EN中文

Regional Sustainability >

2026 , Vol. 7 >Issue 1: 100294

DOI: https://doi.org/10.1016/j.regsus.2026.100294

Full Length Article

Vascular plant diversity and distribution pattern in Tajikistan: A global hotspot of diversity

  • ZHOU Yixin a, b, c ,
  • MA Suliya d ,
  • LI Wenjun , a, b, c, * ,
  • Parvina KURBONOVA e ,
  • Mariyo BOBOEV f ,
  • LI Yufan g ,
  • Hikmat HISORIEV h ,
  • MA Keping b ,
  • YANG Weikang a, b, c ,
  • ZHANG Yuanming a, b, c
Expand
  • aChina-Tajikistan Belt and Road Joint Laboratory on Biodiversity Conservation and Sustainable Use, Chinese Academy of Sciences, Urumqi, 830011, China
  • bXinjiang Institute of Ecology and Geography, Chinese Academy of Sciences, Urumqi, 830011, China
  • cCollege of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China
  • dInner Mongolia University of Technology, Hohhot, 010051, China
  • eTajik National University, Rudaki Avenue, Dushanbe, 734025, Tajikistan
  • fKhatlon Scientific Center of the National Academy of Tajikistan, Kulob Botanic Garden, Kuljab, 735360, Tajikistan
  • gNortheast Forestry University, Harbin, 150040, China
  • hInstitute of Botany, Plant Physiology and Genetics, National Academy of Tajikistan, Dushanbe, 734017, Tajikistan
* E-mail address: (LI Wenjun).

The first and second authors contributed equally to this work

Received date: 2025-08-31

  Revised date: 2025-12-08

  Accepted date: 2026-01-05

  Online published: 2026-03-11

Fold

Abstract

Tajikistan represents a core region of the biodiversity hotspot in Central Asian mountains and has exceptional vascular plant diversity. However, the species diversity of the country faces urgent conservation challenges. There has been a lack of a comprehensive and multidimensional assessment to inform strategic conservation planning. Therefore, this study integrated 4 key biodiversity indices including species richness (SR), phylogenetic diversity (PD), threatened species richness (TSR), and endemic species richness (ESR) to map species diversity distribution patterns, identify conservation gaps, and elucidate their effects of climatic factors. This study revealed that species diversity shows a clear trend of decreasing from the western region to the eastern region of Tajikistan. The central-western mountains (specifically the Gissar-Darvasian and Zeravshanian regions) emerge as irreplaceable biodiversity hotspots. However, we found a severe spatial mismatch between these priority areas and the existing protected areas (PAs). Protection coverage for all hotspots was alarmingly low, ranging from 31.00% to 38.00%. Consequently, a critical 64.80% of integrated priority areas fall outside of the current PAs, representing a major conservation gap. This study identified precipitation seasonality and isothermality as the principal drivers, collectively explaining over 50.00% of the diversity variation and suggesting high vulnerability to hydrological shifts. Furthermore, we detected significant geographic sampling bias in the public biodiversity databases, with the most critical hotspot being systematically under-sampled. This study provides a robust scientific basis for conservation action, highlighting the urgent need to strategically expand PAs in the under-protected southwestern region and to mitigate critical sampling gaps through targeted data digitization and field surveys. These measures are indispensable for securing Tajikistan’s unique biodiversity and achieving the Kunming-Montreal Global Biodiversity Framework Target 3 (“30×30 Protection”).

Key words： Vascular plant; Species diversity; Distribution pattern; Conservation gaps; Tajikistan

Cite this article

ZHOU Yixin , MA Suliya , LI Wenjun , Parvina KURBONOVA , Mariyo BOBOEV , LI Yufan , Hikmat HISORIEV , MA Keping , YANG Weikang , ZHANG Yuanming . Vascular plant diversity and distribution pattern in Tajikistan: A global hotspot of diversity[J]. Regional Sustainability, 2026 , 7(1) : 100294 . DOI: 10.1016/j.regsus.2026.100294

1. Introduction

Central Asia is a globally significant biodiversity nexus, distinguished by unique arid zone biota and recognized as a major center of origin and differentiation for many important plant lineages (Li et al., 2020). The mountains of Central Asia harbor approximately 7000 vascular plant species, representing 75.00% of the region’s total vascular flora (Marchese, 2015; Zhang et al., 2020). Tajikistan, as the core country of the biodiversity hotspot, comprises over 90.00% mountainous terrain and supports remarkable plant diversity. The country sustains exceptional plant diversity, hosting between 4300 and 4542 species distributed across more than 800 genera and 110 families, reflecting a floristic composition that combines both temperate and tropical elements (Nowak et al., 2020a; Ma et al., 2024). The complex topography of the Pamir-Alay and Tianshan mountains has further driven high rates of speciation, yielding a remarkable 1486 endemic species (comprising 35.00% of the total flora). The exceptional species richness (SR) and high rate of endemism make Tajikistan as a critical region for global biodiversity conservation.
Tajikistan’s biodiversity faces significant threats from climate change and anthropogenic pressures. As a sentinel region for the impact of climate change, the country experiences glacier retreating and rising temperature, which endangers high-mountain endemic species adapted to stable hydrothermal conditions (Raduła et al., 2021). Concurrently, human pressures including overgrazing, plant collection, and mining activities further exacerbate habitat degradation. Notably, 38.11% of vascular plant species have already been classified as threatened species according to the International Union for Conservation of Nature (IUCN) Red List criteria (Nowak et al., 2020b; Song et al., 2025). Although the government has established a protected area (PA) network covering 22.00% of the national territory, these existing conservation measures show spatial mismatches with biodiversity hotspots. Critical habitats, such as cliff ecosystems and wetland ecosystems of the Pamir Plateau wetlands, remain inadequately protected (Volkova et al., 2024). Furthermore, the existing PA network disproportionately emphasizes high-mountain regions, resulting in insufficient coverage from a comprehensive biodiversity perspective (Squires et al., 2018).
Despite these conservation challenges, significant knowledge gaps hinder effective, evidence-based conservation planning. Foundational studies have outlined endemic plant distributions (Nowak and Nobis, 2010), and recent sophisticated analyses have successfully identified hotspots for threatened species (Nowak et al., 2020b). However, a comprehensive, multi-dimensional assessment integrating key biodiversity facets remains lacking for Tajikistan. Specifically, the spatial pattern of phylogenetic diversity (PD) remains entirely unexplored. A systematic, quantitative analysis of the overall SR is absent, and the existing knowledge of endemism hotspots is outdated. Critically, no previous study has simultaneously analyzed these dimensions including SR, PD, endemic species richness (ESR), and threatened species richness (TSR) within a single, unified framework to identify regions of spatial congruence and to prioritize holistic conservation planning. Furthermore, the Wallacean shortfall, which focuses on species distribution information, is particularly pronounced in biodiversity hotspots like Tajikistan, where only 5.30% of threatened plant species have been adequately documented (Li et al., 2025). This information gap severely constrains the accuracy of spatial conservation planning (Ondo et al., 2024). Species distribution models based on sparse field surveys exist spatial extrapolation errors, a problem exacerbated in topographically complex mountainous systems. Furthermore, a pervasive sampling bias in databases like the Global Biodiversity Information Facility (GBIF) can systematically distort perceived biodiversity and subsequent conservation assessments.
Based on the above analyses, we systematically assessed Tajikistan’s vascular plant diversity patterns and conservation gaps by integrating multiple biodiversity dimensions. This study aims to achieve the following objectives: (1) compiling a comprehensive vascular plant checklist for Tajikistan; (2) modeling and mapping the spatial patterns of SR, PD, ESR, and TSR at both phytogeographic region and fine-scale grid levels; and (3) identifying conservation gaps by evaluating spatial congruence between biodiversity hotspots and existing PAs. Additionally, we explicitly quantified the impact of sampling bias to validate the robustness of our conservation priorities.

2. Materials and methods

2.1. Study area and species data compilation

This study was conducted across the entire territory of Tajikistan. To analyze plant diversity patterns, we adopted a phytogeographic framework based on 26 distinct regions (Fig. 1). All geographic information system (GIS) operations, including the digitization of the phytogeographic map and subsequent all spatial analysis, were performed using the Quantum GIS (QGIS) 3.40 software.
Fig. 1. Spatial distribution of phytogeographic regions (a) and quality-controlled vascular plant occurrence records (b). 1, Kuraminian; 2, Mogoltausian; 3, Prisyrdarian; 4A, Turkestanian A; 4B, Turkestanian B; 5A, Zeravshanian A; 5B, Zeravshanian B; 5C, Zeravshanian C; 6A, Gissar-Darvasian A; 6B, Gissar-Darvasian B; 6C, Gissar-Darvasian C; 6D, Gissar-Darvasian D; 6E, Gissar-Darvasian E; 6F, Gissar-Darvasian F; 7A, South Tadzhikistanian A; 7B, South Tadzhikistanian B; 7C, South Tadzhikistanian C; 7D, South Tadzhikistanian D; 8A, East Tadzhikistanian A; 8B, East Tadzhikistanian B; 8C, East Tadzhikistanian C; 9A, West Pamirian A; 9B, West Pamirian B; 9C, West Pamirian C; 10, East Pamirian; 11, Alajian. These regions were defined following the classification system established by Goncharov in the Flora Tadzhikskoi SSR (Ovchinnikov, 1957-1991). GBIF, Global Biodiversity Information Facility; IUCN, International Union for Conservation of Nature. Abbreviations are defined in the figure and are consistent in subsequent figures. Note that the figure is based on the standard map (GS(2025)1508) of the Map World (https://map.tianditu.gov.cn) marked by the National Platform for Common GeoSpatial Information Services, and the boundary of the standard map has not been modified.

2.1.1. Regional species checklist

A comprehensive vascular plant checklist for each of 26 phytogeographic regions was compiled by systematic integration of multiple data sources. The foundational taxonomy was derived from the Flora Tadzhikskoi SSR (Ovchinnikov, 1957-1991), which was updated and cross-validated with recent, comprehensive datasets (Nowak et al., 2020b). Species conservation status was identified by cross-referencing the Red Book of the Republic of Tajikistan (National Academy of Sciences of Tajikistan, 2024) and IUCN Red List criteria (v.2025-1). This study determined endemic status based on established criteria following previous research (Ma et al., 2024). All scientific names were standardized using the World Flora Online (Borsch et al., 2020) and World Checklist of Vascular Plants (WCVP) (Govaerts et al., 2021) as primary taxonomic references to ensure global nomenclatural consistency. The method was automated and verified using the ‘Taxonomic Name Resolution Service’ (Boyle et al., 2013) package and ‘rWCVP’ package (Brown et al., 2023). We performed the final taxonomic validation following the Plants of the World Online (https://powo.science.kew.org) and World Flora Online (https://www.worldfloraonline.org).

2.1.2. Georeferenced occurrence data

This study compiled georeferenced occurrence records for the species on our checklist to analyze fine-scale spatial patterns. A total of 49,048 records for 3186 species were retrieved from the GBIF downloaded on 14 August 2025 and the IUCN. This raw dataset underwent a thorough quality control and cleaning pipeline. Records lacking geographic coordinates and duplicate records for the same species at identical coordinates were removed. A final spatial filter was applied to retain only records within the national boundary of Tajikistan. This process resulted a final, high-quality dataset of 11,937 occurrence records representing 2330 species (Fig. 1b). This dataset included 769 occurrence records corresponding to 223 endemic species, and 484 occurrence records representing 125 threatened species.

2.2. Diversity indices and spatial analysis

This study quantified 4 biodiversity dimensions at 2 spatial scales based on the verified species checklist. At the phytogeographic region scale, SR, PD, TSR, and ESR were calculated for each of 26 phytogeographic regions using the comprehensive checklist. SR was calculated using the total number of vascular plant species. PD was calculated using the ‘V.PhyloMaker2’ package (Jin and Qian, 2022) in Scenario 3 method, which constructs a phylogenetic tree based on the Plant List (TPL). TSR was the number of threatened species listed in either the IUCN Red List or the Red Book of the Republic of Tajikistan. ESR was the number of endemic species in Tajikistan.
At the grid scale, using the cleaned georeferenced occurrence data, this study calculated 4 diversity indices (SR, PD, TSR, and ESR) for each cell with a 25 km×25 km grid system covering Tajikistan. Hotspot grid cells were identified as those falling within the top 30.00% of values for each index. For an integrated assessment, Z-score standardized diversity indices were averaged to generate a composite priority score, with the top 30.00% of cells classified as priority zones (Tucker et al., 2017). Conservation gap analysis was performed by overlaying all these hotspots with PA boundaries from the World Database on Protected Areas (WDPA), classifying grid cells with spatial intersection as PA. The analysis was done using the ‘terra’ (Hijmans, 2025) and ‘sf’ packages (Pebesma and Bivand, 2023) in R (version 4.4.3).

2.3. Sampling bias assessment

The cleaned occurrence points were spatially joined to the 26 phytogeographic regions, with record species per region quantified. Sampling bias was quantified through negative binomial regression (Venables and Ripley, 2002), with the number of the GBIF and IUCN species occurrence records as the response variable, using the Pearson residual as the quantitative measure of sampling deviation.

2.4. Environmental data and statistical analysis

Climate data were obtained from the WorldClim 2.1 database (https://worldclim.org) at 30′′, encompassing 19 bioclimatic variables, which were extracted at the centroids of each phytogeographic region using ‘sf’ package in R (version 4.4.3). This study screened all variables, retaining only those with a variance inflation factor (VIF)<10 to mitigate multicollinearity (Naimi et al., 2014). This process yielded 6 key bioclimatic variables including isothermality, temperature seasonality, mean temperature of the wettest quarter, precipitation seasonality, precipitation of the warmest quarter, and precipitation of the coldest quarter.
Multivariate relationships were analyzed through the redundancy analysis (RDA) (Oksanen et al., 2025). Prior to analysis, both response variables (SR, PD, TSR, and ESR) and predictor variables (selected climatic variables) were standardized using the decostand function to eliminate scale effects. The RDA model was constructed with the 4 diversity indices as the response matrix and 6 selected climatic factors as explanatory variables. Model significance was tested with 999 permutations, and the proportion of variance explained was quantified by the adjusted coefficient of determination (R2). Spatial dependence was evaluated using the Mantel tests (Pearson correlation). Biodiversity distance, environmental distance, and geographic distance for each phytogeographic region were compared. The Mantel test was used to assess the correlation between biodiversity distance and environmental distance, while the partial Mantel test was used to evaluate this correlation while controlling for the effects of geographic distance.
Multiple linear regression (MLR) models combined with model selection strategies identified climatic factors for each diversity index using the dredge function from the ‘MuMIn’ package in R version 4.4.3 (Bartoń, 2025). Model selection was based on the corrected Akaike information criterion (AICc), with models having a delta corrected Akaike information criterion (ΔAICc)<2 considered as the candidate model set. The model with the lowest AICc value was designated as the best model for each diversity index. Hierarchical partitioning analysis using the ‘rdacca.hp’ package in R version 4.4.3 (Lai, 2022) was conducted to quantify the independent contribution of environmental distance, decomposing the total explained variance (adjusted R2) into independent and joint effects of each explanatory variable. This study performed model diagnostics to validate model assumptions and assessed variable importance based on both standardized regression coefficients and hierarchical partitioning results.
All visualizations were performed using the ‘ggplot2’ (Wickham, 2016) and ‘tidyterra’ packages (Hernangómez, 2023). Phylogenetic tree visualization was conducted using the Interactive Tree of Life (ITOL; Letunic and Bork, 2024).

3. Results

3.1. Taxonomic composition of vascular plants

The vascular plants of Tajikistan comprised 4320 taxa, among which there are 204 infraspecific taxa (subspecies and varieties), 118 families, and 867 genera. The top 10 families with the highest richness were Asteraceae, Fabaceae, Poaceae, Brassicaceae, Lamiaceae, Apiaceae, Amaranthaceae, Caryophyllaceae, Rosaceae, and Boraginaceae (Fig. 2). At the genus level, the top 20 genera with the most SR were Astragalus, Cousinia, Allium, Taraxacum, Oxytropis, Silene, Artemisia, Carex, Ranunculus, Stipa, Ferula, Potentilla, Scutellaria, Nepeta, Euphorbia, Gagea, Draba, Jurinea, Cuscuta, and Eremurus (Fig. 3a).
Fig. 2. Genus-level phylogenetic tree of vascular plants in Tajikistan. Genera highlighted in yellow color represent the top 20 genera with the most species richness (SR). Blue circles indicate genera containing endemic species, while brown-red stars denote genera containing threatened species. The colored sectors in the inner ring represent the top 10 families with the greatest SR. The outermost ring displays the logarithmic (log10) transformed the number of species within each genus.
Fig. 3. Top 20 genera (a), top 20 endemic genera (b), and top 20 threatened genera (c) with the most number of species in Tajikistan. Each bar is color-coded by family, demonstrating the taxonomic distribution of diversity across different plant lineages. Genera are ranked in descending order of species number within each category.
A total of 505 endemic taxa including 6 infraspecific taxa were identified, representing 11.70% of the total vascular flora. These endemic taxa were distributed across 47 families and 162 genera. The top 5 families with the number of endemic species were Asteraceae, Fabaceae, Lamiaceae, Brassicaceae, and Apiaceae (Fig. 3b), collectively accounting for 59.60% of all endemic species. At the genus level, the top 5 endemic genera with the number of species were Astragalus, Cousinia, Taraxacum, Oxytropis, and Allium, comprising 28.80% of the total endemic species (Fig. 3b).
The threatened vascular flora of Tajikistan consisted of 287 taxa (including 9 infraspecific taxa) belonging to 54 families and 141 genera, representing 6.60% of the total vascular flora. The top 5 families containing the highest number of threatened species were Fabaceae, Amaryllidaceae, Lamiaceae, Apiaceae, and Liliaceae, which collectively contain 43.40% of all threatened species (Fig. 3c). The top 5 genera with the highest number of threatened species were Allium, Tulipa, Astragalus, Ferula, and Eremurus. According to the 2024 edition of the Red Book of the Republic of Tajikistan, 283 taxa were officially recognized as threatened taxa, including 2 species classified as extinct (EX; Ranunculus chodzhamastonicus and Crataegus darvasica), 1 species as extinct in the wild (EW; Populus cataracti), 83 species as critically endangered (CR), 126 species as endangered (EN), and 75 species as vulnerable (VU). The IUCN Red List criteria assessed 21 taxa, classifying 8 species as CR, 4 species as EN, and 9 species as VU.

3.2. Spatial distribution patterns and conservation gap analysis

3.2.1. Regional-scale diversity patterns

At the phytogeographic region scale, our analysis of 4072 species (including 439 endemic species and 256 threatened species) revealed distinct spatial patterns across the 26 phytogeographic regions of Tajikistan. The distribution patterns of SR, PD, and TSR were remarkably congruent, showing a clear decreasing trend from southwestern region to northeast region of Tajikistan (Fig. 4a-c). Diversity hotspots were predominantly concentrated in the central-western mountains and southwestern lowlands, particularly in the Gissar-Darvasian, Zeravshanian, and South Tajikistan. In contrast, the northern basins and Pamir Plateau, including Kuraminian, Mogoltausian, Prisyrdarian, and East Pamirian, demonstrated significantly lower diversity index values. While ESR generally followed a similar trend from the southwestern region to the northeastern region of Tajikistan, it displayed more concentrated hotspots in the Gissar-Darvasian B, South Tajikistan D, and West Pamirian B (Fig. 4d).
Fig. 4. Geographic distribution patterns of SR (a), PD (b), TSR (c), and ESR (d) in Tajikistan. SR, species richness; PD, phylogenetic diversity; TSR, threatened species richness; ESR, endemic species richness.

3.2.2. Fine-scale grid-based diversity patterns

Species occurrence data within a 25 km×25 km grid system was analyzed to achieve a higher spatial resolution. The spatial analysis revealed significant congruence between SR and PD hotspots (Fig. 5a and b). Grid cells with the highest diversity values (SR>200 and PD>5000) were predominantly distributed across the central-western region of Tajikistan, with 6-8 grid cells of the maximum diversity index value concentrating in Gissar-Darvasian and Zeravshanian. The spatial overlap between SR and PD hotspots demonstrated substantial co-occurrence, with the highest-value grid cells exhibiting nearly identical spatial positioning.
Fig. 5. Geographic distribution patterns of SR hotspot (a), PD hotspot (b), TSR hotspot (c), ESR hotspot (d), and composite conservation priority index (CCPI) hotspot (e) in Tajikistan. PA, protected area.
In contrast, TSR hotspot exhibited markedly different spatial patterns characterized by restricted distribution, which was consistent with the spatial distribution patterns of SR and PD hotspots (Fig. 5c). Grid cells with high concentrations of threatened species (TSR>15) were not located in regions with the maximum SR. Instead, these high-value cells formed a spatially aggregated cluster primarily confined to the southwestern region of South Tajikistan, with one additional high-value grid cell occurring in Mogoltausian. TSR hotspot complex comprised 3-4 core grid cells (TSR>15), which were surrounded by a peripheral zone of moderately high-value cells (TSR>10).
ESR hotspot displayed a moderate spatial pattern and partially bridged the distributions observed for other diversity indices (Fig. 5d). While ESR hotspot (ESR>30) was principally located in the central-western regions of Tajikistan, demonstrating partial spatial overlap with SR and PD hotspots, their core distributions were displaced southwards, suggesting a transitional spatial gradient towards TSR hotspot location. The most significant endemic species concentration, represented by a single grid cell with ESR>40, was prominently positioned in the center of Gissar-Darvasian.

3.2.3. Integrated conservation priority assessment

This study established a composite conservation priority index (CCPI) by averaging the Z-score-standardized values of all 4 diversity indices (SR, PD, TSR, and ESR) to comprehensively evaluate conservation priorities across multiple biodiversity dimensions. The resulting integrated priority map clearly identified the southwestern region of Tajikistan as the highest conservation value (Fig. 5e). These regions demonstrated exceptional importance consistently across multiple diversity dimensions, highlighting their irreplaceable role in maintaining the country’s vascular plant diversity.

3.2.4. Conservation gap analysis

A quantitative assessment of the spatial relationship between the identified biodiversity hotspots (the top 30.00% of grid cells for each index) and the existing PA network revealed substantial conservation gaps (Table 1). The current PA network provided inadequate coverage of key vascular plant diversity regions, with protection rates remaining below 40.00% for all diversity indices (Fig. 5e). PD hotspots received the lowest protection coverage, with only 31.30% of these evolutionarily unique regions falling within existing PAs, leaving 68.70% of critical PD hotspots unprotected. TSR hotspot showed the highest relative protection rate at 38.10%, yet 61.90% of TSR hotspot still remained outside the current PA network. Most critically, 64.80% of the highest-ranked grid cells remained formally unprotected, highlighting a severe mismatch between conservation priority and the established PA network.
Table 1 Conservation gap analysis for biodiversity hotspots of vascular plants in Tajikistan.
Diversity index Total hotspot grids Protected hotspot grids Unprotected hotspot grids Percentage of protection area (%) Percentage of conservation gap area (%)
SR 72 26 46 36.10 63.90
ESR 53 20 33 37.70 62.30
TSR 42 16 26 38.10 61.90
PD 67 21 46 31.30 68.70
CCPI 71 25 46 35.20 64.80

Note: SR, species richness; ESR, endemic species richness; TSR, threatened species richness; PD, phylogenetic diversity; CCPI, composite conservation priority index. Hotspot grids represent the top 30.00% of grid cells for each diversity index. Percentage of protection area and conservation gap area were calculated by overlaying hotspot grids with protected area (PA) boundary from the World Database on Protected Areas (WDPA).

3.3. Sampling bias assessment

A comprehensive assessment of sampling bias, extent, and spatial structure was conducted to evaluate the suitability of the GBIF and IUCN species occurrence records for conservation gap analysis. Using phytogeographic regions as the basic analytical unit, this study compared the observed SR from the GBIF and IUCN species occurrence records with SR from the checklist, which is used as the baseline reference (Fig. 6). A negative binomial regression model provided the best fit to the data, revealing a significant positive correlation between SR from the GBIF and IUCN species occurrence records and SR from the checklist (P≤0.001; Fig. 6b). This result indicated that, despite potential biases, publicly available species occurrence data reflected actual biodiversity patterns at macroecological scales. However, the model’s pseudo R2 value of 0.38 suggested that SR from the checklist explains only 38.00% of the variation in SR from the GBIF and IUCN species occurrence records, with the remaining variation likely attributable to spatial sampling heterogeneity.
Fig. 6. Spatial distribution of sampling deviation of GBIF and IUCN species occurrence records (a) and fitted negative binomial regression curve between SR from the GBIF and IUCN species occurrence records and SR from the checklist (b) for vascular plants in Tajikistan. In Figure 6a, red region indicates under-sampled region (negative residual) where SR from the GBIF and IUCN species the occurrence record is significantly lower than SR from the checklist, while blue zone shows relatively well-sampled region (positive residual).
Visualization of the spatial distribution of the Pearson residuals from the regression model revealed obvious spatial characteristics of the sampling bias across Tajikistan (Fig. 6a). The western and southwestern regions of Tajikistan, particularly the biodiversity hotspot regions of Gissar-Darvasian, South Tajikistan, and Zeravshanian, exhibited systematic under-sampling (negative residuals). This pattern indicated that the number of digital collection records in these known species-rich zones is significantly lower than the expected records based on their documented floristic richness. Conversely, East Pamirian region and parts of West Pamirian regions showed over-sampling patterns (positive residuals), where the number of the GBIF and IUCN species occurrence records substantially exceeded expectations based on the checklist.
These findings confirmed the presence of significant and spatially structured sampling deviation in GBIF occurrence records for vascular plants in Tajikistan. Such spatial bias may introduce a significant risk to fine-scale, grid-based environmental interpretation analyses, as the results of these analyses may reflect sampling accessibility rather than the genuine ecological niche requirements. The observed bias pattern, particularly the under-sampling of biodiversity hotspots, could lead to the misidentification of conservation priorities and the inaccurate assessment of the effectiveness of PAs. Consequently, this study adopted a more robust and biologically meaningful phytogeographic regionalization for subsequent analyses.

3.4. Climatic factors of biodiversity indices

3.4.1. Climatic variable selection

A combined analytical framework incorporating the RDA and MLR models was implemented to investigate the effect of climate change on vascular plant diversity patterns in Tajikistan (Fig. 7). Following a VIF analysis of 19 WorldClim bioclimatic variables, this study selected 6 key climatic factors including isothermality, temperature seasonality, mean temperature of wettest quarter, precipitation seasonality, precipitation of the warmest quarter, and precipitation of the coldest quarter.
Fig. 7. Relationships among diversity index, climatic factor, and phytogeographic region for vascular plants in Tajikistan. (a), relationship among diversity index, climatic factor, and phytogeographic regions based on redundancy analysis (RDA); (b), best-fit multiple linear regression (MLR) model of diversity indices and climatic factors; (c), contribution of climatic factors to each diversity index. bio3, isothermality; bio4, temperature seasonality; bio8, mean temperature of the wettest quarter; bio15, precipitation seasonality; bio18, precipitation of the warmest quarter; bio19, precipitation of the coldest quarter. In Figure 7a, arrow length and direction indicate the strength and direction of correlations with ordination axes, respectively. Percentage in parentheses represents the proportion of the total variation explained by each axis. In Figure 7b, point represents coefficient values. Error bar indicates ±1 standard error. The vertical dashed line denotes the zero-reference line. *, P≤0.050; ***, P≤0.001. Bar length represents the percentage of independently explained variation, with different colors corresponding to different climatic factors.

3.4.2. Multivariate relationships between climatic factors and diversity indices

The RDA revealed that the 6 selected climatic factors collectively explained 63.40% of the total variation in plant diversity patterns, with an adjusted R2 of 51.80%. The constrained model demonstrated high statistical significance (permutation test: F=5.48 and P=0.001), confirming that climate change is the primary factor shaping the spatial distribution of plant diversity across Tajikistan.
Figure 7a illustrates the relationships among diversity indices, climatic factors, and phytogeographic regions. The first 2 constrained axes (RDA1 and RDA2) together explained 60.70% of the total variation. RDA1 emerged as the dominant gradient, explaining 54.90% of the variation and primarily representing a precipitation seasonality. It showed strong positive correlations between 4 diversity indices and precipitation seasonality, while exhibiting negative correlations between 4 diversity indices and isothermality. This pattern indicated that regions with the top 2 high diversity, such as Gissar-Darvasian and South Tajikistan, are characterized by climate with high precipitation seasonality but low isothermality (i.e., pronounced intra-annual temperature variation). Although RDA2 only explained 5.80% of the variation, it effectively separated ESR from other diversity indices. This axis was strongly associated with the precipitation of the warmest and coldest quarters, suggesting that the distributions of endemic species are particularly influenced by extreme seasonal moisture conditions.

3.4.3. Independent contributions of climatic factors

The MLR models were developed to examine the specific effects of individual climatic factors on each diversity index. Hierarchical partitioning analysis was then applied to quantify the independent explanatory contribution of each climatic factor. The results consistently identified precipitation seasonality as the primary driver of SR, PD, and TSR (Fig. 7c). This single variable accounted for 52.70%, 57.40%, and 59.20% of the explained variance in SR, PD, and TSR, respectively. Model coefficients revealed a strong positive effect of precipitation seasonality on these 3 indices (P≤0.001). Isothermality was the second most important climatic factor, exhibiting a significant negative effect (P≤0.001). These findings confirmed that plant diversity in terms of taxonomic, phylogenetic, and threatened species in Tajikistan is concentrated in regions with highly seasonal precipitation patterns and pronounced intra-annual temperature variation.
The driving mechanisms for ESR were more complex. While precipitation seasonality remained the strongest positive driver (contributing 39.40% of the explained variance), the mean temperature of the wettest quarter was identified as a significant negative factor in the optimal model (Fig. 7b and c). This suggested that endemic species prefer environments with high precipitation seasonality and tend to occur in regions where the wettest season (typically the growing season) has relatively low temperature, potentially reflecting adaptations to specific mountainous microclimates. Furthermore, precipitation of the coldest quarter exhibited substantial independent explanatory power for ESR (contributing to 26.30%), reinforcing the particular sensitivity of endemic species distributions to cold-season moisture availability and constraints.

4. Discussion

4.1. Phylogenetic clustering of endemic and threatened species

This study revealed significant phylogenetic clustering of endemic species and threatened species in Tajikistan (Figs. 2 and 3), with critical implications for conservation prioritization. Notably, families including Fabaceae, Lamiaceae, and Amaryllidaceae emerged as pivotal, containing the highest number of both endemic and threatened species. At the genus level, the top 4 genera with high endemic and threatened species are Astragalus, Allium, Tulipa, and Ferula. This phylogenetic concentration suggested that certain evolutionary lineages possess intrinsic biological traits that predispose them to elevated rates of endemic species and increased vulnerability to anthropogenic threats and climate change (Molina-Venegas et al., 2020).
The prominence of Fabaceae, particularly Astragalus in both endemic and threatened species likely reflects its evolutionary history and ecological adaptation to Central Asia’s complex, mountainous terrain. These lineages have undergone extensive adaptive radiation in response to the region’s heterogeneous topography and climatic gradients, fostering high rates of allopatric speciation and localization (Qian et al., 2024). However, the same evolutionary adaptations that led to their success, including specialized habitat requirements, narrow ecological tolerances, and limited dispersal capabilities, also make them particularly vulnerable to contemporary anthropogenic disturbances and the impact of climate change. The concentration of threatened species in bulbous genera such as Allium and Tulipa further supports the association between evolutionary specialization and heightened vulnerability to environmental change, as these geophytic lineages typically exhibit highly specialized habitat requirements and life history strategies that increase their susceptibility to environmental change (Wilson et al., 2021).
This phylogenetic clustering has profound implications for conservation planning in Tajikistan. Firstly, it suggests that conservation efforts focused on these key evolutionary lineages could protect both endemic and threatened species simultaneously, thereby maximizing conservation efficiency and resource allocation (Fattorini, 2017). Secondly, the shared vulnerabilities between closely related species indicate that threats may have cascading effects across entire evolutionary lineages, potentially leading to the disproportionate loss of PD if not proactively addressed (Leão et al., 2020). This phylogenetic clustering of endemic and threatened species underscores the importance of incorporating phylogenetic information into conservation planning, as protecting phylogenetically clustered species may require different strategies compared to protecting more randomly distributed biodiversity components.

4.2. Multidimensional diversity pattern and biogeographical significance

This study provided the comprehensive assessment of multidimensional vascular plant diversity patterns in Tajikistan, identifying conservation gaps and their climatic factors. Tajikistan lies in the arid southern zone and the Eurasian continental desert high-mountain system, exhibiting remarkable vascular plant diversity patterns. All measured diversity indices including SR, PD, TSR, and ESR, showed a consistent southwest-northeast decreasing gradient and heterogeneous distribution across the country’s complex mountainous terrain (Figs. 4 and 5), where 93.00% of the territory consists of mountains and plateaus, with approximately half exceeding 3000 m in altitude (Nowak et al., 2020b).
The central-western mountainous regions emerge as the core zones for Tajikistan’s vascular plant diversity, particularly for threatened and endemic species. The Gissar-Darvasian region, which spatially corresponds to the prominent Gissar, Karateginian, Vakhsh, and Darvaz mountainous ranges, emerges as a primary biodiversity center. This region’s extremely complex topography supports a mosaic of habitats, from rich broadleaf forests to unique juniper woodlands, and is recognized as an origin center for wild relatives of temperate fruit trees (Spengler, 2019; Kass et al., 2022). The Zeravshanian region corresponds directly to the great mountainous chains of the Zeravshan and Turkestan ranges, which include the famous Fann Mountains, exhibiting comparably high diversity levels. These 2 regions collectively form a “hotspot within a hotspot”, representing the most critical zone for conservation.
Surrounding this core, other regions display transitional or distinct characteristics. Turkestanian and Alajian regions, located in the northern region of Tajikistan, corresponding to the Turkestan and Alai ranges, form a drier transitional zone. The South Tajikistan encompasses the lower-elevation foothills and plains (e.g., Babatag and Aktau mountains) and is characterized by a distinct subtropical desert and semi-desert flora. While lower in the overall richness compared to the core mountains, this region is a key center for threatened and endemic species adapted to arid conditions (Nowak et al., 2020b).
The eastern half of the country is dominated by the Pamirs Mountains, which is internally heterogeneous. The West Pamirian region, corresponding to a dense network of high, rugged ranges such as the Yazgulem, Rushan, and Shugnan mountains, harbors a unique and highly specialized flora. The Pamir Plateau, with an average elevation exceeding 3800 m and harsh climate, acts as a powerful environmental filter, resulting in a low SR but high endemism.
The distribution of threatened and endemic species is strongly linked to both natural landscapes and human pressures. The complex topography of the central-western ranges, with its deep valleys and high ridges, has historically impeded gene flow, fostering high rates of endemism (Rahmonov et al., 2013). However, these same regions, along with the southwestern lowlands, have more intensive human activities. Endemic species in Tajikistan face primary threats from intensive grazing, soil erosion, and desertification (Nowak and Nobis, 2010). Long-term overgrazing and agricultural practices have severely damaged species habitats, causing significant changes in vegetation composition (Akhmadov et al., 2006). Anthropogenic factors such as illegal logging and overgrazing are causing irreversible damage to Tajikistan’s biodiversity (Squires and Safarov, 2013). Meanwhile, the concentration of endemic species in the southwestern region of Tajikistan does not diminish the importance of northern zones, which may harbor regionally endemic species shared among two or more Central Asian countries, suggesting the need for a comprehensive regional perspective when developing conservation strategies specifically for endemic species (Ma et al., 2024).

4.3. Climatic factors and ecological mechanisms

Through the RDA and MLR methods, our study clearly identified precipitation seasonality and isothermality as the key climatic factors shaping vascular plant diversity patterns in Tajikistan (Fig. 7). There are significant interactions exist among precipitation seasonality, isothermality, and mean temperature of the wettest quarter. These interactions reflect the continental climate and topographical characteristics of Tajikistan. Precipitation is primarily transported by westerly air masses in winter and spring, creating a climate with prominent seasonal precipitation variation. The continental climate with hot summers and cold winters results in lower isotherm smoothness. Simultaneously, the wettest season (winter and spring) coincides with the coldest period, leading to lower average temperatures during this season. Mantel tests revealed no significant correlation between the overall environmental distance and biodiversity distance (Mantel: r=0.119, P=0.111; and Partial Mantel: r=0.084, P=0.156), further confirming that the diversity patterns are driven by specific ecological gradients formed by a few key factors rather than general environmental differences. Climatic factors serve as the core drivers of SR and PD, with SR primarily determined by the coupling status of water and energy available for biological utilization (Hawkins et al., 2003).
In arid and semi-arid regions, where water resources are the primary limiting factor, water availability in specific seasons is far more important than the total annual precipitation (Yao et al., 2021; Kass et al., 2022). Our research demonstrated that precipitation seasonality and isothermality are the most critical factors determining the spatial distribution of SR, PD, and TSR in Tajikistan. The significant positive effect of precipitation seasonality indicated that regions with distinct wet and dry seasons harbor higher plant diversity, which aligns perfectly with the climatic characteristics of the southwestern region of Tajikistan. This environmental filtering has promoted the flourishing of numerous ephemerals, perennial herbs, and shrubs that are adapted to winter-wet and summer-dry environments, constituting the country’s richest species. Isothermality, representing temperature fluctuation, was identified as an important environmental filter, with intense annual temperature variations potentially constituting physiological stress that filters out intolerant species. In the western Himalaya Mountains of Indian, analysis has found that SR peaks when isothermality ranges between 10°C and 12°C, directly demonstrating the importance of temperature stability (Thakur et al., 2022). It also shows close relationships with mean temperature of the wettest quarter and precipitation of the coldest quarter. This reveals the dependence of endemic species survival and differentiation on winter water supplementation. Winter precipitation (typically snowfall) serves as the primary source of spring meltwater in Tajikistan’s mountainous regions, which is crucial for many endemic species (such as various Allium and Tulipa species), as these species will germinate, grow, and flower in early spring. Adequate winter precipitation ensures these species completing critical life cycle stages.
The Pamir-Alai region is predominantly influenced by the westerlies, resulting in precipitation concentrated in winter and spring, revealing precipitation seasonality as the primary driver of biodiversity patterns (Knoche et al., 2017). As the result, the highest diversity occurred in the central-western mountains (Gissar-Darvasian and Zeravshanian regions). In contrast, the East Pamir Plateau region, characterized by a relatively flat and open terrain with low microhabitat differentiation, exhibits extreme aridity and cold conditions, higher isothermality, and lower biodiversity. As a high-altitude frigid desert ecosystem, SR on the Pamir Plateau is generally lower than that on the Qinghai-Xizang Plateau at equivalent elevations, which is largely dominated by the South Asian monsoon. Consequently, plant diversity across the Qinghai-Xizang Plateau is generally driven by a coupling of water and energy (Liu et al., 2021). Moving westward, the Iranian Plateau presents another contrasting system. Its biodiversity is shaped by continental climate and Mediterranean precipitation regime, with wet-cold winters and warm-dry summers (Noroozi et al., 2023). The Caucasus Mountains, while also influenced by westerlies, exhibit a more complex precipitation pattern due to their proximity to the Black Sea and the Caspian Sea, resulting in a less pronounced precipitation seasonality compared to Tajikistan (Noroozi et al., 2021).
The formation of contemporary species distribution patterns results from interactions between modern climatic conditions and long-term geological history (Sun et al., 2020). For mountainous ecosystems, the repeated advances and retreats of Quaternary ice sheets represent a particularly powerful shaping force. During the ice age, as ice sheet significantly expanded, mountains at lower latitudes or higher elevations became “refugia” where species could survive (Hewitt, 2000). During interglacial periods, when climate warmed, species dispersed from these “refugia” to recolonize broader regions. This process not only preserves ancient genetic reservoirs, but also promotes new differentiation events, profoundly influencing current floristic composition and diversity patterns, especially in regions like the Himalaya-Hengduan Mountains and Pamir-Alay Mountains. These regions are considered biodiversity hotspots and evolutionary hubs due to their complex topography and geographic positions connecting different biogeographic regions. It is precisely that these historical biogeographical processes, alongside contemporary climate, likely contribute to the unexplained variance in our models.
Our climatic models explained approximately 60.00% of the variation in diversity patterns, leaving about 40.00% unexplained (Fig. 7a). This unexplained variance likely reflects the influence of additional factors not captured by our models. Beyond factors mentioned above, topographic heterogeneity is a prime candidate. At finer scales, factors such as slope, aspect, and terrain ruggedness create microclimates and habitat heterogeneity that coarse-resolution climatic factors cannot capture (Coelho et al., 2023). Soil properties, including type, pH, and nutrient availability, are also direct determinants of plant community assembly (Ohdo and Takahashi, 2020). Finally, anthropogenic pressures, including grazing intensity and land use change, can directly alter species’ richness and composition, creating patterns that purely deviate from climatic expectations (Kusumoto et al., 2023). Integrating these multifaceted variables into future assessments will be crucial for developing a more holistic and precise understanding of the complex mechanisms driving Tajikistan’s exceptional biodiversity.

4.4. Conservation gaps and data challenges

Tajikistan’s PAs cover 22.00% of territory but protect only 38.10% of threatened species hotspots and 35.20% of composite biodiversity hotspots (Table 1), failing to meet the Target 7 of the Global Strategy for Plant Conservation of protecting 75.00% threatened species (Wyse Jackson and Kennedy, 2009) and the 2030 target of protecting 30.00% of land area. GIS overlay analysis revealed unreasonable distribution of plant PAs in Tajikistan (Fig. 5), which is consistent with the finding from Nowak et al. (2020a). Although the Gissar-Darvasian and Zeravshanian regions have relatively scarce data collection, they are hotspots of biodiversity and unprotected areas that require more attention. Threatened species are mainly distributed in relatively arid regions such as grasslands, alpine zones, semi-deserts, and deserts. Generally, plant growth and reproduction require sufficient water, with water being the primary limiting factor for survival and growth in desert plants. As precipitation in the driest month decreases, water resources become scarce, leading to a reduce in distribution ranges of species and affecting TSR. Against the backdrop of intense climate change, Central Asia is experiencing significant climate change impact, causing 97.00% of glaciers in the Tianshan Mountains to continuously retreat since the 1980s and reducing snow depth by 20.00% (Chen et al., 2016). These changes have profound impacts on water resource systems, with seasonal runoff imbalances exacerbating flood risks while threatening glacier-fed river ecosystems (Fallah et al., 2024). Extreme climate events intensify the interactive effects of climatic factors. For instance, drought and high-temperature extremes may increase extinction risks for montane species (Wu et al., 2025). Similarly, the Himalaya Mountains have been affected by habitat changes driven by shifts in precipitation seasonality and the movement of isotherm lines, resulting in degraded ecosystem service functions (Kumar and Sharma, 2023).
Sampling deficiencies are particularly evident in the phytogeographic regions bordering neighboring countries and the biodiversity hotspot zones in the southwestern and central regions of Tajikistan. Specifically, the Southern Tajikistan, most sub-regions of the Gissar-Darvasian, and West Pamirian C showed occurrence data quantities far below what would be expected given their extremely high SR (Figs. 4 and 6). Possible reasons include the difficulty of comprehensive sampling in these highly biodiverse zones and potential limitations on sampling activities in border or historically unstable regions, thus strengthening international cooperation. Notably, Tajikistan’s biodiversity hotspots (primarily located in the southwestern region of Tajikistan) are data coldspots of the GBIF and IUCN species occurrence records, while biodiversity coldspots (the eastern region of the Pamir Plateau) are data hotspots (relatively well-sampled). This discrepancy may arise because although Pamir Plateau has sparse species, its unique geomorphology has attracted disproportionate sampling activity as a focus for many international scientific expeditions. Additionally, research projects targeting specific ecosystems (such as alpine deserts) or charismatic species may generate intensive sampling records in species-poor areas, resulting in relatively excessive collection. Finally, much of Tajikistan’s rich botanical heritage originates from the Soviet period, with these valuable specimens and records primarily preserved in herbaria across Central Asia and Russia (Ball et al., 2025). The slow digitization and sharing process of these historical data has created significant data gaps. Some of the data were published in local databases (Central Asian and Russian repositories), but they were not shared through iNaturalist and GBIF, requiring extensive data mining efforts.

5. Conclusions

This study provides a comprehensive multidimensional assessment of vascular plant diversity patterns and conservation gaps in Tajikistan. Our analyses revealed consistent decreasing gradients of SR, PD, TSR, and ESR from the southwestern region to northeastern region of Tajikistan. Gissar-Darvasian and Zeravshanian emerged as irreplaceable biodiversity cores. Crucially, we found that a significant portion of these endemic and threatened species are phylogenetically clustered within key evolutionary lineages, particularly Astragalus, Tulipa, and Allium. However, current PAs encompassed less than 40.00% of these priority zones, leaving a 64.80% conservation gap for integrated priority regions. This spatial mismatch underscores the urgent need to expand PA networks, particularly in the mountainous regions of the southwestern Tajikistan, to achieve Kunming-Montreal Global Biodiversity Framework Target 3 (“30×30 Protection”).
Climatic factor analyses identified precipitation seasonality and isothermality as primary determinants of species diversity patterns, collectively explaining 52.70%-59.20% of the variations in SR, PD, and TSR. The significant positive correlation between precipitation seasonality and diversity indices highlighted the vulnerability of vascular plant adapted to winter wet and summer dry environments to climate change-induced hydrological shifts. Endemic species further demonstrated dependence on winter precipitation, suggesting that these climate-diversity linkages necessitate climate-resilient conservation strategies to mitigate warming-driven range contractions.
Based on these findings, we propose explicit and actionable conservation policy recommendations. Firstly, prioritizing the establishment of new PAs or Other Effective Area-based Conservation Measures in the underprotected hotspots identified, particularly within Gissar-Darvazian B and South Tajikistanian C. Secondly, conservation strategies must move beyond mere PA designation to encompass targeted management actions. This includes implementing sustainable grazing management plans to alleviate pressure on fragile alpine meadows, as well as enforcing stricter regulations against the illegal collection of threatened ornamental and medicinal plants. Thirdly, to mitigate the significant sampling bias identified, we strongly advocate that future botanical surveys and monitoring efforts should be concentrated in the data-deficient western and southwestern regions of Tajikistan, as comprehensive data from these regions are fundamental for accurately assessing conservation effectiveness.
While our environmental models are robust, they are constrained to macroclimatic variables and do not incorporate fine-scale topographic heterogeneity, soil properties, or direct anthropogenic pressures such as land use change and infrastructure development. These factors likely contribute to the remaining 40.00% of unexplained variance. Furthermore, the accuracy of our diversity patterns is inherently limited by the spatial sampling bias in occurrence records. Future research should focus on integrating these multifaceted variables to build more comprehensive predictive models. Specifically, applying species distribution models to forecast the dynamics of key species and hotspots under various climate change scenarios would be invaluable for proactive conservation. This will require a concerted effort to digitize and geolocate the vast collections of herbarium specimens held within Central Asian institutions.
In conclusion, by integrating phylogenetic dimensions, spatial bias quantification, and climate-vegetation interactions, this study establishes a robust scientific foundation for implementing evidence-based conservation strategies in Tajikistan, contributing to both the theoretical understanding of arid mountainous biodiversity dynamics and the practical implementation of the Post-2020 Global Biodiversity Framework, particularly Target 3 and Target 4 (“Halt Human-induced Extinctions”).

Authorship contribution statement

ZHOU Yixin: methodology, software, formal analysis, investigation, data curation, visualization, and writing - original draft; MA Suliya: validation, investigation, resources, data curation, and writing - review & editing; LI Wenjun: conceptualization, validation, investigation, resources, writing - review & editing, supervision, project administration, and funding acquisition; Parvina KURBONOVA: investigation, resources, and supervision; Mariyo BOBOEV: investigation, resources, and supervision; LI Yufan: methodology, formal analysis, investigation, and data curation; Hikmat HISORIEV: conceptualization, investigation, resources, and supervision; MA Keping: conceptualization and supervision; YANG Weikang: conceptualization and supervision; and ZHANG Yuanming: conceptualization and supervision. All authors approved the manuscript.

Declaration of conflict of interest

ZHANG Yuanming is a Chief Editor of Regional Sustainability and a Guest Editor in Chief of the Special Issue “Green Sustainability in Tajikistan: Bridging Science, Policy, and Community Action” of Regional Sustainability, and was not involved in the editorial review or the decision to publish this article. All authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors would like to acknowledge the Chinese Academy of Sciences Research Center for Ecology and Environment of Central Asia (RCEECA), the construction and joint research for the China-Tajikistan “Belt and Road” Joint Laboratory on Biodiversity Conservation and Sustainable Use (2024YFE0214200), the Shanghai Cooperation Organization Partnership and International Technology Cooperation Plan of Science and Technology Projects (2023E01018, 2025E01056), and the Chinese Academy of Sciences President’s International Fellowship Initiative (PIFI) (2024VBC0006).

References

Publishing order | Descend order by publishing year | Descend order by cited within
[1]
Akhmadov K.M., Breckle S.W., Breckle U., 2006. Effects of grazing on biodiversity, productivity, and soil erosion of alpine pastures in Tajik mountains. In: SpehnE.M., LibermanM., KörnerC., (eds.). Land Use Change and Mountain Biodiversity. Boca Raton: CRC Press, 239-248.

[2]
Ball A., Daru B.H., Lughadha E.M., et al., 2025. What ‘unexplored’ means: Mapping regions with digitized natural history records to look for ‘biodiversity blindspots’. PeerJ. 13, e18511, doi: 10.7717/peerj.18511.

[3]
Bartoń K., 2025. MuMIn: Multi-Model Inference. R Package Version 1.48.11. [2025-07-11]. https://CRAN.R-project.org/package=MuMIn.

[4]
Borsch T., Berendsohn W., Dalcin E., et al., 2020. World Flora Online: Placing taxonomists at the heart of a definitive and comprehensive global resource on the world’s plants. Taxon. 69(6), 1311-1341.

DOI

[5]
Boyle B., Hopkins N., Lu Z., et al., 2013. The taxonomic name resolution service: An online tool for automated standardization of plant names. BMC Bioinformatics. 14, 16, doi: 10.1186/1471-2105-14-16.

PMID

[6]
Brown M.J., Walker B.E., Black N., et al., 2023. rWCVP: A companion R package for the World Checklist of vascular plants. New Phytologist. 240, 1355-1365.

DOI PMID

[7]
Chen Y.N., Li W.H., Deng H.J., et al., 2016. Changes in Central Asia’s water tower: Past, present and future. Scientific Reports. 6, 35458, doi: 10.1038/srep35458.

[8]
Coelho M.T.P., Barreto E., Rangel T.F., et al., 2023. The geography of climate and the global patterns of species diversity. Nature. 622(7983), 537-544.

DOI

[9]
Fallah B., Didovets I., Rostami M., et al., 2024. Climate change impacts on Central Asia: Trends, extremes and future projections. International Journal of Climatology. 44(10), 3191-3213.

DOI

[10]
Fattorini S., 2017. Endemism in historical biogeography and conservation biology: Concepts and implications. Biodiversity and Evolution. 32(1), 47-75.

[11]
Govaerts R., Nic Lughadha E., Black N., et al., 2021. The World Checklist of Vascular Plants, a continuously updated resource for exploring global plant diversity. Scientific Data. 8, 215, doi: 10.1038/s41597-021-00997-6.

PMID

[12]
Hawkins B.A., Field R., Cornell H.V., et al., 2003. Energy, water, and broad-scale geographic patterns of species richness. Ecology. 84(11), 3105-3117.

DOI

[13]
Hernangómez D., 2023. Using the tidyverse with terra objects: The tidyterra package. Journal of Open Source Software. 8(91), 5751, doi: 10.21105/joss.05751.

[14]
Hewitt G., 2000. The genetic legacy of the Quaternary ice ages. Nature. 405(6789), 907-913.

DOI

[15]
Hijmans R.J., 2025. terra: Spatial Data Analysis. R Package Version 1. [2025-07-11]. https://CRAN.R-project.org/package=terra.

[16]
Jin Y., Qian H., 2022. V.PhyloMaker2: An updated and enlarged R package that can generate very large phylogenies for vascular plants. Plant Diversity. 44(5), 335-339.

DOI

[17]
Kass J.M., Guénard B., Dudley K.L., et al., 2022. The global distribution of known and undiscovered ant biodiversity. Science Advances. 8(34), eabp9908, doi: 10.1126/sciadv.abp9908.

[18]
Knoche M., Merz R., Lindner M., et al., 2017. Bridging glaciological and hydrological trends in the Pamir Mountains, Central Asia. Water. 9(6), 422, doi: 10.3390/w9060422.

[19]
Kumar P., Sharma K., 2023. Snowfall shift and precipitation variability over Sikkim Himalaya attributed to elevation-dependent warming. Journal of Atmospheric Science Research. 6(4), 5854, doi: 10.30564/jasr.v6i4.5854.

[20]
Kusumoto B., Chao A., Eiserhardt W.L., et al., 2023. Occurrence-based diversity estimation reveals macroecological and conservation knowledge gaps for global woody plants. Science Advances. 9, eadh9719, doi: 10.1126/sciadv.adh9719.

[21]
Lai J.S., 2022. Generalizing hierarchical and variation partitioning in multiple regression and canonical analysis using the rdacca.hp R package. Methods in Ecology and Evolution. 13(4), 782-788.

DOI

[22]
Leão T.C.C., Lughadha E.N., Reich P.B., 2020. Evolutionary patterns in the geographic range size of Atlantic Forest plants. Ecography. 43(10), 1510-1520.

DOI

[23]
Letunic I., Bork P., 2024. Interactive Tree of Life (iTOL) v6: recent updates to the phylogenetic tree display and annotation tool. Nucleic Acids Research. 52(W1), W78-W82.

[24]
Li W.J., Tojibaev K.S., Hisoriev H., et al., 2020. Mapping Asia Plants: Current status of floristic information for Central Asian flora. Global Ecology and Conservation. 24, e01220, doi: 10.1016/j.gecco.2020.e01220.

[25]
Li Y.F., Li W.J., Ma S.L.Y., et al., 2025. Diversity and conservation gaps of threatened vascular plant species in Tajikistan. Arid Land Geography. 48(1), 43-52 (in Chinese).

DOI

[26]
Liu J.Q., Li J.L., Lai Y.J., 2021. Plant diversity and ecology on the Qinghai-Tibet Plateau. Journal of Systematics and Evolution. 59(6), 1139-1141.

DOI

[27]
Ma S.L.Y., Li W.J., Tojibaev K.S., et al., 2024. Regionwide and nationwide floristic richness reveal vascular plant diversity in Central Asia. Plants. 13(16), 2275, doi: 10.3390/plants13162275.

[28]
Marchese C., 2015. Biodiversity hotspots: A shortcut for a more complicated concept. Global Ecology and Conservation. 3, 297-309.

DOI

[29]
Molina-Venegas R., Ramos-Gutiérrez I., Moreno-Saiz J.C., 2020. Phylogenetic patterns of extinction risk in the endemic flora of a Mediterranean hotspot as a guiding tool for preemptive conservation actions. Frontiers in Ecology and Evolution. 8, 571587, doi: 10.3389/fevo.2020.571587.

[30]
Naimi B., Hamm N.A.S., Groen T.A., et al., 2014. Where is positional uncertainty a problem for species distribution modelling. Ecography. 37(2), 191-203.

DOI

[31]
National Academy of Sciences of Tajikistan, 2024. The Red Book of the Republic of Tajikistan (3rd edition). Dushanbe: National Academy of Sciences of Tajikistan.

[32]
Noroozi J., Khalvati S., Nafisi H., et al., 2021. Endemics determine bioregionalization in the alpine zone of the Irano-Anatolian biodiversity hotspot (South-West Asia). Alpine Botany. 131(2), 177-186.

DOI PMID

[33]
Noroozi J., Minaei M., Khalvati S., et al., 2023. Hotspots of (sub)alpine plants in the Irano‐Anatolian global biodiversity hotspot are insufficiently protected. Diversity and Distributions. 29(2), 244-253.

DOI

[34]
Nowak A., Nobis M., 2010. Tentative list of endemic vascular plants of the Zeravshan Mts in Tajikistan: Distribution, habitat preferences and conservation status of species. Biodiversity: Research and Conservation. 19, 65-80.

DOI

[35]
Nowak A., Nowak S., Nobis M., 2020a. The Pamir-Alai Mountains (Middle Asia:Tajikistan). In: Noroozi, J., (ed.). Plant Biogeography and Vegetation of High Mountains of Central and South-West Asia. Cham: Springer International Publishing, 1-42.

[36]
Nowak A., Świerszcz S., Nowak S., et al., 2020b. Red List of vascular plants of Tajikistan—the core area of the Mountains of Central Asia global biodiversity hotspot. Scientific Reports. 10, 6235, doi: 10.1038/s41598-020-63333-9.

[37]
Ohdo T., Takahashi K., 2020. Plant species richness and community assembly along gradients of elevation and soil nitrogen availability. AoB Plants. 12(3), plaa014, doi: 10.1093/aobpla/plaa014.

[38]
Oksanen J., Simpson G.L., Blanchet F.G., et al., 2025. vegan: Community Ecology Package R Package Version 2.7-2. [2025-06-11]. https://CRAN.R-project.org/package=vegan.

[39]
Ondo I., Dhanjal-Adams K.L., Pironon S., et al., 2024. Plant diversity darkspots for global collection priorities. New Phytologist. 244(2), 719-733.

DOI

[40]
Ovchinnikov P.N., 1957-1991. Flora Tadzhikskoi SSR. Moscow & Leningrad: Izdatel’stvo Akademii Nauk SSSR (in Russian).

[41]
Pebesma E., Bivand R., 2023. Spatial Data Science:With Applications in R (1st edition). New York: Chapman and Hall/CRC Press.

[42]
Qian H., Kessler M., Qian S., et al., 2024. Patterns and drivers of taxonomic and phylogenetic endemism in regional fern floras across the world. Biological Conservation. 291, 110506, doi: 10.1016/j.biocon.2024.110506.

[43]
Raduła M., Świerszcz S., Nobis M., et al., 2021. Palaeoclimate has a major effect on the diversity of endemic species in the hotspot of mountain biodiversity in Tajikistan. Scientific Reports. 11, 18684, doi: 10.1038/s41598-021-98027-3.

PMID

[44]
Rahmonov O., Majgier L., Andrejczuk W., et al., 2013. Landscape diversity and biodiversity of Fann Mountains (Tajikistan). Ekológia. 32(4), 388-395.

[45]
Song S.L., Chen X., Zan C., et al., 2025. Integrated spatial priority assessment in Central Asia: Bridging biodiversity, ecosystem services, and human activities. Geography and Sustainability. 6, 100231, doi: 10.1016/j.geosus.2024.08.010.

[46]
Spengler R.N., 2019. Origins of the apple: The role of megafaunal mutualism in the domestication of Malus and rosaceous trees. Frontiers in Plant Science. 10, 617, doi: 10.3389/fpls.2019.00617.

[47]
Squires V.R., Safarov N., 2013. High-altitude ecosystems and biodiversity of Tajikistan:Conservation and management. In: Akhmadov, K.M., Khamidov, J.K., Razykov, M.R., (eds.). Rational Use of Natural Resources and Sustainable Development of the Pamir Mountain Region. Dushanbe: Donish, 15-21.

[48]
Squires V.R., Dengler J., Hua L., et al., 2018. Grasslands of the World:Diversity, Management and Conservation (1st edition). Boca Raton: CRC Press.

[49]
Sun L., Luo J., Qian L.L., et al., 2020. The relationship between elevation and seed-plant species richness in the Mt. Namjagbarwa region (Eastern Himalayas) and its underlying determinants. Global Ecology and Conservation. 23, e01053, doi: 10.1016/j.gecco.2020.e01053.

[50]
Thakur S., Dhyani R., Negi V.S., et al., 2022. Water-energy, climate, and habitat heterogeneity mutually drive spatial pattern of tree species richness in the Indian Western Himalaya. Frontiers in Forests and Global Change. 5, 1022082, doi: 10.3389/ffgc.2022.1022082.

[51]
Tucker C.M., Cadotte M.W., Carvalho S.B., et al., 2017. A guide to phylogenetic metrics for conservation, community ecology and macroecology. Biological Reviews. 92(2), 698-715.

DOI

[52]
Venables W.N., Ripley B.D., 2002. Modern Applied Statistics with S (4th edition). New York: Springer.

[53]
Volkova P.A., Ivanova M.O., Boboev M.T., et al., 2024. Are aquatic plants really endangered in Tajikistan (core area of the Mountains of Central Asia global biodiversity hotspot)? Journal of Asia-Pacific Biodiversity. 17(4), 769-779.

DOI

[54]
Wickham H., 2016. ggplot2: Elegant Graphics for Data Analysis (2nd edition). New York: Springer-Verlag.

[55]
Wilson B., Dolotbakov A., Burgess B.J., et al., 2021. Central Asian wild tulip conservation requires a regional approach, especially in the face of climate change. Biodiversity and Conservation. 30(6), 1705-1730.

DOI

[56]
Wu Y.L., Shen J.Q., Deane D.C., et al., 2025. Future extreme climate events threaten alpine and subalpine woody plants in China. Earth’s Future. 13(6), e2024EF005147, doi: 10.1029/2024EF005147.

[57]
Wyse Jackson P., Kennedy K., 2009. The global strategy for plant conservation: A challenge and opportunity for the international community. Trends in Plant Science. 14(11), 578-580.

DOI PMID

[58]
Yao S.T., Akram M.A., Hu W.D., et al., 2021. Effects of water and energy on plant diversity along the aridity gradient across dryland in China. Plants. 10(4), 636, doi: 10.3390/plants10040636.

[59]
Zhang Y.M., Zhang D.Y., Li W.J., et al., 2020. Characteristics and utilization of plant diversity and resources in Central Asia. Regional Sustainability. 1(1), 1-10.

DOI

Options
Outlines

/

〈 〉