Conserving Red List plant species by managing landscape fragmentation and permeability

Conserving Red List plant species by managing landscape fragmentation and permeability | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Conserving Red List plant species by managing landscape fragmentation and permeability Nicu Alexandru Gîlea, Ileana Pătru-Stupariu, Athanasios Gavrilidis, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7957482/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 19 Feb, 2026 Read the published version in Landscape Ecology → Version 1 posted 9 You are reading this latest preprint version Abstract Context : Habitat fragmentation contributes to population isolation and threatens the survival of plant species listed on the Red List. Effective conservation measures require considering both suitable habitats and landscape permeability. Species Distribution Models (SDMs) are valuable for identifying suitable habitats. These models gain utility when combined with landscape permeability analyses for conservation planning. Objectives : This study aims to (1) model the potentially suitable habitat distribution for five threatened plant species in Romania, (2) assess landscape permeability, and (3) identify priority areas for conservation based on the overlay of the habitat and permeability. Methodology : We applied an integrated approach that combined SDM modelling (MaxEnt) and GIS analysis. We used the Continuum Suitability Index (CSI) to quantify landscape permeability. By mapping the spatial overlap of habitat suitability and permeability, we classified areas into four categories for ecological interventions. Results : A significant proportion of the identified suitable habitats have low permeability, which limits species dispersal and persistence. Areas combining high suitability and high permeability are limited and are mostly outside protected areas (PAs). This highlights the urgent need for conservation actions. Measures such as optimal grazing management, reduced fragmentation, and maintaining traditional agro-pastoral mosaics should be prioritized in these critical zones beyond formal protected area boundaries. Conclusion : Including landscape permeability in habitat analyses gives a more realistic view of the requirements of threatened species. It supports more explicit decision-making for conservation management. The study also proposes a replicable method for other fragmented regions, with direct applications in land-use policy and biodiversity conservation. Threatened species Continuum Suitability Index Environmental variables MaxEnt Grasslands Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Introduction Habitat fragmentation and loss represent critical threats to global biodiversity (Haddad et al., 2015; Scholes et al., 2018). These processes, driven by agricultural expansion, urbanisation, and infrastructure development, reduce the area of natural habitats and disrupt the ecological connectivity of the landscape (Babí Almenar et al., 2019; Luo et al., 2020). Simultaneously, climate change forces species to adapt and disperse in an increasingly fragmented landscape (Van Daele et al., 2024). Conservation measures often utilize the ecological permeability of landscapes to address these threats, which are projected to worsen in the future (Costanza & Terando, 2019; Laner et al., 2024). Such measures are a key mechanism for restoring landscape connectivity, which is essential for species dispersal and for enhancing ecosystem resilience . The process of habitat fragmentation, dividing a continuous natural area into smaller, isolated units, significantly affects biodiversity and ecosystem functioning (Haddad et al., 2015; Henle et al., 2004; Liu et al., 2018; Wilson et al., 2016). As fragmentation progresses, it reduces species dispersal and genetic flow, thereby decreasing their capacity to adapt to environmental changes (Fahrig, 2003, 2017). While fragmentation impacts a wide range of species, the effects are more severe for threatened species, which typically have restricted geographic ranges, small populations, and limited dispersal capabilities (Henle et al., 2004). These traits, in turn, increase vulnerability to genetic drift and stochastic events, amplifying the risk of extinction (Lande, 1988). Consequently, the loss of such species has major ecological consequences, affecting the stability of trophic networks and the functioning of ecosystem services (Hassan et al., 2005). Assessments of the Red Lists in Europe show that habitat conversion and population isolation threaten a significant proportion of vascular plants (Bilz et al., 2011). In Central and Eastern Europe, these threats are exacerbated by agricultural conversion, intensive grazing, and land abandonment, which are the main drivers of decline in semi-natural ecosystems (Dahlström et al., 2013; Enyedi et al., 2008; Plieninger et al., 2015; Yezzi et al., 2023). As a consequence, these processes have caused severe fragmentation of grassland habitats, which, in turn, threatens endemic and relict species. Many of these species are Red List taxa and critically depend on these ecosystems (Cremene et al., 2005; Török, Dembicz, et al., 2020). These trends are also strongly reflected in Romania. Threatened species characteristic of steppe and forest-steppe grasslands exhibit a fragmented geographic distribution and are highly sensitive to land-use changes and overgrazing (Chirilă, 2021). Moreover, recent research indicates that a large proportion of these species' populations exist outside protected areas, where they face higher risks (Chirilă, 2021; Hurdu et al., 2022). Consequently, studies report pronounced population declines and increasing isolation (Chirilă, 2021; Chirilă, Bădărău, et al., 2025; Chirilă, Doroftei, et al., 2025). This isolation increases the risk of local extinction and negatively affects ecological balance and ecosystem services. Ultimately, the loss of these species disrupts trophic networks and ecosystem processes, generating imbalances that may accelerate habitat degradation (Bilz et al., 2011; Kalista, 2017; Mykhailenko et al., 2020, 2023; Nowak et al., 2020). To support the viability and genetic diversity of these populations, habitat conservation is essential. However, as landscape fragmentation increases, protecting the remaining habitats alone is not enough. Conservation efforts must also enhance the landscape matrix's permeability (Da Silva et al., 2015; Guiller et al., 2023; Hadley et al., 2018; Silveira dos Santos et al., 2022) and restore ecological connectivity (Kimberley et al., 2021; Magrach et al., 2012; Szitár et al., 2023). Landscape connectivity characterizes the capacity of species to move between habitat fragments via corridors and linkage areas. In contrast, permeability refers to how favourable the landscape matrix—composed of natural, semi-natural, and anthropogenic land-cover types—is for species dispersal and essential ecological processes (Meiklejohn et al., 2010; Taylor et al., 2006). For plants, permeability enables the flow of propagules—pollen, seeds, spores—between populations. This flow maintains genetic diversity, enables colonisation, and supports reproduction. A permeable landscape promotes gene flow, population resilience, and long-term persistence, which reduces extinction risk (Beckman & Sullivan, 2023; Cruzan & Hendrickson, 2020; Török, Bullock, et al., 2020). For effective conservation planning based on the principle of increasing landscape permeability, it is necessary to identify suitable habitats and assess their accessibility to the target species. In this context, species distribution models (SDMs) provide an essential methodological framework for mapping and evaluating potential habitat suitability. These tools, particularly the MaxEnt algorithm, are recognised for their efficiency in identifying areas with optimal ecological conditions for species (Baldwin, 2009; Remya et al., 2015). This forms the scientific basis for numerous conservation interventions (Chauhan et al., 2022; Duan et al., 2025). Most studies focus on identifying bioclimatically suitable habitats (Duan et al., 2025; Mathur & Mathur, 2025; T. Wang et al., 2024). Although useful, this approach does not provide insight into the actual accessibility of favourable habitats for species. Some suitable habitats may be harder to colonise due to landscape barriers, even if bioclimatic conditions are favourable. Moreover, the practice of unsustainable agricultural activities (e.g., overgrazing, burning, and mechanised mowing) increases habitat hostility for target species. In this context, landscape permeability becomes a key variable in identifying priority areas for conservation and restoration (Favilli et al., 2023). The approach based on species distribution modelling (SDM) and landscape permeability is a method validated in numerous studies, particularly on mammals (Acharya et al., 2023; Cerreta et al., 2023; Muthiuru et al., 2024). However, studies assessing landscape permeability specifically for plants are still relatively scarce (Gîlea & Pătru-Stupariu, 2025). Here, we apply this combined framework to define targeted conservation measures for threatened plant species listed on Romania's Red List, within a fragmented, multi-use landscape. In this study, we aim to propose an approach to define spatially explicit conservation measures for threatened plant species listed on the Red List, based on the management of fragmentation and landscape permeability. We focus on the spatial analysis of habitats and landscape permeability, adopting a multispecies approach and using Romania as a case study. The research questions we address are: (1) how much of the habitat considered suitable is permeable for the threatened species? (2) to which extent are the potentially suitable habitats covered by the current network of protected areas (PAs). To answer these research questions we first identify potentially suitable habitats for five target plant species using SDM modelling. Second, we assess landscape permeability using key environmental variables and finally we identify priority areas for the effective conservation of threatened plant species both within and outside the existing network of protected areas (PAs). We expect that integrating landscape permeability into distribution models increases the ecological relevance of potentially suitable habitats for threatened plant species. By integrating spatial and ecological data, the study provides essential scientific support for the formulation of clearly delineated, sustainable conservation measures that are adapted to regional specifics and capable of addressing current challenges in protecting plant biodiversity. Methods Study Area The study was conducted in the Continental Biogeographic Region of Transylvania (CBRT), located in central Romania, covering an area of 34,289 km² (Fig. 1). CBRT is characterised by a moderate continental climate, with altitudinal and soil variations that influence the distribution of species sensitive to habitat changes. The region includes semi-natural habitats of particularly high ecological value, including priority habitats (according to the EU Habitats Directive) such as xerophilous steppe and forest-steppe grasslands, that host rare and vulnerable plant species, which are affected by fragmentation and overgrazing (Chirilă, 2021; Chirilă, Doroftei, et al., 2025; Chirilă & Kiril, 2024). Among the most important habitats are Pannonian subcontinental steppe grasslands (6240*), semi-natural mesoxerophilous grasslands (6210*), Pontic-Sarmatic steppe grasslands (62C0*), and Pannonian loess steppe grasslands (6250*). The landscape is dominated by agricultural lands, deciduous forests, pastures, and dispersed urban areas, (Grădinaru et al., 2020). Additionally, 6,884 km² (20.07%) of the region are included in protected natural areas (Natura 2000 and others), providing essential refuges for biodiversity (Ministry of Environment, 2025a). Target Red List plant species Five threatened plant taxa of high conservation value are associated with semi-natural habitats in the CBRT: Crambe tataria Sebeók, Pontechium maculatum (L.) Böhle & Hilger, Iris aphylla L. , Klasea lycopifolia (Vill.) Á. Löve & D. Löve and Paeonia tenuifolia L.. These species are characteristic of xerophile grasslands (steppe and forest-steppe), preferring well-drained soils and a continental climate. They are sensitive to habitat degradation, and are affected by anthropogenic activities such as overgrazing or the conversion of land to agricultural or urban areas (Chirilă et al., 2022; Sava et al., 2019). The five target species are included in the national Red List and are considered vulnerable and rare (Dihoru & Dihoru, 1994; Oltean et al., 1994; Oprea, 2005), with the exception of P. tenuifolia which is classified as endangered (Dihoru & Dihoru, 1994). They are protected under national and European legislation, being listed in Annexes II and IV of the Habitats Directive and in Annex I of the Bern Convention (Bilz et al., 2011). The species C. tataria and P. maculatum are valuable indicators of steppe grassland conservation status. Both are melliferous plants that support pollinator populations and thus contribute to maintenance of local biodiversity (Corbet & Delfosse, 1984; Martín Arroyo et al., 2017). C. tataria is related to agricultural crops and has potential applications in animal feed, biodiesel production, and phytomedicine (Bilz et al., 2011; Kalista, 2017). P. maculatum is valued for its ornamental, sanitary, and medicinal properties (Nowak et al., 2020). Additionally, the species demonstrates tolerance to heavy metals, indicating potential for phytoremediation and adaptability to harsh edaphic conditions (Jakovljevic et al., 2019). P. tenuifolia is notable for its striking flowers and associated ornamental and medicinal value (Fateryga, 2015; Zanina & Smirnova, 2020). I. aphylla exhibits promising anti-inflammatory, antioxidant, and antiviral activities, with potential applications in treatments against influenza and enteroviral infections (Mykhailenko et al., 2020, 2023). The vulnerability of these species is amplified by their limited dispersal strategies. C. tataria is anemochorous (tumbleweed, mean dispersal distance: 40–150 m), while the remaining taxa have unspecialised dispersal, predominantly over short distances (namely ballochory, blastochory, boleochory; mean dispersal distance: 0.1–5 m) (FloraVeg.EU, 2025). This limited colonisation capacity reduces their potential to respond to disturbances, highlighting the importance of habitat conservation and connectivity for the maintenance of populations. Species presence data We constructed a database of presence points for the threatened plant species by combining information from multiple sources. First, we included official presence data (n = 1,098) obtained from the Romanian Ministry of Environment 2025b), restricted to protected areas. These data were collected through direct field observations and monitoring by biologists and ecologists between 2007 and 2017. Second, we supplemented the database with our own presence points (n = 247), collected through yearly field observations between April and August during the 2017–2025 period. We recorded GPS locations of individual plants using the OsmAnd application (v.4.7.17) (https://osmand.net/). Third, we accessed the Global Biodiversity Information Facility (GBIF) platform (https://doi.org/10.15468/dl.jsgdqy) and retrieved presence records of threatened species from the CBRT (n = 74), collected between 1816 and 2025 (GBIF.org, 2025). The resulting database comprises 1,419 presence points, of which 1,115 are located within protected areas and 304 outside them. The distribution of points by species is as follows: C. tataria – 813, P. maculatum – 354, I. aphylla – 178, K. lycopifolia – 60, and P. tenuifolia – 14. To mitigate the spatial sampling bias caused by the uneven distribution of presence points, we applied a spatial thinning procedure (Duan et al., 2025). This adjustment is essential to avoid SDM overfitting and to improve prediction accuracy. Presence points were filtered so that only those separated by a minimum distance of 30 m were retained. The procedure was carried out in ArcGIS Pro v3.5.1, resulting in a final set of 483 points distributed as uniformly as possible (Fig. 1). This final dataset was used for modelling the potential habitat of the threatened species. A total of 25 environmental variables, selected based on the reviewed literature, were used to predict the potentially suitable habitat of the threatened species. They cover bioclimatic condictions, topography, soil and land use. Variable and sources of data are decribed in Table 1. Raster layers were clipped to the boundaries of the CBRT and reprojected into the same coordinate system. To ensure compatibility between layers and with the MaxEnt model, all variables were rescaled to a common resolution of 30 × 30 m, using bilinear resampling for continuous variables (e.g., elevation) and nearest-neighbour resampling for categorical variables (e.g., land use/cover). The layers were exported in ASCII format for use in MaxEnt, using ArcGIS Pro v3.5.1. To avoid collinearity and obtain a robust set of explanatory variables, selection was carried out in two steps (Duan et al., 2025). First, an initial MaxEnt model evaluated the contribution and importance of each variable, and a jackknife test identified variables with significant unique information. In the second step, a Pearson correlation matrix was calculated among all variables using the cor function in R (v. 4.5.0), and strongly correlated pairs (|r| ≥ 0.7) were removed (Dormann et al., 2013) (see Online Resource 1). Following the application of these criteria, the final set of variables used in the MaxEnt model included: Soil, LULC, WR, Bio15, Bio14, Elevation , and Bio11 , selected for their relevant contribution and lack of excessive collinearity. Table 1. Variables used for SDM modelling Variables Code Description Resolution Source Bioclimatic Bio1 Annual mean temperature 1 km WorldClim v2.1 Bio2 Mean diurnal range Bio3 Isothermality Bio4 Temperature seasonality Bio5 Maximum temperature of the warmest month Bio6 Min temperature of the coldest month Bio7 Range of annual temperature variation Bio8 Mean temperature of the wettest quarter Bio9 Mean temperature of driest quarter Bio10 Mean temperature of warmest quarter Bio11 Mean temperature of coldest quarter Bio12 Annual precipitation Bio13 Precipitation of wettest month Bio14 Precipitation of driest month Bio15 Precipitation seasonality Bio16 Precipitation of wettest quarter Bio17 Precipitation of driest quarter Bio18 Precipitation of warmest quarter Bio19 Precipitation of coldest quarter Topographic Elevation Elevation 25 m EU-DEM v1.1 (Copernicus) Slope Slope Aspect Aspect Soil Soil Soil class 1 km European Soil Database v2 (JRC) WR Dominant annual average soil water regime class of the soil profile Land use/cover LULC Land cover/use classes 100 m Copernicus Land Monitoring Service Identifying potentially suitable habitat (HS) using the SDM We implemented a maximum entropy modelling approach using the MaxEnt software (version 3.4.4) to identify the potential habitat of threatened plant species in the CBRT (Phillips et al., 2006, 2025). The five species were included in a single model due to their similar ecological requirements, which allowed the use of the same environmental parameters to identify areas favourable for conservation. The MaxEnt model, built using seven environmental variables, was run with 10 replicates and with the cloglog output format, generating ASCII files. To evaluate the robustness of the model and reduce the risk of overfitting, we applied a 10-fold cross-validation scheme, which provides a more realistic estimation of model performance on datasets not used for training. Only linear and quadratic features were used (Phillips, et al., 2004). Predictions were expressed in logistic format, with values ranging from 0 to 1, indicating the probability that each raster cell represents a suitable habitat (Phillips & Dudík, 2008). Default software settings were retained, except where specified above. To assess the importance of environmental variables and their influence on species distribution, we applied the Jackknife test and analysed response curves. Overall model accuracy was evaluated using the area under the ROC curve (AUC), with values ranging from 0 (random prediction) to 1 (perfect prediction) (Duan et al., 2025; Phillips & Dudík, 2008). Based on the SDM, we determined habitat suitability (HS), used to interpret and classify the degree of habitat favourability within the CBRT. For this, we loaded the resulting file with the suffix “average” into ArcGIS Pro v3.5.1, which contains the spatial distribution of species occurrence probability in cloglog format (values between 0 and 1). The decision threshold used to separate suitable from unsuitable habitats was the maximum sensitivity-specificity cloglog threshold ( p = 0.2 ), derived from ROC analysis (Qi et al., 2025). We then applied a raster reclassification process using the Reclassify tool to transform the continuous raster into a categorical raster, reflecting distinct levels of ecological suitability. Areas were classified into four categories (Acharya et al., 2023; Duan et al., 2025) as follows: <0.2 — unsuitable, 0.2–0.4 — low suitability, 0.4–0.6 — medium suitability, ≥0.6 — high suitability. Assessing landscape permeability for threatened plant species To assess landscape permeability, we applied the Continuum Suitability Index (CSI) (Affolter et al., 2011; Favilli et al., 2023; Laner et al., 2024), adapted to the regional context and the ecological requirements of the analysed species. This method estimates permeability based on the landscape's physical characteristics and the level of anthropogenic pressure (Swiss National Park, 2019). CSI operates through a weighted multicriteria overlay analysis. In this method, relevant environmental variables are assigned predefined scores. These scores indicate the degree of permeability of each landscape element (Laner et al., 2024). In this study, we used seven environmental variables (Table 2), selected and weighted based on the reviewed literature (Chirilă, 2021; Gîlea & Pătru-Stupariu, 2025; Laner et al., 2024). Permeability scores were assigned on a scale from 0 (low permeability) to 10 (high permeability) using a mixed approach that combined statistical methods, empirical data, and literature information. Specifically, for the LULC, Soil, Aspect , and Slope variables, permeability scores were calculated using Resource Selection Functions (RSF), following the principles of Boyce et al. (2002). These compare the environmental characteristics at species presence points with those at randomly generated background points. For each habitat class, a selection ratio ( wi ) was calculated, with preferred classes receiving high scores, indicating high permeability, while avoided classes received low scores. The resulting values were subsequently validated and calibrated using additional literature data on species ecological requirements (see Online Resource 2). To estimate permeability scores based on livestock stocking rate (STK) (LU/ha), a Gaussian (normal) function was used, centred on the range considered optimal for grazing, 0.4–0.6 LU/ha (Marușca et al., 2014), corresponding to the maximum permeability value (see Online Resource 3). Permeability values attributed to fragmentation and protected areas were obtained from previous studies (Laner et al., 2024; Lüthi et al., 2018a, 2018b). All variables were converted to rasters at a fine resolution of 5 × 5 m, sufficiently detailed to capture small landscape elements (e.g., roads) that influence permeability for the target species. These were reclassified according to the established permeability scores (see Online Resource 4) and integrated into a weighted mean. The calculation was performed in ArcGIS Pro v3.5.1 using the Raster Calculator tool and the following formula: Permeability was assessed in four regional windows (Plots), each covering 600 km². These windows were selected to encompass different levels of potentially suitable habitat for the target species. This approach reduced processing time for the large data volume and allowed for results that are more applicable at the local scale. At the same time, it maintained consistency in the landscape-scale assessment. Table 2. Environmental variables for assessing landscape permeability Environmental variable Abbreviation Source Resolution / Scale Land use/cover LAN Corine Land Cover Plus Backbone + APIA data (Agency for Payments and Intervention in Agriculture) + OpenStreetMap 10 m Habitat fragmentation FRA Derived from Land Use 1 × 1 km (grid) Grazing pressure /Livestock density STK Local Authorities + National Institute of Statistics (TEMPOOnline) 5 m Soil SOIL Romanian Soil Database 1:200.000 Slope SLOPE EU-DEM v1.1 – Copernicus 25 m Aspect ASPECT Protected Areas ENV Ministry of Environment 5 m Identifying priority areas for conservation and ecological restoration Priority areas for conservation were identified through the spatial overlay of Habitat Suitability (HS) and the Continuum Suitability Index (CSI) within the four plots (Fig. 2). This approach allowed the identification not only of optimal habitats but also of areas with ecological potential, which can be enhanced through targeted management of landscape permeability. To implement this strategy and facilitate ecological interpretation, the HS and CSI rasters were first reclassified into binary form. For HS, all cells with values ≥ 0.2 were considered suitable and coded as 1, while values < 0.2 were coded as 0. The threshold used to separate suitable from unsuitable habitats was the maximum sensitivity-specificity cloglog threshold ( p = 0.2 ), derived from ROC analysis. For CSI, the classification threshold was set at 6.0, justified by the observation that approximately 71% of species presence records occurred in areas with CSI ≥ 6 (see Online Resource 5). Accordingly, cells with CSI ≥ 6 were coded as 1 (high permeability), and cells < 6 were coded as 0 (low permeability). By applying a conditional function using the Raster Calculator in ArcGIS Pro (v3.5.1), four spatial categories relevant for planning ecological interventions were identified: Conservation areas: cells with HS ≥ 0.2 and CSI ≥ 6 (coded 1/1). High-quality, permeable habitats ideal for direct conservation and active management. Restoration areas: cells with HS ≥ 0.2 and CSI < 6 (coded 1/0). Suitable habitats degraded by fragmentation or other pressures, requiring interventions to improve permeability . Dispersal/colonisation potential areas: cells with HS < 0.2 and CSI ≥ 6 (coded 0/1). Permeable areas where habitat can be restored or that can facilitate natural dispersal, recommended as buffer zones . Unsuitable areas: cells with HS < 0.2 and CSI < 6 (coded 0/0). Low suitability and permeability; low priority for immediate intervention but potential targets for long-term permeability enhancement. Results SDM performance and variable contributions The potential habitat distribution model for threatened species demonstrated excellent predictive performance (AUC ≥ 0.9), with a mean test AUC of 0.900 (Fig. 3) and very little variation between replicates (standard deviation, SD = 0.011). The analysis of variable contributions revealed that the model is primarily driven by edaphic factors (Soil) and land use/cover (LULC) (Table 3). According to the percentage contribution metric, the most important predictors were soil (44.8%), followed by land use/cover (LULC, 26.8%) and the annual soil water regime (wr, 9.3%). In contrast, the Permutation Importance metric indicated a relatively high importance of the variables reflecting precipitation of the driest month (Bio14, 27.7%) and the mean temperature of the coldest quarter (Bio11, 14.4%), suggesting that these climatic variables contain critical predictive information. The Jackknife test (Fig. 4) confirmed these results, showing that soil is the variable with the highest individual predictive power, while land use/cover holds the most unique information not captured by the other predictors. Table 3. Estimates of relative contribution and permutation importance for environmental predictors Variable Percent contribution Permutation importance Soil 44.8 19.4 LULC 26.8 20.6 WR 9.3 6.4 Bio15 8.5 0.1 Bio11 6.4 14.4 Elevation 2 11.4 Bio14 2 27.7 The analysis of the response curves (Fig. 5, 6) provides detailed insight into the species' ecological niche. The land use/cover variable (LULC) (Fig. 5a) shows a probability of presence reaching maximum values (above 0.90) for classes 2 (grasslands), 7 (bare or sparsely vegetated areas), and 8 (shrublands). In contrast, the probability of presence is significantly lower in classes 1 (deciduous forests) and 3 (arable land). The species exhibit a clear preference for Mollisols and Vertisols (classes 2 and 11), whereas hydromorphic soils (class 3) are the least suitable (Fig. 5b). The analysis of the marginal response curve for the annual soil water regime (WR) illustrates a strong ecological preference (above 0.93) for well-drained or dry soil conditions (class 1) (Fig. 5c). Climatic and topographic factors define the species' tolerance limits. For Bio11 (Fig. 6a), the response curve shows a peak probability of presence—almost 0.95—within the optimal temperature range of approximately −2.5°C to −1.5°C. In the case of Bio14 (Fig. 6b), the maximum probability of presence, nearly 1.0, is observed at values below 25 mm; this probability drops rapidly to 0.0 when monthly precipitation exceeds 40 mm. When examining Bio15 (Fig. 6c), areas with low values (<30 mm) exhibit the highest probability, almost 0.95. With elevation (Fig. 6d), the probability of presence climbs above 0.98 at roughly 600 metres and remains high at greater elevations. Habitat suitability for threatened plant species in the CBRT In the CBRT, 21.5% of the territory (≈7,357.4 km²) provides potential habitat for threatened plant species (HS ≥ 0.2). Of this habitat, 22.0% (≈1,584 km²) lies within protected natural areas. The other 78.0% (≈5,764.4 km²) is outside them (Table 4). Regarding suitability, 5.4% (≈1,854.9 km²) of the territory has high suitability, 5.5% has medium suitability, and 10.5% has low suitability. The remaining 78.5% does not provide favourable conditions and is considered unsuitable. Favourable habitat is mainly in the western and central-southern region (Fig. 7), while non-suitable areas are mostly in the north, southwest, and east. Table 4. Habitat suitability in the CBRT and within protected natural areas Habitat Suitability CBRT Within PAs Outside PAs km² % km² % km² % Suitable HS ≥ 0.2 7.357,4 21,5 1.584 22,0 5.764,4 78,0 High suitability HS ≥ 0.6 1.854,9 5,4 371,8 20,0 1483,1 80,0 Medium suitability HS 0.4–0.6 1.887,5 5,5 466,8 24,7 1420,74 75,3 Low suitability HS 0.2–0.4 3.615,0 10,6 745,4 20,6 2860,6 79,4 Unsuitable HS ≤ 0.2 26.914,9 78,5 5.294,9 19,7 21620,1 80,3 Landscape permeability and habitat suitability in the investigated plots The mean CSI value varies among the four plots, ranging from 4.8 in Plot 2 to 5.4 in Plot 1. Variables associated with topographic conditions, such as slope (SLOPE) and aspect (ASPECT), along with soil class (SOIL), exhibit high mean values, ranging from 7.9 to 9.1. In contrast, variables reflecting landscape fragmentation (FRA), grazing pressure (STK), land use (LAN), and the presence of protected areas (ENV) show low mean values, ranging from 0.3 to 5.9 (Fig. 8). The landscape permeability map (Fig. 10) provides an integrated representation of CSI value distribution across the four plots, highlighting contrasts between areas of high ecological connectivity and fragmented zones affected by anthropogenic pressure. The area covered by suitable habitat (HS ≥ 0.2) varies among the four plots, from ≈241.4 km² in Plot 4, where the lowest value was recorded, to ≈435.2 km² in Plot 2, with the largest extent of suitable habitat (Fig. 9a). Areas with CSI ≤ 6 dominate all four plots, with the highest extent in Plot 2 (≈474.8 km²). Areas with CSI ≥ 6 are the largest in Plot 1, followed by Plots 4 and 3 (Fig. 9b). Priority areas for conservation and ecological restoration Four types of priority areas for the conservation of threatened species were identified, corresponding to varying levels of habitat suitability and landscape permeability: conservation, restoration, dispersal/colonisation, and unsuitable areas . The distribution of these zones varies among the investigated plots, reflecting local-scale differences (Fig. 13). Specifically, the largest conservation area was identified in Plot 1 (≈196.8 km²), while the largest restoration area was in Plot 2 (≈321.4 km²). In addition, Plot 4 contains the area with the highest dispersal potential, totaling ≈85 km². Significant areas of unsuitable zones were identified in Plot 3 (≈265.5 km²) and Plot 4 (≈263.2 km²) (Fig. 11). From the perspective of overlap with protected areas (Fig. 12, 13), we found that Plot 1 and Plot 3 show the largest areas of conservation and colonisation/dispersion included within protected areas, whereas in Plot 2 and Plot 4, these zones are less well covered by protected areas. In contrast, restoration zones are largely located outside protected areas, a pattern consistent across all four plots (Plot 1–Plot 4). This suggests that restoration activities should be prioritised in unprotected areas, where anthropogenic pressure is likely to be higher. Discussion Our findings, obtained through habitat distribution modelling and landscape permeability analysis, highlight critical areas for the survival and dispersal of threatened plant species. The analysis indicates that a significant portion of suitable habitat lies outside protected areas and is poorly permeable to the target species. This underscores the need for urgent interventions in targeted areas of the landscape. Such measures can include maintaining optimal grazing, supporting the traditional agro-pastoral mosaic, and reducing habitat fragmentation in priority conservation areas. These actions are essential to ensure the survival of existing populations and to facilitate dispersal and colonisation of new habitats. Overall, our findings emphasise the crucial role of managing landscape permeability in conserving threatened plant species, particularly in highly fragmented regions outside protected areas. Potential habitat distribution of threatened plant species The results indicate that soil characteristics and land cover type primarily drive potential habitat distribution. Climatic variables and altitude define the ecological limits of the species’ niche. The response curve analysis highlighted species-specific ecological traits and outlined the profile of potentially favourable habitat for the threatened species. The results show that the species prefers open grassland and silvosteppe habitats with well-drained soils, particularly molisols types, and avoids hydromorphic or excessively wet soils. This pattern indicates the species group’s adaptation to mesoxerophilous-xerophilous conditions and moderate moisture regimes. Climatic factors, such as the mean temperature of the coldest quarter (Bio11) and precipitation of the driest month (Bio14), suggest a shared affinity for regions with moderately cold winters and low rainfall, characteristic of semi-arid continental areas. The preference for medium- and high-altitude landscapes reflects a desire to avoid low-altitude areas with intense anthropogenic influence. Overall, these results describe a common ecological pattern among steppe and silvosteppe plant species, helping define priority habitats for conservation. These findings are consistent with other studies analysing the habitat preferences of these species (Chirilă, 2021; Chirilă et al., 2022; Chirilă & Kiril, 2024) and provide new insights into the environmental conditions that influence these priority species in Romania. For threatened plant species, it has been noted that, despite their conservation status, data about their environmental requirements are insufficient, and populations remain fragmented and declining (Chirilă et al., 2022). This emphasises the importance of our study in filling knowledge gaps about the habitat and ecological factors governing the persistence of these species. The spatial distribution of potential habitat indicates that the majority of favourable conditions for the five threatened plant species are outside protected areas. This situation is illustrated by Chirilă's (2021) study on the spatial distribution of Crambe tataria, which found that its populations occupy a relatively large area (approximately 75%) outside protected areas. This uneven distribution highlights the high vulnerability of species to anthropogenic pressures and emphasises the need for conservation measures, particularly in areas outside the protected network where they are generally more exposed to human activities (Hansen & DeFries, 2007; Radeloff et al., 2010). Overall, these results emphasise that conservation strategies must go beyond mere hotspot protection and adopt an integrated approach, combining the maintenance of high-quality habitats with the enhancement of ecological permeability. Landscape permeability and priority actions for conservation management Plot-level analysis revealed a significant discrepancy between the habitat potential identified by the SDM and the landscape's actual permeability. Although the SDM modelling indicated extensive areas of favourable habitat, the CSI assessment showed that many of these areas exhibit low permeability. The reduced permeability values are generated by anthropogenic pressures such as inappropriate grazing (STK), fragmentation (FRA), and unsuitable land use (LAN). This situation aligns with numerous studies that identify overgrazing, fragmentation, and unsuitable land use as primary threats to threatened plant species and their grassland habitats, which are in various stages of degradation due to human activities (Bernhardt et al., 2011; Bilz, 2011a, 2011b; Chirilă, 2021; Kell, 2011). These pressures limit species survival and dispersal and may lead to long-term isolation, even within areas of ecologically suitable habitat. Nevertheless, topographic and edaphic variables scored highly, suggesting that associated microrefugia may play a critical role in maintaining populations (Bennie et al., 2008; Moeslund et al., 2013). Overall, the results confirm that a significant portion of potentially suitable habitat is located in areas of reduced permeability, limiting the dispersal and persistence of threatened species. This study provides the first spatially explicit evaluation of landscape permeability for threatened plant species in Romania, highlighting critical areas where habitat suitability does not translate into ecological connectivity. Priority conservation areas The spatial overlay of habitat suitability (HS) and permeability (CSI) identified four priority zones, providing concrete guidance for conservation management. These results are consistent with previous studies that have integrated habitat suitability and landscape permeability to identify priority areas for conservation (Acharya et al., 2023; Cerreta et al., 2023; Muthiuru et al., 2024). Similar approaches have been successfully applied to delineate connectivity zones and guide targeted management actions in fragmented landscapes (Cable et al., 2021; Favilli et al., 2023). Our findings therefore support the inclusion of landscape permeability as a complementary dimension to habitat suitability in conservation planning. In fragmented landscapes, such as those in Romania, where habitat is often isolated by agricultural and infrastructural barriers, integrating CSI provides an enhanced framework for prioritising conservation actions. By explicitly mapping landscape permeability, our approach captures areas where key limiting factors threaten the survival of threatened plant species. Our integrated perspective goes beyond habitat-centred conservation and highlights that maintaining high permeability is essential even within favourable habitats. Overall, our results emphasise the importance of a landscape-scale management strategy that combines actions to reduce fragmentation, optimise grazing pressure, and maintain the traditional agro-pastoral mosaic in priority areas. Moreover, maintaining high permeability is essential even within ecologically favourable habitats, and the spatial prioritisation of interventions should be based on the degree of overlap between habitat suitability and permeability. Implications for integrated landscape management and Common Agricultural Policy Our spatially explicit results provide a valuable decision-support tool for optimizing the ecological effectiveness of the Common Agricultural Policy (CAP), particularly in non-protected areas where proactive management is vital for maintaining landscape permeability and biodiversity. The priority areas identified in our study offer a strategic blueprint for directing agri-environment interventions, shifting the focus from generalized subsidies towards precision conservation. Specifically, Measure 10 (Agri-environment and Climate) should be proactively promoted among landholders within these high-connectivity areas, especially those located outside strictly protected sites (MADR, 2019). This targeted approach is essential to incentivize the continuation of traditional, moderate-intensity practices (such as extensive mowing and moderate grazing) that underpin both habitat potential and landscape permeability, as confirmed by our findings regarding the optimal livestock stocking rates and the highest CSI values. Simultaneously, Measure 13 (Payments for Areas with Natural Constraints) must be prioritized in these vulnerable, high-priority landscapes to counter the risk of land abandonment and the resulting loss of open habitats from scrub encroachment (MADR, 2019a). By aligning CAP investments with functional ecological mapping, we ensure public funds are strategically allocated to maintain and enhance the interconnected habitat network, which is crucial to the survival of threatened species. Traditional land management practices, such as extensive mowing and moderate grazing, are key to maintaining the mosaic landscape structure that is essential for habitat connectivity and species diversity (Craioveanu et al., 2021; Janišová et al., 2024). Evidence indicates that when applied at moderate intensity, traditional grazing prevents vegetation closure and supports the persistence of rare species, whereas overgrazing degrades plant cover, and abandonment promotes the dominance of competitive species (Pykälä, 2003). Maintaining large, connected grassland fragments, complemented by functional corridors, is essential for species dispersal and for mitigating the effects of fragmentation (Soons et al., 2005; Yan et al., 2022). A key finding of our study—that moderate grazing plays a crucial role in maintaining grassland species diversity and contributing to landscape permeability—confirms the importance of adaptive pasture management in areas prioritized for conservation, restoration, and dispersal. Limiting inappropriate anthropogenic development and restoring degraded habitats, would strategically complement these CAP measures, reinforcing the overall network of habitats favourable to threatened plant species. Relevance, limitations, and future directions Although the present analysis provides valuable insights into the conservation of threatened plant species, it does not comprehensively account for all anthropogenic and natural pressures that influence the quality and stability of their habitats. In this context, it is necessary to extend the assessment to additional disturbance factors, particularly those related to agricultural practices and invasive species, including: the burning of vegetation residues for land clearing; application of chemical fertilisers; proliferation of invasive species (e.g., Robinia pseudoacacia , Ailanthus altissima (Mill.) Swingle, Elaeagnus angustifolia L., Eriogaster lanestris ); mechanised mowing with heavy machinery; off-road vehicular traffic; excessive plant harvesting; and vandalism. Investigating these pressures requires a complementary approach that incorporates detailed field assessments and direct consultations with landowners, managers, or custodians of protected areas. Furthermore, future research could evaluate the potential impacts of climate change on the distribution of species of interest, as alterations in temperature and precipitation regimes may significantly affect habitat conditions and the survival capacity of these threatened plant species (C. Wang et al., 2016). Conclusion Species distribution models (SDMs) are fundamental tools in conservation ecology. They provide essential insights into species distributions, identify optimal habitats (HS), and inform protection strategies by rigorously predicting species presence and their environmental relationships. Consequently, these methods contribute significantly to the effective planning of conservation and ecological restoration measures, offering an integrated approach to the complexity of biodiversity. However, our study highlights that the effectiveness of these models in guiding conservation actions could be substantially enhanced by incorporating landscape permeability. Although we identified extensive areas of habitat with high ecological potential, we demonstrated that their actual value is considerably reduced in contexts of low permeability, which constrains essential ecological processes. Therefore, not only the quantity but, more importantly, the functional quality of habitats determines species persistence. The results indicate that large areas of suitable habitat require targeted measures to improve permeability to effectively support the conservation of threatened species. High permeability remains crucial even in ecologically favourable habitats, and interventions should be spatially prioritised based on the overlap between habitat suitability and connectivity. Accordingly, careful management of land use, grazing, and landscape fragmentation is not merely a best-practice recommendation but a strategic priority for the long-term conservation of threatened plant species. The proposed methodology could be replicated in other fragmented landscapes, providing a clear framework for prioritising conservation interventions. Declarations Acknowledgements The authors are grateful for the institutional and financial support received during this research, particularly from the Romanian Ministry of Education and the Research Institute of the University of Bucharest (ICUB), as well as to all those who contributed valuable ideas to the development of this study. Funding This work was supported by the Romanian Ministry of Education through a doctoral scholarship and by the Research Institute of the University of Bucharest (ICUB) through Grant [7420] for the project “Design of Landscape Networks with Machine Learning.” Conflict of Interest The authors declare no potential conflict of interest. References Acharya, P. M., Thainiramit, P., Techato, K., Baral, S., Rimal, N., Savage, M., Campos-Arceiz, A., & Neupane, D. (2023). Predicting the distribution and habitat suitability of the smooth-coated otter ( Lutrogale perspicillata ) in lowland Nepal. Global Ecology and Conservation, 46. https://doi.org/10.1016/j.gecco.2023.e02578 Affolter, D., Abderhalden, W., & Haller, R. (2011). The Continuum Suitability Index Technical Report. https:// www.parcs.ch/snp/pdf_public/3362_affolter_Technical_Report_JECAMI.pdf Babí Almenar, J., Bolowich, A., Elliot, T., Geneletti, D., Sonnemann, G., & Rugani, B. (2019). Assessing habitat loss, fragmentation and ecological connectivity in Luxembourg to support spatial planning. Landscape and Urban Planning, 189, 335–351. https://doi.org/10.1016/j.landurbplan.2019.05.004 Baldwin, R. A. (2009). Use of Maximum Entropy Modeling in Wildlife Research. Entropy, 11(4), 854–866. https://doi.org/10.3390/e11040854 Beckman, N. G., & Sullivan, L. L. (2023). The Causes and Consequences of Seed Dispersal. Annual Review of Ecology, Evolution, and Systematics, 54(1), 403–427. https://doi.org/10.1146/annurev-ecolsys-102320-104739 Bernhardt, K. G., Dostalova, A., Király, G., & Petrova, A. (2011). Echium russicum (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T162105A5538499. Accessed on 21 October 2025. https://www.iucnredlist.org/species/162105/5538499 Bilz, M. (2011a). Klasea lycopifolia . The IUCN Red List of Threatened Species 2011: e.T195348A8957743. https://www.iucnredlist.org/species/195348/8957743Accessed on 21 October 2025. Bilz, M. (2011b). Paeonia tenuifolia (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T165143A5981869. https://www.iucnredlist.org/species/165143/5981869 Accessed on 21 October 2025. Bilz, M., Kell, S. P., Maxted, N., & Lansdown, R. V. (2011). European Red List of Vascular Plants. Boyce, M. S., Vernier, P. R., Nielsen, S. E., & Schmiegelow, F. K. A. (2002). Evaluating resource selection functions. Ecological Modelling, 157(2–3), 281–300. https://doi.org/10.1016/S0304-3800(02)00200-4 Cable, A. B., O’Keefe, J. M., Deppe, J. L., Hohoff, T. C., Taylor, S. J., & Davis, M. A. (2021). Habitat suitability and connectivity modeling reveal priority areas for Indiana bat ( Myotis sodalis ) conservation in a complex habitat mosaic. Landscape Ecology, 36(1), 119–137. https://doi.org/10.1007/s10980-020-01125-2 Cerreta, A. L., McCarthy, K. P., & Fowles, G. (2023). Habitat suitability and landscape connectivity for an expanding population of bobcats. Landscape Ecology, 38(6), 1571–1589. https://doi.org/10.1007/s10980-023-01639-5 Chauhan, S., Ghoshal, S., Kanwal, K. S., Sharma, V., & Ravikanth, G. (2022). Ecological niche modelling for predicting the habitat suitability of endangered tree species Taxus contorta Griff. in Himachal Pradesh (Western Himalayas, India). Tropical Ecology, 63(2), 300–313. https://doi.org/10.1007/s42965-021-00200-2 Chirilă, S. D. (2021). Ecological And Chorological Studies Of The Species Crambe Tataria Sebeók From Romania. Rom. J. Biol. – Plant Biol, Bucharest, 66, 39–54. Chirilă, S. D., Bădărău, A., & Milanovici, S. (2025). Ecology and distribution of Pontechium maculatum in Romania. Comprehensive Plant Biology, 49(1), 23–38. https://doi.org/10.2298/CPB2501023C Chirilă, S. D., Cara, I. G., & Motrescu, I. (2022). Habitat preference of the endangered species Crambe tataria (Brassicaceae) from Romania. Tuexenia, 42, 275–296. https://doi.org/10.14471/2022.42.009 Chirilă, S. D., Doroftei, M., & Gigea, G. (2025). Ecology and distribution of Paeonia tenuifolia species in Romania. Biologia. https://doi.org/10.1007/s11756-025-01902-x Chirilă, S. D., & Kiril, V. (2024). Habitat preference for the populations of the endangered species Pontechium maculatum (Boraginaceae) in Romania. Tuexenia, 44, 131–157. https://doi.org/10.14471/2024.44.005 Corbet, S. A., & Delfosse, E. S. (1984). Honeybees and the nectar of Echium plantagineum L. in southeastern Australia. Australian Journal of Ecology, 9(2), 125–139. https://doi.org/10.1111/j.1442-9993.1984.tb01351.x Costanza, J. K., & Terando, A. J. (2019). Landscape Connectivity Planning for Adaptation to Future Climate and Land-Use Change. Current Landscape Ecology Reports, 4(1), 1–13. https://doi.org/10.1007/s40823-019-0035-2 Craioveanu, C., Muntean, I., Ruprecht, E., Băncilă, R.-I., Crișan, A., & Rákosy, L. (2021). Factors affecting butterfly and plant diversity in basiphilous dry grasslands of Transylvania, Romania. Community Ecology, 22(3), 295–308. https://doi.org/10.1007/s42974-021-00055-6 Cremene, C., Groza, G., Rakosy, L., Schileyko, A. A., Baur, A., Erhardt, A., & Baur, B. (2005). Alterations of Steppe‐Like Grasslands in Eastern Europe: a Threat to Regional Biodiversity Hotspots. Conservation Biology, 19(5), 1606–1618. https://doi.org/10.1111/j.1523-1739.2005.00084.x Cruzan, M. B., & Hendrickson, E. C. (2020). Landscape Genetics of Plants: Challenges and Opportunities. Plant Communications, 1(6), 100100. https://doi.org/10.1016/j.xplc.2020.100100 Da Silva, L. G., Ribeiro, M. C., Hasui, É., Da Costa, C. A., & Da Cunha, R. G. T. (2015). Patch size, functional isolation, visibility and matrix permeability influences neotropical primate occurrence within highly fragmented landscapes. PLoS ONE, 10(2). https://doi.org/10.1371/journal.pone.0114025 Dahlström, A., Iuga, A.-M., & Lennartsson, T. (2013). Managing biodiversity rich hay meadows in the EU: a comparison of Swedish and Romanian grasslands. Environmental Conservation, 40(2), 194–205. https://doi.org/10.1017/S0376892912000458 Dihoru, G., & Dihoru, A. (1994). Plante rare, periclitate şi endemice în flora României. Lista roşie. [Rare, Threatened and Endemic Plants in the Flora of Romania. Red List]. Acta Bot. Hort. Bucurestiensis. Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J. R. G., Gruber, B., Lafourcade, B., Leitão, P. J., Münkemüller, T., McClean, C., Osborne, P. E., Reineking, B., Schröder, B., Skidmore, A. K., Zurell, D., & Lautenbach, S. (2013). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36(1), 27–46. https://doi.org/10.1111/j.1600-0587.2012.07348.x Duan, Y., Bai, H., Du, Z., Liu, Y., Li, L., Lu, K., Wang, Z., Di, L., Qin, Y., & Liu, A. P. M. (2025). Maxent modeling for predicting the potential distribution of Hippophae Linn species. Tropical Ecology, 66(1), 132–145. https://doi.org/10.1007/s42965-025-00372-1 Enyedi, Z. M., Ruprecht, E., & Deák, M. (2008). Long‐term effects of the abandonment of grazing on steppe‐like grasslands. Applied Vegetation Science, 11(1), 55–62. https://doi.org/10.1111/j.1654-109X.2008.tb00204.x Fahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics, 34(1), 487–515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419 Fahrig, L. (2017). Ecological Responses to Habitat Fragmentation Per Se. Annual Review of Ecology, Evolution, and Systematics, 48(1), 1–23. https://doi.org/10.1146/annurev-ecolsys-110316-022612 Fateryga. (2015). Red book of the Republic of Crimea. Plants, algae and fungi. https://www.researchgate.net/publication/291829868_Red_book_of_the_Republic_of_Crimea_Plants_algae_and_fungi_in_Russian Favilli, F., Laner, P., & Bertoncelj, I. (2023). Application of the continuum suitability index (csi) model to display the permeability of the alpine-dinaric landscape and to define intervention priorities for ecological linkages. Biodiversity and Conservation, 32(10), 3237–3254. https://doi.org/10.1007/s10531-023-02651-6 FloraVeg.EU. (2025, September 28). FloraVeg. EU: Information about species. https://floraveg.eu/taxon/ GBIF.org. (2025, February 25). GBIF Occurrence Download. https://doi.org/10.15468/dl.jsgdqy Gîlea, N. A., & Pătru-Stupariu, I. (2025). Environmental variables: key elements in the assessment of landscape permeability — a review. Landscape Research, 1–16. https://doi.org/10.1080/01426397.2025.2502552 Grădinaru, S. R., Fan, P., Iojă, C. I., Niță, M. R., Suditu, B., & Hersperger, A. M. (2020). Impact of national policies on patterns of built-up development: an assessment over three decades. Land Use Policy, 94, 104510. https://doi.org/10.1016/j.landusepol.2020.104510 Guiller, A., Decocq, G., Kichey, T., Poli, P., Vandepitte, K., Dubois, F., Honnay, O., & Closset-Kopp, D. (2023). Spatial genetic structure of two forest plant metapopulations in dynamic agricultural landscapes. Landscape and Urban Planning, 231, 104648. https://doi.org/10.1016/j.landurbplan.2022.104648 Haddad, N. M., Brudvig, L. A., Clobert, J., Davies, K. F., Gonzalez, A., Holt, R. D., Lovejoy, T. E., Sexton, J. O., Austin, M. P., Collins, C. D., Cook, W. M., Damschen, E. I., Ewers, R. M., Foster, B. L., Jenkins, C. N., King, A. J., Laurance, W. F., Levey, D. J., Margules, C. R., Townshend, J. R. (2015). Habitat fragmentation and its lasting impact on Earth’s ecosystems. Science Advances, 1(2). https://doi.org/10.1126/sciadv.1500052 Hadley, A. S., Frey, S. J. K., Robinson, W. D., & Betts, M. G. (2018). Forest fragmentation and loss reduce richness, availability, and specialization in tropical hummingbird communities. Biotropica, 50(1), 74–83. https://doi.org/10.1111/btp.12487 Hansen, A. J., & DeFries, R. (2007). Ecological mechanisms linking protected areas to surrounding lands. Ecological Applications, 17(4), 974–988. https://doi.org/10.1890/05-1098 Hassan, R. M. ., Scholes, R. J. ., & Ash, Neville. (2005). Ecosystems and human well-being : current state and trends : findings of the Condition and Trends Working Group of the Millennium Ecosystem Assessment. Island Press. Henle, K., Davies, K. F., Kleyer, M., Margules, C., & Settele, J. (2004). Predictors of Species Sensitivity to Fragmentation. Biodiversity & Conservation, 13(1), 207–251. https://doi.org/10.1023/B:BIOC.0000004319.91643.9e Hurdu, B.-I., Coste, A., Halmagyi, A., Szatmari, P.-M., Farkas, A., Pușcaș, M., Dan Turtureanu, P., Roșca-Casian, O., Tănase, C., Oprea, A., Mardari, C., Răduțoiu, D., Camen-Comănescu, P., Sîrbu, I.-M., Stoie, A., Lupoae, P., Cristea, V., Jarda, L., Holobiuc, I., Butiuc-Keul, A. (2022). Ex situ conservation of plant diversity in Romania: A synthesis of threatened and endemic taxa. Journal for Nature Conservation, 68, 126211. https://doi.org/10.1016/j.jnc.2022.126211 Jakovljevic, K., Djurovic, S., Antusevic, M., Mihailovic, N., Buzurovic, U., & Tomovic, G. (2019). Heavy metal tolerance of Pontechium maculatum (Boraginaceae) from several ultramafic localities in Serbia. Botanica Serbica, 43(1), 73–83. https://doi.org/10.2298/BOTSERB1901073J Janišová, M., Magnes, M., Bojko, I., Borsukevych, L. M., Budzhak, V. V., Chorney, I., Iuga, A., Ivașcu, C. M., Kish, R., Kuzemko, A., Palpurina, S., Skokanová, K., Širka, P., Tokaryuk, A., & Dayneko, P. (2024). Agricultural legacy shapes plant diversity patterns in mountain grasslands of Maramureș and Bukovina: A cross‐border perspective (Ukraine, Romania). People and Nature, 6(6), 2283–2299. https://doi.org/10.1002/pan3.10698 Kalista, M. (2017). Underutilized Medicinal Species Of Crambe L . Of The Flora Of Ukraine (pp. 216–220). https://doi.org/10.15414/agrobiodiversity.2017.2585-8246.216-220 Kell, S. P. (2011). Crambe tataria (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T162122A5544350. Accessed on 21 October 2025. https://www.iucnredlist.org/species/162122/5544350 Kimberley, A., Hooftman, D., Bullock, J. M., Honnay, O., Krickl, P., Lindgren, J., Plue, J., Poschlod, P., Traveset, A., & Cousins, S. A. O. (2021). Functional rather than structural connectivity explains grassland plant diversity patterns following landscape scale habitat loss. Landscape Ecology, 36(1), 265–280. https://doi.org/10.1007/s10980-020-01138-x Lande, R. (1988). Genetics and Demography in Biological Conservation. Science, 241(4872), 1455–1460. https://doi.org/10.1126/science.3420403 Laner, P., Rossi, C., Luethi, R., Favilli, F., Bertoncelj, I., Plassmann, G., & Haller, R. M. (2024). Landscape permeability for ecological connectivity at the macro-regional level: The Continuum Suitability Index and its practical implications. Ecological Indicators, 164. https://doi.org/10.1016/j.ecolind.2024.112145 Liu, J., Wilson, M., Hu, G., Liu, J., Wu, J., & Yu, M. (2018). How does habitat fragmentation affect the biodiversity and ecosystem functioning relationship? Landscape Ecology, 33(3), 341–352. https://doi.org/10.1007/s10980-018-0620-5 Luo, Y., Wu, J., Wang, X., Wang, Z., & Zhao, Y. (2020). Can policy maintain habitat connectivity under landscape fragmentation? A case study of Shenzhen, China. Science of The Total Environment, 715, 136829. https://doi.org/10.1016/j.scitotenv.2020.136829 Lüthi, R., Rapp, M., & Haller, R. (2018a). Environmental protection indicator ENV (EUSALP). https://www.jecami.eu/static/mapViewer/docu/docu_env_eusalp.pdf Lüthi, R., Rapp, M., & Haller, R. (2018b). Fragmentation indicator FRA (EUSALP) . https://www.jecami.eu/static/mapViewer/docu/docu_fra_eusalp.pdf MADR. (2019). Ghidul solicitantului Măsurii 10 - agro-mediu şi climă - pachetele 7, 8 și 10 - din Programul National De Dezvoltare Rurala (PNDR) 2014 – 2020. [Applicant’s Guide for Measure 10 - agri-environment and climate - packages 7, 8 and 10 - of the National Rural Development Programme (NRDP) 2014-2020]. https://www.madr.ro/docs/dezvoltare-rurala/agro-mediu/2025/Ghid-Masura-10-2025.pdf Magrach, A., Larrinaga, A. R., & Santamaría, L. (2012). Effects of Matrix Characteristics and Interpatch Distance on Functional Connectivity in Fragmented Temperate Rainforests. Conservation Biology, 26(2), 238–247. https://doi.org/10.1111/j.1523-1739.2011.01818.x Martín Arroyo, T., González-Porto, A. V., & Bartolomé Esteban, C. (2017). Viper’s bugloss (Echium spp.) honey typing and establishing the pollen threshold for monofloral honey. PLOS ONE, 12(10), e0185405. https://doi.org/10.1371/journal.pone.0185405 Marușca, T., Mocanu, V., Haș, E. C., Tod, M. A., Andreoiu, A. C., Dragoș, M. M., Blaj, V. A., Ene, T. A., Silistru, D., Ichim, E., Zevedei, P. M., Constantinescu, C. S., & Tod, S. V. (2014). Ghid de intocmire a amenajamentelor pastorale [Guide for the Elaboration of Pastoral Management Plans]. Mathur, M., & Mathur, P. (2025). Habitat suitability modelling of a critically endangered plant species -Commiphora wightii (Arn.) Bhandari: comparative assessment using various machine learning algorithms. Tropical Ecology, 66(2), 181–203. https://doi.org/10.1007/s42965-025-00374-z Meiklejohn, K., Ament, R., & Tabor, G. (2010). Habitat Corridors & Landscape Connectivity: Clarifying the Terminology. www.climateconservation.org Ministry of Environment, W. and F. (2025a). Arii naturale protejate de interes național și internațional. Limite arii naturale protejate [Protected natural areas of national and international interest. Boundaries of Protected Natural Areas]. https://mmediu.ro/portal-gis/date-spatiale/arii-naturale-protejate-de-interes-national-si-international/ Ministry of Environment, W. and F. (2025b, February 25). Arii naturale protejate de interes național și internațional. Distribuția speciilor și habitatelor din planurile de management [Protected natural areas of national and international interest. Distribution of Species and Habitats in Management Plans]. https://mmediu.ro/portal-gis/date-spatiale/arii-naturale-protejate-de-interes-national-si-international/ Muthiuru, A. C., Crego, R. D., Simbauni, J. A., Muruthi, P. M., Waiguchu, G., Lala, F., Millington, J. D. A., & Kairu, E. W. (2024). Human footprint and rainfall shape Masai giraffe’s habitat suitability and connectivity in a multiple‐use landscape. Ecosphere, 15(7). https://doi.org/10.1002/ecs2.4933 Mykhailenko, O., Hsieh, C.-F., El-Shazly, M., Nikishin, A., Kovalyov, V., Shynkarenko, P., Ivanauskas, L., Chen, B.-H., Horng, J.-T., Hwang, T.-L., Georgiyants, V., & Korinek, M. (2023). Anti-viral and Anti-inflammatory Isoflavonoids from Ukrainian Iris aphylla Rhizomes: Structure-Activity Relationship Coupled with ChemGPS-NP Analysis. Planta Medica, 89(11), 1063–1073. https://doi.org/10.1055/a-2063-5265 Mykhailenko, O., Korinek, M., Ivanauskas, L., Bezruk, I., Myhal, A., Petrikaitė, V., El-Shazly, M., Lin, G.-H., Lin, C.-Y., Yen, C.-H., Chen, B.-H., Georgiyants, V., & Hwang, T.-L. (2020). Qualitative and Quantitative Analysis of Ukrainian Iris Species: A Fresh Look on Their Antioxidant Content and Biological Activities. Molecules, 25(19), 4588. https://doi.org/10.3390/molecules25194588 Nowak, B., Sitek, E., & Augustynowicz, J. (2020). Sourcing and Propagation of Pontechium maculatum for Horticulture and Species Restoration. Biology, 9(10), 317. https://doi.org/10.3390/biology9100317 Oltean, M., Negrean, G., Popescu, A., Roman, N., Dihoru, G., & Sanda, V. & M. S. (1994). Studii, sinteze, documentaţii de ecologie. Lista roşie a plantelor superioare din România [Studies, Syntheses, Ecology Documentation. Red List of Higher Plants of Romania]. Academia Română, Institutul de Biologie, București. Oprea, A. (2005). Lista critică a plantelor vasculare din România [Critical List of Vascular Plants of Romania]. Editura Universității „Alexandru Ioan Cuza”, Iaşi. Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190(3–4), 231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026 Phillips, S. J., & Dudík, M. (2008). Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. https://doi.org/10.1111/j.2007.0906-7590.05203.x Phillips, S. J., Dudík, M., & Schapire, R. E. (2004). A Maximum Entropy Approach to Species Distribution Modeling. Phillips, S. J., Dudík, M., & Schapire, R. E. (2025). Maxent software for modeling species niches and distributions (Version 3.4.4). 2025. Plieninger, T., Hartel, T., Martín-López, B., Beaufoy, G., Bergmeier, E., Kirby, K., Montero, M. J., Moreno, G., Oteros-Rozas, E., & Van Uytvanck, J. (2015). Wood-pastures of Europe: Geographic coverage, social–ecological values, conservation management, and policy implications. Biological Conservation, 190, 70–79. https://doi.org/10.1016/j.biocon.2015.05.014 Pykälä, J. (2003). Effects of restoration with cattle grazing on plant species composition and richness of semi-natural grasslands. Biodiversity & Conservation, 12(11), 2211–2226. https://doi.org/10.1023/A:1024558617080 Qi, Y., Zhang, Y., Xue, J., Zhang, Z., Cao, J., Yang, N., Wan, F., Xian, X., & Liu, W. (2025). Future Climate Change Increases the Risk of Suitable Habitats for the Invasive Macrophyte Elodea nuttallii . Biology, 14(5), 504. https://doi.org/10.3390/biology14050504 Radeloff, V. C., Stewart, S. I., Hawbaker, T. J., Gimmi, U., Pidgeon, A. M., Flather, C. H., Hammer, R. B., & Helmers, D. P. (2010). Housing growth in and near United States protected areas limits their conservation value. Proceedings of the National Academy of Sciences, 107(2), 940–945. https://doi.org/10.1073/pnas.0911131107 Remya, K., Ramachandran, A., & Jayakumar, S. (2015). Predicting the current and future suitable habitat distribution of Myristica dactyloides Gaertn. using MaxEnt model in the Eastern Ghats, India. Ecological Engineering, 82, 184–188. https://doi.org/10.1016/j.ecoleng.2015.04.053 Sava, A.-R., Boz, I., & Costică, N. (2019). Contribution to the knowledge of vegetative organs structure at Echium russicum j.f. gmelin. Analele Ştiinţifice Ale Universităţii „Al. I. Cuza” Iaşi s. II a. Biologie Vegetală, 65, 44–54. Scholes, Robert., Montanarella, L. ., Brainich, Anastasia., Barger, Nichole., & Brink, B. ten. (2018). The assessment report on land degradation and restoration : summary for policymakers. Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES). Silveira dos Santos, J., Vitorino, L. C., Fabrega Gonçalves, R., Corrêa Côrtes, M., Souza Cruz Alves, R., Cezar Ribeiro, M., & Garcia Collevatti, R. (2022). Matrix dominance and landscape resistance affect genetic variability and differentiation of an Atlantic Forest pioneer tree. Landscape Ecology, 37(10), 2481–2501. https://doi.org/10.1007/s10980-022-01498-6 Soons, M. B., Messelink, J. H., Jongejans, E., & Heil, G. W. (2005). Habitat fragmentation reduces grassland connectivity for both short‐distance and long‐distance wind‐dispersed forbs. Journal of Ecology, 93(6), 1214–1225. https://doi.org/10.1111/j.1365-2745.2005.01064.x Swiss National Park. (2019). Jecami 2.0. User Guide. ALPBIONET2030 Project. https://www.jecami.eu/static/mapViewer/docu/jecami_userguide.pdf Szitár, K., Tölgyesi, C., Deák, B., Gallé, R., Korányi, D., & Batáry, P. (2023). Connectivity and fragment size drive plant dispersal and persistence traits in forest steppe fragments. Frontiers in Ecology and Evolution, 11. https://doi.org/10.3389/fevo.2023.1155885 Taylor, P. D., Fahrig, L., & With, K. A. (2006). Landscape connectivity: a return to the basics. In Connectivity Conservation (pp. 29–43). Cambridge University Press. https://doi.org/10.1017/CBO9780511754821.003 Török, P., Bullock, J. M., Jiménez‐Alfaro, B., & Sonkoly, J. (2020). The importance of dispersal and species establishment in vegetation dynamics and resilience. Journal of Vegetation Science, 31(6), 935–942. https://doi.org/10.1111/jvs.12958 Török, P., Dembicz, I., Dajić-Stevanović, Z., & Kuzemko, A. (2020). Grasslands of Eastern Europe. In Encyclopedia of the World’s Biomes (pp. 703–713). Elsevier. https://doi.org/10.1016/B978-0-12-409548-9.12042-1 Van Daele, F., Honnay, O., Janssens, S., & De Kort, H. (2024). Habitat fragmentation affects climate adaptation in a forest herb. Journal of Ecology, 112(2), 246–264. https://doi.org/10.1111/1365-2745.14225 Wang, C., Liu, C., Wan, J., & Zhang, Z. (2016). Climate change may threaten habitat suitability of threatened plant species within Chinese nature reserves. PeerJ, 4, e2091. https://doi.org/10.7717/peerj.2091 Wang, T., Li, W., Cui, H., Song, Y., Liu, C., Yan, Q., Wu, Y., Jia, Y., Fang, L., & Qi, L. (2024). Predicting the Potential Habitat Distribution of Relict Plant Davidia involucrata in China Based on the MaxEnt Model. Forests, 15(2), 272. https://doi.org/10.3390/f15020272 Wilson, M. C., Chen, X.-Y., Corlett, R. T., Didham, R. K., Ding, P., Holt, R. D., Holyoak, M., Hu, G., Hughes, A. C., Jiang, L., Laurance, W. F., Liu, J., Pimm, S. L., Robinson, S. K., Russo, S. E., Si, X., Wilcove, D. S., Wu, J., & Yu, M. (2016). Habitat fragmentation and biodiversity conservation: key findings and future challenges. Landscape Ecology, 31(2), 219–227. https://doi.org/10.1007/s10980-015-0312-3 Yan, Y., Jarvie, S., Liu, Q., & Zhang, Q. (2022). Effects of fragmentation on grassland plant diversity depend on the habitat specialization of species. Biological Conservation, 275, 109773. https://doi.org/10.1016/j.biocon.2022.109773 Yezzi, A., Nebbia, A., & Zalba, S. (2023). Grassland Fragmentation: Introduction to the Special Issue. Diversity, 15(4), 489. https://doi.org/10.3390/d15040489 Zanina, M. A., & Smirnova, E. B. (2020). Structure of coenopopulations and phytocoenotic confinement of Paeonia tenuifolia L. in floristic complexes of Oka-Don lowland. Plant Science Today, 7(4). https://doi.org/10.14719/pst.2020.7.4.978 Additional Declarations No competing interests reported. Supplementary Files Supplementarefile1.pdf Supplementarefile2.pdf Supplementarefile3.pdf Supplementarefile4.pdf Supplementarefile5.pdf Cite Share Download PDF Status: Published Journal Publication published 19 Feb, 2026 Read the published version in Landscape Ecology → Version 1 posted Editorial decision: Revision requested 18 Dec, 2025 Reviews received at journal 14 Dec, 2025 Reviews received at journal 01 Dec, 2025 Reviewers agreed at journal 22 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers invited by journal 10 Nov, 2025 Editor assigned by journal 29 Oct, 2025 Submission checks completed at journal 29 Oct, 2025 First submitted to journal 26 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7957482","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":546346031,"identity":"4744093b-ded8-407b-a2f3-d094bfcaa56d","order_by":0,"name":"Nicu Alexandru Gîlea","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIiWNgGAWjYDACZjApIQNkHQCxeAyI1cLDwMCWQKQWKOCBKyaoxZyd+ZkEwy8LHn72M58/MLbdkTGXSH7AXFGBW4tlM5uZBGOfBI9kT+42Cca2ZzyWM9IMGM+cwa3F4DAPmwRjjwSPwYHcbQyMbYd5DG7kMDA2thGhxf78m8cfiNfC8ANoi0QOgwSRWtiMLRIbJHgkbjwzk0g4d5jHsueZwcEGfH45f/jhjQ9/6uT4+5Mff/hQdtjenD354cMGPCEGBokwZyRA6QMENADBH8JKRsEoGAWjYAQDAOTjR+DLTgJLAAAAAElFTkSuQmCC","orcid":"","institution":"University of Bucharest","correspondingAuthor":true,"prefix":"","firstName":"Nicu","middleName":"Alexandru","lastName":"Gîlea","suffix":""},{"id":546346032,"identity":"e8b61302-ac33-4fd8-a705-0faa8e990c02","order_by":1,"name":"Ileana Pătru-Stupariu","email":"","orcid":"","institution":"University of Bucharest","correspondingAuthor":false,"prefix":"","firstName":"Ileana","middleName":"","lastName":"Pătru-Stupariu","suffix":""},{"id":546346033,"identity":"65bf903b-7454-4425-8971-2f8a99b1edcd","order_by":2,"name":"Athanasios Gavrilidis","email":"","orcid":"","institution":"University of Bucharest","correspondingAuthor":false,"prefix":"","firstName":"Athanasios","middleName":"","lastName":"Gavrilidis","suffix":""},{"id":546346036,"identity":"e72462ea-abce-4647-a372-ec4dca94cb5e","order_by":3,"name":"Simona R. Gradinaru","email":"","orcid":"","institution":"Swiss Federal Research Intitute WSL","correspondingAuthor":false,"prefix":"","firstName":"Simona","middleName":"R.","lastName":"Gradinaru","suffix":""},{"id":546346037,"identity":"5bd8f229-a095-4e88-ad34-4216004634a1","order_by":4,"name":"Simona Dumitrița Chirilă","email":"","orcid":"","institution":"Danube Delta National Institute for Research and Development","correspondingAuthor":false,"prefix":"","firstName":"Simona","middleName":"Dumitrița","lastName":"Chirilă","suffix":""},{"id":546346038,"identity":"66e881fe-b8c8-41d8-ba8d-c1b81ae70f5d","order_by":5,"name":"Ana-Maria Florescu","email":"","orcid":"","institution":"University of Bucharest","correspondingAuthor":false,"prefix":"","firstName":"Ana-Maria","middleName":"","lastName":"Florescu","suffix":""},{"id":546346039,"identity":"786d4dfb-d434-443d-838a-7b56fceea9d3","order_by":6,"name":"Andrei Ionuț Mărghescu","email":"","orcid":"","institution":"University of Bucharest","correspondingAuthor":false,"prefix":"","firstName":"Andrei","middleName":"Ionuț","lastName":"Mărghescu","suffix":""}],"badges":[],"createdAt":"2025-10-27 12:12:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7957482/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7957482/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10980-026-02314-1","type":"published","date":"2026-02-19T15:58:39+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":96364985,"identity":"3fdeeb91-9476-4974-944f-9524fd89c2d1","added_by":"auto","created_at":"2025-11-20 10:09:52","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":4519824,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/a515874b910713dc1f65ef66.docx"},{"id":96364393,"identity":"05ffaf8a-e893-4255-a9f7-466a79c5cbdd","added_by":"auto","created_at":"2025-11-20 10:09:15","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8824,"visible":true,"origin":"","legend":"","description":"","filename":"6de13742c4a442c99d9327ab72fdaaf1.json","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/b02805bb6c31f29ac6f0e9cf.json"},{"id":96294018,"identity":"cfedf54b-c124-4c08-8491-276b63903278","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":530873,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/cf03afad25c7b7169dcabb2f.pdf"},{"id":96364509,"identity":"0f530b3a-164a-40bc-8d11-689eea96f1b5","added_by":"auto","created_at":"2025-11-20 10:09:22","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":295700,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/1aca9879cce83d332e1b688a.pdf"},{"id":96365428,"identity":"a94e0fbd-c511-42b4-83c3-c9d13ff223c7","added_by":"auto","created_at":"2025-11-20 10:10:21","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":318222,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/b975d0aa88f48ac771e510bc.pdf"},{"id":96364227,"identity":"ebf4c3d1-ac42-4fb0-afa3-87c3b6478475","added_by":"auto","created_at":"2025-11-20 10:09:04","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":154635,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/8383270d3bc7a748948654ff.pdf"},{"id":96294028,"identity":"d292d7f0-1bf4-4bc8-949f-a2e51f1b857d","added_by":"auto","created_at":"2025-11-19 13:15:09","extension":"pdf","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":198264,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/1c314d4a7dab737cd7a53618.pdf"},{"id":96294005,"identity":"24424701-3fb4-4a3f-bc06-0c2a674ad584","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1189526,"visible":true,"origin":"","legend":"\u003cp\u003eLocation map of the CBRT study area and the distribution of presence points of threatened plant species\u003c/p\u003e\n\u003cp\u003eSources and selection of environmental variables for species distribution modelling (SDM)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/d673604c3ba6bcc4bf6014fc.png"},{"id":96294007,"identity":"db1ec5e7-3894-4623-96a0-f47fe7a07fd0","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":245212,"visible":true,"origin":"","legend":"\u003cp\u003eWorkflow diagram for identifying priority conservation areas\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/440eaad4bbb0c02c92841c7c.png"},{"id":96294004,"identity":"b611437d-0251-4a24-93e5-169f643d3d7e","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":105345,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver Operating Characteristic (ROC) curve for species distribution prediction\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/c1aba4eb2b9ebc3b70937991.png"},{"id":96364490,"identity":"6717584f-1053-47d7-bf89-8179719d11bd","added_by":"auto","created_at":"2025-11-20 10:09:21","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":94884,"visible":true,"origin":"","legend":"\u003cp\u003eJackknife test of regularised training gain used to assess the relative importance of environmental variables in habitat suitability modelling\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/ca7c3a5d3567ff0bf248d004.png"},{"id":96363911,"identity":"c10e0e84-1c05-4a4b-ae46-e4a7fcf6835f","added_by":"auto","created_at":"2025-11-20 10:08:25","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":82200,"visible":true,"origin":"","legend":"\u003cp\u003eResponse curves for the categorical variables of the SDM model: response of species to (a) land use/cover, (b) soil type, and (c) dominant annual average soil water regime class of the soil profile\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/1b32b120595e21fc47584bd4.png"},{"id":96364332,"identity":"d8372742-35dd-43b7-8c07-9f3d15f47a34","added_by":"auto","created_at":"2025-11-20 10:09:13","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":129027,"visible":true,"origin":"","legend":"\u003cp\u003eResponse curves for the continuous variables of the SDM model: response of species to (a) mean temperature of the coldest quarter, (b) precipitation of the driest month, (c) precipitation seasonality, and (d) elevation\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/b38f4020b60fdcf8a20b4583.png"},{"id":96294027,"identity":"11a29086-7f5b-4510-9102-e61143d40ec7","added_by":"auto","created_at":"2025-11-19 13:15:09","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":376969,"visible":true,"origin":"","legend":"\u003cp\u003eHabitat suitability map of the CBRT for threatened plant species\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/0075fe1468e5113e4582044a.png"},{"id":96363790,"identity":"f8bc8f3b-fb4f-42ba-abb1-7ff516b9c30e","added_by":"auto","created_at":"2025-11-20 10:08:03","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":64798,"visible":true,"origin":"","legend":"\u003cp\u003eComparative analysis of the Continuum Suitability Index (CSI) and its component indicators in the investigated plots\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/74f3c932b967bab91f4f5041.png"},{"id":96363801,"identity":"29eb87e5-78af-4305-8bb1-426fcc5d8a89","added_by":"auto","created_at":"2025-11-20 10:08:04","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":23360,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of Habitat Suitability (HS) (a) and Continuum Suitability Index (CSI) (b) in the investigated plots\u003c/p\u003e","description":"","filename":"9.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/8d3afe1cce07d187aa07a41f.png"},{"id":96364391,"identity":"d04a1db8-b40f-417e-9670-0d4e63e9bb12","added_by":"auto","created_at":"2025-11-20 10:09:15","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":285291,"visible":true,"origin":"","legend":"\u003cp\u003eMaps of landscape permeability for the threatened species in the investigated plots\u003c/p\u003e","description":"","filename":"10.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/3da07cd7885896630b170fd5.png"},{"id":96294023,"identity":"c1a210cb-b646-4192-8898-a50e043425cd","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"png","order_by":11,"title":"Figure 11","display":"","copyAsset":false,"role":"figure","size":19860,"visible":true,"origin":"","legend":"\u003cp\u003eDistribution of priority areas in the investigated plots\u003c/p\u003e","description":"","filename":"11.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/dd5901ac4c1faa6e07b8e204.png"},{"id":96365220,"identity":"d2f3dc3c-e831-4ed7-badd-fef1dad8d9db","added_by":"auto","created_at":"2025-11-20 10:10:06","extension":"png","order_by":12,"title":"Figure 12","display":"","copyAsset":false,"role":"figure","size":76795,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Priority Area (km\u003csup\u003e2\u003c/sup\u003e): total and within protected Areas (PAs), across four study plots\u003c/p\u003e","description":"","filename":"12.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/a14d82b21f4c2262ff51f942.png"},{"id":96294025,"identity":"7bc75f66-c072-4dc1-9253-c291d6bccc30","added_by":"auto","created_at":"2025-11-19 13:15:09","extension":"png","order_by":13,"title":"Figure 13","display":"","copyAsset":false,"role":"figure","size":686981,"visible":true,"origin":"","legend":"\u003cp\u003eMaps of priority areas for conservation and ecological restoration overlain with the protected areas network.\u003c/p\u003e","description":"","filename":"13.png","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/41fe2c8cbc84bb1103246632.png"},{"id":103251161,"identity":"0393e400-1fae-4953-8843-2c64d9138928","added_by":"auto","created_at":"2026-02-23 16:05:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3760830,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/9d759930-5ffa-42a7-893c-6cc3aa774856.pdf"},{"id":96364935,"identity":"c0eb73f3-afc8-4d2c-bd07-f91a3358b7fc","added_by":"auto","created_at":"2025-11-20 10:09:48","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":530873,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/caadb569e4391b19b6a61ff3.pdf"},{"id":96294011,"identity":"05a48e10-48c5-4a04-9ba3-c48b6db3fbef","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":295700,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/c4a5538f94341fc5adbd9308.pdf"},{"id":96364433,"identity":"a9a7f9a1-dc92-41d2-8bd8-b63c3725edba","added_by":"auto","created_at":"2025-11-20 10:09:18","extension":"pdf","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":318222,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/f601da8a4461ab140b74cc33.pdf"},{"id":96294020,"identity":"2b0ad0f6-caa4-4afc-b329-8a8afaa554a5","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"pdf","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":154635,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile4.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/876c602351936fda3590f67b.pdf"},{"id":96294015,"identity":"df25a485-bbb5-4998-9de5-0b3b70b991c2","added_by":"auto","created_at":"2025-11-19 13:15:08","extension":"pdf","order_by":5,"title":"","display":"","copyAsset":false,"role":"supplement","size":198264,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarefile5.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7957482/v1/24b9d5424ee4017fd33c5218.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Conserving Red List plant species by managing landscape fragmentation and permeability","fulltext":[{"header":"Introduction","content":"\u003cp\u003eHabitat fragmentation and loss represent critical threats to global biodiversity (Haddad et al., 2015; Scholes et al., 2018). These processes, driven by agricultural expansion, urbanisation, and infrastructure development, reduce the area of natural habitats and disrupt the ecological connectivity of the landscape (Bab\u0026iacute; Almenar et al., 2019; Luo et al., 2020). Simultaneously, climate change forces species to adapt and disperse in an increasingly fragmented landscape (Van Daele et al., 2024). Conservation measures often utilize the ecological permeability of landscapes to address these threats, which are projected to worsen in the future\u0026nbsp;(Costanza \u0026amp; Terando, 2019; Laner et al., 2024).\u0026nbsp;\u003cstrong\u003eSuch measures are\u003c/strong\u003e a key mechanism for restoring landscape connectivity, which is essential for species dispersal and\u0026nbsp;\u003cstrong\u003efor enhancing ecosystem resilience\u003c/strong\u003e.\u0026nbsp;The process of habitat fragmentation, dividing a continuous natural area into smaller, isolated units, significantly affects biodiversity and ecosystem functioning\u0026nbsp;(Haddad et al., 2015; Henle et al., 2004; Liu et al., 2018; Wilson et al., 2016). As fragmentation progresses, it reduces species dispersal and genetic flow, thereby decreasing their capacity to adapt to environmental changes (Fahrig, 2003, 2017). While fragmentation impacts a wide range of species, the effects are more severe for threatened species, which typically have restricted geographic ranges, small populations, and limited dispersal capabilities (Henle et al., 2004). These traits, in turn, increase vulnerability to genetic drift and stochastic events, amplifying the risk of extinction (Lande, 1988). Consequently, the loss of such species has major ecological consequences, affecting the stability of trophic networks and the functioning of ecosystem services (Hassan et al., 2005).\u003c/p\u003e\n\u003cp\u003eAssessments of the Red Lists in Europe show that habitat conversion and population isolation threaten a significant proportion of vascular plants (Bilz et al., 2011). In Central and Eastern Europe, these threats are exacerbated by agricultural conversion, intensive grazing, and land abandonment, which are the main drivers of decline in semi-natural ecosystems (Dahlstr\u0026ouml;m et al., 2013; Enyedi et al., 2008; Plieninger et al., 2015; Yezzi et al., 2023). As a consequence, these processes have caused severe fragmentation of grassland habitats, which, in turn, threatens endemic and relict species. Many of these species are Red List taxa and critically depend on these ecosystems (Cremene et al., 2005; T\u0026ouml;r\u0026ouml;k, Dembicz, et al., 2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThese trends are also strongly reflected in Romania. Threatened species characteristic of steppe and forest-steppe grasslands exhibit a fragmented geographic distribution and are highly sensitive to land-use changes and overgrazing (Chirilă, 2021). Moreover, recent research indicates that a large proportion of these species\u0026apos; populations exist outside protected areas, where they face higher risks (Chirilă, 2021; Hurdu et al., 2022). Consequently, studies report pronounced population declines and increasing isolation (Chirilă, 2021; Chirilă, Bădărău, et al., 2025; Chirilă, Doroftei, et al., 2025). This isolation increases the risk of local extinction and negatively affects ecological balance and ecosystem services. Ultimately, the loss of these species disrupts trophic networks and ecosystem processes, generating imbalances that may accelerate habitat degradation (Bilz et al., 2011; Kalista, 2017; Mykhailenko et al., 2020, 2023; Nowak et al., 2020).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo support the viability and genetic diversity of these populations, habitat conservation is essential. However, as landscape fragmentation increases, protecting the remaining habitats alone is not enough. Conservation efforts must also enhance the landscape matrix\u0026apos;s permeability (Da Silva et al., 2015; Guiller et al., 2023; Hadley et al., 2018; Silveira dos Santos et al., 2022) and restore ecological connectivity (Kimberley et al., 2021; Magrach et al., 2012; Szit\u0026aacute;r et al., 2023).\u0026nbsp;Landscape connectivity characterizes the capacity of species to move between habitat fragments via corridors and linkage areas. In contrast, permeability refers to how favourable the landscape matrix\u0026mdash;composed of natural, semi-natural, and anthropogenic land-cover types\u0026mdash;is for species dispersal and essential ecological processes (Meiklejohn et al., 2010; Taylor et al., 2006). For plants, permeability enables the flow of propagules\u0026mdash;pollen, seeds, spores\u0026mdash;between populations. This flow maintains genetic diversity, enables colonisation, and supports reproduction. A permeable landscape promotes gene flow, population resilience, and long-term persistence, which reduces extinction risk (Beckman \u0026amp; Sullivan, 2023; Cruzan \u0026amp; Hendrickson, 2020; T\u0026ouml;r\u0026ouml;k, Bullock, et al., 2020).\u003c/p\u003e\n\u003cp\u003eFor effective conservation planning based on the principle of increasing landscape permeability, it is necessary to identify suitable habitats and assess their accessibility to the target species. In this context, species distribution models (SDMs) provide an essential methodological framework for mapping and evaluating potential habitat suitability. These tools, particularly the MaxEnt algorithm, are recognised for their efficiency in identifying areas with optimal ecological conditions for species (Baldwin, 2009; Remya et al., 2015). This forms the scientific basis for numerous conservation interventions (Chauhan et al., 2022; Duan et al., 2025). Most studies focus on identifying bioclimatically suitable habitats (Duan et al., 2025; Mathur \u0026amp; Mathur, 2025; T. Wang et al., 2024). Although useful, this approach does not provide insight into the actual accessibility of favourable habitats for species. Some suitable habitats may be harder to colonise due to landscape barriers, even if bioclimatic conditions are favourable. Moreover, the practice of unsustainable agricultural activities (e.g., overgrazing, burning, and mechanised mowing) increases habitat hostility for target species. In this context, landscape permeability becomes a key variable in identifying priority areas for conservation and restoration (Favilli et al., 2023). The approach based on species distribution modelling (SDM) and landscape permeability is a method validated in numerous studies, particularly on mammals (Acharya et al., 2023; Cerreta et al., 2023; Muthiuru et al., 2024). However, studies assessing landscape permeability specifically for plants are still relatively scarce (G\u0026icirc;lea \u0026amp; Pătru-Stupariu, 2025). Here, we apply this combined framework to define targeted conservation measures for threatened plant species listed on Romania\u0026apos;s Red List, within a fragmented, multi-use landscape.\u003c/p\u003e\n\u003cp\u003eIn this study, we aim to propose an approach to define spatially explicit conservation measures for threatened plant species listed on the Red List, based on the management of fragmentation and landscape permeability. We focus on the spatial analysis of habitats and landscape permeability, adopting a multispecies approach and using Romania as a case study.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe research questions we address are: (1) how much of the habitat considered suitable is permeable for the threatened species? (2) to which extent are the potentially suitable habitats covered by the current network of protected areas (PAs). To answer these research questions we first identify potentially suitable habitats for five target plant species using SDM modelling. Second, we assess landscape permeability using key environmental variables and finally we identify priority areas for the effective conservation of threatened plant species both within and outside the existing network of protected areas (PAs). We expect that integrating landscape permeability into distribution models increases the ecological relevance of potentially suitable habitats for threatened plant species. By integrating spatial and ecological data, the study provides essential scientific support for the formulation of clearly delineated, sustainable conservation measures that are adapted to regional specifics and capable of addressing current challenges in protecting plant biodiversity.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eStudy Area\u003c/p\u003e\n\u003cp\u003eThe study was conducted in the Continental Biogeographic Region of Transylvania (CBRT), located in central Romania, covering an area of 34,289 km\u0026sup2; (Fig. 1). CBRT is characterised by a moderate continental climate, with altitudinal and soil variations that influence the distribution of species sensitive to habitat changes. The region includes semi-natural habitats of particularly high ecological value, including priority habitats (according to the EU Habitats Directive) such as xerophilous steppe and forest-steppe grasslands, that host rare and vulnerable plant species, which are affected by fragmentation and overgrazing (Chirilă, 2021; Chirilă, Doroftei, et al., 2025; Chirilă \u0026amp; Kiril, 2024). Among the most important habitats are\u003cem\u003e\u0026nbsp;Pannonian subcontinental steppe grasslands\u0026nbsp;\u003c/em\u003e(6240*), \u003cem\u003esemi-natural mesoxerophilous grasslands\u003c/em\u003e (6210*), \u003cem\u003ePontic-Sarmatic steppe grasslands\u003c/em\u003e (62C0*), and \u003cem\u003ePannonian loess steppe grasslands\u003c/em\u003e (6250*). The landscape is dominated by agricultural lands, deciduous forests, pastures, and dispersed urban areas, (Grădinaru et al., 2020). Additionally, 6,884 km\u0026sup2; (20.07%) of the region are included in protected natural areas (Natura 2000 and others), providing essential refuges for biodiversity (Ministry of Environment, 2025a).\u003c/p\u003e\n\u003cp\u003eTarget Red List plant species\u003c/p\u003e\n\u003cp\u003eFive threatened plant taxa of high conservation value are associated with semi-natural habitats in the CBRT: \u003cem\u003eCrambe tataria\u0026nbsp;\u003c/em\u003eSebe\u0026oacute;k, \u003cem\u003ePontechium maculatum\u0026nbsp;\u003c/em\u003e(L.) B\u0026ouml;hle \u0026amp; Hilger, \u003cem\u003eIris aphylla L.\u003c/em\u003e, \u003cem\u003eKlasea lycopifolia (Vill.) \u0026Aacute;. L\u0026ouml;ve \u0026amp; D. L\u0026ouml;ve\u0026nbsp;\u003c/em\u003eand \u003cem\u003ePaeonia tenuifolia\u003c/em\u003e\u003cem\u003e\u0026nbsp;\u003c/em\u003eL.. These species are characteristic of xerophile grasslands (steppe and forest-steppe), preferring well-drained soils and a continental climate. They are sensitive to habitat degradation, and are affected by anthropogenic activities such as overgrazing or the conversion of land to agricultural or urban areas (Chirilă et al., 2022; Sava et al., 2019). The five target species are included in the national Red List and are considered vulnerable and rare (Dihoru \u0026amp; Dihoru, 1994; Oltean et al., 1994; Oprea, 2005), with the exception of \u003cem\u003eP. tenuifolia\u003c/em\u003e which is classified as endangered (Dihoru \u0026amp; Dihoru, 1994). They are protected under national and European legislation, being listed in Annexes II and IV of the Habitats Directive and in Annex I of the Bern Convention (Bilz et al., 2011).\u003c/p\u003e\n\u003cp\u003eThe species \u003cem\u003eC. tataria\u003c/em\u003e and \u003cem\u003eP. maculatum\u003c/em\u003e are valuable indicators of steppe grassland conservation status. Both are melliferous plants that support pollinator populations and thus contribute to maintenance of local biodiversity (Corbet \u0026amp; Delfosse, 1984; Mart\u0026iacute;n Arroyo et al., 2017). \u003cem\u003eC. tataria\u0026nbsp;\u003c/em\u003eis related to agricultural crops and has potential applications in animal feed, biodiesel production, and phytomedicine\u003cem\u003e\u0026nbsp;\u003c/em\u003e(Bilz et al., 2011; Kalista, 2017). \u003cem\u003eP. maculatum\u0026nbsp;\u003c/em\u003eis valued for its ornamental, sanitary, and medicinal properties (Nowak et al., 2020). Additionally, the species demonstrates tolerance to heavy metals, indicating potential for phytoremediation and adaptability to harsh edaphic conditions (Jakovljevic et al., 2019). \u003cem\u003eP. tenuifolia\u003c/em\u003e is notable for its striking flowers and associated ornamental and medicinal value (Fateryga, 2015; Zanina \u0026amp; Smirnova, 2020). \u003cem\u003eI. aphylla\u003c/em\u003e exhibits promising anti-inflammatory, antioxidant, and antiviral activities, with potential applications in treatments against influenza and enteroviral infections (Mykhailenko et al., 2020, 2023).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe vulnerability of these species is amplified by their limited dispersal strategies. \u003cem\u003eC. tataria\u003c/em\u003e is anemochorous (tumbleweed, mean dispersal distance: 40\u0026ndash;150 m), while the remaining taxa have unspecialised dispersal, predominantly over short distances (namely ballochory, blastochory, boleochory; mean dispersal distance: 0.1\u0026ndash;5 m) (FloraVeg.EU, 2025). This limited colonisation capacity reduces their potential to respond to disturbances, highlighting the importance of habitat conservation and connectivity for the maintenance of populations.\u003c/p\u003e\n\u003cp\u003eSpecies presence data\u003c/p\u003e\n\u003cp\u003eWe constructed a database of presence points for the threatened plant species by combining information from multiple sources. First, we included official presence data (n = 1,098) obtained from the Romanian Ministry of Environment 2025b), restricted to protected areas. These data were collected through direct field observations and monitoring by biologists and ecologists between 2007 and 2017. Second, we supplemented the database with our own presence points (n = 247), collected through yearly field observations between April and August during the 2017\u0026ndash;2025 period. We recorded GPS locations of individual plants using the OsmAnd application (v.4.7.17) (https://osmand.net/). Third, we accessed the Global Biodiversity Information Facility (GBIF) platform (https://doi.org/10.15468/dl.jsgdqy) and retrieved presence records of threatened species from the CBRT (n = 74), collected between 1816 and 2025 (GBIF.org, 2025). The resulting database comprises 1,419 presence points, of which 1,115 are located within protected areas and 304 outside them. The distribution of points by species is as follows: \u003cem\u003eC. tataria\u003c/em\u003e \u0026ndash; 813, \u003cem\u003eP. maculatum\u003c/em\u003e \u0026ndash; 354, \u003cem\u003eI. aphylla\u003c/em\u003e \u0026ndash; 178, \u003cem\u003eK. lycopifolia\u003c/em\u003e \u0026ndash; 60, and \u003cem\u003eP. tenuifolia\u003c/em\u003e \u0026ndash; 14.\u003c/p\u003e\n\u003cp\u003eTo mitigate the spatial sampling bias caused by the uneven distribution of presence points, we applied a spatial thinning procedure\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e(Duan et al., 2025). This adjustment is essential to avoid SDM overfitting and to improve prediction accuracy. Presence points were filtered so that only those separated by a minimum distance of 30 m were retained. The procedure was carried out in ArcGIS Pro v3.5.1, resulting in a final set of 483 points distributed as uniformly as possible (Fig. 1). This final dataset was used for modelling the potential habitat of the threatened species.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003cimg src=\"data:image/png;base64,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\" alt=\"image\" width=\"764\" height=\"50\"\u003e\u003c/p\u003e\n\u003cp\u003eA total of 25 environmental variables, selected based on the reviewed literature, were used to predict the potentially suitable habitat of the threatened species. They cover bioclimatic condictions, topography, soil and land use. Variable and sources of data are decribed in Table 1. Raster layers were clipped to the boundaries of the CBRT and reprojected into the same coordinate system. To ensure compatibility between layers and with the MaxEnt model, all variables were rescaled to a common resolution of 30 \u0026times; 30 m, using bilinear resampling for continuous variables (e.g., elevation) and nearest-neighbour resampling for categorical variables (e.g., land use/cover). The layers were exported in ASCII format for use in MaxEnt, using ArcGIS Pro v3.5.1.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo avoid collinearity and obtain a robust set of explanatory variables, selection was carried out in two steps (Duan et al., 2025). First, an initial MaxEnt model evaluated the contribution and importance of each variable, and a jackknife test identified variables with significant unique information. In the second step, a Pearson correlation matrix was calculated among all variables using the cor function in R (v. 4.5.0), and strongly correlated pairs (|r| \u0026ge; 0.7) were removed (Dormann et al., 2013) (see Online Resource 1). Following the application of these criteria, the final set of variables used in the MaxEnt model included: \u003cem\u003eSoil, LULC, WR, Bio15, Bio14, Elevation\u003c/em\u003e, and \u003cem\u003eBio11\u003c/em\u003e, selected for their relevant contribution and lack of excessive collinearity.\u003c/p\u003e\n\u003cp\u003eTable 1. Variables used for SDM modelling\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 94px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariables\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCode\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 240px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eDescription\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 81px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eResolution\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 119px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"19\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eBioclimatic\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAnnual mean temperature\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"19\" valign=\"top\"\u003e\n \u003cp\u003e1 km\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"19\" valign=\"top\"\u003e\n \u003cp\u003eWorldClim v2.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean diurnal range\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eIsothermality\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eTemperature seasonality\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMaximum temperature of the warmest month\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMin temperature of the coldest month\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRange of annual temperature variation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean temperature of the wettest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean temperature of driest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean temperature of warmest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMean temperature of coldest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAnnual precipitation\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of wettest month\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of driest month\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation seasonality\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of wettest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of driest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of warmest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eBio19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003ePrecipitation of coldest quarter\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eTopographic\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eElevation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eElevation\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003e25 m\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"3\" valign=\"top\"\u003e\n \u003cp\u003eEU-DEM v1.1 (Copernicus)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSlope\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSlope\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAspect\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAspect\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003eSoil\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSoil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSoil class\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e1 km\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eEuropean Soil Database v2 (JRC)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eWR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDominant annual\u0026nbsp;average soil water regime class of the soil profile\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLand use/cover\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLULC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLand cover/use classes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e100 m\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCopernicus Land Monitoring Service\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIdentifying potentially suitable habitat (HS) using the SDM\u003c/p\u003e\n\u003cp\u003eWe implemented a maximum entropy modelling approach using the MaxEnt software (version 3.4.4) to identify the potential habitat of threatened plant species in the CBRT (Phillips et al., 2006, 2025). The five species were included in a single model due to their similar ecological requirements, which allowed the use of the same environmental parameters to identify areas favourable for conservation.\u003c/p\u003e\n\u003cp\u003eThe MaxEnt model, built using seven environmental variables, was run with 10 replicates and with the cloglog output format, generating ASCII files. To evaluate the robustness of the model and reduce the risk of overfitting, we applied a 10-fold cross-validation scheme, which provides a more realistic estimation of model performance on datasets not used for training. Only linear and quadratic features were used (Phillips, et al., 2004). Predictions were expressed in logistic format, with values ranging from 0 to 1, indicating the probability that each raster cell represents a suitable habitat (Phillips \u0026amp; Dud\u0026iacute;k, 2008). Default software settings were retained, except where specified above. To assess the importance of environmental variables and their influence on species distribution, we applied the Jackknife test and analysed response curves. Overall model accuracy was evaluated using the area under the ROC curve (AUC), with values ranging from 0 (random prediction) to 1 (perfect prediction) \u0026nbsp;(Duan et al., 2025; Phillips \u0026amp; Dud\u0026iacute;k, 2008).\u003c/p\u003e\n\u003cp\u003eBased on the SDM, we determined habitat suitability (HS), used to interpret and classify the degree of habitat favourability within the CBRT. For this, we loaded the resulting file with the suffix \u0026ldquo;average\u0026rdquo; into ArcGIS Pro v3.5.1, which contains the spatial distribution of species occurrence probability in cloglog format (values between 0 and 1). The decision threshold used to separate suitable from unsuitable habitats was the maximum sensitivity-specificity cloglog threshold (\u003cem\u003ep = 0.2\u003c/em\u003e), derived from ROC analysis (Qi et al., 2025). We then applied a raster reclassification process using the Reclassify tool to transform the continuous raster into a categorical raster, reflecting distinct levels of ecological suitability. Areas were classified into four categories (Acharya et al., 2023; Duan et al., 2025) as follows: \u0026lt;0.2 \u0026mdash; unsuitable, 0.2\u0026ndash;0.4 \u0026mdash; low suitability, 0.4\u0026ndash;0.6 \u0026mdash; medium suitability, \u0026ge;0.6 \u0026mdash; high suitability.\u003c/p\u003e\n\u003cp\u003eAssessing landscape permeability for threatened plant species\u003c/p\u003e\n\u003cp\u003eTo assess landscape permeability, we applied the Continuum Suitability Index (CSI) (Affolter et al., 2011; Favilli et al., 2023; Laner et al., 2024), adapted to the regional context and the ecological requirements of the analysed species. This method estimates permeability based on the landscape\u0026apos;s physical characteristics and the level of anthropogenic pressure (Swiss National Park, 2019). CSI operates through a weighted multicriteria overlay analysis. In this method, relevant environmental variables are assigned predefined scores. These scores indicate the degree of permeability of each landscape element (Laner et al., 2024).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn this study, we used seven environmental variables (Table 2), selected and weighted based on the reviewed literature (Chirilă, 2021; G\u0026icirc;lea \u0026amp; Pătru-Stupariu, 2025; Laner et al., 2024). Permeability scores were assigned on a scale from 0 (low permeability) to 10 (high permeability) using a mixed approach that combined statistical methods, empirical data, and literature information.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSpecifically, for the \u003cem\u003eLULC, Soil, Aspect\u003c/em\u003e, and \u003cem\u003eSlope\u003c/em\u003e variables, permeability scores were calculated using Resource Selection Functions (RSF), following the principles of Boyce et al. (2002). These compare the environmental characteristics at species presence points with those at randomly generated background points. For each habitat class, a selection ratio (\u003cem\u003ewi\u003c/em\u003e) was calculated, with preferred classes receiving high scores, indicating high permeability, while avoided classes received low scores. The resulting values were subsequently validated and calibrated using additional literature data on species ecological requirements (see Online Resource 2).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTo estimate permeability scores based on livestock stocking rate (STK) (LU/ha), a Gaussian (normal) function was used, centred on the range considered optimal for grazing, 0.4\u0026ndash;0.6 LU/ha (Marușca et al., 2014), corresponding to the maximum permeability value (see Online Resource 3). Permeability values attributed to fragmentation and protected areas were obtained from previous studies (Laner et al., 2024; L\u0026uuml;thi et al., 2018a, 2018b).\u003c/p\u003e\n\u003cp\u003eAll variables were converted to rasters at a fine resolution of 5 \u0026times; 5 m, sufficiently detailed to capture small landscape elements (e.g., roads) that influence permeability for the target species. These were reclassified according to the established permeability scores (see Online Resource 4) and integrated into a weighted mean. The calculation was performed in ArcGIS Pro v3.5.1 using the \u003cem\u003eRaster Calculator\u003c/em\u003e tool and the following formula:\u003c/p\u003e\n\u003cp\u003ePermeability was assessed in four regional windows (Plots), each covering 600 km\u0026sup2;. These windows were selected to encompass different levels of potentially suitable habitat for the target species. This approach reduced processing time for the large data volume and allowed for results that are more applicable at the local scale. At the same time, it maintained consistency in the landscape-scale assessment.\u003c/p\u003e\n\u003cp\u003eTable 2. Environmental variables for assessing landscape permeability\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eEnvironmental variable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviation\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSource\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eResolution / Scale\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLand use/cover \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLAN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eCorine Land Cover Plus Backbone + APIA data (Agency for Payments and Intervention in Agriculture) + OpenStreetMap\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e10 m\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eHabitat fragmentation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eFRA\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eDerived from Land Use\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1 \u0026times; 1 km (grid)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eGrazing pressure /Livestock density\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSTK\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eLocal Authorities + National Institute of Statistics (TEMPOOnline)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 m\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSoil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSOIL\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eRomanian Soil Database\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1:200.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSlope\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eSLOPE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003eEU-DEM v1.1 \u0026ndash; Copernicus\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" valign=\"top\"\u003e\n \u003cp\u003e25 m\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eAspect\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eASPECT\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eProtected Areas\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eENV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003eMinistry of Environment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5 m\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eIdentifying priority areas for conservation and ecological restoration\u003c/p\u003e\n\u003cp\u003ePriority areas for conservation were identified through the spatial overlay of Habitat Suitability (HS) and the Continuum Suitability Index (CSI) within the four plots (Fig. 2). This approach allowed the identification not only of optimal habitats but also of areas with ecological potential, which can be enhanced through targeted management of landscape permeability.\u003c/p\u003e\n\u003cp\u003eTo implement this strategy and facilitate ecological interpretation, the HS and CSI rasters were first reclassified into binary form. For HS, all cells with values \u0026ge; 0.2 were considered suitable and coded as 1, while values \u0026lt; 0.2 were coded as 0. The threshold used to separate suitable from unsuitable habitats was the maximum sensitivity-specificity cloglog threshold (\u003cem\u003ep = 0.2\u003c/em\u003e), derived from ROC analysis.\u003c/p\u003e\n\u003cp\u003eFor CSI, the classification threshold was set at 6.0, justified by the observation that approximately 71% of species presence records occurred in areas with CSI \u0026ge; 6 (see Online Resource 5). Accordingly, cells with CSI \u0026ge; 6 were coded as 1 (high permeability), and cells \u0026lt; 6 were coded as 0 (low permeability).\u003c/p\u003e\n\u003cp\u003eBy applying a conditional function using the \u003cem\u003eRaster Calculator\u003c/em\u003e in ArcGIS Pro (v3.5.1), four spatial categories relevant for planning ecological interventions were identified:\u003c/p\u003e\n\u003col start=\"1\" type=\"1\"\u003e\n \u003cli\u003e\u003cem\u003eConservation areas:\u0026nbsp;\u003c/em\u003ecells with HS \u0026ge; 0.2 and CSI \u0026ge; 6 (coded 1/1). High-quality, permeable habitats ideal for direct conservation and active management.\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eRestoration areas:\u0026nbsp;\u003c/em\u003ecells with HS \u0026ge; 0.2 and CSI \u0026lt; 6 (coded 1/0). Suitable habitats degraded by fragmentation or other pressures, requiring interventions to improve permeability\u003cem\u003e.\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eDispersal/colonisation potential areas:\u0026nbsp;\u003c/em\u003ecells with HS \u0026lt; 0.2 and CSI \u0026ge; 6 (coded 0/1). Permeable areas where habitat can be restored or that can facilitate natural dispersal, recommended as buffer zones\u003cem\u003e.\u003c/em\u003e\u003c/li\u003e\n \u003cli\u003e\u003cem\u003eUnsuitable areas:\u0026nbsp;\u003c/em\u003ecells with HS \u0026lt; 0.2 and CSI \u0026lt; 6 (coded 0/0). Low suitability and permeability; low priority for immediate intervention but potential targets for long-term permeability enhancement.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Results","content":"\u003cp\u003eSDM performance and variable contributions\u003c/p\u003e\n\u003cp\u003eThe potential habitat distribution model for threatened species demonstrated excellent predictive performance (AUC \u0026ge; 0.9), with a mean test AUC of 0.900 (Fig. 3) and very little variation between replicates (standard deviation, SD = 0.011).\u003c/p\u003e\n\u003cp\u003eThe analysis of variable contributions revealed that the model is primarily driven by edaphic factors (Soil) and land use/cover (LULC) (Table 3). According to the percentage contribution metric, the most important predictors were soil (44.8%), followed by land use/cover (LULC, 26.8%) and the annual soil water regime (wr, 9.3%). In contrast, the Permutation Importance metric indicated a relatively high importance of the variables reflecting precipitation of the driest month (Bio14, 27.7%) and the mean temperature of the coldest quarter (Bio11, 14.4%), suggesting that these climatic variables contain critical predictive information. The Jackknife test (Fig. 4) confirmed these results, showing that soil is the variable with the highest individual predictive power, while land use/cover holds the most unique information not captured by the other predictors.\u003c/p\u003e\n\u003cp\u003eTable 3. Estimates of relative contribution and permutation importance for environmental predictors\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eVariable\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePercent contribution\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ePermutation importance\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eSoil\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e44.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e19.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eLULC\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e26.8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e20.6\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eWR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e9.3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e6.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eBio15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e8.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eBio11\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e6.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e14.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eElevation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e11.4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 75px;\"\u003e\n \u003cp\u003eBio14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 170px;\"\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 189px;\"\u003e\n \u003cp\u003e27.7\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eThe analysis of the response curves (Fig. 5, 6) provides detailed insight into the species\u0026apos; ecological niche. The land use/cover variable (LULC) (Fig. 5a) shows a probability of presence reaching maximum values (above 0.90) for classes 2 (grasslands), 7 (bare or sparsely vegetated areas), and 8 (shrublands). In contrast, the probability of presence is significantly lower in classes 1 (deciduous forests) and 3 (arable land). The species exhibit a clear preference for Mollisols and Vertisols (classes 2 and 11), whereas hydromorphic soils (class 3) are the least suitable (Fig. 5b). The analysis of the marginal response curve for the annual soil water regime (WR) illustrates a strong ecological preference (above 0.93) for well-drained or dry soil conditions (class 1) (Fig. 5c).\u003c/p\u003e\n\u003cp\u003eClimatic and topographic factors define the species\u0026apos; tolerance limits. For Bio11 (Fig. 6a), the response curve shows a peak probability of presence\u0026mdash;almost 0.95\u0026mdash;within the optimal temperature range of approximately \u0026minus;2.5\u0026deg;C to \u0026minus;1.5\u0026deg;C. In the case of Bio14 (Fig. 6b), the maximum probability of presence, nearly 1.0, is observed at values below 25 mm; this probability drops rapidly to 0.0 when monthly precipitation exceeds 40 mm. When examining Bio15 (Fig. 6c), areas with low values (\u0026lt;30 mm) exhibit the highest probability, almost 0.95. With elevation (Fig. 6d), the probability of presence climbs above 0.98 at roughly 600 metres and remains high at greater elevations.\u003c/p\u003e\n\u003cp\u003eHabitat suitability for threatened plant species in the CBRT\u003c/p\u003e\n\u003cp\u003eIn the CBRT, 21.5% of the territory (\u0026asymp;7,357.4 km\u0026sup2;) provides potential habitat for threatened plant species (HS \u0026ge; 0.2). Of this habitat, 22.0% (\u0026asymp;1,584 km\u0026sup2;) lies within protected natural areas. The other 78.0% (\u0026asymp;5,764.4 km\u0026sup2;) is outside them (Table 4). Regarding suitability, 5.4% (\u0026asymp;1,854.9 km\u0026sup2;) of the territory has high suitability, 5.5% has medium suitability, and 10.5% has low suitability. The remaining 78.5% does not provide favourable conditions and is considered unsuitable.\u003c/p\u003e\n\u003cp\u003eFavourable habitat is mainly in the western and central-southern region (Fig. 7), while non-suitable areas are mostly in the north, southwest, and east.\u003c/p\u003e\n\u003cp\u003eTable 4. Habitat suitability in the CBRT and within protected natural areas\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHabitat Suitability\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 103px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCBRT\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 110px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eWithin PAs\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 104px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eOutside PAs\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ekm\u0026sup2;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ekm\u0026sup2;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ekm\u0026sup2;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e%\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003eSuitable HS \u0026ge; 0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e7.357,4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e21,5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e1.584\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e22,0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e5.764,4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e78,0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003eHigh suitability HS \u0026ge; 0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e1.854,9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e5,4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e371,8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e20,0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e1483,1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e80,0\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003eMedium suitability HS 0.4\u0026ndash;0.6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e1.887,5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e5,5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e466,8\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e24,7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e1420,74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e75,3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003eLow suitability HS 0.2\u0026ndash;0.4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e3.615,0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e10,6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e745,4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e20,6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e2860,6\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e79,4\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 208px;\"\u003e\n \u003cp\u003eUnsuitable HS \u0026le; 0.2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 61px;\"\u003e\n \u003cp\u003e26.914,9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 42px;\"\u003e\n \u003cp\u003e78,5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 57px;\"\u003e\n \u003cp\u003e5.294,9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 54px;\"\u003e\n \u003cp\u003e19,7\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 64px;\"\u003e\n \u003cp\u003e21620,1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 40px;\"\u003e\n \u003cp\u003e80,3\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLandscape permeability and habitat suitability in the investigated plots\u003c/p\u003e\n\u003cp\u003eThe mean CSI value varies among the four plots, ranging from 4.8 in Plot 2 to 5.4 in Plot 1. Variables associated with topographic conditions, such as slope (SLOPE) and aspect (ASPECT), along with soil class (SOIL), exhibit high mean values, ranging from 7.9 to 9.1. In contrast, variables reflecting landscape fragmentation (FRA), grazing pressure (STK), land use (LAN), and the presence of protected areas (ENV) show low mean values, ranging from 0.3 to 5.9 (Fig. 8). The landscape permeability map (Fig. 10) provides an integrated representation of CSI value distribution across the four plots, highlighting contrasts between areas of high ecological connectivity and fragmented zones affected by anthropogenic pressure.\u003c/p\u003e\n\u003cp\u003eThe area covered by suitable habitat (HS \u0026ge; 0.2) varies among the four plots, from \u0026asymp;241.4 km\u0026sup2; in Plot 4, where the lowest value was recorded, to \u0026asymp;435.2 km\u0026sup2; in Plot 2, with the largest extent of suitable habitat (Fig. 9a). Areas with CSI \u0026le; 6 dominate all four plots, with the highest extent in Plot 2 (\u0026asymp;474.8 km\u0026sup2;). Areas with CSI \u0026ge; 6 are the largest in Plot 1, followed by Plots 4 and 3 (Fig. 9b).\u003c/p\u003e\n\u003cp\u003ePriority areas for conservation and ecological restoration\u003c/p\u003e\n\u003cp\u003eFour types of priority areas for the conservation of threatened species were identified, corresponding to varying levels of habitat suitability and landscape permeability: \u003cem\u003econservation, restoration, dispersal/colonisation, and unsuitable areas\u003c/em\u003e. The distribution of these zones varies among the investigated plots, reflecting local-scale differences (Fig. 13). Specifically, the largest conservation area was identified in Plot 1 (\u0026asymp;196.8 km\u0026sup2;), while the largest restoration area was in Plot 2 (\u0026asymp;321.4 km\u0026sup2;). In addition, Plot 4 contains the area with the highest dispersal potential, totaling \u0026asymp;85 km\u0026sup2;. Significant areas of unsuitable zones were identified in Plot 3 (\u0026asymp;265.5 km\u0026sup2;) and Plot 4 (\u0026asymp;263.2 km\u0026sup2;) (Fig. 11).\u003c/p\u003e\n\u003cp\u003eFrom the perspective of overlap with protected areas (Fig. 12, 13), we found that Plot 1 and Plot 3 show the largest areas of conservation and colonisation/dispersion included within protected areas, whereas in Plot 2 and Plot 4, these zones are less well covered by protected areas. In contrast, restoration zones are largely located outside protected areas, a pattern consistent across all four plots (Plot 1\u0026ndash;Plot 4). This suggests that restoration activities should be prioritised in unprotected areas, where anthropogenic pressure is likely to be higher.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur findings, obtained through habitat distribution modelling and landscape permeability analysis, highlight critical areas for the survival and dispersal of threatened plant species. The analysis indicates that a significant portion of suitable habitat lies outside protected areas and is poorly permeable to the target species. This underscores the need for urgent interventions in targeted areas of the landscape. Such measures can include maintaining optimal grazing, supporting the traditional agro-pastoral mosaic, and reducing habitat fragmentation in priority conservation areas. These actions are essential to ensure the survival of existing populations and to facilitate dispersal and colonisation of new habitats. Overall, our findings emphasise the crucial role of managing landscape permeability in conserving threatened plant species, particularly in highly fragmented regions outside protected areas.\u003c/p\u003e\n\u003cp\u003ePotential habitat distribution of threatened plant species\u003c/p\u003e\n\u003cp\u003eThe results indicate that soil characteristics and land cover type primarily drive potential habitat distribution. Climatic variables and altitude define the ecological limits of the species\u0026rsquo; niche. The response curve analysis highlighted species-specific ecological traits and outlined the profile of potentially favourable habitat for the threatened species. The results show that the species prefers open grassland and silvosteppe habitats with well-drained soils, particularly molisols types, and avoids hydromorphic or excessively wet soils. This pattern indicates the species group\u0026rsquo;s adaptation to mesoxerophilous-xerophilous conditions and moderate moisture regimes. Climatic factors, such as the mean temperature of the coldest quarter (Bio11) and precipitation of the driest month (Bio14), suggest a shared affinity for regions with moderately cold winters and low rainfall, characteristic of semi-arid continental areas. The preference for medium- and high-altitude landscapes reflects a desire to avoid low-altitude areas with intense anthropogenic influence. Overall, these results describe a common ecological pattern among steppe and silvosteppe plant species, helping define priority habitats for conservation. These findings are consistent with other studies analysing the habitat preferences of these species \u0026nbsp;(Chirilă, 2021; Chirilă et al., 2022; Chirilă \u0026amp; Kiril, 2024) and provide new insights into the environmental conditions that influence these priority species in Romania. For threatened plant species, it has been noted that, despite their conservation status, data about their environmental requirements are insufficient, and populations remain fragmented and declining (Chirilă et al., 2022). This emphasises the importance of our study in filling knowledge gaps about the habitat and ecological factors governing the persistence of these species.\u003c/p\u003e\n\u003cp\u003eThe spatial distribution of potential habitat indicates that the majority of favourable conditions for the five threatened plant species are outside protected areas. This situation is illustrated by Chirilă\u0026apos;s (2021) study on the spatial distribution of Crambe tataria, which found that its populations occupy a relatively large area (approximately 75%) outside protected areas. This uneven distribution highlights the high vulnerability of species to anthropogenic pressures and emphasises the need for conservation measures, particularly in areas outside the protected network where they are generally more exposed to human activities (Hansen \u0026amp; DeFries, 2007; Radeloff et al., 2010). Overall, these results emphasise that conservation strategies must go beyond mere \u003cem\u003ehotspot\u003c/em\u003e protection and adopt an integrated approach, combining the maintenance of high-quality habitats with the enhancement of ecological permeability.\u003c/p\u003e\n\u003cp\u003eLandscape permeability and priority actions for conservation management\u003c/p\u003e\n\u003cp\u003ePlot-level analysis revealed a significant discrepancy between the habitat potential identified by the SDM and the landscape\u0026apos;s actual permeability. Although the SDM modelling indicated extensive areas of favourable habitat, the CSI assessment showed that many of these areas exhibit low permeability. The reduced permeability values are generated by anthropogenic pressures such as inappropriate grazing (STK), fragmentation (FRA), and unsuitable land use (LAN). This situation aligns with numerous studies that identify overgrazing, fragmentation, and unsuitable land use as primary threats to threatened plant species and their grassland habitats, which are in various stages of degradation due to human activities (Bernhardt et al., 2011; Bilz, 2011a, 2011b; Chirilă, 2021; Kell, 2011). These pressures limit species survival and dispersal and may lead to long-term isolation, even within areas of ecologically suitable habitat. Nevertheless, topographic and edaphic variables scored highly, suggesting that associated microrefugia may play a critical role in maintaining populations (Bennie et al., 2008; Moeslund et al., 2013). Overall, the results confirm that a significant portion of potentially suitable habitat is located in areas of reduced permeability, limiting the dispersal and persistence of threatened species. This study provides the first spatially explicit evaluation of landscape permeability for threatened plant species in Romania, highlighting critical areas where habitat suitability does not translate into ecological connectivity.\u003c/p\u003e\n\u003cp\u003ePriority conservation areas\u003c/p\u003e\n\u003cp\u003eThe spatial overlay of habitat suitability (HS) and permeability (CSI) identified four priority zones, providing concrete guidance for conservation management. These results are consistent with previous studies that have integrated habitat suitability and landscape permeability to identify priority areas for conservation (Acharya et al., 2023; Cerreta et al., 2023; Muthiuru et al., 2024). Similar approaches have been successfully applied to delineate connectivity zones and guide targeted management actions in fragmented landscapes (Cable et al., 2021; Favilli et al., 2023). Our findings therefore support the inclusion of landscape permeability as a complementary dimension to habitat suitability in conservation planning.\u003c/p\u003e\n\u003cp\u003eIn fragmented landscapes, such as those in Romania, where habitat is often isolated by agricultural and infrastructural barriers, integrating CSI provides an enhanced framework for prioritising conservation actions. By explicitly mapping landscape permeability, our approach captures areas where key limiting factors threaten the survival of threatened plant species. Our integrated perspective goes beyond habitat-centred conservation and highlights that maintaining high permeability is essential even within favourable habitats.\u003c/p\u003e\n\u003cp\u003eOverall, our results emphasise the importance of a landscape-scale management strategy that combines actions to reduce fragmentation, optimise grazing pressure, and maintain the traditional agro-pastoral mosaic in priority areas. Moreover, maintaining high permeability is essential even within ecologically favourable habitats, and the spatial prioritisation of interventions should be based on the degree of overlap between habitat suitability and permeability.\u003c/p\u003e\n\u003cp\u003eImplications for integrated landscape management and Common Agricultural Policy\u003c/p\u003e\n\u003cp\u003eOur spatially explicit results provide a valuable decision-support tool for optimizing the ecological effectiveness of the Common Agricultural Policy (CAP), particularly in non-protected areas where proactive management is vital for maintaining landscape permeability and biodiversity. The priority areas identified in our study offer a strategic blueprint for directing agri-environment interventions, shifting the focus from generalized subsidies towards precision conservation. Specifically, Measure 10 (Agri-environment and Climate) should be proactively promoted among landholders within these high-connectivity areas, especially those located outside strictly protected sites (MADR, 2019). This targeted approach is essential to incentivize the continuation of traditional, moderate-intensity practices (such as extensive mowing and moderate grazing) that underpin both habitat potential and landscape permeability, as confirmed by our findings regarding the optimal livestock stocking rates and the highest CSI values. Simultaneously, Measure 13 (Payments for Areas with Natural Constraints) must be prioritized in these vulnerable, high-priority landscapes to counter the risk of land abandonment and the resulting loss of open habitats from scrub encroachment (MADR, 2019a). By aligning CAP investments with functional ecological mapping, we ensure public funds are strategically allocated to maintain and enhance the interconnected habitat network, which is crucial to the survival of threatened species.\u003c/p\u003e\n\u003cp\u003eTraditional land management practices, such as extensive mowing and moderate grazing, are key to maintaining the mosaic landscape structure that is essential for habitat connectivity and species diversity (Craioveanu et al., 2021; Jani\u0026scaron;ov\u0026aacute; et al., 2024). Evidence indicates that when applied at moderate intensity, traditional grazing prevents vegetation closure and supports the persistence of rare species, whereas overgrazing degrades plant cover, and abandonment promotes the dominance of competitive species (Pyk\u0026auml;l\u0026auml;, 2003). Maintaining large, connected grassland fragments, complemented by functional corridors, is essential for species dispersal and for mitigating the effects of fragmentation (Soons et al., 2005; Yan et al., 2022). A key finding of our study\u0026mdash;that moderate grazing plays a crucial role in maintaining grassland species diversity and contributing to landscape permeability\u0026mdash;confirms the importance of adaptive pasture management in areas prioritized for conservation, restoration, and dispersal. Limiting inappropriate anthropogenic development and restoring degraded habitats, would strategically complement these CAP measures, reinforcing the overall network of habitats favourable to threatened plant species.\u003c/p\u003e\n\u003cp\u003eRelevance, limitations, and future directions\u003c/p\u003e\n\u003cp\u003eAlthough the present analysis provides valuable insights into the conservation of threatened plant species, it does not comprehensively account for all anthropogenic and natural pressures that influence the quality and stability of their habitats. In this context, it is necessary to extend the assessment to additional disturbance factors, particularly those related to agricultural practices and invasive species, including: the burning of vegetation residues for land clearing; application of chemical fertilisers; proliferation of invasive species (e.g., \u003cem\u003eRobinia pseudoacacia\u003c/em\u003e, \u003cem\u003eAilanthus altissima\u003c/em\u003e (Mill.) Swingle, \u003cem\u003eElaeagnus angustifolia\u003c/em\u003e L., \u003cem\u003eEriogaster lanestris\u003c/em\u003e); mechanised mowing with heavy machinery; off-road vehicular traffic; excessive plant harvesting; and vandalism. Investigating these pressures requires a complementary approach that incorporates detailed field assessments and direct consultations with landowners, managers, or custodians of protected areas. Furthermore, future research could evaluate the potential impacts of climate change on the distribution of species of interest, as alterations in temperature and precipitation regimes may significantly affect habitat conditions and the survival capacity of these threatened plant species (C. Wang et al., 2016).\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eSpecies distribution models (SDMs) are fundamental tools in conservation ecology. They provide essential insights into species distributions, identify optimal habitats (HS), and inform protection strategies by rigorously predicting species presence and their environmental relationships. Consequently, these methods contribute significantly to the effective planning of conservation and ecological restoration measures, offering an integrated approach to the complexity of biodiversity.\u003c/p\u003e\n\u003cp\u003eHowever, our study highlights that the effectiveness of these models in guiding conservation actions could be substantially enhanced by incorporating landscape permeability. Although we identified extensive areas of habitat with high ecological potential, we demonstrated that their actual value is considerably reduced in contexts of low permeability, which constrains essential ecological processes. Therefore, not only the quantity but, more importantly, the functional quality of habitats determines species persistence. The results indicate that large areas of suitable habitat require targeted measures to improve permeability to effectively support the conservation of threatened species. High permeability remains crucial even in ecologically favourable habitats, and interventions should be spatially prioritised based on the overlap between habitat suitability and connectivity.\u003c/p\u003e\n\u003cp\u003eAccordingly, careful management of land use, grazing, and landscape fragmentation is not merely a best-practice recommendation but a strategic priority for the long-term conservation of threatened plant species. The proposed methodology could be replicated in other fragmented landscapes, providing a clear framework for prioritising conservation interventions.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors are grateful for the institutional and financial support received during this research, particularly from the Romanian Ministry of Education and the Research Institute of the University of Bucharest (ICUB), as well as to all those who contributed valuable ideas to the development of this study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Romanian Ministry of Education through a doctoral scholarship and by the Research Institute of the University of Bucharest (ICUB) through Grant [7420] for the project \u0026ldquo;Design of Landscape Networks with Machine Learning.\u0026rdquo;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of Interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no potential conflict of interest.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAcharya, P. M., Thainiramit, P., Techato, K., Baral, S., Rimal, N., Savage, M., Campos-Arceiz, A., \u0026amp; Neupane, D. (2023). Predicting the distribution and habitat suitability of the smooth-coated otter (\u003cem\u003eLutrogale perspicillata\u003c/em\u003e) in lowland Nepal. Global Ecology and Conservation, 46. https://doi.org/10.1016/j.gecco.2023.e02578\u003c/li\u003e\n\u003cli\u003eAffolter, D., Abderhalden, W., \u0026amp; Haller, R. (2011). The Continuum Suitability Index Technical Report. https:// www.parcs.ch/snp/pdf_public/3362_affolter_Technical_Report_JECAMI.pdf\u003c/li\u003e\n\u003cli\u003eBab\u0026iacute; Almenar, J., Bolowich, A., Elliot, T., Geneletti, D., Sonnemann, G., \u0026amp; Rugani, B. (2019). Assessing habitat loss, fragmentation and ecological connectivity in Luxembourg to support spatial planning. Landscape and Urban Planning, 189, 335\u0026ndash;351. https://doi.org/10.1016/j.landurbplan.2019.05.004\u003c/li\u003e\n\u003cli\u003eBaldwin, R. A. (2009). Use of Maximum Entropy Modeling in Wildlife Research. Entropy, 11(4), 854\u0026ndash;866. https://doi.org/10.3390/e11040854\u003c/li\u003e\n\u003cli\u003eBeckman, N. G., \u0026amp; Sullivan, L. L. (2023). The Causes and Consequences of Seed Dispersal. Annual Review of Ecology, Evolution, and Systematics, 54(1), 403\u0026ndash;427. https://doi.org/10.1146/annurev-ecolsys-102320-104739\u003c/li\u003e\n\u003cli\u003eBernhardt, K. G., Dostalova, A., Kir\u0026aacute;ly, G., \u0026amp; Petrova, A. (2011). \u003cem\u003eEchium russicum\u003c/em\u003e (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T162105A5538499. Accessed on 21 October 2025. https://www.iucnredlist.org/species/162105/5538499\u003c/li\u003e\n\u003cli\u003eBilz, M. (2011a). \u003cem\u003eKlasea lycopifolia\u003c/em\u003e. The IUCN Red List of Threatened Species 2011: e.T195348A8957743. https://www.iucnredlist.org/species/195348/8957743Accessed on 21 October 2025.\u003c/li\u003e\n\u003cli\u003eBilz, M. (2011b). \u003cem\u003ePaeonia tenuifolia\u003c/em\u003e (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T165143A5981869. https://www.iucnredlist.org/species/165143/5981869 Accessed on 21 October 2025.\u003c/li\u003e\n\u003cli\u003eBilz, M., Kell, S. P., Maxted, N., \u0026amp; Lansdown, R. V. (2011). European Red List of Vascular Plants.\u003c/li\u003e\n\u003cli\u003eBoyce, M. S., Vernier, P. R., Nielsen, S. E., \u0026amp; Schmiegelow, F. K. A. (2002). Evaluating resource selection functions. Ecological Modelling, 157(2\u0026ndash;3), 281\u0026ndash;300. https://doi.org/10.1016/S0304-3800(02)00200-4\u003c/li\u003e\n\u003cli\u003eCable, A. B., O\u0026rsquo;Keefe, J. M., Deppe, J. L., Hohoff, T. C., Taylor, S. J., \u0026amp; Davis, M. A. (2021). Habitat suitability and connectivity modeling reveal priority areas for Indiana bat (\u003cem\u003eMyotis sodalis\u003c/em\u003e) conservation in a complex habitat mosaic. Landscape Ecology, 36(1), 119\u0026ndash;137. https://doi.org/10.1007/s10980-020-01125-2\u003c/li\u003e\n\u003cli\u003eCerreta, A. L., McCarthy, K. P., \u0026amp; Fowles, G. (2023). Habitat suitability and landscape connectivity for an expanding population of bobcats. Landscape Ecology, 38(6), 1571\u0026ndash;1589. https://doi.org/10.1007/s10980-023-01639-5\u003c/li\u003e\n\u003cli\u003eChauhan, S., Ghoshal, S., Kanwal, K. S., Sharma, V., \u0026amp; Ravikanth, G. (2022). Ecological niche modelling for predicting the habitat suitability of endangered tree species Taxus contorta Griff. in Himachal Pradesh (Western Himalayas, India). Tropical Ecology, 63(2), 300\u0026ndash;313. https://doi.org/10.1007/s42965-021-00200-2\u003c/li\u003e\n\u003cli\u003eChirilă, S. D. (2021). Ecological And Chorological Studies Of The Species \u003cem\u003eCrambe Tataria\u003c/em\u003e Sebe\u0026oacute;k From Romania. Rom. J. Biol. \u0026ndash; Plant Biol, Bucharest, 66, 39\u0026ndash;54.\u003c/li\u003e\n\u003cli\u003eChirilă, S. D., Bădărău, A., \u0026amp; Milanovici, S. (2025). Ecology and distribution of \u003cem\u003ePontechium maculatum\u003c/em\u003e in Romania. Comprehensive Plant Biology, 49(1), 23\u0026ndash;38. https://doi.org/10.2298/CPB2501023C\u003c/li\u003e\n\u003cli\u003eChirilă, S. D., Cara, I. G., \u0026amp; Motrescu, I. (2022). Habitat preference of the endangered species \u003cem\u003eCrambe tataria\u003c/em\u003e (Brassicaceae) from Romania. Tuexenia, 42, 275\u0026ndash;296. https://doi.org/10.14471/2022.42.009\u003c/li\u003e\n\u003cli\u003eChirilă, S. D., Doroftei, M., \u0026amp; Gigea, G. (2025). Ecology and distribution of \u003cem\u003ePaeonia tenuifolia\u003c/em\u003e species in Romania. Biologia. https://doi.org/10.1007/s11756-025-01902-x\u003c/li\u003e\n\u003cli\u003eChirilă, S. D., \u0026amp; Kiril, V. (2024). Habitat preference for the populations of the endangered species \u003cem\u003ePontechium maculatum\u003c/em\u003e (Boraginaceae) in Romania. Tuexenia, 44, 131\u0026ndash;157. https://doi.org/10.14471/2024.44.005\u003c/li\u003e\n\u003cli\u003eCorbet, S. A., \u0026amp; Delfosse, E. S. (1984). Honeybees and the nectar of \u003cem\u003eEchium plantagineum\u003c/em\u003e L. in southeastern Australia. Australian Journal of Ecology, 9(2), 125\u0026ndash;139. https://doi.org/10.1111/j.1442-9993.1984.tb01351.x\u003c/li\u003e\n\u003cli\u003eCostanza, J. K., \u0026amp; Terando, A. J. (2019). Landscape Connectivity Planning for Adaptation to Future Climate and Land-Use Change. Current Landscape Ecology Reports, 4(1), 1\u0026ndash;13. https://doi.org/10.1007/s40823-019-0035-2\u003c/li\u003e\n\u003cli\u003eCraioveanu, C., Muntean, I., Ruprecht, E., Băncilă, R.-I., Crișan, A., \u0026amp; R\u0026aacute;kosy, L. (2021). Factors affecting butterfly and plant diversity in basiphilous dry grasslands of Transylvania, Romania. Community Ecology, 22(3), 295\u0026ndash;308. https://doi.org/10.1007/s42974-021-00055-6\u003c/li\u003e\n\u003cli\u003eCremene, C., Groza, G., Rakosy, L., Schileyko, A. A., Baur, A., Erhardt, A., \u0026amp; Baur, B. (2005). Alterations of Steppe‐Like Grasslands in Eastern Europe: a Threat to Regional Biodiversity Hotspots. Conservation Biology, 19(5), 1606\u0026ndash;1618. https://doi.org/10.1111/j.1523-1739.2005.00084.x\u003c/li\u003e\n\u003cli\u003eCruzan, M. B., \u0026amp; Hendrickson, E. C. (2020). Landscape Genetics of Plants: Challenges and Opportunities. Plant Communications, 1(6), 100100. https://doi.org/10.1016/j.xplc.2020.100100\u003c/li\u003e\n\u003cli\u003eDa Silva, L. G., Ribeiro, M. C., Hasui, \u0026Eacute;., Da Costa, C. A., \u0026amp; Da Cunha, R. G. T. (2015). Patch size, functional isolation, visibility and matrix permeability influences neotropical primate occurrence within highly fragmented landscapes. PLoS ONE, 10(2). https://doi.org/10.1371/journal.pone.0114025\u003c/li\u003e\n\u003cli\u003eDahlstr\u0026ouml;m, A., Iuga, A.-M., \u0026amp; Lennartsson, T. (2013). Managing biodiversity rich hay meadows in the EU: a comparison of Swedish and Romanian grasslands. Environmental Conservation, 40(2), 194\u0026ndash;205. https://doi.org/10.1017/S0376892912000458\u003c/li\u003e\n\u003cli\u003eDihoru, G., \u0026amp; Dihoru, A. (1994). Plante rare, periclitate şi endemice \u0026icirc;n flora Rom\u0026acirc;niei. Lista roşie. [Rare, Threatened and Endemic Plants in the Flora of Romania. Red List]. Acta Bot. Hort. Bucurestiensis.\u003c/li\u003e\n\u003cli\u003eDormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carr\u0026eacute;, G., Marqu\u0026eacute;z, J. R. G., Gruber, B., Lafourcade, B., Leit\u0026atilde;o, P. J., M\u0026uuml;nkem\u0026uuml;ller, T., McClean, C., Osborne, P. E., Reineking, B., Schr\u0026ouml;der, B., Skidmore, A. K., Zurell, D., \u0026amp; Lautenbach, S. (2013). Collinearity: a review of methods to deal with it and a simulation study evaluating their performance. Ecography, 36(1), 27\u0026ndash;46. https://doi.org/10.1111/j.1600-0587.2012.07348.x\u003c/li\u003e\n\u003cli\u003eDuan, Y., Bai, H., Du, Z., Liu, Y., Li, L., Lu, K., Wang, Z., Di, L., Qin, Y., \u0026amp; Liu, A. P. M. (2025). Maxent modeling for predicting the potential distribution of \u003cem\u003eHippophae Linn\u003c/em\u003e species. Tropical Ecology, 66(1), 132\u0026ndash;145. https://doi.org/10.1007/s42965-025-00372-1\u003c/li\u003e\n\u003cli\u003eEnyedi, Z. M., Ruprecht, E., \u0026amp; De\u0026aacute;k, M. (2008). Long‐term effects of the abandonment of grazing on steppe‐like grasslands. Applied Vegetation Science, 11(1), 55\u0026ndash;62. https://doi.org/10.1111/j.1654-109X.2008.tb00204.x\u003c/li\u003e\n\u003cli\u003eFahrig, L. (2003). Effects of Habitat Fragmentation on Biodiversity. Annual Review of Ecology, Evolution, and Systematics, 34(1), 487\u0026ndash;515.\u003c/li\u003e\n\u003cli\u003ehttps://doi.org/10.1146/annurev.ecolsys.34.011802.132419\u003c/li\u003e\n\u003cli\u003eFahrig, L. (2017). Ecological Responses to Habitat Fragmentation Per Se. Annual Review of Ecology, Evolution, and Systematics, 48(1), 1\u0026ndash;23. https://doi.org/10.1146/annurev-ecolsys-110316-022612\u003c/li\u003e\n\u003cli\u003eFateryga. (2015). Red book of the Republic of Crimea. Plants, algae and fungi. https://www.researchgate.net/publication/291829868_Red_book_of_the_Republic_of_Crimea_Plants_algae_and_fungi_in_Russian\u003c/li\u003e\n\u003cli\u003eFavilli, F., Laner, P., \u0026amp; Bertoncelj, I. (2023). Application of the continuum suitability index (csi) model to display the permeability of the alpine-dinaric landscape and to define intervention priorities for ecological linkages. Biodiversity and Conservation, 32(10), 3237\u0026ndash;3254. https://doi.org/10.1007/s10531-023-02651-6\u003c/li\u003e\n\u003cli\u003eFloraVeg.EU. (2025, September 28). FloraVeg. EU: Information about species. https://floraveg.eu/taxon/\u003c/li\u003e\n\u003cli\u003eGBIF.org. (2025, February 25). GBIF Occurrence Download. https://doi.org/10.15468/dl.jsgdqy\u003c/li\u003e\n\u003cli\u003eG\u0026icirc;lea, N. A., \u0026amp; Pătru-Stupariu, I. (2025). Environmental variables: key elements in the assessment of landscape permeability \u0026mdash; a review. Landscape Research, 1\u0026ndash;16. https://doi.org/10.1080/01426397.2025.2502552\u003c/li\u003e\n\u003cli\u003eGrădinaru, S. R., Fan, P., Iojă, C. I., Niță, M. R., Suditu, B., \u0026amp; Hersperger, A. M. (2020). Impact of national policies on patterns of built-up development: an assessment over three decades. Land Use Policy, 94, 104510. https://doi.org/10.1016/j.landusepol.2020.104510\u003c/li\u003e\n\u003cli\u003eGuiller, A., Decocq, G., Kichey, T., Poli, P., Vandepitte, K., Dubois, F., Honnay, O., \u0026amp; Closset-Kopp, D. (2023). Spatial genetic structure of two forest plant metapopulations in dynamic agricultural landscapes. Landscape and Urban Planning, 231, 104648. https://doi.org/10.1016/j.landurbplan.2022.104648\u003c/li\u003e\n\u003cli\u003eHaddad, N. M., Brudvig, L. A., Clobert, J., Davies, K. F., Gonzalez, A., Holt, R. D., Lovejoy, T. E., Sexton, J. O., Austin, M. P., Collins, C. D., Cook, W. M., Damschen, E. I., Ewers, R. M., Foster, B. L., Jenkins, C. N., King, A. J., Laurance, W. F., Levey, D. J., Margules, C. R., Townshend, J. R. (2015). Habitat fragmentation and its lasting impact on Earth\u0026rsquo;s ecosystems. Science Advances, 1(2). https://doi.org/10.1126/sciadv.1500052\u003c/li\u003e\n\u003cli\u003eHadley, A. S., Frey, S. J. K., Robinson, W. D., \u0026amp; Betts, M. G. (2018). Forest fragmentation and loss reduce richness, availability, and specialization in tropical hummingbird communities. Biotropica, 50(1), 74\u0026ndash;83. https://doi.org/10.1111/btp.12487\u003c/li\u003e\n\u003cli\u003eHansen, A. J., \u0026amp; DeFries, R. (2007). Ecological mechanisms linking protected areas to surrounding lands. Ecological Applications, 17(4), 974\u0026ndash;988. https://doi.org/10.1890/05-1098\u003c/li\u003e\n\u003cli\u003eHassan, R. M. ., Scholes, R. J. ., \u0026amp; Ash, Neville. (2005). Ecosystems and human well-being : current state and trends : findings of the Condition and Trends Working Group of the Millennium Ecosystem Assessment. Island Press.\u003c/li\u003e\n\u003cli\u003eHenle, K., Davies, K. F., Kleyer, M., Margules, C., \u0026amp; Settele, J. (2004). Predictors of Species Sensitivity to Fragmentation. Biodiversity \u0026amp; Conservation, 13(1), 207\u0026ndash;251. https://doi.org/10.1023/B:BIOC.0000004319.91643.9e\u003c/li\u003e\n\u003cli\u003eHurdu, B.-I., Coste, A., Halmagyi, A., Szatmari, P.-M., Farkas, A., Pușcaș, M., Dan Turtureanu, P., Roșca-Casian, O., Tănase, C., Oprea, A., Mardari, C., Răduțoiu, D., Camen-Comănescu, P., S\u0026icirc;rbu, I.-M., Stoie, A., Lupoae, P., Cristea, V., Jarda, L., Holobiuc, I., Butiuc-Keul, A. (2022). Ex situ conservation of plant diversity in Romania: A synthesis of threatened and endemic taxa. Journal for Nature Conservation, 68, 126211. https://doi.org/10.1016/j.jnc.2022.126211\u003c/li\u003e\n\u003cli\u003eJakovljevic, K., Djurovic, S., Antusevic, M., Mihailovic, N., Buzurovic, U., \u0026amp; Tomovic, G. (2019). Heavy metal tolerance of \u003cem\u003ePontechium maculatum\u003c/em\u003e (Boraginaceae) from several ultramafic localities in Serbia. Botanica Serbica, 43(1), 73\u0026ndash;83. https://doi.org/10.2298/BOTSERB1901073J\u003c/li\u003e\n\u003cli\u003eJani\u0026scaron;ov\u0026aacute;, M., Magnes, M., Bojko, I., Borsukevych, L. M., Budzhak, V. V., Chorney, I., Iuga, A., Ivașcu, C. M., Kish, R., Kuzemko, A., Palpurina, S., Skokanov\u0026aacute;, K., \u0026Scaron;irka, P., Tokaryuk, A., \u0026amp; Dayneko, P. (2024). Agricultural legacy shapes plant diversity patterns in mountain grasslands of Maramureș and Bukovina: A cross‐border perspective (Ukraine, Romania). People and Nature, 6(6), 2283\u0026ndash;2299. https://doi.org/10.1002/pan3.10698\u003c/li\u003e\n\u003cli\u003eKalista, M. (2017). Underutilized Medicinal Species Of \u003cem\u003eCrambe L\u003c/em\u003e. Of The Flora Of Ukraine (pp. 216\u0026ndash;220). https://doi.org/10.15414/agrobiodiversity.2017.2585-8246.216-220\u003c/li\u003e\n\u003cli\u003eKell, S. P. (2011). \u003cem\u003eCrambe tataria\u003c/em\u003e (Europe assessment). The IUCN Red List of Threatened Species 2011: e.T162122A5544350. Accessed on 21 October 2025. https://www.iucnredlist.org/species/162122/5544350\u003c/li\u003e\n\u003cli\u003eKimberley, A., Hooftman, D., Bullock, J. M., Honnay, O., Krickl, P., Lindgren, J., Plue, J., Poschlod, P., Traveset, A., \u0026amp; Cousins, S. A. O. (2021). Functional rather than structural connectivity explains grassland plant diversity patterns following landscape scale habitat loss. Landscape Ecology, 36(1), 265\u0026ndash;280. https://doi.org/10.1007/s10980-020-01138-x\u003c/li\u003e\n\u003cli\u003eLande, R. (1988). Genetics and Demography in Biological Conservation. Science, 241(4872), 1455\u0026ndash;1460. https://doi.org/10.1126/science.3420403\u003c/li\u003e\n\u003cli\u003eLaner, P., Rossi, C., Luethi, R., Favilli, F., Bertoncelj, I., Plassmann, G., \u0026amp; Haller, R. M. (2024). Landscape permeability for ecological connectivity at the macro-regional level: The Continuum Suitability Index and its practical implications. Ecological Indicators, 164. https://doi.org/10.1016/j.ecolind.2024.112145\u003c/li\u003e\n\u003cli\u003eLiu, J., Wilson, M., Hu, G., Liu, J., Wu, J., \u0026amp; Yu, M. (2018). How does habitat fragmentation affect the biodiversity and ecosystem functioning relationship? Landscape Ecology, 33(3), 341\u0026ndash;352. https://doi.org/10.1007/s10980-018-0620-5\u003c/li\u003e\n\u003cli\u003eLuo, Y., Wu, J., Wang, X., Wang, Z., \u0026amp; Zhao, Y. (2020). Can policy maintain habitat connectivity under landscape fragmentation? A case study of Shenzhen, China. Science of The Total Environment, 715, 136829. https://doi.org/10.1016/j.scitotenv.2020.136829\u003c/li\u003e\n\u003cli\u003eL\u0026uuml;thi, R., Rapp, M., \u0026amp; Haller, R. (2018a). Environmental protection indicator ENV (EUSALP). https://www.jecami.eu/static/mapViewer/docu/docu_env_eusalp.pdf\u003c/li\u003e\n\u003cli\u003eL\u0026uuml;thi, R., Rapp, M., \u0026amp; Haller, R. (2018b). Fragmentation indicator FRA (EUSALP) . https://www.jecami.eu/static/mapViewer/docu/docu_fra_eusalp.pdf\u003c/li\u003e\n\u003cli\u003eMADR. (2019). Ghidul solicitantului Măsurii 10 - agro-mediu şi climă - pachetele 7, 8 și 10 - din Programul National De Dezvoltare Rurala (PNDR) 2014 \u0026ndash; 2020. [Applicant\u0026rsquo;s Guide for Measure 10 - agri-environment and climate - packages 7, 8 and 10 - of the National Rural Development Programme (NRDP) 2014-2020]. https://www.madr.ro/docs/dezvoltare-rurala/agro-mediu/2025/Ghid-Masura-10-2025.pdf\u003c/li\u003e\n\u003cli\u003eMagrach, A., Larrinaga, A. R., \u0026amp; Santamar\u0026iacute;a, L. (2012). Effects of Matrix Characteristics and Interpatch Distance on Functional Connectivity in Fragmented Temperate Rainforests. Conservation Biology, 26(2), 238\u0026ndash;247. https://doi.org/10.1111/j.1523-1739.2011.01818.x\u003c/li\u003e\n\u003cli\u003eMart\u0026iacute;n Arroyo, T., Gonz\u0026aacute;lez-Porto, A. V., \u0026amp; Bartolom\u0026eacute; Esteban, C. (2017). Viper\u0026rsquo;s bugloss (Echium spp.) honey typing and establishing the pollen threshold for monofloral honey. PLOS ONE, 12(10), e0185405. https://doi.org/10.1371/journal.pone.0185405\u003c/li\u003e\n\u003cli\u003eMarușca, T., Mocanu, V., Haș, E. C., Tod, M. A., Andreoiu, A. C., Dragoș, M. M., Blaj, V. A., Ene, T. A., Silistru, D., Ichim, E., Zevedei, P. M., Constantinescu, C. S., \u0026amp; Tod, S. V. (2014). Ghid de intocmire a amenajamentelor pastorale [Guide for the Elaboration of Pastoral Management Plans].\u003c/li\u003e\n\u003cli\u003eMathur, M., \u0026amp; Mathur, P. (2025). Habitat suitability modelling of a critically endangered plant species -Commiphora wightii (Arn.) Bhandari: comparative assessment using various machine learning algorithms. Tropical Ecology, 66(2), 181\u0026ndash;203. https://doi.org/10.1007/s42965-025-00374-z\u003c/li\u003e\n\u003cli\u003eMeiklejohn, K., Ament, R., \u0026amp; Tabor, G. (2010). Habitat Corridors \u0026amp; Landscape Connectivity: Clarifying the Terminology. www.climateconservation.org\u003c/li\u003e\n\u003cli\u003eMinistry of Environment, W. and F. (2025a). Arii naturale protejate de interes național și internațional. Limite arii naturale protejate [Protected natural areas of national and international interest. Boundaries of Protected Natural Areas]. https://mmediu.ro/portal-gis/date-spatiale/arii-naturale-protejate-de-interes-national-si-international/\u003c/li\u003e\n\u003cli\u003eMinistry of Environment, W. and F. (2025b, February 25). Arii naturale protejate de interes național și internațional. Distribuția speciilor și habitatelor din planurile de management [Protected natural areas of national and international interest. Distribution of Species and Habitats in Management Plans]. https://mmediu.ro/portal-gis/date-spatiale/arii-naturale-protejate-de-interes-national-si-international/\u003c/li\u003e\n\u003cli\u003eMuthiuru, A. C., Crego, R. D., Simbauni, J. A., Muruthi, P. M., Waiguchu, G., Lala, F., Millington, J. D. A., \u0026amp; Kairu, E. W. (2024). Human footprint and rainfall shape Masai giraffe\u0026rsquo;s habitat suitability and connectivity in a multiple‐use landscape. Ecosphere, 15(7). https://doi.org/10.1002/ecs2.4933\u003c/li\u003e\n\u003cli\u003eMykhailenko, O., Hsieh, C.-F., El-Shazly, M., Nikishin, A., Kovalyov, V., Shynkarenko, P., Ivanauskas, L., Chen, B.-H., Horng, J.-T., Hwang, T.-L., Georgiyants, V., \u0026amp; Korinek, M. (2023). Anti-viral and Anti-inflammatory Isoflavonoids from Ukrainian Iris aphylla Rhizomes: Structure-Activity Relationship Coupled with ChemGPS-NP Analysis. Planta Medica, 89(11), 1063\u0026ndash;1073. https://doi.org/10.1055/a-2063-5265\u003c/li\u003e\n\u003cli\u003eMykhailenko, O., Korinek, M., Ivanauskas, L., Bezruk, I., Myhal, A., Petrikaitė, V., El-Shazly, M., Lin, G.-H., Lin, C.-Y., Yen, C.-H., Chen, B.-H., Georgiyants, V., \u0026amp; Hwang, T.-L. (2020). Qualitative and Quantitative Analysis of Ukrainian Iris Species: A Fresh Look on Their Antioxidant Content and Biological Activities. Molecules, 25(19), 4588. https://doi.org/10.3390/molecules25194588\u003c/li\u003e\n\u003cli\u003eNowak, B., Sitek, E., \u0026amp; Augustynowicz, J. (2020). Sourcing and Propagation of \u003cem\u003ePontechium maculatum\u003c/em\u003e for Horticulture and Species Restoration. Biology, 9(10), 317. https://doi.org/10.3390/biology9100317\u003c/li\u003e\n\u003cli\u003eOltean, M., Negrean, G., Popescu, A., Roman, N., Dihoru, G., \u0026amp; Sanda, V. \u0026amp; M. S. (1994). Studii, sinteze, documentaţii de ecologie. Lista roşie a plantelor superioare din Rom\u0026acirc;nia [Studies, Syntheses, Ecology Documentation. Red List of Higher Plants of Romania]. Academia Rom\u0026acirc;nă, Institutul de Biologie, București.\u003c/li\u003e\n\u003cli\u003eOprea, A. (2005). Lista critică a plantelor vasculare din Rom\u0026acirc;nia [Critical List of Vascular Plants of Romania]. Editura Universității \u0026bdquo;Alexandru Ioan Cuza\u0026rdquo;, Iaşi.\u003c/li\u003e\n\u003cli\u003ePhillips, S. J., Anderson, R. P., \u0026amp; Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190(3\u0026ndash;4), 231\u0026ndash;259. https://doi.org/10.1016/j.ecolmodel.2005.03.026\u003c/li\u003e\n\u003cli\u003ePhillips, S. J., \u0026amp; Dud\u0026iacute;k, M. (2008). Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. https://doi.org/10.1111/j.2007.0906-7590.05203.x\u003c/li\u003e\n\u003cli\u003ePhillips, S. J., Dud\u0026iacute;k, M., \u0026amp; Schapire, R. E. (2004). A Maximum Entropy Approach to Species Distribution Modeling.\u003c/li\u003e\n\u003cli\u003ePhillips, S. J., Dud\u0026iacute;k, M., \u0026amp; Schapire, R. E. (2025). Maxent software for modeling species niches and distributions (Version 3.4.4). 2025.\u003c/li\u003e\n\u003cli\u003ePlieninger, T., Hartel, T., Mart\u0026iacute;n-L\u0026oacute;pez, B., Beaufoy, G., Bergmeier, E., Kirby, K., Montero, M. J., Moreno, G., Oteros-Rozas, E., \u0026amp; Van Uytvanck, J. (2015). Wood-pastures of Europe: Geographic coverage, social\u0026ndash;ecological values, conservation management, and policy implications. Biological Conservation, 190, 70\u0026ndash;79. https://doi.org/10.1016/j.biocon.2015.05.014\u003c/li\u003e\n\u003cli\u003ePyk\u0026auml;l\u0026auml;, J. (2003). Effects of restoration with cattle grazing on plant species composition and richness of semi-natural grasslands. Biodiversity \u0026amp; Conservation, 12(11), 2211\u0026ndash;2226. https://doi.org/10.1023/A:1024558617080\u003c/li\u003e\n\u003cli\u003eQi, Y., Zhang, Y., Xue, J., Zhang, Z., Cao, J., Yang, N., Wan, F., Xian, X., \u0026amp; Liu, W. (2025). Future Climate Change Increases the Risk of Suitable Habitats for the Invasive Macrophyte \u003cem\u003eElodea nuttallii\u003c/em\u003e. Biology, 14(5), 504. https://doi.org/10.3390/biology14050504\u003c/li\u003e\n\u003cli\u003eRadeloff, V. C., Stewart, S. I., Hawbaker, T. J., Gimmi, U., Pidgeon, A. M., Flather, C. H., Hammer, R. B., \u0026amp; Helmers, D. P. (2010). Housing growth in and near United States protected areas limits their conservation value. Proceedings of the National Academy of Sciences, 107(2), 940\u0026ndash;945. https://doi.org/10.1073/pnas.0911131107\u003c/li\u003e\n\u003cli\u003eRemya, K., Ramachandran, A., \u0026amp; Jayakumar, S. (2015). Predicting the current and future suitable habitat distribution of Myristica dactyloides Gaertn. using MaxEnt model in the Eastern Ghats, India. Ecological Engineering, 82, 184\u0026ndash;188. https://doi.org/10.1016/j.ecoleng.2015.04.053\u003c/li\u003e\n\u003cli\u003eSava, A.-R., Boz, I., \u0026amp; Costică, N. (2019). Contribution to the knowledge of vegetative organs structure at Echium russicum j.f. gmelin. Analele Ştiinţifice Ale Universităţii \u0026bdquo;Al. I. Cuza\u0026rdquo; Iaşi s. II a. Biologie Vegetală, 65, 44\u0026ndash;54.\u003c/li\u003e\n\u003cli\u003eScholes, Robert., Montanarella, L. ., Brainich, Anastasia., Barger, Nichole., \u0026amp; Brink, B. ten. (2018). The assessment report on land degradation and restoration : summary for policymakers. Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES).\u003c/li\u003e\n\u003cli\u003eSilveira dos Santos, J., Vitorino, L. C., Fabrega Gon\u0026ccedil;alves, R., Corr\u0026ecirc;a C\u0026ocirc;rtes, M., Souza Cruz Alves, R., Cezar Ribeiro, M., \u0026amp; Garcia Collevatti, R. (2022). Matrix dominance and landscape resistance affect genetic variability and differentiation of an Atlantic Forest pioneer tree. Landscape Ecology, 37(10), 2481\u0026ndash;2501. https://doi.org/10.1007/s10980-022-01498-6\u003c/li\u003e\n\u003cli\u003eSoons, M. B., Messelink, J. H., Jongejans, E., \u0026amp; Heil, G. W. (2005). Habitat fragmentation reduces grassland connectivity for both short‐distance and long‐distance wind‐dispersed forbs. Journal of Ecology, 93(6), 1214\u0026ndash;1225. https://doi.org/10.1111/j.1365-2745.2005.01064.x\u003c/li\u003e\n\u003cli\u003eSwiss National Park. (2019). Jecami 2.0. User Guide. ALPBIONET2030 Project. https://www.jecami.eu/static/mapViewer/docu/jecami_userguide.pdf\u003c/li\u003e\n\u003cli\u003eSzit\u0026aacute;r, K., T\u0026ouml;lgyesi, C., De\u0026aacute;k, B., Gall\u0026eacute;, R., Kor\u0026aacute;nyi, D., \u0026amp; Bat\u0026aacute;ry, P. (2023). Connectivity and fragment size drive plant dispersal and persistence traits in forest steppe fragments. Frontiers in Ecology and Evolution, 11. https://doi.org/10.3389/fevo.2023.1155885\u003c/li\u003e\n\u003cli\u003eTaylor, P. D., Fahrig, L., \u0026amp; With, K. A. (2006). Landscape connectivity: a return to the basics. In Connectivity Conservation (pp. 29\u0026ndash;43). Cambridge University Press. https://doi.org/10.1017/CBO9780511754821.003\u003c/li\u003e\n\u003cli\u003eT\u0026ouml;r\u0026ouml;k, P., Bullock, J. M., Jim\u0026eacute;nez‐Alfaro, B., \u0026amp; Sonkoly, J. (2020). The importance of dispersal and species establishment in vegetation dynamics and resilience. Journal of Vegetation Science, 31(6), 935\u0026ndash;942. https://doi.org/10.1111/jvs.12958\u003c/li\u003e\n\u003cli\u003eT\u0026ouml;r\u0026ouml;k, P., Dembicz, I., Dajić-Stevanović, Z., \u0026amp; Kuzemko, A. (2020). Grasslands of Eastern Europe. In Encyclopedia of the World\u0026rsquo;s Biomes (pp. 703\u0026ndash;713). Elsevier. https://doi.org/10.1016/B978-0-12-409548-9.12042-1\u003c/li\u003e\n\u003cli\u003eVan Daele, F., Honnay, O., Janssens, S., \u0026amp; De Kort, H. (2024). Habitat fragmentation affects climate adaptation in a forest herb. Journal of Ecology, 112(2), 246\u0026ndash;264. https://doi.org/10.1111/1365-2745.14225\u003c/li\u003e\n\u003cli\u003eWang, C., Liu, C., Wan, J., \u0026amp; Zhang, Z. (2016). Climate change may threaten habitat suitability of threatened plant species within Chinese nature reserves. PeerJ, 4, e2091. https://doi.org/10.7717/peerj.2091\u003c/li\u003e\n\u003cli\u003eWang, T., Li, W., Cui, H., Song, Y., Liu, C., Yan, Q., Wu, Y., Jia, Y., Fang, L., \u0026amp; Qi, L. (2024). Predicting the Potential Habitat Distribution of Relict Plant Davidia involucrata in China Based on the MaxEnt Model. Forests, 15(2), 272. https://doi.org/10.3390/f15020272\u003c/li\u003e\n\u003cli\u003eWilson, M. C., Chen, X.-Y., Corlett, R. T., Didham, R. K., Ding, P., Holt, R. D., Holyoak, M., Hu, G., Hughes, A. C., Jiang, L., Laurance, W. F., Liu, J., Pimm, S. L., Robinson, S. K., Russo, S. E., Si, X., Wilcove, D. S., Wu, J., \u0026amp; Yu, M. (2016). Habitat fragmentation and biodiversity conservation: key findings and future challenges. Landscape Ecology, 31(2), 219\u0026ndash;227. https://doi.org/10.1007/s10980-015-0312-3\u003c/li\u003e\n\u003cli\u003eYan, Y., Jarvie, S., Liu, Q., \u0026amp; Zhang, Q. (2022). Effects of fragmentation on grassland plant diversity depend on the habitat specialization of species. Biological Conservation, 275, 109773. https://doi.org/10.1016/j.biocon.2022.109773\u003c/li\u003e\n\u003cli\u003eYezzi, A., Nebbia, A., \u0026amp; Zalba, S. (2023). Grassland Fragmentation: Introduction to the Special Issue. Diversity, 15(4), 489. https://doi.org/10.3390/d15040489\u003c/li\u003e\n\u003cli\u003eZanina, M. A., \u0026amp; Smirnova, E. B. (2020). Structure of coenopopulations and phytocoenotic confinement of Paeonia tenuifolia L. in floristic complexes of Oka-Don lowland. Plant Science Today, 7(4). https://doi.org/10.14719/pst.2020.7.4.978\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"landscape-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"land","sideBox":"Learn more about [Landscape Ecology](https://www.springer.com/journal/10980)","snPcode":"10980","submissionUrl":"https://submission.nature.com/new-submission/10980/3","title":"Landscape Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Threatened species, Continuum Suitability Index, Environmental variables, MaxEnt, Grasslands","lastPublishedDoi":"10.21203/rs.3.rs-7957482/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7957482/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eContext\u003c/strong\u003e: Habitat fragmentation contributes to population isolation and threatens the survival of plant species listed on the Red List. Effective conservation measures require considering both suitable habitats and landscape permeability. Species Distribution Models (SDMs) are valuable for identifying suitable habitats. These models gain utility when combined with landscape permeability analyses for conservation planning.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eObjectives\u003c/strong\u003e: This study aims to (1) model the potentially suitable habitat distribution for five threatened plant species in Romania, (2) assess landscape permeability, and (3) identify priority areas for conservation based on the overlay of the habitat and permeability.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethodology\u003c/strong\u003e: We applied an integrated approach that combined SDM modelling (MaxEnt) and GIS analysis. We used the Continuum Suitability Index (CSI) to quantify landscape permeability. By mapping the spatial overlap of habitat suitability and permeability, we classified areas into four categories for ecological interventions.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: A significant proportion of the identified suitable habitats have low permeability, which limits species dispersal and persistence. Areas combining high suitability and high permeability are limited and are mostly outside protected areas (PAs). This highlights the urgent need for conservation actions. Measures such as optimal grazing management, reduced fragmentation, and maintaining traditional agro-pastoral mosaics should be prioritized in these critical zones beyond formal protected area boundaries.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: Including landscape permeability in habitat analyses gives a more realistic view of the requirements of threatened species. It supports more explicit decision-making for conservation management. The study also proposes a replicable method for other fragmented regions, with direct applications in land-use policy and biodiversity conservation.\u003c/p\u003e","manuscriptTitle":"Conserving Red List plant species by managing landscape fragmentation and permeability","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-19 13:15:03","doi":"10.21203/rs.3.rs-7957482/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-18T16:04:33+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-15T00:00:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-01T11:32:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"165711954508730674209893019867029924614","date":"2025-11-22T20:38:54+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"170376400191042458744511718941941040570","date":"2025-11-11T07:56:18+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-10T10:44:09+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-29T09:05:43+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-29T09:05:40+00:00","index":"","fulltext":""},{"type":"submitted","content":"Landscape Ecology","date":"2025-10-26T12:09:29+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"landscape-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"land","sideBox":"Learn more about [Landscape Ecology](https://www.springer.com/journal/10980)","snPcode":"10980","submissionUrl":"https://submission.nature.com/new-submission/10980/3","title":"Landscape Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"735e9bad-5dc1-4212-83d9-c47436e1dee8","owner":[],"postedDate":"November 19th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-02-23T16:02:07+00:00","versionOfRecord":{"articleIdentity":"rs-7957482","link":"https://doi.org/10.1007/s10980-026-02314-1","journal":{"identity":"landscape-ecology","isVorOnly":false,"title":"Landscape Ecology"},"publishedOn":"2026-02-19 15:58:39","publishedOnDateReadable":"February 19th, 2026"},"versionCreatedAt":"2025-11-19 13:15:03","video":"","vorDoi":"10.1007/s10980-026-02314-1","vorDoiUrl":"https://doi.org/10.1007/s10980-026-02314-1","workflowStages":[]},"version":"v1","identity":"rs-7957482","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7957482","identity":"rs-7957482","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}