Next Article in Journal
Plant-Derived Bioactive Compounds in Inflammation-Related Cancers: Mechanisms and Therapeutic Potential
Next Article in Special Issue
Floral Traits, Pollination and Reproductive Differentiation in Gynodioecious Minuartia nifensis (Caryophyllaceae)
Previous Article in Journal
Conventional and Rock-Based Potassium Fertilization Improves Fennel (Foeniculum vulgare Mill.) Performance Under Extended Irrigation Intervals
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Plants
Volume 15
Issue 4
10.3390/plants15040574
plants-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Assessing Potential Habitat Suitability of the Endangered Endo-Holoparasitic Sapria himalayana and Its Multiple Hosts in China Under Global Warming

by
Weiyi Hang
1,†,
Yan Li
2,† and
Guangfu Zhang
1,*
1
Key Laboratory of Biodiversity and Biotechnology, School of Life Sciences, Nanjing Normal University, Nanjing 210023, China
2
Nanjing Institute of Environmental Science, Ministry of Ecology and Environment of the People’s Republic of China, Nanjing 210042, China
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Plants 2026, 15(4), 574; https://doi.org/10.3390/plants15040574
Submission received: 26 December 2025 / Revised: 8 February 2026 / Accepted: 9 February 2026 / Published: 11 February 2026
(This article belongs to the Special Issue Conservation, Ecology and Management of Rare and Endangered Plant Species)
Browse Figures
Review Reports Versions Notes

Abstract

Global warming severely threatens parasitic plants worldwide. However, little is known about how a parasite with multiple hosts responds to climate change in its distribution. Sapria himalayana is an endangered endo-holoparasite, obligately parasitizing Tetrastigma species. We employed MaxEnt to predict suitable habitats for S. himalayana and its five hosts, and determined key environmental factors. Then, we calculated niche overlaps for the five parasite-host pairs. Currently, it covers a suitable area of 1.35 × 104 km2, accounting for 0.14% of China’s total territory. Temperature-related variables were identified as the key factors shaping potential distribution for this parasite and three hosts (i.e., T. planicaule, T. obovatum, and T. cruciatum), while precipitation-related ones were identified for the other hosts (i.e., T. obtectum and T. serrulatum). Collectively, the five pairs presented low niche overlaps under current and future scenarios. While S. himalayana will increase by 37.78% in future suitable habitat, the two host categories show contrasting trends in potential habitat shifts. Divergent climatic sensitivities across host species, along with parasite–host suitability mismatches, could shape the survival and distribution of S. himalayana. Consequently, this research offers valuable insights for the conservation of S. himalayana in China, highlighting the necessity of safeguarding its distinct hosts under global warming.

1. Introduction

Global warming is altering temperature and precipitation patterns worldwide, thereby affecting species’ growth, reproduction, and distribution [1,2]. Parasitic plants are impacted by global warming more severely compared to non-parasitic plants [3]. The impact of climate change on parasitic plants operates through two primary pathways. Firstly, it can directly modify their habitat conditions, necessitating range shifts [4,5]. Secondly, because parasitic plants depend functionally on hosts via haustoria for water and nutrient acquisition [6], their distribution is further constrained by the responses of their host species to climate change. According to the host-quality hypothesis, the distribution of parasitic plants is limited by the quality of their hosts—specifically, the availability of resources such as water and nutrients that are typically limiting the hosts themselves [7]. Consequently, global warming can also indirectly impact the survival of parasitic plants by changing the strength of host–parasite interactions [8]. Endangered plants often have characteristics of small population sizes, narrow geographic distributions, and low genetic diversity [9]. For endangered parasitic plants, their distribution is further restricted by the environmental requirements of their hosts, making them more sensitive to climate change. Based on the definition by [10], parasitic plants can be classified into four functional categories: root hemiparasites, root holoparasites, stem parasites, and endophytic parasites. Endoparasitic plants exhibit a suite of distinct adaptations, including obligate parasitism with immediate haustorial development, a dominant endophytic phase, a vegetative body reduced to a mycelium-like structure, and transient exophytes only during flowering and fruiting [10,11]. Consequently, endangered species within this group are likely to be disproportionately vulnerable to global warming.
Currently, numerous studies have predicted the distribution of parasitic plants [12,13,14], yet only a few have simultaneously addressed how both parasitic plants and their hosts may respond to future climate change [15,16]. In fact, hosts play a significant role in shaping the distribution and survival of parasitic plants, particularly for those with high host specificity [3]. For parasitic plants that depend on obligate hosts, the number of hosts can range from one to several (n ≥ 2). However, current research has largely focused on single-host systems, with multi-host parasites receiving comparatively little attention [5,17].
Endophytic parasites represent a small subset of all parasitic plants, restricted to the families Apodanthaceae, Cytinaceae, Rafflesiaceae, and several genera of Santalales. Most of these are obligate parasites and can rely on multiple host species [10,18]. For example, all species of the genus Rafflesia (Rafflesiaceae) are obligate parasites on vines of the genus Tetrastigma and are known to utilize multiple species within this genus as hosts [19,20,21].
Sapria himalayana Griff. is a fleshy herb from the family Rafflesiaceae. As an obligate endoparasite, it lacks leaves, roots, and stems, and persists predominantly as thread-like filaments within its host tissues. The only emergent stage occurs during flowering, which presents ten bright red bracts marked with sulphur-yellow spots [22,23]. This species parasitizes several members of the genus Tetrastigma (Vitaceae) [24]. Although earlier records reported its occurrence on Vitis species [25], such accounts are likely attributable to historical misidentification of certain Tetrastigma taxa as Vitis in the same family [26]. Current research on S. himalayana primarily focuses on genetic [27], biochemical [28], and physiological [29] aspects, but little is known about its potential distribution.
S. himalayana has a narrow distribution in China. According to Flora of China (FOC), it only occurs in Yunnan Province and Xizang Autonomous Region (i.e., Tibet) [25], with no updated reports from other provinces in the country [18]. Based on the existing literature, five Tetrastigma species are recognized as hosts for S. himalayana: Tetrastigma planicaule (Hook.) Gagnep., T. obtectum (Wall.) Planch., T. serrulatum (Roxb.) Planch., T. obovatum Gagnep., and T. cruciatum Craib & Gagnep. [30,31,32,33,34]. Furthermore, according to Flora of China, all these climbers occur in Yunnan, the primary distribution area of S. himalayana [35]. Therefore, these five species are considered the major hosts of S. himalayana.
Nowadays, S. himalayana is threatened primarily by habitat destruction and illegal excavation within its known range. Such a threat is exacerbated by the species’ inherent reproductive challenges—including high bud mortality and strongly skewed sex ratio—coupled with its dependence on specific host plants [36,37,38]. These factors jointly render the species endangered in China. Therefore, this species has been listed as a Grade II National Key Protected Wild Plant in China and categorized as a “Vulnerable (VU)” species in the IUCN Red List [39]. However, it remains unclear what the potential geographic distribution of S. himalayana is, and how the endangered parasite and its five primary hosts respond in the context of climate change.
This study employed a species distribution modeling framework (MaxEnt) to project the potential geographic distributions of S. himalayana and its five obligate host species, utilizing known occurrence data and environmental variables. The work aimed to (1) identify the key environmental factors and shared drivers shaping their distributions, (2) predict shifts in their suitable habitats under current and future climate scenarios, and (3) quantify the niche overlap between each parasite–host pair across these scenarios. Ultimately, our findings are intended to inform conservation strategies for S. himalayana and other endangered multi-host endophytic parasites.

2. Results

2.1. Model Performance and Key Environmental Factors

AUC values of prediction models for Sapria himalayana and its five hosts (i.e., Tetrastigma planicaule, T. obtectum, T. obovatum, T. cruciatumc and T. serrulatum) under different climate scenarios were all greater than 0.90 (Table S1), indicating excellent model performance for all species. In addition, TSS values of the six species across different climate scenarios were all above 0.75 (Table S1). Specifically, S. himalayana and T. cruciatum had TSS values exceeding 0.9; T. planicaule and T. serrulatum had TSS values exceeding 0.8; while T. obtectum and T. obovatum had TSS values exceeding 0.75 (Table S1). Generally, a TSS value above 0.75 suggests a model with strong predictive performance [40]. Overall, these results indicated high accuracy and credibility of MaxEnt models for each species under both current and future climate scenarios.
The results showed that extrapolation risks were collectively much lower for both S. himalayana and its five hosts, although there were certain differences in this respect for these six species (Table 1). For S. himalayana, its future average proportion of pixels with multivariate environmental similarity surface (MESS) value < 0 was 1.6512%. For the five hosts, T. planicaule had the highest average proportion of pixels with MESS < 0 under future conditions (i.e., 2.5877%). For T. obtectum, T. obovatum, T. cruciatum, and T. serrulatum, their corresponding proportions were 0.7791%, 0.8555%, 1.1138%, and 1.1750%, respectively (Table 1). Moreover, for both S. himalayana and its five hosts, the proportion of pixels with MESS < 0 varied noticeably across different future climate scenarios (Table 1).
Among the six species, S. himalayana had the greatest extent of niche unfilling, reaching 90.38%, followed by T. planicaule, T. obtectum, and T. cruciatum with percentages of 65.00%, 63.76%, and 60.61%, respectively. In contrast, T. serrulatum only had the niche unfilling of 37.65%; T. obovatum had the lowest value of 18.18% (Table 2).
For each species, the top five environmental variables, whose cumulative contribution exceeded 90%, were identified as key predictors (Table 3). For S. himalayana, the key factors affecting its potential distribution were isothermality (bio3, 88.4%), temperature seasonality (bio4, 4.8%), annual mean temperature (bio1, 3.3%), precipitation of warmest quarter (bio18, 1.2%) and aspect (0.6%), together accounting for 98.3% of the total.
For T. planicaule, the key factors affecting its potential distribution were mean temperature of coldest quarter (bio11, 67.4%), annual precipitation (bio12, 16.2%), temperature seasonality (bio4, 5.7%), human influence (HI, 4.4%) and mean temperature of wettest quarter (bio8, 1.6%), totaling 95.3%.
For T. obtectum, the key factors affecting its potential distribution were annual precipitation (bio12, 40.1%), min temperature of coldest month (bio6, 19.9%), precipitation of driest month (bio14, 15.2%), elevation (12.7%) and human influence (HI, 7.8%), totaling 92.8%.
For T. obovatum, the key factors affecting its potential distribution were min temperature of coldest month (bio6, 51.4%), isothermality (bio3, 17.9%), human influence (HI, 11.2%), annual precipitation (bio12, 7.8%) and precipitation of driest quarter (bio17, 7.5%), totaling 95.8%.
For T. cruciatum, the key factors affecting its potential distribution were temperature seasonality (bio4, 72.9%), isothermality (bio3, 16.1%), annual precipitation (bio12, 3.2%), precipitation of wettest month (bio13, 2.2%) and human influence (HI, 2.2%), totaling 96.6%.
For T. serrulatum, the key factors affecting its potential distribution were precipitation of driest quarter (bio17, 33.8%), temperature seasonality (bio4, 33.7%), elevation (11.4%), annual precipitation (bio12, 9.3%) and human influence (HI, 6.1%), totaling 94.3%.
Three temperature-related variables (i.e., bio3, bio4, and bio1) collectively accounted for 96.5% of the explained contribution among the five key factors shaping the distribution of S. himalayana, underscoring their dominant role. Likewise, temperature-related variables constituted the primary explanatory factors for T. planicaule, T. obovatum, and T. cruciatum, with cumulative contributions of 74.7%, 69.3%, and 89.0%, respectively. In contrast, precipitation exerted a greater influence on distribution for T. obtectum and T. serrulatum, with their precipitation-related variables contributing 55.3% and 43.1% in total, respectively (Table 3).

2.2. Potential Suitable Distribution of Sapria himalayana and Its Five Hosts

Under the current climate scenario, suitable habitat (moderately and highly suitable areas) for S. himalayana was mainly located in southern Yunnan (Figure 1), with a total area of 1.35 × 104 km2, accounting for approximately 0.14% of China’s territory. Across the nine future climate scenarios, the suitable habitat for S. himalayana was projected to expand by an average of 37.78%, reaching a total area of 1.86 × 104 km2 (Table 4). Only under SSP5-8.5 2050s and SSP5-8.5 2090s did its suitable habitat show a slight decrease of 4.44% and 0.74%, respectively.
Under the current climate scenario, the suitable habitat for T. planicaule was mainly distributed in southeastern coastal regions of China (Guangxi, Guangdong, Hainan, southern Fujian, and western Taiwan), as well as southwestern China (western and southern Yunnan, southern Guizhou) (Figure 2). The total area was 36.68 × 104 km2, accounting for approximately 3.82% of China’s territory (Table 4). Additionally, its overlap suitable area with S. himalayana was 0.93 × 104 km2, accounting for 68.89% of the total suitable habitat for S. himalayana (Table S2).
Under future climate scenarios, the suitable habitat area for T. planicaule was predicted to expand by an average of 2.51%, reaching a total area of 37.60 × 104 km2 (Table 4). Only under SSP1-2.6 2090s and SSP2-4.5 2090s did the suitable habitat area show a slight decrease of 2.86% and 1.66%, respectively. Furthermore, its overlap with the suitable area of S. himalayana was projected to increase by an average of 34.41%, reaching a total area of 1.25 × 104 km2 (Table S2).
Under the current climate scenario, the suitable habitat for T. obtectum was mainly distributed in southwestern China (Chongqing, Guizhou, eastern and southern Sichuan, southeastern Tibet, Yunnan) and surrounding provinces (southern Shaanxi, southern Henan, western Guangxi, western Hubei, western Hunan), with several fragmented suitable habitats scattered in southeastern coastal regions (Fujian, Guangdong, Hainan, Zhejiang, western Taiwan) (Figure 3). The total area was 71.70 × 104 km2, accounting for approximately 7.47% of China’s territory (Table 4). Additionally, its overlap suitable area with S. himalayana was 0.89 × 104 km2, constituting 65.93% of the latter’s total suitable habitat (Table S2).
Under future climate scenarios, the suitable habitat area for T. obtectum was predicted to reduce by an average of 3.26%, dropping to an area of 69.36 × 104 km2 (Table 4). Only under SSP1-2.6 2070s did the suitable habitat area increase by 6.60%. Furthermore, its overlap with the suitable area of S. himalayana was projected to decrease by an average of 20.22%, dropping to an area of 0.71 × 104 km2 (Table S2).
Under the current climate scenario, the suitable habitat for T. obovatum was mainly distributed in southwestern China (Chongqing, Guizhou, eastern Sichuan, southeastern Tibet, Yunnan) and surrounding provinces (western Hubei), with several suitable habitats distributed in southeastern coastal regions (central Guangdong, northern Hainan, southwestern Fujian, western Taiwan) (Figure 4). The total area was 30.42 × 104 km2, accounting for approximately 3.17% of China’s territory (Table 4). Additionally, its overlap with the suitable area of S. himalayana was 1.35 × 104 km2, accounting for 100% of the total suitable habitat area for the endangered parasite (Table S2).
Under future climate scenarios, the suitable habitat for T. obovatum was predicted to expand by an average of 0.95%, reaching an area of 30.71 × 104 km2 (Table 4). Across all SSP2-4.5 scenarios, the suitable habitat area of T. obovatum decreased, with an average decline of 6.28%. Under the SSP1-2.6 2070s and SSP5-8.5 2090s, the suitable habitat area for T. obovatum also showed a decrease of 3.52% and 13.02%, respectively. Furthermore, its overlap with the suitable area of S. himalayana was projected to increase by an average of 22.96%, reaching a total area of 1.66 × 104 km2 (Table S2).
Under the current climate scenario, the suitable habitat for T. cruciatum was mainly distributed in southern Yunnan, southern Hainan, and southwestern Taiwan (Figure 5). The total area was 7.35 × 104 km2, accounting for approximately 0.77% of China’s territory (Table 4). Additionally, its overlap with the suitable area of S. himalayana was 1.35 × 104 km2, accounting for 100% of the total suitable habitat area for the parasite (Table S2).
Under future climate scenarios, the suitable habitat area for T. cruciatum was predicted to expand by an average of 8.16%, reaching a total area of 7.95 × 104 km2 (Table 4). Under four future scenarios, the suitable habitat area of T. cruciatum exhibited a slight decrease (i.e., SSP1-2.6 2090s, 5.99%; SSP2-4.5 2070s, 1.77%; SSP2-4.5 2090s, 1.36%; SSP5-8.5 2070s, 0.95%). Furthermore, its overlap with the suitable area of S. himalayana was projected to increase by an average of 36.30%, reaching a total area of 1.84 × 104 km2 (Table S2).
Under the current climate scenario, the suitable habitat for T. serrulatum was mainly distributed in southwestern China (southeastern Guizhou, southern Sichuan, southeastern Tibet, Yunnan) and surrounding provinces (western Guangxi), with several fragmented suitable habitats scattered in southeastern Fujian (Figure 6). The total area was 38.09 × 104 km2, accounting for approximately 3.97% of China’s territory (Table 4). Additionally, its overlap with the suitable area of S. himalayana was 0.04 × 104 km2, accounting for only 2.96% of the total suitable habitat area for the parasite (Table S2).
Under future climate scenarios, the suitable habitat area for T. serrulatum was predicted to reduce by an average of 4.25%, dropping to an area of 36.47 × 104 km2 (Table 4). Only under SSP1-2.6 2050s and SSP1-2.6 2070s did the suitable habitat area of T. serrulatum show a slight increase of 2.86% and 2.13%, respectively. Furthermore, its overlap suitable area with S. himalayana was projected to increase by an average of 272.00%, reaching a total area of 0.15 × 104 km2 (Table S2).
Overall, the endangered parasite and its five hosts showed divergent future trends in suitable habitat, and they could be divided into two groups: increasing type (S. himalayana, T. planicaule, T. obovatum, and T. cruciatum) and decreasing type (T. obtectum and T. serrulatum).

2.3. Future Centroid Shifts of Sapria himalayana and Its Five Hosts

The suitable habitat centroid for S. himalayana under the current climate was located in the southern part of Yunnan Province (22.37° N, 100.42° E). Under future climate scenarios, its centroid was projected to shift predominantly southeastward (Figure 7). The longest shifting distance was observed under SSP1-2.6 2050s, with a southeastward movement of 363.67 km to 21.15° N, 103.69° E.
The centroid of T. planicaule under the current climate was located in the central part of Guangxi Province (24.06° N, 109.01° E). Under future climate scenarios, no noticeable trend was identified in its centroid shift direction (Figure 7). The longest shifting distance occurred under SSP5-8.5 2090s, with a northwestward movement of 122.90 km to 24.11° N, 107.80° E.
The centroid of T. obtectum under the current climate was located in the eastern part of Guizhou Province (27.16° N, 108.39° E). Under future climate scenarios, its centroid was projected to shift predominantly southwestward (Figure 7). The longest shifting distance occurred under SSP2-4.5 2090s, with a southwestward movement of 62.47 km to 27.04° N, 107.77° E.
The centroid of T. obovatum under the current climate was located in the central part of Guizhou Province (26.94° N, 106.42° E). Under future climate scenarios, its centroid was projected to shift predominantly southwestward (Figure 7). The longest shifting distance occurred under SSP1-2.6 2070s, with a southwestward movement of 73.13 km to 26.37° N, 106.05° E.
The centroid of T. cruciatum under the current climate was located in the southeastern part of Yunnan Province (22.64° N, 103.21° E). Under future climate scenarios, no noticeable trend was identified in its centroid shift direction (Figure 7). The longest shifting distance occurred under SSP5-8.5 2050s, with a northwestward movement of 155.59 km to 23.02° N, 101.75° E.
The centroid of T. serrulatum under the current climate was located in the northeastern part of Yunnan Province (26.00° N, 103.43° E). Under future climate scenarios, its centroid was projected to shift predominantly northwestward (Figure 7). The longest shifting distance occurred under SSP1-2.6 2090s, with a northwestward movement of 86.69 km to 26.08° N, 102.57° E.
In brief, the five hosts exhibited different future centroid shifts and can be categorized into three types: shifting southwestward (i.e., T. obtectum and T. obovatum), shifting northwestward (i.e., T. serrulatum), and shifting in various directions (i.e., T. planicaule and T. cruciatum). Accordingly, these hosts differed from S. himalayana in centroid shift.

2.4. Niche Overlap of Sapria himalayana and Its Five Hosts

The niche overlap indices D and I between S. himalayana and T. planicaule under the current scenario were 0.1299 and 0.5118, respectively (Table 5), indicating a very limited niche overlap between them. Under future scenarios, both D and I values were predicted to increase, with their mean values reaching 0.1923 and 0.5493, respectively (Table 5). Notably, the level of niche overlap still remained in the “very limited overlap” category.
The niche overlap indices D and I between S. himalayana and T. obtectum under the current scenario were 0.1435 and 0.4899, respectively (Table 5), indicating a very limited niche overlap. Under future scenarios, both D and I values were predicted to increase, with their mean values reaching 0.1965 and 0.5093, respectively (Table 5). Notably, the level of niche overlap still remained in the “very limited overlap” category.
The niche overlap indices D and I between S. himalayana and T. obovatum under the current scenario were 0.2162 and 0.5851, respectively (Table 5), indicating a low niche overlap. Under future scenarios, both D and I values were predicted to increase, with their mean values reaching 0.2925 and 0.6265, respectively (Table 5). Notably, the level of niche overlap still remained in the “low overlap” category.
The niche overlap indices D and I between S. himalayana and T. cruciatum under the current scenario were 0.4413 and 0.7960, respectively (Table 5), indicating a moderate niche overlap level. Under future scenarios, both D and I values were predicted to increase, with their mean values reaching 0.5556 and 0.8498, respectively (Table 5). Notably, the level of niche overlap still remained in the “moderate overlap” category.
The niche overlap indices D and I between S. himalayana and T. serrulatum under the current scenario were 0.1903 and 0.5300, respectively (Table 5), indicating a very limited niche overlap. Under future scenarios, both D and I values were predicted to increase, with their mean values reaching 0.2601 and 0.5802, respectively (Table 5). In this case, the level of niche overlap increased to the “low overlap” category.
Taken together, both niche overlap indices revealed consistent patterns across the five species pairs. They can be classified into three categories: moderate overlap (i.e., S. himalayana vs. T. cruciatum), low overlap (i.e., S. himalayana vs. T. obovatum; S. himalayana vs. T. serrulatum), and very limited overlap (i.e., S. himalayana vs. T. planicaule; S. himalayana vs. T. obtectum) (Table 5).

3. Discussion

3.1. Model Evaluation and Key Factors

The potential distributions of the endangered S. himalayana and its five host species were projected using MaxEnt models. Prior to final modeling, each species-specific model was optimized through parameter tuning of the regularization multiplier (RM) and feature class (FC) to enhance robustness and mitigate overfitting. The optimized models demonstrate high predictive performance, as reflected in strong evaluation metrics. For S. himalayana, both the AUC and TSS values indicate excellent model fit (Table S1), and its predicted current distribution aligns closely with existing occurrence records. Similarly, all host plant models performed reliably, with AUC and TSS scores consistently within the range indicative of high accuracy (Table S1). These results confirm that the tuned models are suitable for assessing current and future habitat suitability across the targeted species. Furthermore, our results show that the MESS values for most of the predicted areas are positive (Table 1), indicating that these areas are similar to the model calibration area in environmental conditions, and that the projection outcomes are mainly based on the simulation of the observed environmental range.
Our model indicates that the top three key factors influencing the current distribution of S. himalayana are bio3, bio4, and bio1, with their combined contribution reaching 96.5% (Table 3). It is noteworthy that all three factors are temperature-related climatic variables, suggesting that temperature, rather than topography or anthropogenic influence, is the primary limiting factor for the endangered holoparasite in distribution. S. himalayana belongs to the genus Sapria from the family Rafflesiaceae, and now this genus only comprises four species [41]. Following Wu Zhengyi’s classification for the genus areal-types of Chinese seed plants, this genus exhibits a tropical Asian distribution pattern, with its members spanning from tropical India to southern China [18]. According to Flora of China, there is just one species, namely S. himalayana, in China. It is confined to Zayü of Tibet, as well as Lancang, Mengla, and Jinghong of Yunnan Province, China [25,38]. The known records of this species align with the outcome of our MaxEnt modeling (Figure 1). Currently, only some studies have analyzed the ecological requirements of S. himalayana. Chaisung et al. [42] reported that its three populations in Chiang Mai Province, Thailand, were affected by soil temperature, air temperature, and light intensity, and they contended that temperature affected its population mortality rate, because lower temperatures corresponded to higher mortality. Bänziger et al. [29] discovered that even during the dry season, the humidity in the perianth tube of this species in Chiang Mai Province, northern Thailand, remained consistently close to saturation. This was because the water transpired from the perianth tube could readily be replenished from the host Tetrastigma. As a result, it seems likely that precipitation may not be a key factor limiting its population distribution therein. Combined with our model projections, the results indicate that S. himalayana favors warm, humid climates and that its distribution is more strongly constrained by temperature than by other variables.
The five host species of S. himalayana can be divided into two categories based on their primary influencing factors. T. planicaule, T. obovatum, and T. cruciatum are mainly influenced by temperature-related climatic variables in distribution, while T. obtectum and T. serrulatum are primarily affected by precipitation-related climatic variables (Table 3). Key environmental factors shared between S. himalayana and its hosts varied among species. Bio3 emerged as a common key factor for T. obovatum (17.9%) and T. cruciatum (16.1%) in relation to S. himalayana (88.4%). Also, bio4 was identified as a shared factor for T. serrulatum (33.7%), T. planicaule (5.7%) and S. himalayana (4.8%). In contrast, T. obtectum and S. himalayana have no key common factors (Table 3). Furthermore, a notable pattern emerged regarding future habitat trends between two host groups. Host species whose distribution was primarily influenced by temperature-related factors exhibited an expansion in suitable habitat, whereas those governed mainly by precipitation-related factors showed a contraction. This may suggest that although both S. himalayana and its five hosts are primarily influenced by climatic factors, there is a considerable difference in the key climate factors for these five hosts. Furthermore, the Human Influence (HI) is identified as a key common factor shaping the suitable distribution for the five hosts, rather than for the parasite, highlighting a greater anthropogenic pressure on these hosts.

3.2. Suitable Range and Future Change for Sapria himalayana and Its Hosts

This study provides the first delineation of the suitable habitat for S. himalayana in China under current climatic conditions, covering approximately 1.35 × 104 km2 (0.14% of China’s land area). Projections suggest this area may expand to 1.86 × 104 km2 in future climates (Table 4). Notably, although some parts of Taiwan fall into the suitable habitat range of S. himalayana and its hosts like T. cruciatum, we do not think that S. himalayana will shift to Taiwan in the near future. First of all, S. himalayana is a root holoparasitic plant with limited distribution in southern China, such as Yunnan Province, and in some mountainous areas of neighboring countries, including India, Myanmar, Thailand, and Vietnam [41]. Secondly, according to the Flora of Taiwan and related literature, up to now, no occurrence record of S. himalayana has been reported in Taiwan [43,44]. Moreover, the Chinese mainland and Taiwan Province are separated by the Taiwan Strait, which has an average width of 200 km [45]. The fruit of S. himalayana is a spherical berry [25]. Although the dispersal mechanism of its fruit or seed is unclear, we think that it is unlikely for this parasite to spread and reach Taiwan in the coming decades.
Habitat overlap for S. himalayana with its hosts varies markedly, from the minimal overlap with T. serrulatum (0.04 × 104 km2, 2.96%) to the complete overlap (100%) with both T. obovatum and T. cruciatum, matching the parasite’s entire current suitable area (Table S2). Niche overlap indices (D and I) also corroborate this variation. T. serrulatum, T. planicaule, and T. obtectum show only negligible overlap with S. himalayana, while T. cruciatum shows a moderate overlap tier, which is the highest overlap among all pairs (Table 5). Although future projections indicate an increase in the numerical overlap values for all pairs, the overall tiered structure remains largely unchanged. The only notable shift is for S. himalayana vs. T. serrulatum, which moves from negligible to low overlap (Table 5). This pattern may be attributed to the currently restricted distribution of S. himalayana in China, as its broader geographic range extends into other regions such as India, Myanmar, and Vietnam [41].
Notably, future projections show a decline for hosts with the largest current ranges (i.e., T. obtectum and T. serrulatum), but an expansion for those with smaller ranges (i.e., T. planicaule, T. obovatum, and T. cruciatum) (Table 4). This inverse relationship suggests that the responses of S. himalayana and its hosts to climate change may be partially mediated by the hosts’ extant distributional extents. One major reason for this is that while the suitable habitats of these hosts are primarily governed by climatic conditions, their sensitivity to climate change varies across species (see Table 3 and Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6). Consequently, climate change may directly alter the future ranges of both the holoparasitic plant and its hosts, and indirectly shape the distribution of the endangered holoparasite S. himalayana by differentially modifying the available habitat of its various hosts. These shifts could also potentially lead to changes in the parasite’s host preference.

3.3. Implications for Sapria himalayana Conservation

This study is the first to identify key limiting factors and delineate suitable habitat range of S. himalayana in China, providing a reference for its future conservation. Firstly, we think that more conservation efforts should be paid to the suitable habitats of S. himalayana, particularly highly suitable areas. Considering that part of the predicted suitable habitat for S. himalayana in southern Yunnan is already located within the Xishuangbanna National Nature Reserve, southern China [38], while some other suitable areas lie just outside the boundary of the reserve, we propose considering enlarging the range of reserve appropriately or establishing a new plant conservation site to better protect wild S. himalayana populations. Notably, some areas in southeastern Tibet are expected to become suitable habitats for S. himalayana under several future scenarios in this study (e.g., SSP1-2.6 2070s, SSP1-2.6 2090s, SSP5-8.5 2090s) (Figure 1). Therefore, we recommend conducting supplementary field surveys and dynamic monitoring in these potential areas. Furthermore, S. himalayana is an obligate root holoparasite, forming an essential symbiotic association with multiple Tetrastigma host species [24,46]. Consequently, effective conservation strategies should simultaneously take into account the conservation of these host plants across the parasite’s distribution range and the conservation of S. himalayana itself as well.
In contrast to the overall expansion projected for S. himalayana, the five hosts exhibit divergent future habitat trends. In future scenarios, T. obtectum and T. serrulatum are expected to contract. While T. planicaule, T. obovatum, and T. cruciatum are projected to expand, the extent of increases is limited. T. planicaule and T. obovatum increase by only 2.49% and 0.95% on average, respectively. Although T. cruciatum shows a larger relative increase (8.26%), its average suitable area in the future (7.95 × 104 km2) remains the smallest among all hosts (Table 4). Moreover, the future suitable areas for these three hosts fluctuate across climate scenarios, often asynchronously with S. himalayana. As exemplified by T. obovatum, its suitable area decreases under four scenarios (i.e., SSP1-2.6 2070s, SSP2-4.5 2050s, SSP2-4.5 2070s, SSP2-4.5 2090s) while S. himalayana’s increases, yet under SSP5-8.5 2050s, the trend reverses (Table 4). Furthermore, future habitat overlaps between S. himalayana and these hosts are generally projected to decline (e.g., from 68.69% to 67.20% with T. planicaule; from 100% to 89.78% with T. obovatum), or remain minimal despite an increase (e.g., from 2.96% to 8.06%, averaging only 0.15 × 104 km2 with T. cruciatum) (Table S4). In summary, considering the divergent responses of multiple hosts, climate change will likely have adverse effects on S. himalayana in its distribution.
As an endophytic herb, S. himalayana is an obligate root holoparasite with high dependence on its hosts [23,24]. Furthermore, this parasitic plant is dioecious, and it has difficulties in pollination, resulting in low fruit set in the wild [36,47]. Our analysis indicates that an asynchrony exists in suitable habitat shift between S. himalayana and its hosts under varying climate scenarios. As a result, this mismatch could pose challenges to the future survival and distribution of S. himalayana in a warming climate.

3.4. The Limitations of This Study

Our study has certain limitations. Firstly, given that S. himalayana has few occurrence records, resulting from the geographically restricted distribution in China, we assessed its risk of niche truncation, although we did not employ a spatially nested hierarchical SDMs (n-SDMs) framework [48]. The results suggest that S. himalayana and its five hosts appear to have a high degree of niche truncation (Table 2), which indicate that the resulting projections may not reflect the entire ecological niche of the targeted species. They are better suited for “relative habitat suitability ranking” and “survey and conservation priority area screening” within China. Therefore, the key influencing factors identified in this study may only reflect the environmental situation of S. himalayana within China, and it is difficult to determine its ecological niches in other countries in Southeast Asia. Although developing n-SDMs contributes to improving model accuracy, we did not employ such a method because our research objective is to identify national-scale priority conservation areas and provide management recommendations for this endangered holoparasite in China.
Moreover, as a correlative model, MaxEnt prediction is based on known occurrence (or absence) points and associated environmental variables [49], which do not incorporate species-specific biological traits such as dispersal capability. In fact, although parasitic plants’ distribution is often constrained by their hosts, it is a great challenge to quantify host influence as a model variable due to the complexity of interspecific relationships [50,51]. In addition, this study did not take into account the preference of S. himalayana for different hosts.

4. Materials and Methods

4.1. Occurrence Records

The occurrence data of Sapria himalayana and its hosts were sourced through the following approaches. (1) Literature search: We used specific names, Latin names, and synonyms of each species as keywords to search for their occurrence records from Flora of China, provincial floras, relevant checklists, and academic papers. (2) Herbarium retrieval: We obtained specimen data with precise latitude–longitude coordinates or detailed location descriptions for each species from the Chinese Virtual Herbarium (CVH, https://www.cvh.ac.cn, accessed on 18 April 2025), National Specimen Information Infrastructure (NSII, http://www.nsii.org.cn, accessed on 18 April 2025), and Global Biodiversity Information Facility (GBIF, https://www.gbif.org/, accessed on 16 January 2026). (3) Image library: We searched for images of each species in the Plant Photo Bank of China (PPBC, http://ppbc.iplant.cn, accessed on 18 April 2025) and extracted corresponding detailed place names. (4) Field survey: We documented the spatial locations of these species’ wild populations via field investigations in eastern, southern, and southwestern China from 2023 to 2025.
For records with specific location descriptions but lacking coordinate data, we used Google Earth (https://earth.google.com/web/, accessed on 26 April 2025) to obtain their latitude and longitude information, which was precise to two decimal places. Subsequently, we excluded duplicate records and those representing cultivated plants in schools, parks, and botanical gardens and so on. Then we spatially rarefied the occurrence data by using the Spatial Rare Occurrence Data for SDMs tool in SDMtoolbox v2.6 to ensure that only one occurrence point was retained per 1 km × 1 km grid cell [52]. Ultimately, we collected the coordinate information for each species with the following numbers: 22 for S. himalayana, 317 for Tetrastigma planicaule, 495 for T. obtectum, 93 for T. obovatum, 84 for T. cruciatum, and 329 for T. serrulatum (Table S3). The occurrence records for these six species are shown in Figure 8 and Table S3.

4.2. Environmental Variables

At regional scales, climate acts as a primary factor governing the geographic distribution of plants [53,54]. The six species spanned a three-step ladder of China in terms of occurrence records (Figure 8) and were mainly located in densely populated areas of southern China. Therefore, we selected topographic and anthropogenic variables besides climate. In total, 23 environmental variables were used as candidate predictors, which can be categorized into three groups: (1) Bioclimatic variables: We downloaded 19 bioclimatic variables from WorldClim, including both current and future climate data. Current bioclimatic data were based on WorldClim v2.1. Future climate data were derived from the Beijing Climate Center Climate System Model version 2 with Medium Resolution (BCC-CSM2-MR) in the Coupled Model Intercomparison Project Phase 6 (CMIP6). This model is considered suitable for Asia, particularly for China [40]. Because the scope of this research is in China, where the parasite and its multiple hosts occur, we only selected the BCC-CSM2-MR climate model in this study. We selected three Shared Socioeconomic Pathways (SSPs), namely SSP1-2.6 (optimistic scenario), SSP2-4.5 (moderate scenario), and SSP5-8.5 (pessimistic scenario), to represent different future climate change scenarios [55]. Each pathway contained three time periods (i.e., 2041–2060, 2061–2080, and 2081–2100). (2) Topographic variables: We downloaded elevation data from WorldClim (https://www.worldclim.org/, accessed on 3 May 2025) and extracted slope and aspect data from the Digital Elevation Model (DEM) (http://www.tuxingis.com, accessed on 3 May 2025). (3) Anthropogenic variables: We downloaded the global Human Influence (HI) dataset from the NASA Socioeconomic Data and Applications Center (SEDAC) (https://sedac.ciesin.columbia.edu, accessed on 3 May 2025) which comprised nine layers: population density, built-up areas, night-time light, land use/land cover, and human access features (roads, railways, coastlines, and navigable rivers) [56]. Detailed descriptions of each environmental variable are provided in Table S4.
To constrain modeling to the study area, we uniformly cropped all environmental variables to match China’s territory and resampled them to 30 arc-second spatial resolution. Subsequently, to avoid overfitting of models caused by multicollinearity among variables, we screened all 23 variables for each species (including S. himalayana and its five hosts) by using Spearman’s correlation coefficient. For two variables with an absolute coefficient value > 0.7, the one with the lower contribution was removed. If the value between two variables was ≤0.7, both were retained [57].

4.3. Model Construction

In this study, we once tried to use Biomod2 to construct an ensemble model for the targeted species. Unfortunately, we noted that the model’s performance was below expectations. The MaxEnt model is extensively employed due to its strong predictive performance, user-friendly operation, and ability to generate reliable projections at low sample sizes [58,59]. Existing studies have shown that for endangered plants, which often have limited distributions and few records, it seems a good choice to use the MaxEnt model [60]. Therefore, we used MaxEnt v3.4.4 to predict potential suitable habitats for S. himalayana and its five hosts under both current and future climate scenarios rather than ensemble models in Biomod2. Firstly, we used the ENMeval package in R v4.4.2 to select the optimal combination of RM and FC parameters for each model. Specifically, the RM value ranged from 0.5 to 4 (increments of 0.5) while the five FC features combined into six classes (i.e., L, H, LQ, LQH, LQHP, LQHPT; L = linear, Q = quadratic, H = hinge, P = product, T = threshold). The combination with ΔAICc = 0 was identified as the best modeling parameters [61,62]. During modeling, 75% of the occurrence data were selected as the training set and the remaining 25% as the testing set. To ensure projection accuracy, 10,000 background points were randomly generated and 10 bootstrap replicates were performed for each model. Meanwhile, we set corresponding RM and FC parameters while modeling for different species based on optimization results (Table 6). Additionally, we used a jackknife test to evaluate the contribution of each environmental variable to each species’ distribution.
We used the area under the receiver operating characteristic curve (AUC) and true skill statistic (TSS) to evaluate model performance [63]. The AUC value ranged from 0 to 1 and can be categorized into five levels: failing (0.5–0.6), poor (0.6–0.7), fair (0.7–0.8), good (0.8–0.9), and excellent (0.9–1.0) [64,65]. The TSS value ranged from –1 to 1, where 1 indicated perfect performance, while 0 or less suggested that the model performed no better than random prediction [3,66].

4.4. Geospatial Analysis

MaxEnt generated distribution predictions based on a logistic output format, which included a continuous habitat suitability index with values ranging from 0 (unsuitable) to 1 (perfectly suitable) [67]. We imported model outputs for S. himalayana and its five hosts into ArcGIS v10.8.1 in “.asc” format. Subsequently, we reclassified habitat suitability into four categories (i.e., highly suitable, moderately suitable, low suitable, and unsuitable) based on maximizing the sum of sensitivity and specificity (maxSSS) [68,69]. Given the considerable variation in maxSSS values among the six targeted species, we set a suitability threshold for each species (Table 7). Meanwhile, we used the Spatial Analyst tool to calculate overlapping areas of suitable habitat between S. himalayana and each of its hosts [19]. To reflect the shift in suitable habitats for S. himalayana and its five hosts under future climate scenarios, we used SDMtoolbox v2.6 in ArcGIS v10.8.1 to calculate the suitable distribution centroid under current and future scenarios for each species [70].
To evaluate niche truncation in S. himalayana and its five hosts, we calculated the niche unfilling index for each species using the “ecospat” package in R v4.4.2 [71,72]. Given that the core contribution of this study is to provide survey priorities and risk alerts within China, we did not further construct spatially nested hierarchical Species Distribution Models (n-SDMs). This approach can tackle the issue of species niche truncation through the integration of large-scale, low-resolution models with small-scale, high-resolution ones. However, the current methodological framework remains fragmented, having various nested strategies (e.g., “covariate”, “multiply”, etc.) [48,73], thus lacking a standardized, widely accepted methodology. Additionally, the multivariate environmental similarity surface (MESS) values for all six species were computed with the “dismo” package to assess extrapolation risks under future climate scenarios [74].
The niche overlap index quantified similarity in resource utilization and habitat preferences between different species [75]. To measure the niche overlap between S. himalayana and its five hosts across different climate scenarios, we selected Schoener’s D [76] and modified Hellinger distance (Hellinger’s I) [77]. These metrics were calculated in ENMTools v1.3.1.
Schoener’s D is defined as:
D p X ,   p Y = 1 − 1 2 ∑ i | p X , i − p Y , i |
In which p X , i (or p Y , i ) is the normalized suitability score for species X (or Y) in grid cell i.
Schoener’s D quantified the similarity of potential distributions by comparing corresponding values in each grid cell. D value ranged from 0 to 1, where 0 indicated no niche overlap and 1 indicated identical potential distribution [71]. Following the method of [78], D value was classified into five levels: (1) extremely high overlap: 0.80–1.00; (2) high overlap: 0.60–0.80; (3) moderate overlap: 0.40–0.60; (4) low overlap: 0.20–0.40; (5) no or very limited overlap: 0.00–0.20.
Modified Hellinger’s I is defined as:
I p X ,   p Y = 1 − 1 2 ∑ i ( p X , i − p Y , i ) 2
I value ranged from 0 to 1, with a value of 0 indicating entirely divergent environmental requirements, and a value of 1 indicating that the environmental factor requirements for both species completely overlapped [79].

5. Conclusions

Our study delineates, for the first time, the suitable habitat range of the endangered endo-holoparasitic plant Sapria himalayana in China and identifies temperature-related climatic variables as key environmental factors influencing its distribution. Optimized MaxEnt modeling indicates that while the suitable habitat for S. himalayana will undergo a moderate increase under future climate scenarios, its multiple hosts show varying trends in potential habitat shifts. Together with the mismatch in suitability between the parasite and its hosts, divergent climatic sensitivities among host species could affect the survival and distribution of S. himalayana. Therefore, we recommend that future conservation efforts should prioritize: (1) supplementary field surveys of S. himalayana populations; (2) expansion and improved management of its protected areas; and (3) maintenance of its distinct ecological interactions with obligate hosts. In summary, this study provides an important reference for the conservation of S. himalayana in China, emphasizing the necessity of safeguarding its distinct hosts in the context of global warming.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/plants15040574/s1, Table S1: The mean value (±SD) of the area under curve (AUC) and true skill statistic (TSS) of Sapria himalayana and its five hosts under different climate scenarios; Table S2: Dynamics of changes in overlap suitable areas of Sapria himalayana and its five hosts under different climate scenarios; Table S3: Occurrence records of Sapria himalayana and its five hosts in China; Table S4: Description of 23 candidate predictors used in MaxEnt model under different climate scenarios.

Author Contributions

W.H.: writing—original draft preparation, visualization, software, investigation, formal analysis, data curation. Y.L.: visualization, investigation, data curation. G.Z.: writing–review and editing, supervision, investigation, conceptualization. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the investigation and assessment of key protected wild plants in Jiangsu Province from Jiangsu Forestry Bureau (No. 2023053SMnull0162).

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Acknowledgments

We thank Z. Lan, T. Liu, X. Lu, L.Y. Pang, H.R. Wang, and Y. Zhang for their valuable advice on an earlier draft of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Zandalinas, S.I.; Fritschi, F.B.; Mittler, R. Global warming, climate change, and environmental pollution: Recipe for a multifactorial stress combination disaster. Trends Plant Sci. 2021, 26, 588–599. [Google Scholar] [CrossRef]
  2. Abbass, K.; Qasim, M.Z.; Song, H.M.; Murshed, M.; Mahmood, H.; Younis, I. A review of the global climate change impacts, adaptation, and sustainable mitigation measures. Environ. Sci. Pollut. Res. 2022, 29, 42539–42559. [Google Scholar] [CrossRef] [PubMed]
  3. Lu, X.; Jiang, R.Y.; Zhang, G.F. Predicting the potential distribution of four endangered holoparasites and their primary hosts in China under climate change. Front. Plant Sci. 2022, 13, 942448. [Google Scholar] [CrossRef]
  4. Waldvogel, A.M.; Feldmeyer, B.; Rolshausen, G.; Exposito-Alonso, M.; Rellstab, C.; Kofler, R.; Mock, T.; Schmid, K.; Schmitt, I.; Bataillon, T.; et al. Evolutionary genomics can improve prediction of species’ responses to climate change. Evol. Lett. 2020, 4, 4–18. [Google Scholar] [CrossRef] [PubMed]
  5. Mkala, E.M.; Jost, M.; Wanke, S.; Ngarega, B.K.; Hughes, A.; Mutinda, E.S.; Waswa, E.N.; Mwanzia, V.M.; Oulo, M.A.; Wang, V.O.; et al. How vulnerable are holoparasitic plants with obligate hosts to negative climate change impacts? Ecol. Inf. 2022, 69, 101636. [Google Scholar] [CrossRef]
  6. Watson, D.M.; McLellan, R.C.; Fontúrbel, F.E. Functional roles of parasitic plants in a warming world. Annu. Rev. Ecol. Evol. Syst. 2022, 53, 25–45. [Google Scholar] [CrossRef]
  7. Watson, D.M. Determinants of parasitic plant distribution: The role of host quality. Botany 2009, 87, 16–21. [Google Scholar] [CrossRef]
  8. Rafferty, N.E.; Agnew, L.; Nabity, P.D. Parasitism modifies the direct effects of warming on a hemiparasite and its host. PLoS ONE 2019, 14, e0224482. [Google Scholar] [CrossRef]
  9. Yang, Z.B.; Bai, Y.; Alatalo, J.M.; Huang, Z.D.; Yang, F.; Pu, X.Y.; Wang, R.B.; Yang, W.; Guo, X.Y. Spatio-temporal variation in potential habitats for rare and endangered plants and habitat conservation based on the maximum entropy model. Sci. Total Environ. 2021, 784, 147080. [Google Scholar] [CrossRef] [PubMed]
  10. Těšitel, J. Functional biology of parasitic plants: A review. Plant Ecol. Evol. 2016, 149, 5–20. [Google Scholar] [CrossRef]
  11. Teixeira-Costa, L.; Davis, C.C. Life history, diversity, and distribution in parasitic flowering plants. Plant Physiol. 2021, 187, 32–51. [Google Scholar] [CrossRef]
  12. Chang, C.; Cai, F.; Shen, L.; Jia, X.; Liu, Z.; Wang, C.; Fu, Y.; Luo, Y. Predicting the potential distribution of Phacellanthus tubiflorus (Orobanchaceae): A modeling approach using MaxEnt and ArcGIS. PeerJ 2025, 13, e19291. [Google Scholar] [CrossRef]
  13. Ji, J.C.; Wei, X.X.; Guan, H.; Jin, Z.K.; Yue, X.; Jiang, Z.R.; Su, Y.L.; Sun, S.Y.; Chen, G.L. Evaluation of the impact of ecological factors on the habitat suitability and bioactive components accumulation of the medicinal holoparasitic plant Cynomorium songaricum using machine learning models. Front. Plant Sci. 2025, 16, 1586682. [Google Scholar] [CrossRef]
  14. Vásquez-Aguilar, A.A.; Morales-Saldaña, S.; de los Santos-Gómez, S.M.; Barraza-Ochoa, A.I.; Ornelas, J.F. Range and elevational shifts of mistletoes under future climate change scenarios. Ecol. Evol. 2025, 15, e72388. [Google Scholar] [CrossRef]
  15. Piwowarczyk, R.; Kolanowska, M. Predicting the effect of global warming on the distribution of a polyphagous tree parasite, Orobanche laxissima, based on climatic and ecological data. Glob. Ecol. Conserv. 2023, 44, e02486. [Google Scholar] [CrossRef]
  16. Tu, G.H.; Guo, X.D.; Xi, S.Y.; Ma, X.H.; Jin, L. Predicting potential suitable habitat of Cistanche deserticola by integrating parasitic constraints and land use data into MaxEnt modeling. Front. Plant Sci. 2025, 16, 10.3389. [Google Scholar] [CrossRef] [PubMed]
  17. Piwowarczyk, R.; Kolanowska, M. Effect of global warming on the potential distribution of a holoparasitic plant (Phelypaea tournefortii): Both climate and host distribution matter. Sci. Rep. 2023, 13, 10741. [Google Scholar] [CrossRef] [PubMed]
  18. Li, D.Z. The Families and Genera of Chinese Vascular Plants; Science Press: Beijing, China, 2020. [Google Scholar]
  19. Renjana, E.; Astuti, I.P.; Munawaroh, E.; Mursidawati, S.; Witono, J.R.; Yuzammi; Fijridiyanto, I.A.; Raharjo, P.D.; Solihah, S.M.; Robiansyah, I.; et al. Assessing potential habitat suitability of parasitic plant: A case study of Rafflesia arnoldii and its host plants. Glob. Ecol. Conserv. 2022, 34, e02063. [Google Scholar] [CrossRef]
  20. Obico, J.J.A.; Lapuz, R.S.C.; Barcelona, J.F.; Pelser, P.B. What explains the high island endemicity of Philippine Rafflesia? A species distribution modeling analysis of three threatened parasitic plant species and their hosts. Am. J. Bot. 2024, 111, e16267. [Google Scholar] [CrossRef] [PubMed]
  21. Malabrigo, P., Jr.; Tobias, A.B.; Witono, J.; Mursidawati, S.; Susatya, A.; Siti-Munirah, M.Y.; Wicaksono, A.; Raihandhany, R.; Edwards, S.; Thorogood, C.J. Most of the world’s largest flowers (genus Rafflesia) are now on the brink of extinction. Plants People Planet 2025, 7, 331–346. [Google Scholar] [CrossRef]
  22. Nikolov, L.A.; Davis, C.C. The big, the bad, and the beautiful: Biology of the world’s largest flowers. J. Syst. Evol. 2017, 55, 516–524. [Google Scholar] [CrossRef]
  23. Osathanunkul, M. eDNA-based monitoring of parasitic plant (Sapria himalayana). Sci. Rep. 2019, 9, 9161. [Google Scholar] [CrossRef]
  24. Westwood, J.H. Plant biology: Genome reveals secrets of the alien within. Curr. Biol. 2021, 31, R241–R243. [Google Scholar] [CrossRef] [PubMed]
  25. Huang, S.; Gilbert, M.G. Rafflesiaceae. In Flora of China; Wu, Z.Y., Raven, P.H., Eds.; Science Press: Beijing, China, 2003; Volume 5. [Google Scholar]
  26. Wicaksono, A.; Trần, H.Đ.; Kamal, S.H.S. Preliminary histoanatomical study of Sapria himalayana Griff. f. albovinosa flower buds and its intersection with its host plant Tetrastigma laoticum Gagnep. J. Plant Dev. 2021, 28, 23–31. [Google Scholar] [CrossRef]
  27. Wicaksono, A.; Meitha, K.; Wan, K.L.; Mat Isa, M.N.M.; Parikesit, A.A.; Molina, J. Hairpin in a haystack: In silico identification and characterization of plant-conserved microRNA in Rafflesiaceae. Open Life Sci. 2025, 20, 20221033. [Google Scholar] [CrossRef] [PubMed]
  28. Ralte, L.; Sailo, H.; Kumar, R.; Khiangte, L.; Kumar, N.S.; Singh, Y.T. Identification of novel AKT1 inhibitors from Sapria himalayana bioactive compounds using structure-based virtual screening and molecular dynamics simulations. BMC Complement. Med. Ther. 2024, 24, 116. [Google Scholar] [CrossRef]
  29. Bänziger, H.; Gigon, A.; Bänziger, S.; Suttiprapan, P. Strong emissions of carbon dioxide and water vapour by Sapria himalayana Griff. (Rafflesiaceae): Waste or necessity in a cool flower? Taiwania 2022, 67, 201–210. [Google Scholar]
  30. Wu, Z.Y. Flora Yunnanica; Science Press: Beijing, China, 1979; Volume 2. [Google Scholar]
  31. Arunachalam, A.; Adhikari, D.; Sarmah, R.; Majumder, M.; Khan, M.L. Population and conservation of Sapria himalayana Griffith in Namdapha National Park, Arunachal Pradesh, India. Biodivers. Conserv. 2004, 13, 2391–2397. [Google Scholar] [CrossRef]
  32. Borah, D.; Ghosh, D. Sapria himalayana: The Indian cousin of world’s largest flower. Resonance 2018, 23, 479–489. [Google Scholar] [CrossRef]
  33. Liu, J.N.; Gong, Y.X. Sapria himalayana Griff. and Mitrastemon yamamotoi Makino. Hum. Nat. 2019, 2, 76–77. [Google Scholar]
  34. Gao, S.W. Sapria himalayana Griff. Bull. Bio. 2025, 60, 24. [Google Scholar]
  35. Chen, Z.D.; Ren, H.; Wen, J. Vitaceae. In Flora of China; Wu, Z.Y., Raven, P.H., Eds.; Science Press: Beijing, China, 2007; Volume 12. [Google Scholar]
  36. Cai, L.; Arnold, B.J.; Xi, Z.X.; Khost, D.E.; Patel, N.; Hartmann, C.B.; Manickam, S.; Sasirat, S.; Nikolov, L.A.; Mathews, S.; et al. Deeply altered genome architecture in the endoparasitic flowering plant Sapria himalayana Griff. (Rafflesiaceae). Curr. Biol. 2021, 31, 1002–1011. [Google Scholar] [CrossRef]
  37. Ye, Z.R. A miniature Rafflesiaceae flower in Yunnan Province—Sapria himalayana. Life World 2025, 8, 70–75. [Google Scholar]
  38. Wei, W.; Li, Y.M.; Liu, D.T.; Sun, W.B. Composition, distribution of national key protected wild plants in Yunnan Province and the cultivation status in Kunming Botanical Garden. Guihaia 2025, 45, 1–61. [Google Scholar]
  39. Lu, Z.L.; Qin, H.N.; Jin, X.H.; Zhang, Z.X.; Yang, Q.W.; Hong, D.Y.; Li, D.Z.; Li, K.F.; Yuan, L.C.; Zhou, Z.H. On the necessity, principle, and process of updating the List of National Key Protected Wild Plants. Biodivers. Sci. 2021, 29, 1577–1582. [Google Scholar] [CrossRef]
  40. Zhou, Y.R.; Lu, X.; Zhang, G.F. Potentially differential impacts on niche overlap between Chinese endangered Zelkova schneideriana and its associated tree species under climate change. Front. Ecol. Evol. 2023, 11, 1218149. [Google Scholar] [CrossRef]
  41. Iwashina, T.; Tanaka, N.; Aung, M.M.; Devkota, H.P.; Mizuno, T. Phenolic compounds from parasitic Sapria himalayana f. albovinosa and Sapria myanmarensis (Rafflesiaceae) in Myanmar. Biochem. Syst. Ecol. 2020, 93, 104179. [Google Scholar] [CrossRef]
  42. Chaisung, N.; Busaban, B.; Panmongkol, A.; Wongnak, W.; Panyadee, P.; Inta, A. Environmental factors affecting mortality rate of Sapria himalayana Griff. KKU Sci. J. 2018, 46, 485–495. [Google Scholar]
  43. Huang, T.C. Flora of Taiwan, 2nd ed.; Department of Botany, National Taiwan University: Taipei, China, 1996; Volume 2. [Google Scholar]
  44. Yang, Y.B.; Liao, J.K.; Tang, M.S. Seed Plants of Taiwan; Forestry Bureau of the Agriculture Commission of the Executive: Taipei, China, 2008.
  45. Lin, S.F.; Tang, T.Y.; Jan, S.; Chen, C.J. Taiwan strait current in winter. Cont. Shelf Res. 2005, 25, 1023–1042. [Google Scholar] [CrossRef]
  46. Zhang, G.F.; Li, Q.; Sun, S.C. Diversity and distribution of parasitic angiosperms in China. Ecol. Evol. 2018, 8, 4378–4386. [Google Scholar] [CrossRef]
  47. Trần, H.Đ.; Lưu, H.T.; Nguyễn, Q.Đ.; Nguyễn1, H.C.; Athen, P.; Wong, K.M. Identification, sexual dimorphism and aspects of the natural history of Sapria himalayana (Rafflesiaceae) on Vietnam’s Lang Biang Plateau. Bot. Stud. 2018, 59, 29. [Google Scholar] [CrossRef]
  48. Goicolea, T.; Adde, A.; Broennimann, O.; García-Viñas, J.I.; Gastón, A.; José Aroca-Fernández, M.; Guisan, A.; Mateo, R.G. Spatially-nested hierarchical species distribution models to overcome niche truncation in national-scale studies. Ecography 2025, 2025, e07328. [Google Scholar] [CrossRef]
  49. Wang, H.R.; Zhi, F.Y.; Zhang, G.F. Predicting impacts of climate change on suitable distribution of critically endangered tree species Yulania zenii (W. C. Cheng) D. L. Fu in China. Forests 2024, 15, 883. [Google Scholar] [CrossRef]
  50. Yousefzadeh, H.; Amirchakhmaghi, N.; Naseri, B.; Shafzadeh, F.; Kozlowski, G.; Walas, Ł. The impact of climate change on the future geographical distribution range of the endemic relict tree Gleditsia caspica (Fabaceae) in Hyrcanian forests. Ecol. Inform. 2022, 71, 101773. [Google Scholar] [CrossRef]
  51. Mambo, W.W.; Zhu, G.F.; Milne, R.I.; Wambulwa, M.C.; Oyebanji, O.O.; Ngarega, B.K.; Carver, D.; Liu, J. Shrinking horizons: Climate-induced range shifts and conservation status of hickory trees (Carya Nutt.). Ecol. Inf. 2024, 84, 102910. [Google Scholar] [CrossRef]
  52. Brown, J.L.; Bennett, J.R.; French, C.M. SDMtoolbox 2.0: The next generation Python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses. PeerJ 2017, 5, e4095. [Google Scholar] [CrossRef]
  53. Scheffers, B.R.; De Meester, L.; Bridge, T.C.L.; Hoffmann, A.A.; Pandolfi, J.M.; Corlett, R.T.; Butchart, S.H.M.; Pearce-Kelly, P.; Kovacs, K.M.; Dudgeon, D.; et al. The broad footprint of climate change from genes to biomes to people. Science 2016, 354, aaf7671. [Google Scholar] [CrossRef] [PubMed]
  54. Zhang, L.; Sun, P.S.; Huettmann, F.; Liu, S.R. Where should China practice forestry in a warming world? Glob. Change Biol. 2022, 28, 2461–2475. [Google Scholar] [CrossRef] [PubMed]
  55. IPCC. Climate Change 2023: Synthesis Report; Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change; IPCC: Geneva, Switzerland, 2023. [Google Scholar]
  56. Liao, D.; Zhou, B.; Xiao, H.Y.; Zhang, Y.X.; Zhang, S.J.; Su, Q.T.; Yan, X.H. MaxEnt modeling of the impacts of human activities and climate change on the potential distribution of Plantago in China. Biology 2025, 14, 564. [Google Scholar] [CrossRef]
  57. Shao, M.H.; Wang, L.; Li, B.W.; Li, S.Y.; Fan, J.L.; Li, C.J. MaxEnt modeling for Identifying the nature reserve of Cistanche deserticola Ma under effects of the host (Haloxylon bunge) forest and climate changes in Xinjiang, China. Forests 2022, 13, 189. [Google Scholar] [CrossRef]
  58. Abdelaal, M.; Fois, M.; Fenu, G.; Bacchetta, G. Using MaxEnt modeling to predict the potential distribution of the endemic plant Rosa arabica Crép. in Egypt. Ecol. Inform. 2019, 50, 68–75. [Google Scholar] [CrossRef]
  59. Tümer, Y.O.; Caglayan, I.; Yesil, A.; Iverson, L. Two decades of species distribution modeling: A systematic review of methods and applications. Ecol. Model. 2026, 513, 111441. [Google Scholar] [CrossRef]
  60. Rathore, M.K.; Sharma, L.K. Efficacy of species distribution models (SDMs) for ecological realms to ascertain biological conservation and practices. Biodivers. Conserv. 2023, 32, 3053–3087. [Google Scholar] [CrossRef]
  61. Phillips, S.J.; Anderson, R.P.; Dudík, M.; Schapire, R.E.; Blair, M.E. Opening the black box: An open-source release of MaxEnt. Ecography 2017, 40, 887–893. [Google Scholar] [CrossRef]
  62. Chen, Q.H.; Yin, Y.J.; Zhao, R.; Yang, Y.; Teixeira da Silva, J.A.; Yu, X.N. Incorporating local adaptation into species distribution modeling of Paeonia mairei, an endemic plant to China. Front. Plant Sci. 2020, 10, 1717. [Google Scholar] [CrossRef] [PubMed]
  63. Calevo, J.; Dixon, K.W.; Fay, M.F.; Duffy, K.J. Climate change and conservation of Caladenia orchids in Western Australia. Biodivers. Conserv. 2025, 34, 5153–5177. [Google Scholar] [CrossRef]
  64. Phillips, S.J.; Dudík, M. Modeling of species distributions with MaxEnt: New extensions and a comprehensive evaluation. Ecography 2008, 31, 161–175. [Google Scholar] [CrossRef]
  65. Yan, G.; Zhang, G.F. Predicting the potential distribution of endangered Parrotia subaequalis in China. Forests 2022, 13, 1595. [Google Scholar] [CrossRef]
  66. 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, 1223–1232. [Google Scholar] [CrossRef]
  67. Wang, D.; Cui, B.C.; Duan, S.S.; Chen, J.J.; Fan, H.; Lu, B.B.; Zheng, J.H. Moving north in China: The habitat of Pedicularis kansuensis in the context of climate change. Sci. Total Environ. 2019, 697, 133979. [Google Scholar] [CrossRef]
  68. Liu, C.; White, M.; Newell, G. Selecting thresholds for the prediction of species occurrence with presence-only data. J. Biogeogr. 2013, 40, 778–789. [Google Scholar] [CrossRef]
  69. Melese, D.; Woldu, Z.; Asfaw, Z.; Warkineh, B. Predicting the invasion potential of two alien plant species (Parthenium hysterophorus L., Asteraceae and Salvia tiliifolia Vahl, Lamiaceae) under the current and future climate change scenarios in Ethiopia. Ecol. Process 2025, 14, 79. [Google Scholar] [CrossRef]
  70. Xiang, Y.; Yang, Q.; Li, S.; Liu, Y.; Li, Y.; Ren, J.; Yao, J.; Luo, X.; Luo, Y.; Yao, B. Climate change drives northwestward migration of Betula alnoides: A multi-Scenario MaxEnt modeling approach. Plants 2025, 14, 2539. [Google Scholar] [CrossRef] [PubMed]
  71. Broennimann, O.; Fitzpatrick, M.C.; Pearman, P.B.; Petitpierre, B.; Pellissier, L.; Yoccoz, N.G.; Thuiller, W.; Fortin, M.J.; Randin, C.; Zimmermann, N.E.; et al. Measuring ecological niche overlap from occurrence and spatial environmental data. Glob. Ecol. Biogeogr. 2012, 21, 481–497. [Google Scholar] [CrossRef]
  72. Di Cola, V.; Broennimann, O.; Petitpierre, B.; Breiner, F.T.; D’Amen, M.; Randin, C.; Engler, R.; Pottier, J.; Pio, D.; Dubuis, A.; et al. Ecospat: An R package to support spatial analyses and modeling of species niches and distributions. Ecography 2017, 40, 774–787. [Google Scholar] [CrossRef]
  73. Guisan, A.; Chevalier, M.; Adde, A.; Zarzo-Arias, A.; Goicolea, T.; Broennimann, O.; Petitpierre, B.; Scherrer, D.; Rey, P.L.; Collart, F.; et al. Spatially nested species distribution models (N-SDM): An effective tool to overcome niche truncation for more robust inference and projections. J. Ecol. 2025, 113, 1588–1605. [Google Scholar] [CrossRef]
  74. Elith, J.; Kearney, M.; Phillips, S. The art of modelling range-shifting species. Methods Ecol. Evol. 2010, 1, 330–342. [Google Scholar] [CrossRef]
  75. Lan, Z.; Zhang, G.F. Range shifts of the endangered Luehdorfia chinensis chinensis (Lepidoptera, Papilionidae) and its specific hosts in China under climate change. Ecol. Evol. 2025, 15, e72057. [Google Scholar] [CrossRef]
  76. Schoener, T.W. The Anolis lizards of Bimini: Resource partitioning in a complex fauna. Ecology 1968, 49, 704–726. [Google Scholar] [CrossRef]
  77. Warren, D.L.; Glor, R.E.; Turelli, M. Environmental niche equivalency versus conservatism: Quantitative approaches to niche evolution. Evolution 2008, 62, 2868–2883. [Google Scholar] [CrossRef]
  78. Rödder, D.; Engler, J.O. Quantitative metrics of overlaps in Grinnellian niches: Advances and possible drawbacks. Glob. Ecol. Biogeogr. 2011, 20, 915–927. [Google Scholar] [CrossRef]
  79. Hang, W.Y.; Yan, G.; Zhang, G.F. Population-level ecological niche models to assess the impact of climate change on endangered and relict tree species: A case study of Parrotia subaequalis in China. Trees For. People 2025, 22, 101049. [Google Scholar] [CrossRef]
Plants 15 00574 g001
Figure 1. Predicted suitable habitat distribution of Sapria himalayana under different climate scenarios in China.
Figure 1. Predicted suitable habitat distribution of Sapria himalayana under different climate scenarios in China.
Plants 15 00574 g001
Plants 15 00574 g002
Figure 2. Predicted suitable habitat distribution of Tetrastigma planicaule under different climate scenarios in China.
Figure 2. Predicted suitable habitat distribution of Tetrastigma planicaule under different climate scenarios in China.
Plants 15 00574 g002
Plants 15 00574 g003
Figure 3. Predicted suitable habitat distribution of Tetrastigma obtectum under different climate scenarios in China.
Figure 3. Predicted suitable habitat distribution of Tetrastigma obtectum under different climate scenarios in China.
Plants 15 00574 g003
Plants 15 00574 g004
Figure 4. Predicted suitable habitat distribution of Tetrastigma obovatum under different climate scenarios in China.
Figure 4. Predicted suitable habitat distribution of Tetrastigma obovatum under different climate scenarios in China.
Plants 15 00574 g004
Plants 15 00574 g005
Figure 5. Predicted suitable habitat distribution of Tetrastigma cruciatum under different climate scenarios in China.
Figure 5. Predicted suitable habitat distribution of Tetrastigma cruciatum under different climate scenarios in China.
Plants 15 00574 g005
Plants 15 00574 g006
Figure 6. Predicted suitable habitat distribution of Tetrastigma serrulatum under different climate scenarios in China.
Figure 6. Predicted suitable habitat distribution of Tetrastigma serrulatum under different climate scenarios in China.
Plants 15 00574 g006
Plants 15 00574 g007
Figure 7. Shift of the core distribution of Sapria himalayana and its five hosts under nine future climate scenarios in China. For each species, the centroid shift map shows a close-up view of the corresponding red rectangle in the upper-right corner.
Figure 7. Shift of the core distribution of Sapria himalayana and its five hosts under nine future climate scenarios in China. For each species, the centroid shift map shows a close-up view of the corresponding red rectangle in the upper-right corner.
Plants 15 00574 g007
Plants 15 00574 g008
Figure 8. Occurrence records of Sapria himalayana and its five hosts in China. (A) S. himalayana; (B) Tetrastigma planicaule; (C) T. obtectum; (D) T. obovatum; (E) T. cruciatum; (F) T. serrulatum.
Figure 8. Occurrence records of Sapria himalayana and its five hosts in China. (A) S. himalayana; (B) Tetrastigma planicaule; (C) T. obtectum; (D) T. obovatum; (E) T. cruciatum; (F) T. serrulatum.
Plants 15 00574 g008
Table 1. Environmental extrapolation risk for Sapria himalayana and its five hosts, quantified by proportion (%) of pixels with multivariate environmental similarity surface (MESS) value < 0 under future climate scenarios.
Table 1. Environmental extrapolation risk for Sapria himalayana and its five hosts, quantified by proportion (%) of pixels with multivariate environmental similarity surface (MESS) value < 0 under future climate scenarios.
ScenariosSpecies
Sapria himalayanaTetrastigma planicauleTetrastigma obtectumTetrastigma obovatumTetrastigma cruciatumTetrastigma serrulatum
SSP1-2.62050s0.59670.52370.33080.35150.43340.4308
2070s1.41871.38621.20310.51720.49361.3672
2090s0.70440.64190.38270.66340.55900.7012
SSP2-4.52050s0.77150.70190.52870.48450.55080.5164
2070s1.61321.50101.17570.85070.60901.6607
2090s2.18863.80920.52981.21021.78931.0131
SSP5-8.52050s1.02010.98470.59711.12990.90291.2994
2070s2.63802.25411.50621.08261.25622.0254
2090s3.910111.48620.75801.40923.43011.5606
Table 2. Niche truncation assessment for Sapria himalayana and its five hosts.
Table 2. Niche truncation assessment for Sapria himalayana and its five hosts.
SpeciesNiche Unfilling (%)Global Occurrences (n)China Occurrences (n)
Sapria himalayana90.3811622
Tetrastigma planicaule65.00349317
Tetrastigma obtectum63.762710495
Tetrastigma obovatum18.189376
Tetrastigma cruciatum60.6113984
Tetrastigma serrulatum37.65577329
Niche unfilling (%) means the proportion of occurrence points of species in China that are not covered within the global ecological niche space.
Table 3. Percent contribution of environmental factors of Sapria himalayana and its five hosts in China.
Table 3. Percent contribution of environmental factors of Sapria himalayana and its five hosts in China.
FactorsPercent Contribution (%)
Sapria
himalayana		Tetrastigma planicauleTetrastigma obtectumTetrastigma obovatumTetrastigma cruciatumTetrastigma serrulatum
Bio13.3---1.4-
Bio20.40.50.3--1.4
Bio388.4--17.916.1-
Bio44.85.7--72.933.7
Bio5------
Bio6--19.951.4--
Bio7--2.4-1.6-
Bio8-1.6----
Bio9-----2.3
Bio10------
Bio11-67.4----
Bio12-16.240.17.83.29.3
Bio130.5---2.2-
Bio140-15.2-0-
Bio15---1.2-1
Bio16------
Bio170.21.5-7.5-33.8
Bio181.21.40.7-0.1-
Bio19----0-
Aspect0.60.30.30.90.10.4
Elevation-0.612.71.5-11.4
Slope0.40.50.70.50.10.6
HI0.14.47.811.22.26.1
“-” represents the variables removed after collinear screening, and variables with the top five contribution rates are in bold.
Table 4. Dynamics of changes in the distribution area of Sapria himalayana and its five hosts under different climate scenarios.
Table 4. Dynamics of changes in the distribution area of Sapria himalayana and its five hosts under different climate scenarios.
SpeciesScenariosLow
Suitable Areas		Moderately Suitable AreasHighly
Suitable Areas		Suitable Areas (Moderately and Highly)
Area
(×104 km2)		Trend (%)Area
(×104 km2)		Trend (%)Area (×104 km2)Trend (%)Area
(×104 km2)		Trend
(%)
Sapria himalayanaCurrent2.28-0.69-0.66-1.35-
SSP1-2.62050s2.87↑25.880.85↑23.190.99↑50.001.84↑36.30
2070s6.91↑203.071.31↑89.861.02↑54.552.33↑72.59
2090s4.25↑86.401.19↑72.461.00↑51.522.19↑62.22
Average4.68↑105.261.12↑62.321.00↑51.522.12↑57.04
SSP2-4.52050s4.86↑113.161.50↑117.390.96↑45.452.46↑82.22
2070s2.02↓11.401.28↑85.510.49↓25.761.77↑31.11
2090s2.82↑23.681.40↑102.900.65↓1.522.05↑51.85
Average3.23↑41.671.39↑101.450.70↑6.062.09↑54.81
SSP5-8.52050s4.35↑90.790.82↑18.840.47↓28.791.29↓4.44
2070s4.78↑109.650.89↑28.990.64↓3.031.53↑13.33
2090s3.57↑56.580.88↑27.540.46↓30.301.34↓0.74
Average4.23↑85.530.86↑24.640.52↓21.211.38↑2.22
Total average4.05↑77.631.12↑62.320.74↑12.121.86↑37.78
Tetrastigma planicauleCurrent33.66 -31.05 -5.63 -36.68 -
SSP1-2.62050s31.15 ↓7.4634.82 ↑12.144.88 ↓13.3239.70 ↑8.23
2070s33.62 ↓0.1232.52 ↑4.735.62 ↓0.1838.14 ↑3.98
2090s38.92 ↑15.6329.64 ↓4.545.99 ↑6.3935.63 ↓2.86
Average34.56 ↑2.6732.33 ↑4.125.50 ↓2.3137.82 ↑3.11
SSP2-4.52050s31.35 ↓6.8633.89 ↑9.155.68 ↑0.8939.57 ↑7.88
2070s34.24 ↑1.7231.57 ↑1.675.26 ↓6.5736.83 ↑0.41
2090s38.08 ↑13.1330.63 ↓1.355.44 ↓3.3736.07 ↓1.66
Average34.56 ↑2.6732.03 ↑3.165.46 ↓3.0237.49 ↑2.21
SSP5-8.52050s35.11 ↑4.3131.17 ↑0.396.40 ↑13.6837.57 ↑2.43
2070s35.45 ↑5.3230.96 ↓0.295.84 ↑3.7336.80 ↑0.33
2090s34.02 ↑1.0732.62 ↑5.065.46 ↓3.0238.08 ↑3.82
Average34.86 ↑3.5731.58 ↑1.715.90 ↑4.8037.48 ↑2.18
Total average34.66 ↑2.9731.98 ↑3.005.62 ↓0.1837.60 ↑2.51
Tetrastigma obtectumCurrent76.08 -63.07 -8.63 -71.70 -
SSP1-2.62050s76.13 ↑0.0762.18 ↓1.418.82 ↑2.2071.00 ↓0.98
2070s72.10 ↓5.2366.81 ↑5.939.62 ↑11.4776.43 ↑6.60
2090s81.69 ↑7.3759.77 ↓5.239.35 ↑8.3469.12 ↓3.60
Average76.64 ↑0.7462.92 ↓0.249.26 ↑7.3072.18 ↑0.67
SSP2-4.52050s79.03 ↑3.8860.82 ↓3.579.37 ↑8.5770.19 ↓2.11
2070s75.24 ↓1.1057.67 ↓8.5610.31 ↑19.4767.98 ↓5.19
2090s73.62 ↓3.2357.42 ↓8.968.82 ↑2.2066.24 ↓7.62
Average75.96 ↓0.1658.64 ↓7.029.50 ↑10.0868.14 ↓4.97
SSP5-8.52050s78.61 ↑3.3361.26 ↓2.879.49 ↑9.9770.75 ↓1.32
2070s79.60 ↑4.6358.77 ↓6.829.82 ↑13.7968.59 ↓4.34
2090s76.55 ↑0.6253.99 ↓14.409.95 ↑15.3063.94 ↓10.82
Average78.25 ↑2.8558.01 ↓8.029.75 ↑12.9867.76 ↓5.50
Total average76.95 ↑1.1459.85 ↓5.119.51 ↑10.2069.36 ↓3.26
Tetrastigma obovatumCurrent74.85 -20.82 -9.60 -30.42 -
SSP1-2.62050s62.22 ↓16.8720.97 ↑0.729.69 ↑0.9430.66 ↑0.79
2070s64.85 ↓13.3621.09 ↑1.308.26 ↓13.9629.35 ↓3.52
2090s61.87 ↓17.3426.19 ↑25.799.80 ↑2.0835.99 ↑18.31
Average62.98 ↓15.8622.75 ↑9.279.25 ↓3.6532.00 ↑5.19
SSP2-4.52050s69.71 ↓6.8721.95 ↑5.438.14 ↓15.2130.09 ↓1.08
2070s72.81 ↓2.7321.12 ↑1.449.05 ↓5.7330.17 ↓0.82
2090s59.70 ↓20.2417.56 ↓15.667.70 ↓19.7925.26 ↓16.96
Average67.41 ↓9.9420.21 ↓2.938.30 ↓13.5428.51 ↓6.28
SSP5-8.52050s75.95 ↑1.4723.87 ↑14.6511.55 ↑20.3135.42 ↑16.44
2070s66.76 ↓10.8123.42 ↑12.499.61 ↑0.1033.03 ↑8.58
2090s67.43 ↓9.9118.80 ↓9.707.66 ↓20.2126.46 ↓13.02
Average70.05 ↓6.4122.03 ↑5.819.61 ↑0.1031.64 ↑4.01
Total average66.81 ↓10.7421.66 ↑4.039.05 ↓5.7330.71 ↑0.95
Tetrastigma cruciatumCurrent26.03 -5.43 -1.92 -7.35 -
SSP1-2.62050s26.92 ↑3.425.77 ↑6.262.28 ↑18.758.05 ↑9.52
2070s28.74 ↑10.417.11 ↑30.942.63 ↑36.989.74 ↑32.52
2090s22.00 ↓15.484.71 ↓13.262.20 ↑14.586.91 ↓5.99
Average25.89 ↓0.545.86 ↑7.922.37 ↑23.448.23 ↑11.97
SSP2-4.52050s29.29 ↑12.526.55 ↑20.631.89 ↓1.568.44 ↑14.83
2070s21.33 ↓18.064.90 ↓9.762.32 ↑20.837.22 ↓1.77
2090s24.32 ↓6.575.36 ↓1.291.89 ↓1.567.25 ↓1.36
Average24.98 ↓4.035.60 ↑3.132.03 ↑5.737.64 ↑3.95
SSP5-8.52050s23.43 ↓9.986.52 ↑20.072.00 ↑4.178.52 ↑15.92
2070s26.60 ↑2.195.10 ↓6.082.18 ↑13.547.28 ↓0.95
2090s25.05 ↓3.766.22 ↑14.551.96 ↑2.088.18 ↑11.29
Average25.03 ↓3.845.95 ↑9.582.05 ↑6.777.99 ↑8.71
Total average25.30 ↓2.805.80 ↑6.812.15 ↑11.987.95 ↑8.16
Tetrastigma serrulatumCurrent64.84 -27.70 -10.39 -38.09 -
SSP1-2.62050s62.11 ↓4.2129.50 ↑6.509.68 ↓6.8339.18 ↑2.86
2070s66.67 ↑2.8228.94 ↑4.489.96 ↓4.1438.90 ↑2.13
2090s65.13 ↑0.4526.55 ↓4.159.46 ↓8.9536.01 ↓5.46
Average64.64 ↓0.3128.33 ↑2.279.70 ↓6.6438.03 ↓0.16
SSP2-4.52050s61.37 ↓5.3527.95 ↑0.909.32 ↓10.3037.27 ↓2.15
2070s59.48 ↓8.2726.35 ↓4.878.34 ↓19.7334.69 ↓8.93
2090s68.23 ↑5.2328.27 ↑2.068.52 ↓18.0036.79 ↓3.41
Average63.03 ↓2.7927.52 ↓0.658.73 ↓15.9836.25 ↓4.83
SSP5-8.52050s68.43 ↑5.5425.37 ↓8.418.96 ↓13.7634.33 ↓9.87
2070s65.81 ↑1.5025.23 ↓8.929.23 ↓11.1634.46 ↓9.53
2090s68.67 ↑5.9126.87 ↓3.009.71 ↓6.5436.58 ↓3.96
Average67.64 ↑4.3225.82 ↓6.799.30 ↓10.4935.12 ↓7.80
Total average65.10 ↑0.4027.23 ↓1.709.24 ↓11.0736.47 ↓4.25
Up arrow (↑) means increase compared to the current; down arrow (↓) means decrease. Total average is the average of the suitable habitat areas under nine future climate scenarios. SSP = Shared Socioeconomic Pathways. SSP1-2.6 indicates optimistic pathways; SSP2-4.5 indicates moderate pathways; SSP5-8.5 indicates pessimistic pathways.
Table 5. Niche overlap measured by Schoener’s D and Hellinger’s I of Sapria himalayana and its five hosts under different climate scenarios.
Table 5. Niche overlap measured by Schoener’s D and Hellinger’s I of Sapria himalayana and its five hosts under different climate scenarios.
ScenariosSpecies
Sapria himalayana vs. Tetrastigma planicauleSapria himalayana vs. Tetrastigma obtectumSapria himalayana vs. Tetrastigma obovatumSapria himalayana vs. Tetrastigma cruciatumSapria himalayana vs. Tetrastigma serrulatum
DIDIDIDIDI
Current0.12990.51180.14350.48990.21620.58510.44130.79600.19030.5300
SSP1-2.62050s↑0.2397↑0.5903↑0.2723↑0.5839↑0.3549↑0.6768↑0.5130↑0.8183↑0.3333↑0.6370
2070s↑0.2379↑0.5677↑0.2440↑0.5485↑0.3462↑0.6672↑0.5921↑0.8471↑0.3105↑0.6132
2090s↑0.2034↑0.5441↑0.2214↑0.5312↑0.3069↑0.6258↑0.5806↑0.8507↑0.2817↑0.5992
Average↑0.2270↑0.5674↑0.2459↑0.5545↑0.3360↑0.6566↑0.5619↑0.8387↑0.3085↑0.6165
SSP2-4.52050s↑0.2064↑0.5646↑0.1979↑0.5083↑0.2938↑0.6253↑0.5987↑0.8744↑0.2648↑0.5826
2070s↑0.2307↑0.5878↑0.2593↑0.5909↑0.3519↑0.6905↑0.5434↑0.8528↑0.3194↑0.6449
2090s↑0.1671↑0.5221↑0.1759↑0.5013↑0.2747↑0.6313↑0.5283↑0.8451↑0.2333↑0.5683
Average↑0.2014↑0.5582↑0.2110↑0.5335↑0.3068↑0.6490↑0.5568↑0.8574↑0.2725↑0.5986
SSP5-8.52050s↑0.1785↑0.5711↑0.2046↑0.5065↑0.2998↑0.6469↑0.5708↑0.8545↑0.2403↑0.5650
2070s↑0.1755↑0.5388↓0.1423↓0.4608↑0.2440↓0.5750↑0.5612↑0.8605↑0.2190↑0.5472
2090s↓0.0919↓0.4574↓0.0508↓0.3522↓0.1605↓0.4999↑0.5126↑0.8444↓0.1388↓0.4640
Average↑0.1486↑0.5224↓0.1326↓0.4398↑0.2348↓0.5739↑0.5482↑0.8531↑0.1994↓0.5254
Future average↑0.1923↑0.5493↑0.1965↑0.5093↑0.2925↑0.6265↑0.5556↑0.8498↑0.2601↑0.5802
Total average ± SD↑0.1861 ± 0.0481↑0.5456 ± 0.0406↑0.1912 ± 0.0665↑0.5074 ± 0.0681↑0.2849 ± 0.0632↑0.6224 ± 0.0567↑0.5442 ± 0.0476↑0.8444 ± 0.0221↑0.2531 ± 0.0613↑0.5751 ± 0.0538
Total average ± SD refers to the average value of Schoener’s D and Hellinger’s I of each pair under ten climate scenarios. Up arrow (↑) means increase compared to the current; down arrow (↓) means decrease.
Table 6. Optimization results of the MaxEnt model for Sapria himalayana and its five hosts.
Table 6. Optimization results of the MaxEnt model for Sapria himalayana and its five hosts.
SpeciesFeature ClassRegularization Multiplier
Sapria himalayanaLQHPT1.5
Tetrastigma planicauleLQHPT1.5
Tetrastigma obtectumLQH1.5
Tetrastigma obovatumLQ0.5
Tetrastigma cruciatumLQH2.5
Tetrastigma serrulatumLQHP1.0
Table 7. Habitat suitability categorization based on maxSSS threshold values for Sapria himalayana and its five hosts.
Table 7. Habitat suitability categorization based on maxSSS threshold values for Sapria himalayana and its five hosts.
SpeciesUnsuitable AreaLow
Suitable Area		Moderately Suitable AreaHighly
Suitable Area
Sapria himalayana0–0.120.12–0.410.41–0.710.71–1.00
Tetrastigma planicaule0–0.190.19–0.460.46–0.730.73–1.00
Tetrastigma obtectum0–0.260.26–0.500.50–0.750.75–1.00
Tetrastigma obovatum0–0.150.15–0.430.43–0.720.72–1.00
Tetrastigma cruciatum0–0.050.05–0.370.37–0.680.68–1.00
Tetrastigma serrulatum0–0.170.17–0.450.45–0.720.72–1.00
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hang, W.; Li, Y.; Zhang, G. Assessing Potential Habitat Suitability of the Endangered Endo-Holoparasitic Sapria himalayana and Its Multiple Hosts in China Under Global Warming. Plants 2026, 15, 574. https://doi.org/10.3390/plants15040574

AMA Style

Hang W, Li Y, Zhang G. Assessing Potential Habitat Suitability of the Endangered Endo-Holoparasitic Sapria himalayana and Its Multiple Hosts in China Under Global Warming. Plants. 2026; 15(4):574. https://doi.org/10.3390/plants15040574

Chicago/Turabian Style

Hang, Weiyi, Yan Li, and Guangfu Zhang. 2026. "Assessing Potential Habitat Suitability of the Endangered Endo-Holoparasitic Sapria himalayana and Its Multiple Hosts in China Under Global Warming" Plants 15, no. 4: 574. https://doi.org/10.3390/plants15040574

APA Style

Hang, W., Li, Y., & Zhang, G. (2026). Assessing Potential Habitat Suitability of the Endangered Endo-Holoparasitic Sapria himalayana and Its Multiple Hosts in China Under Global Warming. Plants, 15(4), 574. https://doi.org/10.3390/plants15040574

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop