, the centroid of suitable habitats shows a southeastward movement followed by a westward shift in the past, while future migration trends appear to be less pronounced. This study will provide an important basis for better understanding the adaptation mechanisms of alpine herbaceous plants like L. meyenii to environmental changes and offer significant guidance for the future establishment of conservation areas for L. meyenii . Declarations Author contributions Zeyu Qin: Conceptualization, Methodology, Investigation, Data curation, Writing - Original Draft, Writing - Review & Editing. Huasheng Huang: Conceptualization, Methodology, Investigation, Writing - Review & Editing, Supervision, Funding acquisition. Xuanqi Liu: Validation, Visualization, Writing - Review & Editing. Xia Meng: Validation, Visualization, Writing - Review & Editing. Minqiao Li: Visualization, Writing - Review & Editing. Funding This study was supported by the Starting Grant for Introduced Talents of Sun Yat-sen University, the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (No. 24qnpy021), and the General Project of Basic and Applied Basic Research of Guangzhou Bureau of Science and Technology (No. 2025A04J4384). Data availability Data is provided within the manuscript or supplementary information files. Clinical trial number Not applicable. Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Acknowledgements Not applicable. Competing Interest The authors declare no competing interests. References Intergovernmental Panel on Climate Change (IPCC). (2012). Managing the risks of extreme events and disasters to advance climate change adaptation (SREX). In C. B. Field editors, Special Report of the Intergovernmental Panel on Climate Change (p. 582). Cambridge University Press. https://doi.org/10.1017/CBO9781139177245 Filonchyk M, Peterson MP, Zhang L, Hurynovich V, He Y. Greenhouse gases emissions and global climate change: Examining the influence of CO 2 , CH 4 , and N 2 O. Sci Total Environ. 2024;935:173359. https://doi.org/10.1016/j.scitotenv.2024.173359 . Intergovernmental Panel on Climate Change (IPCC). Climate Change 2021: The Physical Science Basis: Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. 1st ed. Cambridge University Press; 2021. https://doi.org/10.1017/9781009157896 . Lu M. The research progress on phytogeography: Global plant distribution patterns and environmental driving mechanisms. Geographical Res Bull. 2025;4:3–25. https://doi.org/10.50908/grb.4.0_3 . Wang P, Luo W, Zhang Q, Han S, Jin Z, Liu J, Li Z, Tao J. Assessing the impact of climate change on three Populus species in China: Distribution patterns and implications. Global Ecol Conserv. 2024;50:e02853. https://doi.org/10.1016/j.gecco.2024.e02853 . Zhao G, Cui X, Sun J, Li T, Wang Q, Ye X, Fan B. Analysis of the distribution pattern of Chinese Ziziphus jujuba under climate change based on optimized biomod2 and MaxEnt models. Ecol Ind. 2021;132:108256. https://doi.org/10.1016/j.ecolind.2021.108256 . Kelly AE, Goulden ML. (2008). Rapid shifts in plant distribution with recent climate change. Proceedings of the National Academy of Sciences, 105(33), 11823–11826. https://doi.org/10.1073/pnas.0802891105 Bellard C, Bertelsmeier C, Leadley P, Thuiller W, Courchamp F. Impacts of climate change on the future of biodiversity. Ecol Lett. 2012;15(4):365–77. https://doi.org/10.1111/j.1461-0248.2011.01736.x . Grimm NB, Chapin FS, Bierwagen B, Gonzalez P, Groffman PM, Luo Y, Melton F, Nadelhoffer K, Pairis A, Raymond PA, Schimel J, Williamson CE. The impacts of climate change on ecosystem structure and function. Front Ecol Environ. 2013;11(9):474–82. https://doi.org/10.1890/120282 . Jezkova T, Wiens JJ. (2016). Rates of change in climatic niches in plant and animal populations are much slower than projected climate change. Proceedings of the Royal Society B: Biological Sciences, 283(1843), 20162104. https://doi.org/10.1098/rspb.2016.2104 Kearney M, Porter W. Mechanistic niche modelling: Combining physiological and spatial data to predict species’ ranges. Ecol Lett. 2009;12(4):334–50. https://doi.org/10.1111/j.1461-0248.2008.01277.x . Elith J, Phillips SJ, Hastie T, Dudík M, Chee YE, Yates CJ. A statistical explanation of MaxEnt for ecologists. Divers Distrib. 2011;17(1):43–57. https://doi.org/10.1111/j.1472-4642.2010.00725.x . Cutler DR, Edwards TC, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ. Random forests for classification in ecology. Ecology. 2007;88(11):2783–92. https://doi.org/10.1890/07-0539.1 . Yee TW, Mitchell ND. Generalized additive models in plant ecology. J Veg Sci. 1991;2(5):587–602. https://doi.org/10.2307/3236170 . Nelder JA, Wedderburn RWM. Generalized Linear Models. J Royal Stat Soc Ser (General). 1972;135(3):370. https://doi.org/10.2307/2344614 . Huang D, An Q, Huang S, Tan G, Quan H, Chen Y, Zhou J, Liao H. Biomod2 modeling for predicting the potential ecological distribution of three Fritillaria species under climate change. Sci Rep. 2023;13(1):18801. https://doi.org/10.1038/s41598-023-45887-6 . Wang R, Wu N, Shi Z, Li C, Jiang N, Fu C, Wang M. Biomod2 for evaluating the changes in the spatiotemporal distribution of Locusta migratoria tibetensis Chen in the Qinghai-Tibet Plateau under climate change. Global Ecol Conserv. 2025;58:e03508. https://doi.org/10.1016/j.gecco.2025.e03508 . Thuiller W, Georges D, Gueguen M, Engler R, Breiner F, Lafourcade B, Patin R, Blancheteau H. (2012). biomod2: Ensemble Platform for Species Distribution Modeling (p. 4.2-6-2) [Dataset]. https://doi.org/10.32614/CRAN.package.biomod2 Muscarella R, Galante PJ, Soley-Guardia M, Boria RA, Kass JM, Uriarte M, Anderson RP. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol Evol. 2014;5(11):1198–205. https://doi.org/10.1111/2041-210X.12261 . Al-Shehbaz I. The genera of Lepidieae (Cruciferae; Brassicaceae) in the southeastern United States. J Arnold Arboretum. 1986;67:265–311. https://doi.org/10.5962/bhl.part.27392 . Warwick SI, Al-Shehbaz IA. Brassicaceae: Chromosome number index and database on CD-Rom. Plant Syst Evol. 2006;259(2–4):237–48. https://doi.org/10.1007/s00606-006-0421-1 . Mummenhoff K, Brüggemann H, Müntzing A. Evidence for the origin of the Mediterranean endemic species Lepidium subulatum and L. suffruticosum (Cruciferae) by interspecific hybridization. Plant Syst Evol. 1997;208(3–4):275–86. https://doi.org/10.1007/BF00937738 . Toledo J. Genetic variability of Lepidium meyenii and other Andean Lepidium species (Brassicaceae) assessed by molecular markers. Ann Botany. 1998;82(4):523–30. https://doi.org/10.1006/anbo.1998.0715 . Zhang J, Tian Y, Yan L, Zhang G, Wang X, Zeng Y, Zhang J, Ma X, Tan Y, Long N, Wang Y, Ma Y, He Y, Xue Y, Hao S, Yang S, Wang W, Zhang L, Dong Y, Sheng J. Genome of plant Maca (Lepidium meyenii) illuminates genomic basis for high-altitude adaptation in the Central Andes. Mol Plant. 2016;9(7):1066–77. https://doi.org/10.1016/j.molp.2016.04.016 . Gonzales GF, Gasco M, Cordova A, Chung A, Rubio J, Villegas L. Effect of Lepidium meyenii (Maca) on spermatogenesis in male rats acutely exposed to high altitude (4340 m). J Endocrinol. 2004;180(1):87–95. https://doi.org/10.1677/joe.0.1800087 . Gonzales GF, Miranda S, Nieto J, Fernández G, Yucra S, Rubio J, Yi P, Gasco M. Red maca (Lepidium meyenii) reduced prostate size in rats. Reproductive Biology Endocrinol. 2005;3(1):5. https://doi.org/10.1186/1477-7827-3-5 . Rubio J, Caldas M, Dávila S, Gasco M, Gonzales GF. Effect of three different cultivars of Lepidium meyenii (Maca) on learning and depression in ovariectomized mice. BMC Complement Altern Med. 2006;6(1):23. https://doi.org/10.1186/1472-6882-6-23 . Stone M, Ibarra A, Roller M, Zangara A, Stevenson E. A pilot investigation into the effect of Maca supplementation on physical activity and sexual desire in sportsmen. J Ethnopharmacol. 2009;126(3):574–6. https://doi.org/10.1016/j.jep.2009.09.012 . Rubio J, Qiong W, Liu X, Jiang Z, Dang H, Chen S-L, Gonzales GF. Aqueous Extract of Black Maca (Lepidium meyenii) on Memory Impairment Induced by Ovariectomy in Mice. Evidence-Based Complement Altern Med. 2011;2011(1):253958. https://doi.org/10.1093/ecam/nen063 . Da Silva Leitão Peres N, Parra Bortoluzzi C, Marques LM, Formigoni LL, Fuchs M, Droval RHB, A. A., Cardoso R, F. A. Medicinal effects of Peruvian maca (Lepidium meyenii): A review. Food Funct. 2020;11(1):83–92. https://doi.org/10.1039/C9FO02732G . Yin D, Qian J, Zhu H. Frontier development in the midst of ecological civilization: Unravelling the production of maca in Yunnan, China. Geoforum. 2019;106:144–54. https://doi.org/10.1016/j.geoforum.2019.08.005 . He J, Yang Z, Li Y, He W, He X, Bi H, Yang W, Wang Y, He J. Situation and development strategy of maca industry in Yunnan. Chin Agric Sci Bull. 2016;32(27):93–8. https://doi.org/10.11924/j.issn.1000-6850.casb15120105 . Chen R, Wei J, Gao Y. A review of the study of active components and their pharmacology value in Lepidium meyenii (Maca). Phytother Res. 2021;35(12):6706–19. https://doi.org/10.1002/ptr.7257 . Cui Z, Yang W-S, Cheng Z, Zhang Z, Li S, Zhao J, López-Vicente M, Wu G-L. Top-down degradation of alpine meadow in the Qinghai-Tibetan Plateau: Gravelization initiate hillside surface aridity and meadow community disappearance. CATENA. 2022;210:105933. https://doi.org/10.1016/j.catena.2021.105933 . Sun Q, Li B, Jiang Y, Chen X, Zhou G. Declined trend in herbaceous plant green-up dates on the Qinghai–Tibetan Plateau caused by spring warming slowdown. Sci Total Environ. 2021;772:145039. https://doi.org/10.1016/j.scitotenv.2021.145039 . Montgomery DR, Balco G, Willett SD. Climate, tectonics, and the morphology of the Andes. Geology. 2001;29(7):579. https://doi.org/10.1130/0091-7613(2001)0292.0.CO;2 . Garreaud RD. The Andes climate and weather. Adv Geosci. 2009;22:3–11. https://doi.org/10.5194/adgeo-22-3-2009 . Garreaud RD, Vuille M, Compagnucci R, Marengo J. Present-day South American climate. Palaeogeogr Palaeoclimatol Palaeoecol. 2009;281(3–4):180–95. https://doi.org/10.1016/j.palaeo.2007.10.032 . Toledo O, Palazzi E, Cely Toro IM, Mortarini L. Comparison of elevation-dependent warming and its drivers in the tropical and subtropical Andes. Clim Dyn. 2022;58(11–12):3057–74. https://doi.org/10.1007/s00382-021-06081-4 . GBIF.org, 2025. GBIF occurrence downloaded on 14th February 2025. Available at https://doi.org/10.15468/dl.d36rsc Zizka A, Silvestro D, Andermann T, Azevedo J, Ritter CD, Edler D, Farooq H, Herdean A, Ariza M, Scharn R, Svantesson S, Wengström N, Zizka V, Antonelli A. CoordinateCleaner: Standardized cleaning of occurrence records from biological collection databases. Methods Ecol Evol. 2019;10(5):744–51. https://doi.org/10.1111/2041-210X.13152 . Dini A, Migliuolo G, Rastrelli L, Saturnino P, Schettino O. Chemical composition of Lepidium meyenii. Food Chem. 1994;49(4):347–9. https://doi.org/10.1016/0308-8146(94)90003-5 . Sandoval M, Okuhama NN, Angeles FM, Melchor VV, Condezo LA, Lao J, Miller MJS. Antioxidant activity of the cruciferous vegetable Maca (Lepidium meyenii). Food Chem. 2002;79(2):207–13. https://doi.org/10.1016/S0308-8146(02)00133-4 . Zhang M-Z, Xu Z, Han Y, Guo W. Evaluation of CMIP6 models toward dynamical downscaling over 14 CORDEX domains. Clim Dyn. 2024;62(6):4475–89. https://doi.org/10.1007/s00382-022-06355-5 . O’Neill BC, Tebaldi C, Van Vuuren DP, Eyring V, Friedlingstein P, Hurtt G, Knutti R, Kriegler E, Lamarque J-F, Lowe J, Meehl GA, Moss R, Riahi K, Sanderson BM. (2016). The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6. Geoscientific Model Development, 9(9), 3461–3482. https://doi.org/10.5194/gmd-9-3461-2016 Barbet-Massin M, Jiguet F, Albert CH, Thuiller W. (2012). Selecting pseudo‐absences for species distribution models: How, where and how many? Methods in Ecology and Evolution, 3(2), 327–338. https://doi.org/10.1111/j.2041-210X.2011.00172.x Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29–36. https://doi.org/10.1148/radiology.143.1.7063747 . Allouche O, Tsoar A, Kadmon R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J Appl Ecol. 2006;43(6):1223–32. https://doi.org/10.1111/j.1365-2664.2006.01214.x . Lobo JM, Jiménez-Valverde A, Real R. AUC: A misleading measure of the performance of predictive distribution models. Glob Ecol Biogeogr. 2008;17(2):145–51. https://doi.org/10.1111/j.1466-8238.2007.00358.x . Lasram BR, Guilhaumon F, Albouy F, Somot C, Thuiller S, W., Mouillot D. The Mediterranean Sea as a ‘cul-de‐sac’ for endemic fishes facing climate change. Glob Change Biol. 2010;16(12):3233–45. https://doi.org/10.1111/j.1365-2486.2010.02224.x . Wang B, Liu DL, O’Leary GJ, Asseng S, Macadam I, Lines-Kelly R, Yang X, Clark A, Crean J, Sides T, Xing H, Mi C, Yu Q. Australian wheat production expected to decrease by the late 21st century. Glob Change Biol. 2018;24(6):2403–15. https://doi.org/10.1111/gcb.14034 . Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159. https://doi.org/10.2307/2529310 . Fox EW, Hill RA, Leibowitz SG, Olsen AR, Thornbrugh DJ, Weber MH. Assessing the accuracy and stability of variable selection methods for random forest modeling in ecology. Environ Monit Assess. 2017;189(7):316. https://doi.org/10.1007/s10661-017-6025-0 . Luan J, Zhang C, Xu B, Xue Y, Ren Y. The predictive performances of random forest models with limited sample size and different species traits. Fish Res. 2020;227:105534. https://doi.org/10.1016/j.fishres.2020.105534 . Zhang X, Wu W, Liang Y. Analysis of the potential distribution of shoot blight of larch in China based on the optimized MaxEnt and Biomod2 ensemble models. Forests. 2024;15(8):1313. https://doi.org/10.3390/f15081313 . Yang L, Jia H, Hua Q. 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. 2025;25(1):351. https://doi.org/10.1186/s12870-025-06370-8 . Hao T, Elith J, Guillera-Arroita G, Lahoz‐Monfort JJ. A review of evidence about use and performance of species distribution modelling ensembles like BIOMOD. Divers Distrib. 2019;25(5):839–52. https://doi.org/10.1111/ddi.12892 . Li Y, Li PY, Zhou XT, Zhou LY, Huang LQ, Yang G, Chen M. Research and application progress of Lepidium meyenii (maca) [in Chinese]. China J Chin Materia Med. 2018;43(23):4599–607. https://doi.org/10.19540/j.cnki.cjcmm.2018.0122 . De Carvalho LMV, Jones C, editors. (2016). The Monsoons and Climate Change: Observations and Modeling. Springer International Publishing. https://doi.org/10.1007/978-3-319-21650-8 Espinoza JC, Garreaud R, Poveda G, Arias PA, Molina-Carpio J, Masiokas M, Viale M, Scaff L. Hydroclimate of the Andes Part I: Main Climatic Features. Front Earth Sci. 2020;8:64. https://doi.org/10.3389/feart.2020.00064 . Turner AG, Annamalai H. Climate change and the South Asian summer monsoon. Nat Clim Change. 2012;2(8):587–95. https://doi.org/10.1038/nclimate1495 . Collatz GJ, Bounoua L, Los SO, Randall DA, Fung IY, Sellers PJ. A mechanism for the influence of vegetation on the response of the diurnal temperature range to changing climate. Geophys Res Lett. 2000;27(20):3381–4. https://doi.org/10.1029/1999GL010947 . Boulangeat I, Gravel D, Thuiller W. Accounting for dispersal and biotic interactions to disentangle the drivers of species distributions and their abundances. Ecol Lett. 2012;15(6):584–93. https://doi.org/10.1111/j.1461-0248.2012.01772.x . Suganuma Y, Yamazaki T, Kanamatsu T. South Asian monsoon variability during the past 800 kyr revealed by rock magnetic proxies. Q Sci Rev. 2009;28(9–10):926–38. https://doi.org/10.1016/j.quascirev.2008.12.014 . Sun Y, Kutzbach J, An Z, Clemens S, Liu Z, Liu W, Liu X, Shi Z, Zheng W, Liang L, Yan Y, Li Y. Astronomical and glacial forcing of East Asian summer monsoon variability. Q Sci Rev. 2015;115:132–42. https://doi.org/10.1016/j.quascirev.2015.03.009 . Berman AL, Silvestri GE, Tonello MS. Differences between Last Glacial Maximum and present-day temperature and precipitation in southern South America. Q Sci Rev. 2016;150:221–33. https://doi.org/10.1016/j.quascirev.2016.08.025 . Berman AL, Silvestri GE, Tonello MS. On the differences between Last Glacial Maximum and Mid-Holocene climates in southern South America simulated by PMIP3 models. Q Sci Rev. 2018;185:113–21. https://doi.org/10.1016/j.quascirev.2018.02.003 . Wanga VO, Ngarega BK, Oulo MA, Mkala EM, Ngumbau VM, Onjalalaina GE, Odago WO, Nanjala C, Ochieng CO, Gichua MK, Gituru RW, Hu G-W. Projected impacts of climate change on the habitat of Xerophyta species in Africa. Plant Divers. 2024;46(1):91–100. https://doi.org/10.1016/j.pld.2023.05.001 . Mountain Research Initiative EDW Working Group. Elevation-dependent warming in mountain regions of the world. Nat Clim Change. 2015;5(5):424–30. https://doi.org/10.1038/nclimate2563 . Xiao S, Li S, Wang X, Chen L, Su T. Cedrus distribution change: Past, present, and future. Ecol Ind. 2022;142:109159. https://doi.org/10.1016/j.ecolind.2022.109159 . Pan XD, Zhang L, Huang CL. Future climate projection in northwest china with RegCM4.6. Earth Space Sci. 2020;7(2). https://doi.org/10.1029/2019EA000819 . e2019EA000819. Zhou M, Yu Z, Gu H, Ju Q, Gao Y, Wen L, Huang T, Wang W. Evaluation and projections of surface air temperature over the Tibetan Plateau from CMIP6 and CMIP5: Warming trend and uncertainty. Clim Dyn. 2023;60(11–12):3863–83. https://doi.org/10.1007/s00382-022-06518-4 . Zhou J-Q, Huang W, Zhu Y, Li M, Zhou B-T. Climate comfort distribution, change and projection in Yunnan province. Clim Change Res. 2018;14(2):144–54. https://doi.org/10.12006/j.issn.1673-1719.2017.065 . Ren J, Zhang W, Wan Y, Chen Y. Advances in the Research of Yunnan’s Arid Climate and Extreme Drought. Atmospheric Clim Sci. 2017;7(1):23–35. https://doi.org/10.4236/acs.2017.71003 . Fan Z, Bai X. Scenarios of potential vegetation distribution in the different gradient zones of Qinghai-Tibet Plateau under future climate change. Sci Total Environ. 2021;796:148918. https://doi.org/10.1016/j.scitotenv.2021.148918 . Additional Declarations No competing interests reported. Supplementary Files SupplementaryMaterialQinetal.docx 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-6472833","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":451904890,"identity":"6f185862-4a80-4eff-b5b1-0ec038b69992","order_by":0,"name":"Zeyu Qin","email":"","orcid":"","institution":"School of Geography and Planning, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Zeyu","middleName":"","lastName":"Qin","suffix":""},{"id":451904891,"identity":"67a875ea-6c67-42f3-8910-161da86d612b","order_by":1,"name":"Huasheng Huang","email":"data:image/png;base64,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","orcid":"","institution":"School of Geography and Planning, Sun Yat-sen University","correspondingAuthor":true,"prefix":"","firstName":"Huasheng","middleName":"","lastName":"Huang","suffix":""},{"id":451904892,"identity":"a4d70f80-cd04-4330-91d2-8e3e43b67204","order_by":2,"name":"Xuanqi Liu","email":"","orcid":"","institution":"School of Geography and Planning, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Xuanqi","middleName":"","lastName":"Liu","suffix":""},{"id":451904893,"identity":"b55e932f-7a94-4feb-9a0e-37188ed3c3dc","order_by":3,"name":"Xia Meng","email":"","orcid":"","institution":"School of Geography and Planning, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Xia","middleName":"","lastName":"Meng","suffix":""},{"id":451904894,"identity":"d7ea7bb8-da56-4bd6-9a27-6ee414e2f781","order_by":4,"name":"Minqiao Li","email":"","orcid":"","institution":"School of Geography and Planning, Sun Yat-sen University","correspondingAuthor":false,"prefix":"","firstName":"Minqiao","middleName":"","lastName":"Li","suffix":""}],"badges":[],"createdAt":"2025-04-17 14:53:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6472833/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6472833/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82003852,"identity":"4cf0c045-c574-4452-94de-671d2c91ae2d","added_by":"auto","created_at":"2025-05-05 20:45:04","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":521976,"visible":true,"origin":"","legend":"\u003cp\u003eGeographically valid occurrence records of \u003cem\u003eLepidium meyenii\u003c/em\u003e sourced from GBIF (http://www.gbif.org). The base map and the embedded image of \u003cem\u003eL. meyenii\u003c/em\u003e were obtained from Natural Earth (\u003ca href=\"https://www.naturalearthdata.com/\"\u003ehttps://www.naturalearthdata.com/\u003c/a\u003e) and Wikipedia (\u003ca href=\"https://en.wikipedia.org/\"\u003ehttps://en.wikipedia.org/\u003c/a\u003e, credit: Vahe Martirosyan, under the CC BY 2.0 license, https://creativecommons.org/licenses/by/2.0/), respectively.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/0ba0bd5bac374727c6b59898.png"},{"id":82003271,"identity":"76dbd443-ffee-419b-b467-2869b0259ba6","added_by":"auto","created_at":"2025-05-05 20:37:03","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":25459,"visible":true,"origin":"","legend":"\u003cp\u003eEvaluation of the accuracy of 11 models for simulating the global potential distribution for \u003cem\u003eLepidium meyenii\u003c/em\u003eusing the biomod2 package. The error bars represent the standard deviation of area under the curve (AUC) and true skill statistic (TSS). The models evaluated include artificial neural network (ANN), classification tree analysis (CTA), flexible discriminant analysis (FDA), generalized additive model (GAM), generalized boosting model (GBM), generalized linear model (GLM), multiple adaptive regression splines (MARS), default parameters of the maximum entropy model (MaxEnt_default), optimal parameters of the maximum entropy model (MaxEnt_optimal), random forest (RF), surface range envelope (SRE), and extreme gradient boosting (XGBoost).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/eb6b1f473fe876199d21d474.png"},{"id":82003851,"identity":"8c525fee-d10b-47d1-82a1-ccd45df75238","added_by":"auto","created_at":"2025-05-05 20:45:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":11094,"visible":true,"origin":"","legend":"\u003cp\u003eRelative importance of environmental variables on the current potential distribution of \u003cem\u003eLepidium meyenii\u003c/em\u003e.\u003cem\u003e \u003c/em\u003eThe error bars represent the standard deviation of relative importance for each environmental variable. These variables include annual mean temperature (bio1), mean diurnal range (bio2), temperature annual range (bio7), annual precipitation (bio12), precipitation of driest month (bio14), precipitation seasonality (bio15), precipitation of coldest quarter (bio19), aspect, slope, and elevation.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/e74a59974331e640b0e72b99.png"},{"id":82003274,"identity":"a8dc8263-53c2-4ec3-aa9e-6b7b4eafb3f6","added_by":"auto","created_at":"2025-05-05 20:37:04","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":20641,"visible":true,"origin":"","legend":"\u003cp\u003eResponse curves of the distribution of \u003cem\u003eLepidium meyenii\u003c/em\u003e in relation to the four most important environmental variables. They are elevation (meters), mean diurnal range (bio2) (℃), temperature annual range (bio7) (℃) and slope (°). The grey background indicates the fluctuations observed across multiple random simulations.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/7d989a3c690a0c9f2e972234.png"},{"id":82003276,"identity":"01e59126-55d1-4ea1-8f80-3fdb3151d9b6","added_by":"auto","created_at":"2025-05-05 20:37:04","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":194301,"visible":true,"origin":"","legend":"\u003cp\u003eGlobal potential suitable distribution of \u003cem\u003eLepidium meyenii\u003c/em\u003eacross different time periods using the random forest (RF) model. Abbreviations: LGM = Last Glacial Maximum, MH = Mid-Holocene. Habitat suitability classifications are as follows: unsuitable = 0 \u0026lt; habitat suitability value \u0026lt; 0.2, lowly suitable = 0.2 \u0026lt; habitat suitability value \u0026lt; 0.4, moderately suitable = 0.4 \u0026lt; habitat suitability value \u0026lt; 0.6, highly suitable = 0.6 \u0026lt; habitat suitability value \u0026lt; 1.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/065fe2738902376b399b56b3.png"},{"id":82003277,"identity":"41928d23-de76-4f1e-ab0b-54e15ac787ab","added_by":"auto","created_at":"2025-05-05 20:37:04","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":235080,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal changes in potential suitable areas for \u003cem\u003eLepidium meyenii\u003c/em\u003e under various climate scenarios. Abbreviations: LGM = Last Glacial Maximum, MH = Mid-Holocene.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/fa0d51e0769534cfe9af893e.png"},{"id":82004301,"identity":"f2f3af7b-c0a4-409a-8da4-15b47582d641","added_by":"auto","created_at":"2025-05-05 21:01:04","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":87355,"visible":true,"origin":"","legend":"\u003cp\u003eTemporal changes in the migration trends of the geometric centers of potential suitable areas for \u003cem\u003eLepidium meyenii\u003c/em\u003eunder various climate scenarios in South America (a) and East Asia (b). Abbreviations; LGM = Last Glacial Maximum, MH = Mid-Holocene. The lines represent the migration pathways of the centroids, and indicate centroid movement over time: orange—from the past to the present; green—from the present to the future under the SSP245 scenario; blue—from the present to the future under the SSP585 scenario.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/be1448088471b501ee1d741e.png"},{"id":104568896,"identity":"3b1977e6-d055-4276-acce-f99beda61c25","added_by":"auto","created_at":"2026-03-13 12:12:00","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2092215,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/849b7b6f-1e4b-4d5c-8e10-910ffcbfd7a1.pdf"},{"id":82003853,"identity":"bf49d440-79d0-413f-9f9d-5a23b03401fd","added_by":"auto","created_at":"2025-05-05 20:45:04","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":270064,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryMaterialQinetal.docx","url":"https://assets-eu.researchsquare.com/files/rs-6472833/v1/db2ff0251ae26eddd1048537.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Global temporal distribution of Maca (Lepidium meyenii Walp.) under climate change","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eSince the Industrial Revolution, the surge in greenhouse gas emissions resulting from industrialization has led to a rise in global average temperatures by 0.5 ℃. It is projected that by 2100, under moderate and high emissions scenarios (SSP2-4.5 and SSP5-8.5), global temperatures will increase by 2.3\u0026ndash;4.6 ℃ and 6.6\u0026ndash;14.1 ℃, respectively, compared to pre-industrial levels [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Climate change not only alters weather patterns but also exacerbates the frequency and severity of extreme climatic events, such as heavy rainstorms, droughts, and storms [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. As a major consequence of global warming, climate change is increasingly recognized as a key driver of species distribution shifts [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. Many species are moving to higher altitudes and latitudes as temperatures rise [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. These shifts not only change distribution patterns but also have significant implications for biodiversity and ecosystems [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Furthermore, species niche shifts occur at a much slower rate than the pace of climate change [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. This highlights potential risks of habitat mismatch and biodiversity loss. Therefore, it is essential to understand the relationship between climate change and species distribution for developing effective conservation strategies.\u003c/p\u003e \u003cp\u003eSpecies distribution models (SDMs) are developed to investigate the relationship between climate change and species distribution. They analyze the interactions between species and environmental variables to predict potentially suitable distributions. SDMs also incorporate the physiological responses and adaptation mechanisms of species [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Common algorithms used in species distribution modelling include the maximum entropy model (MaxEnt) [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], random forest (RF) [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], generalized additive model (GAM) [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] generalized linear model (GLM) [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e], among others. The biomod2 package in R is widely adopted for species distribution modelling [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. It integrates various algorithms and allows users to customize parameters for different applications [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Selecting the optimal model is important for accurately predicting potentially suitable areas for species, and this can be achieved through the quantitative evaluation of different SDMs. Among these algorithms, the MaxEnt model, when parameter-tuned using the ENMeval R package, is particularly popular. It effectively reduces overfitting compared to default settings and significantly improves predictive performance [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eLepidium\u003c/em\u003e L. is a genus within the Brassicaceae family, and comprises 265 species (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.powo.science.kew.org\u003c/span\u003e\u003cspan address=\"https://www.powo.science.kew.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). It is widely distributed across all continents except Antarctica and can be found in high-altitude regions [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. The genus originated in Eurasia, evolved, and thrived in the Mediterranean region before its long-distance dispersal to the Americas and Australia during the late Neogene or Quaternary periods [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. \u003cem\u003eL. meyenii\u003c/em\u003e is an alpine herbaceous species native to the Andean region of western South America. It diverged ca. 12\u0026nbsp;million years ago [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], and has adapted to the extreme conditions of high altitudes. It is commonly referred to as \"Peruvian ginseng\" and possesses significant economic value due to its rich nutrient profile and numerous health benefit, including enhancing fertility, reducing fatigue, balancing hormone levels, and improving athletic performance [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. Consequently, products derived from \u003cem\u003eL. meyenii\u003c/em\u003e have gained considerable market popularity [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Since the 1990s, cultivation attempts in the United States, Japan, and Germany have met with limited success, while China successfully domesticated \u003cem\u003eL. meyenii\u003c/em\u003e in 2002 and has established a substantial market. For example, in Yunnan Province, the cultivation area of \u003cem\u003eL. meyenii\u003c/em\u003e has reached 22,000 hm\u003csup\u003e2\u003c/sup\u003e [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eClimate change can lead to alpine gravelization and thus result in the degradation of meadow communities [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e], and significantly affect the phenology of alpine herbaceous species [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. The ecological adaptation mechanism of \u003cem\u003eL. meyenii\u003c/em\u003e may change in response to global warming which could impact its distribution. Therefore, it is crucial to investigate the changes in the distribution patterns of \u003cem\u003eL. meyenii\u003c/em\u003e under future climate change scenarios to develop effective cultivation strategies that ensure its sustainable development and conservation.\u003c/p\u003e \u003cp\u003eIn this study, we address the following questions: 1) What are the critical environmental variables that influence the distribution of \u003cem\u003eL. meyenii\u003c/em\u003e? 2) How has the suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e changed from the past to the present? 3) How will \u003cem\u003eL. meyenii\u003c/em\u003e respond to future climate change? To answer these questions, we quantitatively compare the MaxEnt model, optimized with the ENMeval package, against 10 other models from the biomod2 package to simulate the potential suitable areas for \u003cem\u003eL. meyenii\u003c/em\u003e under various climatic conditions across different time periods. We aim to reconstruct the past, present, and future potential suitable areas for \u003cem\u003eL. meyenii\u003c/em\u003e, understand the key environmental factors that drive its distribution, and provide new insights for the conservation and sustainable development of herbaceous plants in the context of global climate change.\u003c/p\u003e"},{"header":"2. Data and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Natural distribution area\u0026mdash;western North America\u003c/h2\u003e \u003cp\u003e \u003cem\u003eL. meyenii\u003c/em\u003e is primarily distributed along the western coast of South America within the Andes, encompassing parts of northwestern Argentina, Bolivia, Peru, and the western region of Brazil (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The Andes Mountain Range stretches from north to south, with its average elevation that decreasing and then increasing along its length. Within the latitudinal range of 10\u0026ndash;30\u0026deg; S, the average altitude is ca. 3000 m, which coincides with the area where \u003cem\u003eL. meyenii\u003c/em\u003e is most densely populated [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. The climate in western South America is highly complex, shaped by the perennial effects of the Peru Current, the barrier created by the Andes, and the South American monsoon system. These factors contribute to a diverse range of climates from north to south, including tropical arid, tropical humid, savanna, semi-arid, and temperate oceanic climates [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].\u003c/p\u003e\u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Species distribution and climate data, and environmental variables\u003c/h2\u003e \u003cp\u003eThe present-day distribution data for \u003cem\u003eL. meyenii\u003c/em\u003e were obtained from the Global Biodiversity Information Facility (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.gbif.org\u003c/span\u003e\u003cspan address=\"http://www.gbif.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e; GBIF.org, 2025) and subsequently cleaned [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. We first used ENMTools to eliminate duplicate coordinates and performed random gridding, ensuring that only one occurrence per 5\u0026prime; \u0026times; 5\u0026prime; grid cell was retained. We then applied the \u0026ldquo;clean_coordinates\u0026rdquo; function from the \u0026ldquo;CoordinateCleaner\u0026rdquo; R package v3.0.1 [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e] to exclude records with potentially erroneous geographical coordinates. Based on previous studies and information from Plants of the World Online (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.powo.science.kew.org\u003c/span\u003e\u003cspan address=\"https://www.powo.science.kew.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), we confirmed that \u003cem\u003eL. meyenii\u003c/em\u003e is native only to northwestern Argentina, Bolivia, northern Chile, and Peru, thus excluded all records outside these regions [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. We eventually obtained 64 valid natural distribution records for \u003cem\u003eL. meyenii\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eWe selected 22 environmental variables from the WorldClim database (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://worldclim.org\u003c/span\u003e\u003cspan address=\"http://worldclim.org\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), including 19 bioclimatic and 3 topographic variables (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). The bioclimatic variables for these periods were chosen at a spatial resolution of 5\u0026prime; (ca. 85 km\u0026sup2;). Present-day climate data represent average conditions from 1970 to 2020, while past and future climate data were derived from various global climate models (GCMs).\u003c/p\u003e \u003cp\u003eWe simulated the potential distributions of \u003cem\u003eL. meyenii\u003c/em\u003e for the past (Last Glacial Maximum\u0026mdash;LGM, ca. 22,000 years ago; Mid-Holocene\u0026mdash;MH, ca. 6000 years ago), present, and future (for 2050 and 2090, representing averages from 2040\u0026ndash;2060 and 2080\u0026ndash;2100, respectively). Averages of climate data for the LGM were obtained from three available GCMs (CCSM4, MIROC-ESM, and MPI-ESM-P), while averages for the MH were obtained from nine GCMs (BCC-CSM1-1, CCSM4, CNRM-CM5, HadGEM2-CC, HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM, MPI-ESM-P, and MRI-CGCM3) in WorldClim v1.4. Future climate data were averaged from three GCMs (FIO-ESM-2-0, MPI-ESM1-2-HR, and MRI-ESM2-0), which performed better than other models across different regions [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe two future periods were chosen to illustrate both short-term and long-term trends in habitat changes. For each period, we incorporated two climate scenarios from CMIP6 GCM: SSP245 (a moderate-emission scenario) and SSP585 (a high-emission and pessimistic scenario) [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. To minimize the impact of multicollinearity among environmental variables on model accuracy and to prevent overfitting, we conducted a Pearson correlation analysis using ENMTools software. Subsequently, we excluded highly correlated variables (|r| \u0026lt; 0.8), and selected ten environmental variables for simulation: annual mean temperature (bio1), mean diurnal range (bio2), temperature annual range (bio7), annual precipitation (bio12), precipitation of driest month (bio14), precipitation seasonality (bio15), precipitation of coldest quarter (bio19), aspect, slope and elevation (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Model selection\u003c/h2\u003e \u003cp\u003eWe employed both ENMeval and biomod2 packages for model analysis [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. All environmental variable data were converted to ASCII format using ArcGIS 10.8.1. The ENMeval package was utilized to optimize MaxEnt model parameters [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], enabling us to test various feature combination (FC) and regularization multiplier (RM) to identify the most suitable settings. The MaxEnt software offers five types of FC: L (linear), Q (quadratic), H (hinge), P (product), and T (threshold). To refine the model, we tested RM values from 0.5 to 6 in increments of 0.5, resulting in a total of 12 RM values. Additionally, we evaluated six different FC combinations: L, LQ, H, LQH, LQHP, and LQHPT. In total, we assessed 72 parameter combinations using the corrected Akaike Information Criterion (AICc), selecting the configuration with the lowest AICc value to ensure an optimal balance between model fit and complexity. A model is considered optimal when delta.AICc\u0026thinsp;=\u0026thinsp;0. The best-performing configuration is FC\u0026thinsp;=\u0026thinsp;LQHP and RM\u0026thinsp;=\u0026thinsp;3, with delta.AICc\u0026thinsp;=\u0026thinsp;0 (Table S2).\u003c/p\u003e \u003cp\u003eWe then used the biomod2 package to construct models with 11 different algorithms for simulations: artificial neural network (ANN), classification tree analysis (CTA), flexible discriminant analysis (FDA), generalized additive model (GAM), generalized boosting model (GBM), generalized linear model (GLM), multivariate adaptive regression splines (MARS), maximum entropy model (MaxEnt), random forest (RF), surface range envelope (SRE), and extreme gradient boosting (XGBoost). The optimized MaxEnt model parameters from the previous step were implemented in biomod2, while all other models were executed with their default settings. To minimize model uncertainty, the dataset was randomly partitioned into a training set (75%) and a testing set (25%), with each model repeated 20 times. Following the recommendations of the biomod2 team, we generated 200 pseudo-absence (PA) points, which is three times the number of occurrence points [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFinally, the selection of the best-performing model was based on the receiver operating characteristic (ROC) and true skill statistics (TSS) [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. We analyzed the area under the curve (AUC) and TSS values to compare the performance of 11 models in simulating the potential suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e. Model performance was evaluated with the following criteria: (1) poor (AUC\u0026thinsp;\u0026lt;\u0026thinsp;0.6), (2) fair (0.6\u0026thinsp;\u0026le;\u0026thinsp;AUC\u0026thinsp;\u0026lt;\u0026thinsp;0.8), (3) good (0.8\u0026thinsp;\u0026le;\u0026thinsp;AUC\u0026thinsp;\u0026lt;\u0026thinsp;0.9), and (4) excellent (0.9\u0026thinsp;\u0026le;\u0026thinsp;AUC\u0026thinsp;\u0026lt;\u0026thinsp;1) [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e]. Additionally, a TSS value greater than 0.8 is generally considered indicative of excellent model performance [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Classification of habitat suitability and centroid analysis of suitable areas\u003c/h2\u003e \u003cp\u003eWe applied the natural breaks method in ArcGIS 10.8.1 to classify habitat suitability into four categories: 0\u0026ndash;0.2 (suitable), 0.2\u0026ndash;0.4 (lowly suitable), 0.4\u0026ndash;0.6 (moderately suitable), and 0.6\u0026ndash;1 (highly suitable) [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. These categories were subsequently reclassified as follows: 0\u0026ndash;0.2\u0026thinsp;=\u0026thinsp;0, 0.2\u0026ndash;0.4\u0026thinsp;=\u0026thinsp;1, 0.4\u0026ndash;0.6\u0026thinsp;=\u0026thinsp;2, and 0.6\u0026ndash;1\u0026thinsp;=\u0026thinsp;4. Raster calculations were performed by subtracting suitability maps of earlier periods from those of later periods, where values\u0026thinsp;\u0026lt;\u0026thinsp;0, = 0, and \u0026gt;\u0026thinsp;0 indicate areas of contraction, unchange, and expansion, respectively. This analysis provides insights into the temporal dynamics of potentially suitable distribution areas for \u003cem\u003eL. meyenii\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eUsing the centroid calculation function in ArcGIS 10.8.1, we also calculated the areas of habitat suitability classified as moderate or higher and identified the centroids of potentially suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e across three time periods: the past, the present and the future in both South America and East Asia. This analysis enables us to track shifts in the centroids of suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e over time in response to climate change. By examining these centroid movements, we can understand how favorable growing conditions for \u003cem\u003eL. meyenii\u003c/em\u003e have geographically relocated as a result of changing climatic factors.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1 The random forest (RF) has the optimal performance\u003c/h2\u003e \u003cp\u003eBased on the AUC and TSS values, and the actual simulation outcomes, the random forest (RF) model shows the best performance, with an AUC of 0.9888 and a TSS of 0.8726 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). These metrics demonstrate its superior accuracy in predicting the distribution of suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Elevation and temperature annual range (bio7) are the most critical factors\u003c/h2\u003e \u003cp\u003eThe present-day distribution of \u003cem\u003eL. meyenii\u003c/em\u003e is shaped by both climatic and topographic factors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The most significant environmental variables influencing this distribution are elevation (47.6%), temperature annual range (bio7) (24.1%), mean diurnal range (bio2) (10.4%), slope (8.6%), annual mean temperature (bio1) (4.4%), precipitation seasonality (bio15) (2.2%), precipitation of coldest quarter (bio19) (1%), annual precipitation (bio12) (0.8%), precipitation of driest month (bio14) (0.7%), and aspect (0.2%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Consequently, elevation and the temperature annual range (bio7) emerge as the most critical factors that influence the current distribution of \u003cem\u003eL. meyenii\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eThe probability of occurrence for \u003cem\u003eL. meyenii\u003c/em\u003e shows a positive correlation with elevation, bio2, and slope. Specifically, \u003cem\u003eL. meyenii\u003c/em\u003e thrives at elevations above 2000 m, where the mean diurnal range exceeds 13 ℃, and slopes are greater than 1\u0026deg;. Conversely, the probability of occurrence is negatively correlated with temperature annual range (bio7). This indicates that \u003cem\u003eL. meyenii\u003c/em\u003e grows better when temperature annual range is less than 20 ℃.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Potential distribution under past, current, and future climate conditions\u003c/h2\u003e \u003cp\u003eThe simulation results suggest that highly suitable areas are predominantly concentrated in the Andes Mountains of western South America, consistent with current sampling locations, while moderately suitable areas are mainly found in the Tibetan Plateau, and lowly suitable areas are more widely dispersed (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Specifically, suitable habitats in the Tibetan Plateau have undergone significant changes through time: 1) During the LGM and MH, the habitat was predominantly of moderate suitability; 2) At present and in 2050 (under the SSP245 scenario), moderately and lowly suitable areas are interspersed; and 3) By 2050 (SSP585) and 2090 (for both SSP245 and SSP585), a larger areaclassified as highly suitable appears along the southern edge of the Tibetan Plateau. This reaches its maximum extent in 2090 under the SSP585 scenario.\u003c/p\u003e \u003cp\u003eDuring the LGM, the total area of potential suitable habitat was 1190.5\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, the largest among all time periods, although it had the smallest highly suitable area, covering only 68.3\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e. From the past to present, the total suitable habitat has decreased by ca. 12%. From the present to 2090 under the SSP245 scenario, the total suitable habitat area will decrease from 1047.5\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e to 976.1\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, with a reduction of 6.81%. However, the highly suitable area is projected to decline from 70.0\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e to 74.3\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, with an increase of 6.1%, while the moderately suitable area will rise from 170.0\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e to 171.2\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, with an increase of 0.72%. In comparison, from the present to 2090 under the SSP585 scenario, the total suitable habitat area will decrease significantly to 957.4\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, with a decline of 8.6%. In contrast, both the highly suitable and moderately suitable areas are projected to increase substantially, which reach 78.3\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e with an increase of 11.75% and 174.7\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e with an increase of 2.01%, respectively.\u003c/p\u003e \u003cp\u003e \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\u003eTemporal changes in the area (10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e) of suitable habitats for \u003cem\u003eLepidium meyenii\u003c/em\u003e. Abbreviations: LGM\u0026thinsp;=\u0026thinsp;Last Glacial Maximum, MH\u0026thinsp;=\u0026thinsp;Mid-Holocene.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeriod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLowly suitable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eModerately suitable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHighly suitable\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eTotal\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLGM\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e895.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e226.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e68.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1190.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e768.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e231.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1078.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresent\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e807.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e170.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e70.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e1047.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP245 (2050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e742.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e168.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e72.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e983.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP585 (2050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e726.1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e163.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e75.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e965.3\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP245 (2090)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e730.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e171.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e74.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e976.1\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP585 (2090)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e704.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e174.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e78.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e957.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Changes in potential habitat suitability since the LGM\u003c/h2\u003e \u003cp\u003eGenerally, both historical data and future projections indicate that the area of habitat contraction for \u003cem\u003eL. meyenii\u003c/em\u003e globally exceeds the area of habitat expansion. Additionally, the extent of both habitat expansion and contraction was more pronounced in the past (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). Between the LGM and the MH, as well as from the MH to the present, an opposing trend emerged in the areas with expansion and contraction across South America, Africa, and East Asia. Specifically, from the LGM to the MH, the global potential suitable habitat contracted by 346\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e mainly in southern and eastern Africa, and expanded by 260.6\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e primarily along the edges of the Tibetan Plateau, Southeast Asia, and northern South America. From the MH to the present, the global potential suitable habitat contracted by 312.5\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e mainly in the Tibetan Plateau and northern South America, with an expansion of 204\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e distributed without significant concentration.\u003c/p\u003e \u003cp\u003eIn the future, habitat contraction is expected primarily along the southern edge of the Tibetan Plateau, with additional scattered regions of contraction in southern Africa and northwestern China. Under the SSP245 scenario, changes in suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e are moderate. By 2050, the area of contraction is projected to be 126.7\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, while the area undergoing expansion is estimated at 71.8\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e; in 2090, these areas are expected to increase to 149.2\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e and 93.5\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e, respectively. In contrast, under the SSP585 scenario, changes in suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e are expected to be more pronounced. By 2050, the areas with contraction and expansion will be 152.6\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e and 80.7\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e; By 2090, these areas are projected to expand further to 210.9\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e for contraction and 147.7\u0026times;10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e for expansion.\u003c/p\u003e \u003cp\u003e \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\u003eTemporal changes in the distribution area (10\u003csup\u003e4\u003c/sup\u003e km\u003csup\u003e2\u003c/sup\u003e) for \u003cem\u003eLepidium meyenii\u003c/em\u003e. LGM\u0026thinsp;=\u0026thinsp;Last Glacial Maximum, MH\u0026thinsp;=\u0026thinsp;Mid-Holocene.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\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 \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePeriod\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnchanged\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eContracted\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eExpanded\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLGM\u0026ndash;MH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12887.4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e346.0\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e260.6\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMH\u0026ndash;Present\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e12977.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e312.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e204.0\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresent\u0026ndash;SSP245 (2050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13295.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e126.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e71.8\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP245 (2050)\u0026ndash;SSP245 (2090)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13251.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e149.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e93.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePresent\u0026ndash;SSP585 (2050)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13260.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e152.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e80.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSP585 (2050)\u0026ndash;SSP585 (2090)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e13135.3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e210.9\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e147.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Temporal centroid migration of potential suitable areas for \u003cem\u003eL. meyenii\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe migration trend is more pronounced in the past compare to the future. In South America, the centroid of suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e generally shifts northward over time (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). From the LGM to the MH, the centroid moves north by 135 km, from coordinates 9.71\u0026deg; S, 67.44\u0026deg; W to 8.49\u0026deg; S, 67.41\u0026deg; W. By the present day, it has further shifted 50.8 km to 8.03\u0026deg; S, 67.35\u0026deg; W. Under the SSP245 scenario, the centroid migrates northwest by 28.3 km to 7.87\u0026deg; S, 67.55\u0026deg; W by 2050, and then further to 7.35\u0026deg; S, 67.47\u0026deg; W, with a total movement of 58.1 km by 2090. In the SSP585 scenario, the centroid is expected to move to 7.73\u0026deg; S, 67.48\u0026deg; W by 2050, with a shift of 35.8 km, and then slightly to 7.66\u0026deg; S, 67.44\u0026deg; W by 2090, with a distance of 9.6 km.\u003c/p\u003e \u003cp\u003eIn East Asia, the centroid migration generally follows an initial southeastward to westward trend (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eb). From the LGM to the MH, the centroid shifts southeast by 252.8 km from 34.03\u0026deg; N, 93.93\u0026deg; E to 31.96\u0026deg; N, 95.12\u0026deg; E. By the present day, it has moved west to 32.19\u0026deg; N, 92.68\u0026deg; E with 231.9 km. Under the SSP245 scenario, the centroid shifts with 61.3 km further west to 32.11\u0026deg; N, 92.03\u0026deg; E by 2050, and then to 32.33\u0026deg; N, 92.34\u0026deg; E by 2090, with a further movement of 38.1 km. In the SSP585 scenario, the centroid moves to 32\u0026deg; N, 92.25\u0026deg; E with 45 km, then to 32.45\u0026deg; N, 92.57\u0026deg; E with 56 km.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e4.1 The evaluation of model performance\u003c/h2\u003e \u003cp\u003eWe optimized the MaxEnt model with the ENMeval package and compared it with other models available in the biomod2 package. The results indicate that the random forest (RF) model outperforms others in simulating the potential suitable habitats of \u003cem\u003eL. meyenii\u003c/em\u003e through time. The optimized MaxEnt model shows a significant improvement over the default parameters, with an increase of 0.08 in AUC and 0.05 in TSS, though its accuracy remained lower than that of the RF model.\u003c/p\u003e \u003cp\u003eAs an ensemble learning method based on the CART model, RF is widely used in ecology and biogeography due to its stability, insensitivity to parameter tuning, and ability to maintain high accuracy even with small sample sizes [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. While the RF model showed superior performance in this study, the best model may differ among various species. Many studies have opted for ensemble models in biomod2 to overcome the limitations of single models [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. However, ensemble models do not always outperform an individual model, as an optimized single model can sometimes provide better predictive performance [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. Therefore, we emphasize the importance of an integrative evaluation of different models with quantitative methods to identify the optimal model and ensure the accuracy and robustness of the simulation results.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Key environmental variables and potential distribution in the present day\u003c/h2\u003e \u003cp\u003eThe most critical factors the influence the potential suitable habitat distribution of \u003cem\u003eL. meyenii\u003c/em\u003e are elevation, temperature annual range (bio7), mean diurnal range (bio2), and slope (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Elevation and slope are topographic variables, while bio7 and bio2 relate to temperature. Precipitation has a minor role in determining species distribution. \u003cem\u003eL. meyenii\u003c/em\u003e is primarily found in areas with elevations exceeding 2,000 m, a mean diurnal temperature range over 15\u0026deg;C, and an annual temperature range below 20\u0026deg;C (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). This suggests that the species is better adapted to high-altitude environments characterized by significant diurnal temperature variations and relatively low annual temperature fluctuations.\u003c/p\u003e \u003cp\u003eThe Andes Mountains in western South America represent the largest area of natural distribution for plants, while Yunnan in China is known for the largest area of cultivated introduction [\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. These regions share similar climatic characteristics influenced by monsoonal climates, which promote the growth of \u003cem\u003eL. meyenii\u003c/em\u003e [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e, \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e]. Genomic sequencing and analysis of \u003cem\u003eL. meyenii\u003c/em\u003e reveal its molecular mechanisms that enable its adaptation to extreme high-altitude environments characterized by low temperatures and intense ultraviolet radiation [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. This study further confirms earlier results. A low annual temperature range provides a relatively stable environment for growth, while a high diurnal temperature range enables plants to optimize daytime photosynthesis and reduce nighttime respiration, and this thereby promotes organic matter accumulation [\u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e62\u003c/span\u003e]. However, species distribution is influenced not only by climate and topography but also by factors such as sunlight, soil, and water availability [\u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e63\u003c/span\u003e]. Future studies on \u003cem\u003eL. meyenii\u003c/em\u003e distribution should incorporate a wider array of environmental variables.\u003c/p\u003e \u003cp\u003eThe primary potential suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e at present are located in western South America and the southern edge of the Tibetan Plateau. Both regions feature alpine and plateau climates, are situated in subtropical zones, and have elevations exceeding 2,000 m, along with an annual temperature range below 20\u0026deg;C and a mean diurnal temperature range above 15\u0026deg;C, and this makes them highly conductive to \u003cem\u003eL. meyenii\u003c/em\u003e growth. This climate type is characterized by low annual temperature variation, significant diurnal temperature fluctuations, strong ultraviolet radiation, low atmospheric pressure, and high wind speeds. The current potential suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e align well with its natural distribution in western South America and its cultivated areas in Yunnan, China (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Changes in potential suitable habitat from past to present\u003c/h2\u003e \u003cp\u003eDuring the LGM\u0026ndash;MH, the contraction of potential suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e predominantly occurred in central South America and eastern and southern Africa (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea), while expansion was primarily observed along the edges of the Tibetan Plateau, Southeast Asia, and northern South America. This pattern may be associated with global warming and regional monsoon dynamics during the LGM\u0026ndash;MH, as both factors contributed to a decrease in the annual temperature range (bio7). The intensification of the South Asian and East Asian monsoons during this period created a more favorable climatic condition for the survival of \u003cem\u003eL. meyenii\u003c/em\u003e along the edges of the Tibetan Plateau and in Southeast Asia, and thereby this facilitated the expansion of its suitable habitat range [\u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e64\u003c/span\u003e, \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e65\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn the LGM, cooling in the Northern Hemisphere caused a southward shift of the Intertropical Convergence Zone (ITCZ). In contrast, during the MH, the ITCZ shifted northward, and this led to intensified monsoons in northern South America while weakening them in central and southern regions [\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e66\u003c/span\u003e, \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e67\u003c/span\u003e]. Consequently, as northern South America experienced significant habitat expansion, central regions saw a greater contraction (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). By the MH, the centroid of the potential suitable habitat shifted northward (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea). This suggests that, \u003cem\u003eL. meyenii\u003c/em\u003e was inclined to expand toward lower latitudes in South America. During the MH, Africa experienced significantly warmer climatic conditions, which were unfavorable for \u003cem\u003eL. meyenii\u003c/em\u003e, a cold-adapted alpine herb. This resulted in substantial contractions of suitable habitats in eastern and southern Africa. Similar distribution shifts have been observed in \u003cem\u003eXerophyta\u003c/em\u003e species across Africa [\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFrom the MH to the present, the contraction of suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e has primarily occurred in the Tibetan Plateau and northern South America, with only limited expansion observed (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). This contraction may be closely related to elevation-dependent warming (EDW)\u0026mdash;a phenomenon where high-altitude regions experience more rapid temperature increases compared to low-altitude areas due to greenhouse gas emissions [\u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e68\u003c/span\u003e]. Both the Tibetan Plateau and the Andes Mountains are affected by EDW, which accelerates climate change in alpine ecosystems and modifies vegetation distribution [\u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e69\u003c/span\u003e]. This process has led to habitat contraction or shifts, with negative consequences for biodiversity [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWhen comparing the LGM\u0026ndash;MH and MH\u0026ndash;Present periods, the primary areas with contraction and expansion for \u003cem\u003eL. meyenii\u003c/em\u003e show opposite patterns, particularly in the Tibetan Plateau and northern-central South America. Similarly, many high-altitude, cold-adapted plant species have undergone comparable distribution shifts during these two periods [\u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e70\u003c/span\u003e]. We conclude that natural environmental factors primarily drove changes in plant distribution during the LGM\u0026ndash;MH, while human activities have increasingly influenced distribution during the MH\u0026ndash;present period, particularly under the EDW effect in high-altitude regions. Therefore, future studies should integrate human activities as an environmental variable in distribution models to better understand their impact on shifts in species habitat.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e4.4 Changes in potential suitable habitat from the present to the future\u003c/h2\u003e \u003cp\u003eIn the future, the most significant expansion of suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e is projected to occur along the southern margin of the Tibetan Plateau, while the most substantial contraction is observed in the Tarim Basin of northwestern China and southern Africa. Although the suitable habitat in South America shows minimal overall change, its centroid continues to shift northward. Specifically, in the SSP245 scenario, global average temperatures are anticipated to rise by ca. 2.3\u0026ndash;4.6\u0026deg;C above pre-industrial levels, while in the SSP585 scenario, the increase could be 6.6\u0026ndash;14.1\u0026deg;C [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The Tarim Basin is thought to become even hotter under these future climate conditions [\u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e71\u003c/span\u003e]. Given its significant diurnal and annual temperature variations and an average elevation below 2000 m, this region is likely to become unsuitable for \u003cem\u003eL. meyenii\u003c/em\u003e growth.\u003c/p\u003e \u003cp\u003eAlthough southern Africa includes the South African Plateau and the Drakensberg Mountains, its average elevation remains below 2000 m, and it is characterized by a tropical savanna climate, and this makes it a unsuitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e. In contrast, the Tibetan Plateau, with an average elevation exceeding 4000 m, is expected to experience overall warming in the future. This warming trend is particularly pronounced in the northern region, with significant increases in winter temperatures anticipated, and this leads to a further reduction in annual temperature variation [\u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e72\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eYunnan Province in China has a subtropical monsoon climate, marked by large diurnal temperature variations, minimal annual temperature fluctuations, and mild weather throughout the year [\u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e73\u003c/span\u003e], and this makes it highly suitable for \u003cem\u003eL. meyenii\u003c/em\u003e growth. However, under the SSP585 scenario in 2090, a contraction of suitable habitat in northwestern Yunnan was observed. likely attributed to future climate warming and decreased precipitation in the area [\u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e74\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eGiven the topography and projected future climatic conditions of both the southern margin of the Tibetan Plateau and Yunnan Province, we propose that the southern Tibetan Plateau may become a more favorable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e in the future. Additionally, the distribution of certain plant species and the area of alpine humid forests in the southern margin of the Tibetan Plateau were predicted to increase under future climate conditions [\u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e75\u003c/span\u003e]. We therefore suggest that this region could become one of the most suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e and recommend establishing conservation areas to support its future sustainability.\u003c/p\u003e \u003cp\u003eAs the climate warms, shifts in plant niches may alter interspecific competition. In planning these conservation areas, it is essential to consider the interactions between herbaceous plants like \u003cem\u003eL. meyenii\u003c/em\u003e and woody plants such as those found in alpine humid forests. Understanding their competitive dynamics will be crucial for ensuring harmonious coexistence among plant species, ultimately promoting the stability and development of the ecosystem.\u003c/p\u003e \u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eUsing the species distribution data of \u003cem\u003eL. meyenii\u003c/em\u003e alongside multiple environmental variables, we evaluated 11 species distribution models and ultimately selected the Random Forest model as the optimal approach. This model allowed us to pinpoint the key environmental factors that influence the distribution of \u003cem\u003eL. meyenii\u003c/em\u003e and to simulate its potential suitable habitats through time. We find that the distribution of \u003cem\u003eL. meyenii\u003c/em\u003e is primarily influenced by elevation and temperature, while precipitation has a relatively minor impact. Specifically, \u003cem\u003eL. meyenii\u003c/em\u003e favors high elevations, a low temperature annual range (bio7), and a high mean diurnal range (bio2).\u003c/p\u003e \u003cp\u003eDriven by climate change, the total area of suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e has decreased from the past to the future, while the area classified as highly suitable has expanded. Historically, highly suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e were mainly concentrated in western South America, with moderate suitability observed in the Tibetan Plateau. During the LGM\u0026ndash;the MH, climate-driven changes led to the expansion of suitable habitats along the margins of the Tibetan Plateau and in Southeast Asia, while significant contractions occurred in southern and eastern Africa. However, from the MH to the present, this distribution pattern has reversed, with a contraction in suitable habitats on the Tibetan Plateau. In the future, western South America is projected to continue serving as a stable and highly suitable region for \u003cem\u003eL. meyenii\u003c/em\u003e. Meanwhile, as time progresses and greenhouse gas emissions intensify, highly suitable areas along the southern margin of the Tibetan Plateau are expected to expand further. This suggests that the region may emerge as an important potential area for the cultivation and conservation of \u003cem\u003eL. meyenii\u003c/em\u003e, alongside the Andes Mountains. Centroid analysis indicates that in South America, the centroid of suitable habitats for \u003cem\u003eL. meyenii\u003c/em\u003e will continue to shift northward. In East Asia, the centroid of suitable habitats shows a southeastward movement followed by a westward shift in the past, while future migration trends appear to be less pronounced.\u003c/p\u003e \u003cp\u003eThis study will provide an important basis for better understanding the adaptation mechanisms of alpine herbaceous plants like \u003cem\u003eL. meyenii\u003c/em\u003e to environmental changes and offer significant guidance for the future establishment of conservation areas for \u003cem\u003eL. meyenii\u003c/em\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eAuthor contributions\u003c/p\u003e\n\u003cp\u003eZeyu Qin: Conceptualization, Methodology, Investigation, Data curation, Writing - Original Draft, Writing - Review \u0026amp; Editing. Huasheng Huang: Conceptualization, Methodology, Investigation, Writing - Review \u0026amp; Editing, Supervision, Funding acquisition. Xuanqi Liu: Validation, Visualization, Writing - Review \u0026amp; Editing. Xia Meng: Validation, Visualization, Writing - Review \u0026amp; Editing. Minqiao Li: Visualization, Writing - Review \u0026amp; Editing.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eThis study was supported by the Starting Grant for Introduced Talents of Sun Yat-sen University, the Fundamental Research Funds for the Central Universities, Sun Yat-sen University (No. 24qnpy021), and the General Project of Basic and Applied Basic Research of Guangzhou Bureau of Science and Technology (No. 2025A04J4384).\u003c/p\u003e\n\u003cp\u003eData availability\u003c/p\u003e\n\u003cp\u003eData is provided within the manuscript or supplementary information files.\u003c/p\u003e\n\u003cp\u003eClinical trial number\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eConsent for publication\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eAcknowledgements\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003eCompeting Interest\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eIntergovernmental Panel on Climate Change (IPCC). (2012). Managing the risks of extreme events and disasters to advance climate change adaptation (SREX). In C. B. Field editors, Special Report of the Intergovernmental Panel on Climate Change (p. 582). Cambridge University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/CBO9781139177245\u003c/span\u003e\u003cspan address=\"10.1017/CBO9781139177245\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFilonchyk M, Peterson MP, Zhang L, Hurynovich V, He Y. Greenhouse gases emissions and global climate change: Examining the influence of CO\u003csub\u003e2\u003c/sub\u003e, CH\u003csub\u003e4\u003c/sub\u003e, and N\u003csub\u003e2\u003c/sub\u003eO. Sci Total Environ. 2024;935:173359. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2024.173359\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2024.173359\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIntergovernmental Panel on Climate Change (IPCC). Climate Change 2021: The Physical Science Basis: Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. 1st ed. Cambridge University Press; 2021. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/9781009157896\u003c/span\u003e\u003cspan address=\"10.1017/9781009157896\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLu M. The research progress on phytogeography: Global plant distribution patterns and environmental driving mechanisms. Geographical Res Bull. 2025;4:3\u0026ndash;25. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.50908/grb.4.0_3\u003c/span\u003e\u003cspan address=\"10.50908/grb.4.0_3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang P, Luo W, Zhang Q, Han S, Jin Z, Liu J, Li Z, Tao J. Assessing the impact of climate change on three Populus species in China: Distribution patterns and implications. Global Ecol Conserv. 2024;50:e02853. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gecco.2024.e02853\u003c/span\u003e\u003cspan address=\"10.1016/j.gecco.2024.e02853\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhao G, Cui X, Sun J, Li T, Wang Q, Ye X, Fan B. Analysis of the distribution pattern of Chinese Ziziphus jujuba under climate change based on optimized biomod2 and MaxEnt models. Ecol Ind. 2021;132:108256. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolind.2021.108256\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolind.2021.108256\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKelly AE, Goulden ML. (2008). Rapid shifts in plant distribution with recent climate change. Proceedings of the National Academy of Sciences, 105(33), 11823\u0026ndash;11826. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1073/pnas.0802891105\u003c/span\u003e\u003cspan address=\"10.1073/pnas.0802891105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBellard C, Bertelsmeier C, Leadley P, Thuiller W, Courchamp F. Impacts of climate change on the future of biodiversity. Ecol Lett. 2012;15(4):365\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1461-0248.2011.01736.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1461-0248.2011.01736.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrimm NB, Chapin FS, Bierwagen B, Gonzalez P, Groffman PM, Luo Y, Melton F, Nadelhoffer K, Pairis A, Raymond PA, Schimel J, Williamson CE. The impacts of climate change on ecosystem structure and function. Front Ecol Environ. 2013;11(9):474\u0026ndash;82. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/120282\u003c/span\u003e\u003cspan address=\"10.1890/120282\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJezkova T, Wiens JJ. (2016). Rates of change in climatic niches in plant and animal populations are much slower than projected climate change. Proceedings of the Royal Society B: Biological Sciences, 283(1843), 20162104. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1098/rspb.2016.2104\u003c/span\u003e\u003cspan address=\"10.1098/rspb.2016.2104\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKearney M, Porter W. Mechanistic niche modelling: Combining physiological and spatial data to predict species\u0026rsquo; ranges. Ecol Lett. 2009;12(4):334\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1461-0248.2008.01277.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1461-0248.2008.01277.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElith J, Phillips SJ, Hastie T, Dud\u0026iacute;k M, Chee YE, Yates CJ. A statistical explanation of MaxEnt for ecologists. Divers Distrib. 2011;17(1):43\u0026ndash;57. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1472-4642.2010.00725.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1472-4642.2010.00725.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCutler DR, Edwards TC, Beard KH, Cutler A, Hess KT, Gibson J, Lawler JJ. Random forests for classification in ecology. Ecology. 2007;88(11):2783\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/07-0539.1\u003c/span\u003e\u003cspan address=\"10.1890/07-0539.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYee TW, Mitchell ND. Generalized additive models in plant ecology. J Veg Sci. 1991;2(5):587\u0026ndash;602. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/3236170\u003c/span\u003e\u003cspan address=\"10.2307/3236170\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNelder JA, Wedderburn RWM. Generalized Linear Models. J Royal Stat Soc Ser (General). 1972;135(3):370. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2344614\u003c/span\u003e\u003cspan address=\"10.2307/2344614\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuang D, An Q, Huang S, Tan G, Quan H, Chen Y, Zhou J, Liao H. Biomod2 modeling for predicting the potential ecological distribution of three Fritillaria species under climate change. Sci Rep. 2023;13(1):18801. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-023-45887-6\u003c/span\u003e\u003cspan address=\"10.1038/s41598-023-45887-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang R, Wu N, Shi Z, Li C, Jiang N, Fu C, Wang M. Biomod2 for evaluating the changes in the spatiotemporal distribution of Locusta migratoria tibetensis Chen in the Qinghai-Tibet Plateau under climate change. Global Ecol Conserv. 2025;58:e03508. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.gecco.2025.e03508\u003c/span\u003e\u003cspan address=\"10.1016/j.gecco.2025.e03508\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThuiller W, Georges D, Gueguen M, Engler R, Breiner F, Lafourcade B, Patin R, Blancheteau H. (2012). biomod2: Ensemble Platform for Species Distribution Modeling (p. 4.2-6-2) [Dataset]. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.32614/CRAN.package.biomod2\u003c/span\u003e\u003cspan address=\"10.32614/CRAN.package.biomod2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuscarella R, Galante PJ, Soley-Guardia M, Boria RA, Kass JM, Uriarte M, Anderson RP. ENMeval: An R package for conducting spatially independent evaluations and estimating optimal model complexity for Maxent ecological niche models. Methods Ecol Evol. 2014;5(11):1198\u0026ndash;205. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/2041-210X.12261\u003c/span\u003e\u003cspan address=\"10.1111/2041-210X.12261\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAl-Shehbaz I. The genera of Lepidieae (Cruciferae; Brassicaceae) in the southeastern United States. J Arnold Arboretum. 1986;67:265\u0026ndash;311. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5962/bhl.part.27392\u003c/span\u003e\u003cspan address=\"10.5962/bhl.part.27392\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWarwick SI, Al-Shehbaz IA. Brassicaceae: Chromosome number index and database on CD-Rom. Plant Syst Evol. 2006;259(2\u0026ndash;4):237\u0026ndash;48. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00606-006-0421-1\u003c/span\u003e\u003cspan address=\"10.1007/s00606-006-0421-1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMummenhoff K, Br\u0026uuml;ggemann H, M\u0026uuml;ntzing A. Evidence for the origin of the Mediterranean endemic species \u003cem\u003eLepidium subulatum\u003c/em\u003e and \u003cem\u003eL. suffruticosum\u003c/em\u003e (Cruciferae) by interspecific hybridization. Plant Syst Evol. 1997;208(3\u0026ndash;4):275\u0026ndash;86. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF00937738\u003c/span\u003e\u003cspan address=\"10.1007/BF00937738\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eToledo J. Genetic variability of \u003cem\u003eLepidium meyenii\u003c/em\u003e and other Andean \u003cem\u003eLepidium\u003c/em\u003e species (Brassicaceae) assessed by molecular markers. Ann Botany. 1998;82(4):523\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1006/anbo.1998.0715\u003c/span\u003e\u003cspan address=\"10.1006/anbo.1998.0715\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Tian Y, Yan L, Zhang G, Wang X, Zeng Y, Zhang J, Ma X, Tan Y, Long N, Wang Y, Ma Y, He Y, Xue Y, Hao S, Yang S, Wang W, Zhang L, Dong Y, Sheng J. Genome of plant Maca (Lepidium meyenii) illuminates genomic basis for high-altitude adaptation in the Central Andes. Mol Plant. 2016;9(7):1066\u0026ndash;77. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.molp.2016.04.016\u003c/span\u003e\u003cspan address=\"10.1016/j.molp.2016.04.016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzales GF, Gasco M, Cordova A, Chung A, Rubio J, Villegas L. Effect of Lepidium meyenii (Maca) on spermatogenesis in male rats acutely exposed to high altitude (4340 m). J Endocrinol. 2004;180(1):87\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1677/joe.0.1800087\u003c/span\u003e\u003cspan address=\"10.1677/joe.0.1800087\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGonzales GF, Miranda S, Nieto J, Fern\u0026aacute;ndez G, Yucra S, Rubio J, Yi P, Gasco M. Red maca (Lepidium meyenii) reduced prostate size in rats. Reproductive Biology Endocrinol. 2005;3(1):5. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1477-7827-3-5\u003c/span\u003e\u003cspan address=\"10.1186/1477-7827-3-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRubio J, Caldas M, D\u0026aacute;vila S, Gasco M, Gonzales GF. Effect of three different cultivars of Lepidium meyenii (Maca) on learning and depression in ovariectomized mice. BMC Complement Altern Med. 2006;6(1):23. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/1472-6882-6-23\u003c/span\u003e\u003cspan address=\"10.1186/1472-6882-6-23\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStone M, Ibarra A, Roller M, Zangara A, Stevenson E. A pilot investigation into the effect of Maca supplementation on physical activity and sexual desire in sportsmen. J Ethnopharmacol. 2009;126(3):574\u0026ndash;6. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jep.2009.09.012\u003c/span\u003e\u003cspan address=\"10.1016/j.jep.2009.09.012\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRubio J, Qiong W, Liu X, Jiang Z, Dang H, Chen S-L, Gonzales GF. Aqueous Extract of Black Maca (Lepidium meyenii) on Memory Impairment Induced by Ovariectomy in Mice. Evidence-Based Complement Altern Med. 2011;2011(1):253958. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/ecam/nen063\u003c/span\u003e\u003cspan address=\"10.1093/ecam/nen063\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDa Silva Leit\u0026atilde;o Peres N, Parra Bortoluzzi C, Marques LM, Formigoni LL, Fuchs M, Droval RHB, A. A., Cardoso R, F. A. Medicinal effects of Peruvian maca (Lepidium meyenii): A review. Food Funct. 2020;11(1):83\u0026ndash;92. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1039/C9FO02732G\u003c/span\u003e\u003cspan address=\"10.1039/C9FO02732G\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin D, Qian J, Zhu H. Frontier development in the midst of ecological civilization: Unravelling the production of maca in Yunnan, China. Geoforum. 2019;106:144\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.geoforum.2019.08.005\u003c/span\u003e\u003cspan address=\"10.1016/j.geoforum.2019.08.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe J, Yang Z, Li Y, He W, He X, Bi H, Yang W, Wang Y, He J. Situation and development strategy of maca industry in Yunnan. Chin Agric Sci Bull. 2016;32(27):93\u0026ndash;8. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.11924/j.issn.1000-6850.casb15120105\u003c/span\u003e\u003cspan address=\"10.11924/j.issn.1000-6850.casb15120105\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen R, Wei J, Gao Y. A review of the study of active components and their pharmacology value in Lepidium meyenii (Maca). Phytother Res. 2021;35(12):6706\u0026ndash;19. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ptr.7257\u003c/span\u003e\u003cspan address=\"10.1002/ptr.7257\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui Z, Yang W-S, Cheng Z, Zhang Z, Li S, Zhao J, L\u0026oacute;pez-Vicente M, Wu G-L. Top-down degradation of alpine meadow in the Qinghai-Tibetan Plateau: Gravelization initiate hillside surface aridity and meadow community disappearance. CATENA. 2022;210:105933. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.catena.2021.105933\u003c/span\u003e\u003cspan address=\"10.1016/j.catena.2021.105933\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun Q, Li B, Jiang Y, Chen X, Zhou G. Declined trend in herbaceous plant green-up dates on the Qinghai\u0026ndash;Tibetan Plateau caused by spring warming slowdown. Sci Total Environ. 2021;772:145039. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2021.145039\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2021.145039\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMontgomery DR, Balco G, Willett SD. Climate, tectonics, and the morphology of the Andes. Geology. 2001;29(7):579. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1130/0091-7613(2001)029\u0026lt;0579:CTATMO\u0026gt;2.0.CO;2\u003c/span\u003e\u003cspan address=\"10.1130/0091-7613(2001)029%3C0579:CTATMO%3E2.0.CO;2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarreaud RD. The Andes climate and weather. Adv Geosci. 2009;22:3\u0026ndash;11. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/adgeo-22-3-2009\u003c/span\u003e\u003cspan address=\"10.5194/adgeo-22-3-2009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarreaud RD, Vuille M, Compagnucci R, Marengo J. Present-day South American climate. Palaeogeogr Palaeoclimatol Palaeoecol. 2009;281(3\u0026ndash;4):180\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.palaeo.2007.10.032\u003c/span\u003e\u003cspan address=\"10.1016/j.palaeo.2007.10.032\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eToledo O, Palazzi E, Cely Toro IM, Mortarini L. Comparison of elevation-dependent warming and its drivers in the tropical and subtropical Andes. Clim Dyn. 2022;58(11\u0026ndash;12):3057\u0026ndash;74. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-021-06081-4\u003c/span\u003e\u003cspan address=\"10.1007/s00382-021-06081-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGBIF.org, 2025. GBIF occurrence downloaded on 14th February 2025. Available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15468/dl.d36rsc\u003c/span\u003e\u003cspan address=\"10.15468/dl.d36rsc\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZizka A, Silvestro D, Andermann T, Azevedo J, Ritter CD, Edler D, Farooq H, Herdean A, Ariza M, Scharn R, Svantesson S, Wengstr\u0026ouml;m N, Zizka V, Antonelli A. CoordinateCleaner: Standardized cleaning of occurrence records from biological collection databases. Methods Ecol Evol. 2019;10(5):744\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/2041-210X.13152\u003c/span\u003e\u003cspan address=\"10.1111/2041-210X.13152\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDini A, Migliuolo G, Rastrelli L, Saturnino P, Schettino O. Chemical composition of Lepidium meyenii. Food Chem. 1994;49(4):347\u0026ndash;9. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0308-8146(94)90003-5\u003c/span\u003e\u003cspan address=\"10.1016/0308-8146(94)90003-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandoval M, Okuhama NN, Angeles FM, Melchor VV, Condezo LA, Lao J, Miller MJS. Antioxidant activity of the cruciferous vegetable Maca (Lepidium meyenii). Food Chem. 2002;79(2):207\u0026ndash;13. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0308-8146(02)00133-4\u003c/span\u003e\u003cspan address=\"10.1016/S0308-8146(02)00133-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang M-Z, Xu Z, Han Y, Guo W. Evaluation of CMIP6 models toward dynamical downscaling over 14 CORDEX domains. Clim Dyn. 2024;62(6):4475\u0026ndash;89. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-022-06355-5\u003c/span\u003e\u003cspan address=\"10.1007/s00382-022-06355-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eO\u0026rsquo;Neill BC, Tebaldi C, Van Vuuren DP, Eyring V, Friedlingstein P, Hurtt G, Knutti R, Kriegler E, Lamarque J-F, Lowe J, Meehl GA, Moss R, Riahi K, Sanderson BM. (2016). The Scenario Model Intercomparison Project (ScenarioMIP) for CMIP6. Geoscientific Model Development, 9(9), 3461\u0026ndash;3482. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5194/gmd-9-3461-2016\u003c/span\u003e\u003cspan address=\"10.5194/gmd-9-3461-2016\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbet-Massin M, Jiguet F, Albert CH, Thuiller W. (2012). Selecting pseudo‐absences for species distribution models: How, where and how many? Methods in Ecology and Evolution, 3(2), 327\u0026ndash;338. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.2041-210X.2011.00172.x\u003c/span\u003e\u003cspan address=\"10.1111/j.2041-210X.2011.00172.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29\u0026ndash;36. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1148/radiology.143.1.7063747\u003c/span\u003e\u003cspan address=\"10.1148/radiology.143.1.7063747\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllouche O, Tsoar A, Kadmon R. Assessing the accuracy of species distribution models: Prevalence, kappa and the true skill statistic (TSS). J Appl Ecol. 2006;43(6):1223\u0026ndash;32. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2664.2006.01214.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2664.2006.01214.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLobo JM, Jim\u0026eacute;nez-Valverde A, Real R. AUC: A misleading measure of the performance of predictive distribution models. Glob Ecol Biogeogr. 2008;17(2):145\u0026ndash;51. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1466-8238.2007.00358.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1466-8238.2007.00358.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLasram BR, Guilhaumon F, Albouy F, Somot C, Thuiller S, W., Mouillot D. The Mediterranean Sea as a \u0026lsquo;cul-de‐sac\u0026rsquo; for endemic fishes facing climate change. Glob Change Biol. 2010;16(12):3233\u0026ndash;45. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2486.2010.02224.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2486.2010.02224.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang B, Liu DL, O\u0026rsquo;Leary GJ, Asseng S, Macadam I, Lines-Kelly R, Yang X, Clark A, Crean J, Sides T, Xing H, Mi C, Yu Q. Australian wheat production expected to decrease by the late 21st century. Glob Change Biol. 2018;24(6):2403\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.14034\u003c/span\u003e\u003cspan address=\"10.1111/gcb.14034\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLandis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2529310\u003c/span\u003e\u003cspan address=\"10.2307/2529310\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFox EW, Hill RA, Leibowitz SG, Olsen AR, Thornbrugh DJ, Weber MH. Assessing the accuracy and stability of variable selection methods for random forest modeling in ecology. Environ Monit Assess. 2017;189(7):316. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10661-017-6025-0\u003c/span\u003e\u003cspan address=\"10.1007/s10661-017-6025-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLuan J, Zhang C, Xu B, Xue Y, Ren Y. The predictive performances of random forest models with limited sample size and different species traits. Fish Res. 2020;227:105534. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.fishres.2020.105534\u003c/span\u003e\u003cspan address=\"10.1016/j.fishres.2020.105534\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang X, Wu W, Liang Y. Analysis of the potential distribution of shoot blight of larch in China based on the optimized MaxEnt and Biomod2 ensemble models. Forests. 2024;15(8):1313. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/f15081313\u003c/span\u003e\u003cspan address=\"10.3390/f15081313\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang L, Jia H, Hua Q. 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. 2025;25(1):351. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s12870-025-06370-8\u003c/span\u003e\u003cspan address=\"10.1186/s12870-025-06370-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHao T, Elith J, Guillera-Arroita G, Lahoz‐Monfort JJ. A review of evidence about use and performance of species distribution modelling ensembles like BIOMOD. Divers Distrib. 2019;25(5):839\u0026ndash;52. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ddi.12892\u003c/span\u003e\u003cspan address=\"10.1111/ddi.12892\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Y, Li PY, Zhou XT, Zhou LY, Huang LQ, Yang G, Chen M. Research and application progress of \u003cem\u003eLepidium meyenii\u003c/em\u003e (maca) [in Chinese]. China J Chin Materia Med. 2018;43(23):4599\u0026ndash;607. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.19540/j.cnki.cjcmm.2018.0122\u003c/span\u003e\u003cspan address=\"10.19540/j.cnki.cjcmm.2018.0122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Carvalho LMV, Jones C, editors. (2016). The Monsoons and Climate Change: Observations and Modeling. Springer International Publishing. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/978-3-319-21650-8\u003c/span\u003e\u003cspan address=\"10.1007/978-3-319-21650-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEspinoza JC, Garreaud R, Poveda G, Arias PA, Molina-Carpio J, Masiokas M, Viale M, Scaff L. Hydroclimate of the Andes Part I: Main Climatic Features. Front Earth Sci. 2020;8:64. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/feart.2020.00064\u003c/span\u003e\u003cspan address=\"10.3389/feart.2020.00064\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTurner AG, Annamalai H. Climate change and the South Asian summer monsoon. Nat Clim Change. 2012;2(8):587\u0026ndash;95. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nclimate1495\u003c/span\u003e\u003cspan address=\"10.1038/nclimate1495\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCollatz GJ, Bounoua L, Los SO, Randall DA, Fung IY, Sellers PJ. A mechanism for the influence of vegetation on the response of the diurnal temperature range to changing climate. Geophys Res Lett. 2000;27(20):3381\u0026ndash;4. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/1999GL010947\u003c/span\u003e\u003cspan address=\"10.1029/1999GL010947\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoulangeat I, Gravel D, Thuiller W. Accounting for dispersal and biotic interactions to disentangle the drivers of species distributions and their abundances. Ecol Lett. 2012;15(6):584\u0026ndash;93. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1461-0248.2012.01772.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1461-0248.2012.01772.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuganuma Y, Yamazaki T, Kanamatsu T. South Asian monsoon variability during the past 800 kyr revealed by rock magnetic proxies. Q Sci Rev. 2009;28(9\u0026ndash;10):926\u0026ndash;38. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2008.12.014\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2008.12.014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun Y, Kutzbach J, An Z, Clemens S, Liu Z, Liu W, Liu X, Shi Z, Zheng W, Liang L, Yan Y, Li Y. Astronomical and glacial forcing of East Asian summer monsoon variability. Q Sci Rev. 2015;115:132\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2015.03.009\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2015.03.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerman AL, Silvestri GE, Tonello MS. Differences between Last Glacial Maximum and present-day temperature and precipitation in southern South America. Q Sci Rev. 2016;150:221\u0026ndash;33. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2016.08.025\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2016.08.025\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerman AL, Silvestri GE, Tonello MS. On the differences between Last Glacial Maximum and Mid-Holocene climates in southern South America simulated by PMIP3 models. Q Sci Rev. 2018;185:113\u0026ndash;21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.quascirev.2018.02.003\u003c/span\u003e\u003cspan address=\"10.1016/j.quascirev.2018.02.003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWanga VO, Ngarega BK, Oulo MA, Mkala EM, Ngumbau VM, Onjalalaina GE, Odago WO, Nanjala C, Ochieng CO, Gichua MK, Gituru RW, Hu G-W. Projected impacts of climate change on the habitat of Xerophyta species in Africa. Plant Divers. 2024;46(1):91\u0026ndash;100. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.pld.2023.05.001\u003c/span\u003e\u003cspan address=\"10.1016/j.pld.2023.05.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMountain Research Initiative EDW Working Group. Elevation-dependent warming in mountain regions of the world. Nat Clim Change. 2015;5(5):424\u0026ndash;30. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nclimate2563\u003c/span\u003e\u003cspan address=\"10.1038/nclimate2563\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiao S, Li S, Wang X, Chen L, Su T. \u003cem\u003eCedrus\u003c/em\u003e distribution change: Past, present, and future. Ecol Ind. 2022;142:109159. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolind.2022.109159\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolind.2022.109159\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePan XD, Zhang L, Huang CL. Future climate projection in northwest china with RegCM4.6. Earth Space Sci. 2020;7(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2019EA000819\u003c/span\u003e\u003cspan address=\"10.1029/2019EA000819\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. e2019EA000819.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou M, Yu Z, Gu H, Ju Q, Gao Y, Wen L, Huang T, Wang W. Evaluation and projections of surface air temperature over the Tibetan Plateau from CMIP6 and CMIP5: Warming trend and uncertainty. Clim Dyn. 2023;60(11\u0026ndash;12):3863\u0026ndash;83. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-022-06518-4\u003c/span\u003e\u003cspan address=\"10.1007/s00382-022-06518-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou J-Q, Huang W, Zhu Y, Li M, Zhou B-T. Climate comfort distribution, change and projection in Yunnan province. Clim Change Res. 2018;14(2):144\u0026ndash;54. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.12006/j.issn.1673-1719.2017.065\u003c/span\u003e\u003cspan address=\"10.12006/j.issn.1673-1719.2017.065\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRen J, Zhang W, Wan Y, Chen Y. Advances in the Research of Yunnan\u0026rsquo;s Arid Climate and Extreme Drought. Atmospheric Clim Sci. 2017;7(1):23\u0026ndash;35. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.4236/acs.2017.71003\u003c/span\u003e\u003cspan address=\"10.4236/acs.2017.71003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFan Z, Bai X. Scenarios of potential vegetation distribution in the different gradient zones of Qinghai-Tibet Plateau under future climate change. Sci Total Environ. 2021;796:148918. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2021.148918\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2021.148918\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Climate change, Brassicaceae, Lepidium meyenii, Species distribution, Ecological niche model, Biogeography","lastPublishedDoi":"10.21203/rs.3.rs-6472833/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6472833/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eGlobal warming has influenced phenological shifts and community degradation in alpine herbaceous plants. However, the current distribution and future shifts of these plants and their environmental driving factors under climate change are not well known. Here, we focus on \u003cem\u003eLepidium meyenii\u003c/em\u003e, a medicinal herb native to the high-altitude regions of the Andes Mountains in South America, which may face habitat contraction and potential population decline in the future due to changing environmental conditions. We integrate species distribution data with a random forest model to simulate the potential habitats of \u003cem\u003eL. meyenii\u003c/em\u003e across time and identify their key environmental drivers. We find that the most significant environmental variables that impact the distribution of \u003cem\u003eL. meyenii\u003c/em\u003e include elevation, temperature annual range (bio7), and mean diurnal range (bio2). Since the Last Glacial Maximum, the western Andes in South America has consistently offered suitable habitat for \u003cem\u003eL. meyenii\u003c/em\u003e. In contrast, habitat suitability in the Tibetan Plateau (TP) has varied over time, and in the futurer the TP may become a potential area for biodiversity conservation. This study will enhance our understanding of the distribution patterns of alpine herbaceous plants in response to climate change, and contributes to future biodiversity conservation efforts, the establishment of protected areas, and the sustainable management of medicinal plants as biological resources.\u003c/p\u003e","manuscriptTitle":"Global temporal distribution of Maca (Lepidium meyenii Walp.) under climate change","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-05 20:36:59","doi":"10.21203/rs.3.rs-6472833/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":"8fc4d829-34f8-4e50-b93a-9505d58faa8d","owner":[],"postedDate":"May 5th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-13T12:11:35+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-05 20:36:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6472833","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6472833","identity":"rs-6472833","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}