Climate change facilitates fungal pathogen expansion while driving endemic host range contractions in a tropical biodiversity hotspot

Climate change facilitates fungal pathogen expansion while driving endemic host range contractions in a tropical biodiversity hotspot | 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 Climate change facilitates fungal pathogen expansion while driving endemic host range contractions in a tropical biodiversity hotspot Emma L Underwood, Kerry A Brown, Anna Rönnfeldt, Mark Mulligan, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8741587/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 02 May, 2026 Read the published version in Landscape Ecology → Version 1 posted 15 You are reading this latest preprint version Abstract Context. Non-native plant pathogens are reshaping ecosystems globally, yet their spread and potential spatial distributions under future climate and land-use change remains underexplored, particularly in biodiversity hotspots with vulnerable hosts. In Madagascar, a vascular wilt pathogen ( Leptographium calophylli ) has been increasingly observed infecting native forest trees raising conservation concerns. Objectives. We aimed to model future distributions of both a vulnerable host tree Calophyllum paniculatum and incipient wilt pathogen Leptographium calophylli under climate and land cover change scenarios over the next 80 years to assess overlap, divergence, and implications for extinction risk. Methods. Using ensemble SDMs across Madagascar, we forecasted potential distributional ranges for both pathogen and host using the Global Climate Model (GFDL-ESM4) under three scenarios (SSP1–2.6, SSP3–7.0, SSP5–8.5) and time periods (2011–2040, 2041–2070, 2071–2100). We measured future range shifts using a habitat exposure index and potential area of occupancy. Results. The pathogen is predicted to retain 68.5% of the current projected distribution by 2100, with expansion into previously uninhabited regions. C. paniculatum is forecast to experience range contraction (65.9% loss by 2100) with persistent distributional overlap predicted across all scenarios. Conclusion. We demonstrate that future climatic conditions may facilitate fungal pathogen expansion, while simultaneously exposing vulnerable hosts within their native, endemic range to infection. The asymmetric dynamic of host range losses could intensify biodiversity loss in insular ecosystems, particularly when compounded by deforestation. This study highlights the importance of considering disease threats in biodiversity forecasts and conservation strategies, particularly in tropical systems facing rapid environmental change. climate change fungal pathogen tropics endemism biodiversity hotspot sdm Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 INTRODUCTION Emerging plant pathogens are increasingly recognised as a major driver of biodiversity loss (IPBES, 2023 ). In tropical island hotspots, where host species are endemic, often range-restricted, ecologically specialised and already stressed by habitat loss and climate change (Brown, 2022), pathogen invasion can have profound consequences. In such systems, pathogen-driven mortality of dominant or functionally unique trees can rapidly alter community structure and ecosystem function, potentially triggering local extinctions and cascading ecological consequences (Boyd et al., 2013 ). For instance, such infections can truncate seed-shadow and recruitment by removing parent trees (Augspurger and Wilkinson, 2007 ), reduce fruit and nectar availability that in-turn depress frugivores and pollinators (Razafindratsima et al., 2014 ), diminish above-ground carbon stocks (Slik et al., 2013 ), and alter litter inputs (Fonte and Schowalter, 2004 ). In combination, these impacts can shift community composition toward disturbance-tolerant states (e.g., liana-dominant, (Visser et al., 2018 ) and heighten fire susceptibility at forest margins (Metz et al., 2011 ). These changes are difficult to reverse once thresholds are crossed (Reyer et al., 2015 ). Consequently, as global change intensifies, understanding how plant-pathogen dynamics may progress temporally and spatially becomes central to predicting and mitigating biodiversity loss. Climate change influences host-pathogen interactions through changes in temperature and moisture regimes that govern pathogen sporulation, infection efficiency and survival (Elad and Pertot, 2014 , Garrett et al., 2006 ), as well as host stress and defence (Desprez-Loustau et al., 2007 , Sturrock et al., 2011 ). Depending on the pathogen-host system, warming, altered precipitation and seasonality patterns may facilitate pathogens to track hosts and expand into previously unoccupied refugia (Bebber et al., 2013 ). Conversely, some pathogens might encounter environmental thresholds that limit their survival and spread, thereby creating spatial refugia for host species. The development of robust disease risk models that forecast pathogen distribution under altered climatic conditions is therefore crucial, not only for predicting disease outbreaks but also for informing effective conservation strategies and management practices in vulnerable ecosystems. Many conservation strategies in invasion ecology rely on correlative species distribution models (SDMs) to predict how species’ ranges may shift in response to environmental change. These SDMs are widely used to identify climate refugia and prioritise conservation strategies (Li and Wang, 2013 , Peterson et al., 2012 ). However, most SDMs focus solely on abiotic variables and assume species respond independently to environmental gradients. This assumption breaks down when biotic factors, such as host-pathogen dynamics, and human-led dispersal shape a species’ realised niche, e.g., taking consideration of the ‘BAM’ framework (Escobar and Craft, 2016 ; Soberón, 2005). For host species facing pathogen pressure, projected range shifts may not reflect future viability if pathogens are able to track or outpace hosts. Conversely, if pathogen ranges contract or shift differently than their hosts, it may result in spatial refugia for the host species, potentially increasing their resilience. These outcomes have fundamentally different implications for extinction risk, yet few SDM-based conservation studies consider such spatial overlap explicitly (Goberville et al., 2016 , Sopniewski et al., 2022 ). Critically, this omission may lead to over-optimistic or misleading conservation assessments. Yet the methodological and data challenges of doing so are considerable, especially for under-studied pathogens in poorly sampled regions. These knowledge gaps are particularly stark in tropical island ecosystems, where high rates of endemism intersect with high vulnerability to invasive species, habitat loss and climate change (Bellard et al., 2017 , Dawson et al., 2017 ). Madagascar is emblematic of this convergence. The island harbours exceptional biodiversity, including over 12,000 plant species, of which 83% are endemic (Goodman, 2023 ). Yet this richness coexists with accelerating environmental degradation: deforestation, driven by slash-and-burn agriculture and fuelwood harvesting, has already reduced the country’s natural forest cover to less than 10% of its original extent (Harper et al., 2007 , Vieilledent et al., 2018 ). Concurrently, climate models forecast substantial shifts in temperature and precipitation patterns across the island (Brown, 2022; Goodman, 2023 ). These stressors interact with increasing biotic pressures, including the incipient introduction of pathogens not previously known to occur in these systems. However, the extent to which disease dynamics will interact with climate and land-use change to reshape species distributions in Madagascar remains largely unknown. One illustrative example is the recently described vascular wilt disease affecting Calophyllum paniculatum , a native tree on Madagascar. The fungal pathogen thought to be Leptographium calophylli (formerly Verticillium ) has been observed causing mortality in C. paniculatum populations in Ranomafana National Park since 2016 (Wright et al., 2020 ). Although Leptographium species are globally distributed and known to cause disease in other woody hosts, their ecology and epidemiology in Madagascar’s forests remain poorly characterised. The origins of this outbreak are uncertain, which underscores the urgency of proactive monitoring and spatial forecasting under future environmental change. To address these ecological knowledge gaps, we use ensemble SDMs to model current and future suitability for both the vascular wilt pathogen ( Leptographium calophylli ) and a susceptible endemic host ( Calophyllum paniculatum ) across Madagascar under three climate and deforestation scenarios, at three future time slices. The overarching aim of this study is to identify persistent risk hotspots and candidate refugia to inform conservation prioritisation and providing a template for data-poor tropical ecosystems. For the final research objective, we will map host-pathogen spatial coupling and identify mortality and risk hotspots by deriving overlap, divergence and elevational range metrics to classify regions of intensified exposure and potential refugia. One plausible outcome is partial decoupling with persistent overlap, for example, if pathogen suitability remains relatively stable while the host contracts or shifts differently in space or elevation, yielding refugia alongside enduring co-occurrence and sustained mortality risk. The alternatives are equally plausible: (i) tight coupling if both taxa track similar directional and elevational shifts, maintaining high overlap; (ii) risk attenuation if pathogen suitability contracts or shifts away faster than the host, enlarging refugia; or (iii) risk escalation if pathogen suitability expands or shifts faster than the host, increasing exposure. MATERIALS AND METHODS We chose to accompany this methods section with Overview, Data, Model, Assessment and Prediction (ODMAP) protocols and full code for each species modelled (Zurell et al., 2020 ) in the name of openness and transparency ( https://github.com/anon-017/anon-fungal-host-sdm ). These protocols summarise all parameter decisions and inputs made to build and predict future range shifts of both the focal tree and its wilt pathogen. Study site Madagascar (and Indian Ocean Islands) is one of 36 global biodiversity hotspots, combining exceptional endemism with extensive habitat loss (Mittermeier et al., 2011). The island lies ca. 90 km east of mainland Africa (Fig. 1 , extent A) and has lost much of its natural forest cover to deforestation and degradation, primarily through shifting agriculture and fuelwood extraction (Brown, 2022, Goodman, 2023 , Harper et al., 2007 , Vieilledent et al., 2013 ). Because the focal non-native fungal pathogen – Leptographium calophylli – has not yet been systematically recorded in Madagascar but has a documented history on neighbouring Indian Ocean islands and mainland Africa (Webber et al., 1999 ; Wiehe, 1949 ), we treated continental Africa as the putative source region and Madagascar as the potential invasion front. We therefore defined two raster extents at 1-km resolution (Fig. 1 ): (i) a training extent covering mainland Africa (extent A) used to model the pathogen’s climatic niche based on African occurrences and (ii) a projection extent covering Madagascar (extent B) where we projected both pathogen and host suitability through time. Raster templates for both extents were prepared using Global Administrative Areas (GADM.org) data, including only mainland extents and excluding coastal islands to avoid artefacts along shorelines. Ranomafana National Park (extent B) was highlighted as the location where wilt symptoms on Calophyllum paniculatum were first reported and monitored (Wright et al., 2020 ), providing an independent reference for evaluating projected invasion risk. Study species Endemic host plant Calophyllum is a genus of evergreen tropical trees in the family Calophyllaceae (formerly “Clusiaceae”), comprising 87 recognised species globally (IUCN, 2023 ). Many species in the genus show negative population trends, with 29% assessed as decreasing and 57% are data deficient (IUCN, 2023 ). Calophyllum paniculatum ( hereafter C. paniculatum) is endemic to mid-level elevation humid and sub-humid ecoregions of Madagascar, where it forms part of the canopy in evergreen rainforest. It is currently categorised as vulnerable on the IUCN Red List (IUCN, 2019 ), and recent reports of high adult mortality associated with the vascular wilt (Wright et al., 2020 ), raise concern that climate change, land-use change and pathogen invasion may interact to accelerate its decline. Fungal wilt pathogen A vascular wilt disease has been causing increasing adult mortality in C. paniculatum in Ranomafana National Park since 2016 and is attributed to a fungal pathogen in the genus Leptographium , most likely Leptographium calophylli ( L. calophylli ) (Wright et al. 2020 ). The fungal wilt L. calophylli has long been recognised as a serious pathogen of Calophyllum trees in other tropical and subtropical moist broadleaf biomes in the Indian Ocean islands of Seychelles and Mauritius (Wainhouse et al., 1998 , Webber et al., 1999 , Wiehe, 1949 ). The external symptoms include rapid crown wilting and canopy dieback, while internally the disease is characterised by dark brown vascular streaking visible in the xylem tissue (Wiehe, 1949 , Wright et al., 2020 ). Dispersal vector Although aerial dispersal may only contribute locally, long distance spread of L. calophylli across islands separated by hundreds of kilometres is likely to have been facilitated by human movement of infected timber. An ecologically important vector for the dispersal within forests is the bark beetle, Cryphalus Erichson, 1836. This genus of beetle range in size from 0.8–3 mm in length and inhabit tree bark where fungal spores attach to their exoskeleton and are transported to new hosts (Wainhouse et al., 1998 ). At least 25 Cryphalus species occur on the African Continent (GBIF, 2025c ) and Cryphalus trypanus has been identified as the principal vector of the Calophyllum wilt pathogen in Seychelles and Mauritius (Wainhouse et al., 1998 ). Occurrence records To our knowledge, there are no officially reported geolocated occurrences of the fungal wilt L calophylli in Madagascar in any global biodiversity database, aside from surveys from Ranomafana National Park (Fig. 1 , extent B; Wright et al. 2020 ). Due to recent taxonomic and historical uncertainty surrounding the fungal pathogen’s identity (Wainhouse et al., 1998 , Webber et al., 1999 , Wiehe, 1949 ), we broadened our query to the wilt genera Verticillium Nees, 1816 and Leptographium Lagerb. & Melin. All records for these genera restricted to mainland Africa (Fig. 1 , extent A) were downloaded from Global Biodiversity Information Facility (GBIF, 2025a ). We removed records with missing, inconsistent, or obviously incorrect geographic coordinates and excluded records with zero counts. The initial dataset comprised Leptographium (n = 10) occurrences from southern Africa and Verticillium (n = 86) occurrences from northern, western, and eastern Africa. After thinning spatial duplicates within 1 km grid cells, n = 34 unique fungal wilt occurrences remained. For simplicity, we refer to these the genera-level occurrences as L. calophylli . For the endemic host plant C. paniculatum , we again, collated all records (n = 65) with accepted coordinates (GBIF, 2025b ) and supplemented these with field observations from multiple surveys in Madagascar, including the Tropical Ecological Assessment and Monitoring (TEAM) network (Rovero and Ahumada, 2017 ), Centre ValBio (CVB) wilt surveys (Wright et al., 2020 ), a microhabitat study (Ramananjato, 2021 ; Ramananjato and Razafindratsima, 2021 ) and a soil survey in which the host was recorded as present (Armstrong et al., 2018 ). To minimise temporal duplication, we retained one record per plot for a common survey year (2018) within each dataset where appropriate. This yielded a final set of 62 unique C. paniculatum occurrences for terrestrial Madagascar (Fig. 1 , extent B). To accompany the collated species presence points within the ensemble modelling framework, pseudoabsences and background points were generated then combined with the presence data for the wilt and host. The fungal wilt pseudoabsences were constructed using the ‘2° far’ approach (Fig. 1 , extent A (Barbet-Massin et al., 2012 )), where following Li et al. ( 2024 ), we also combined presence records for the bark beetle vector Cryphalus (Fig. 1 , extent A) obtained from GBIF and grouped with the fungal wilt occurrences (GBIF, 2025c ). Host background points were randomly generated across the whole of Madagascar. For more information on these approaches, see Online Resource 3 – Methods. Environmental information To characterise contemporary and future macroclimate, we extracted bioclimatic variables from CHELSA v2.1 (Karger et al., 2017 , Karger et al., 2021 ), at each occurrence location. These represent the 19 standard temperature and precipitation predictors at 30 arc-seconds (~ 1 km at the equator) spatial resolution. These variables capture seasonal variability and extremes to constrain tropical tree distributions and fungal pathogen dynamics. We extracted climate projections from the CMIP6 GFDL-ESM4 general circulation model under three Shared Socioeconomic Pathway scenarios: SSP1-RCP2.6, SSP3-RCP7, and SSP5-RCP8.5 (Krasting et al., 2018 ), spanning low, medium and high emissions futures and thus bracketing plausible climates that could mediate both host range shifts and climate-facilitated pathogen invasion. We assembled these variables across four time periods: 1981–2010 (current baseline), 2011–2040, 2041–2070 and 2071–2100. In addition, we included forest cover projections from the ForestAtRisk model for Madagascar (Vieilledent, 2021 , Vieilledent et al., 2018 ). ForestAtRisk provides 30 m resolution maps of historical forest cover and projected deforestation under a “business as usual” scenario with baseline data for 2010–2020 and projections for 2040, 2070, and 2100 aligned with our climate time slices. These projections suggest that, under continued deforestation, humid forest in Madagascar could be almost entirely lost by 2100, with large uncertainty in the exact timing of collapse but consistent trends of decline across scenarios (Vieilledent et al., 2018 ). These combined data represented temporally matched, dynamic forest cover layers that were used as a proxy for intact humid and sub-humid forest structure relevant to host C. paniculatum establishment and to microclimatic buffering of the wilt pathogen and its bark beetle vector. Species distribution modelling (SDM) We used ensemble species distribution models (SDMs) to relate species occurrences to climatic and land-cover variables and to project habitat suitability under future scenarios (Araújo and New, 2007 ). Our framework models an emerging fungal wilt ( L. calophylli ) and its endemic host tree C. paniculatum to: (i) characterise their current environmental niches, (ii) forecast potential invasion and range shifts under combinations of climate change and deforestation and (iii) quantify changing spatial overlap between pathogen and host. In the context of Madagascar’s highlighted endemic flora, these models provide an effective way to explore how climate-mediated pathogen invasion could interact with ongoing habitat loss to reshape host distributions and extinction risk. Predictor variable selection Because both the pathogen and host were represented by relatively few occurrences, we restricted SDM complexity and minimised multicollinearity among predictors to avoid overfitting (Dormann et al., 2013 ). Initially, we considered the 19 CHELSA bioclimatic variables and ForestAtRisk forest cover and quantified collinearity separately for the pathogen (African training extent) and host (Madagascar) datasets using Spearman’s rank correlation matrices and Variance Inflation Factors (VIF). Highly correlated pairs ( \(\:\left|r\right|\ge\:0.7\) ) were identified and one variable from each pair was removed based on lower VIF, yielding a pool of non-collinear predictors for each species (Online Resource 4, Figures S2-3 ). To keep model complexity, proportionate to sample size, we limited the number of climate predictors to approximately one variable per 8–10 presence records. For the fungal pathogen ( L. calophylli ) this resulted in a maximum of four climate predictors, while for the host plant ( C. paniculatum ) we allowed for eight variables, including the ForestAtRisk forest cover predictor. These were ecologically interpretable and formed the basis for SDM fitting and subsequent ensemble model predictions. For the fungal wilt pathogen, the final non-collinear climate predictors were: (i) precipitation of the driest month, (ii) precipitation seasonality, (iii) mean temperature of the coldest quarter, and (iv) precipitation of the coldest quarter. Together, these variables captured dry-season moisture stress (Desprez-Loustau et al., 2007 ), intra-annual rainfall variability (Thompson et al., 2014 ) and cool season thermal (Aguayo et al., 2014 ) and moisture regimes (Burgess et al., 2017 ) that are known to constrain survival, overwintering and infection windows of vascular wilt fungi. For the endemic host tree, the final predictors comprised: (i) temperature seasonality, (ii) mean temperature of the driest quarter, (iii) annual temperature range, (iv) precipitation of the wettest month, (v) precipitation of the coldest quarter, and (vi) forest cover. These variables summarise thermal stability and extremes (Engelbrecht et al., 2007 , Morellato et al., 2016 , Park Williams et al., 2013 ), water availability during both wet and cool seasons (Esquivel-Muelbert et al., 2017 , Mcdowell et al., 2020 ), and the presence of closed-canopy forest, consistent with C. paniculatum as a humid rainforest tree whose distribution is shaped by relatively stable tropical climates and the persistence of intact forest structure (De Moraes et al., 2006 , Stevens, 1980 ). Ensemble model fitting and validation We used a weighted ensemble approach to reduce bias from individual algorithms and better capture uncertainties in predictions across climate and deforestation scenarios (Araújo and New, 2007 , Thuiller et al., 2019 ). Ensemble models were built using the sdm R package (Naimi and Araújo, 2016 ). For each species, we fitted three algorithms that perform well with limited occurrence data: Boosted Regression Tree (BRT), Random Forest (RF) and MaxEnt (Williams et al., 2009 ). Models were replicated ten times for each algorithm using bootstrap resampling, producing 30 models per species in the final model. For each ensemble, to account for the much larger number of background data relative to presences, we applied a weighting ratio such that each pseudo-absence/background point contributed one-tenth of the weight of a presence, making the total weighted contribution of presences and pseudo-absences/background points approximately equal in each model replicate. We used five-fold cross-validation, with an 80:20 training-testing split in each fold (Araújo et al., 2005 ). Two- and three-dimensional partial response curves were generated for all predictors and suitability predictions were averaged across replicates and algorithms to obtain ensemble mean predictions for subsequent performance evaluation and projection (Online Resource 4, Figures S11-12 ). Ensemble model performance evaluation We assessed the performance of each individual model replicate using threshold-independent discrimination metrics: Area under the receiver operating curve (AUC), True Skill Statistic (TSS) and the standard deviation of both for each cross-validation fold (Allouche et al., 2006 , Liu et al., 2013 ). The top trained, ensembles were inspected for ecologically plausible response curves and spatial distribution patterns before being retained for future projections (Online Resource 4, Figures S4-5 ). For the final wilt ensemble, we chose Matthews Correlation Coefficient (MCC) as the primary evaluation metric due to its robustness in novel environmental conditions and effectiveness in handling class imbalance (Boughorbel et al., 2017 ). For the host tree C. paniculatum , the final ensemble used AUC-weighted averaging, further optimised with respect to maximising TSS to balance omission and commission errors. For each species, we quantified variable importance for the top ensemble model by permuting predictors one at a time and measuring the resulting loss in predictive performance (Dormann et al., 2013 ). In the permutation approach, values of a given predictor were randomly shuffled across sites while all other predictors were kept unchanged, so the greater the reduction in model performance, the higher the inferred importance of that predictor (Online Resource 4, Figures S9-10 ). Importance scores were averaged across algorithms within the top ensemble and rescaled to facilitate comparison among variables within each species (Naimi and Araújo, 2016 ). Thresholding to create binary presence/absence To analyse future range shift and host-pathogen overlap, we converted continuous probability of presence maps (Online Resource 4, Figures S7-8 ) to binary presence-absence maps for each species, climate scenario and time period. We used the threshold function in the sdm package to identify an optimal binary cut-off for each ensemble, weighting towards the best-performing algorithms and replicates. For the wilt pathogen L. calophylli , we used the minimum distance to ROC curve ("minROCdist") optimiser to improve transferability between Africa and Madagascar and reduce overprediction in novel climates (Owens et al., 2013 ). For the host tree C. paniculatum , modelled within its native range, we optimised the weighted ensemble towards the maximum sum of sensitivity and specificity ("max(se + sp)", or TSS) to balance omission and commission error rates (Liu et al., 2013 ). The binary maps were used to calculate distributional changes using area, centroid, altitudinal changes, as well as habitat exposure, and the predicted area of occupancy (pAOO), following Choe et al. ( 2017 ) and the IUCN ( 2022 ). See Online Resource 3, Figure S1 . Mapping novel environments and uncertainty We performed a Multivariate Environmental Similarity Surface (MESS) analysis on the top ensemble for each species to assess the similarity between current and future climatic conditions (Elith et al., 2010 , Zurell et al., 2012 ). Given the model transfer from Africa to Madagascar, this was particularly helpful to analyse any differences in conditions between training and testing (geographic disparity) as well as changes in environmental conditions over time. In addition, we evaluated the spatial uncertainty of predictions, using high and low confidence presence and absence categories, allowing for spatially explicit interpretation of model outputs mapped across Madagascar. For this, we used a percentile-based approach to threshold the uncertainty (entropy) layer created for each ensemble, using 75% and 25% as thresholds for high and low confidence, respectively (Thuiller et al., 2019 ). For high confidence predictions, we applied a threshold at the point where > 75% of models were in agreement for each pixel (the inverse of the entropy layer). The same was applied for low confidence predictions, where a threshold was chosen where < 25% of models agreed. We wrote custom code to make this process dynamic, testing 5% increments to account for any cases of high agreement across model algorithms and replicates in case the default resulted in a zero-value threshold. This ensured the uncertainty measurements were specific to each species and ensemble model. Preliminary environmental data processing was carried out in QGIS (3.40 LTR). All further data preparation, modelling and analysis was performed using R Statistical Software (v4.3.2 (2023-10-31 ucrt); (R Core Team, 2023)) with packages rgbif (Chamberlain S, 2017; Chamberlain and Boettiger, 2017 ) and terra (Hijmans, 2025 ) for spatial data processing of pseudoabsences and background points. Further methods information is found in Online Resources 1 & 2 (ODMAP protocols). RESULTS Projections of host pathogen distributions under current climate When visualising current projections for the wilt pathogen L. calophylli (Fig. 2 a), the pathogen appears to be suitable across a larger and overlapping geographic extent than host C. paniculatum (Fig. 2 b). There are high confidence areas of suitable habitat across the entire north-south humid belt of the island, and high confidence absences to the west, largely encompassing the entire distribution of C. p aniculatum , although the host distribution is sparser within the same humid and sub-humid belts. There are some small patches of uncertainty (low confidence absence – red) within around the outside of the projected distribution for the wilt, and in the south and west which does not overlap with the host (Fig. 2 a). The host presence distribution is small and isolated, with low confidence absences projected along western coastal fringes. Spatial-temporal patterns of future range dynamics Overall, ensemble SDMs for both species project changes in the geographic distributions of suitable habitat of the fungal wilt L. calophylli , and endemic host, C. paniculatum across all climate scenarios examined. These projections, however, reveal markedly different trajectories for the pathogen and its host, suggesting potential disruption of their current spatial relationship by the end of the century. The wilt pathogen L. calophylli , demonstrates considerably greater resilience under future climates, with projected range contractions of only 7.7–31.6% across all scenarios and timeframes (Fig. 3 , Online Resource 4, Table S5 ). Even under the most extreme climate scenario (SSP5) by 2071–2100, the pathogen is projected to retain 68.5% of its current range. In contrast to the wilt pathogen, C. paniculatum is projected to experience severe range shifting, with an overall net contraction across all scenarios, except for the first period (2011–2040 for the low-emission pathway only) (Fig. 4 ). Moderate reductions of 22.5–41.1% under the low emissions pathway become progressively more severe under higher emission scenarios, with the most extreme reductions of 65.9% by the end of the century observed under the fossil-fuelled shared socioeconomic pathway (Fig. 5 , Online Resource 4, Table S5 ). Measuring range differences along a temporal gradient reveals distinctly different trajectories of range stability, contraction and expansion. The fungal pathogen shows consistent range stability across scenarios and time periods, with slight expansion into new range by 2040 under high emissions scenario, and even minor expansion by end of century under low emissions scenario (Fig. 5 ). Meanwhile, a progressive intensification of range contractions across all scenarios (except the low-emissions pathway) are predicted for the host C. paniculatum , with the most dramatic changes occurring in the latter half of the century, equating to more than double range losses compared to the wilt (Fig. 5 ). Towards the end of the century (2071–2100), the most dramatic changes occur, particularly for C. paniculatum , which predicts the highest contractions of original range area, along with the lowest levels of stability and expansion compared to earlier in the century. Equally, the pathogen - although still experiencing some range contractions during this period (8.3–31.5%) - maintains relatively high range stability (68.5–91.7%) (Fig. 5 , Online Resource 4, Table S5 ). L. calophylli exhibits conservative spatial shifts with a consistent directional pattern (Fig. 6 , purple icons). These directional shift results reveal a potential spatial decoupling in the geographic responses of host and pathogen. For C. paniculatum , predominant shifts were observed in southward and eastward directions across most scenarios, with substantial centroid shifts ranging from 29.7-111.2 km (Fig. 6 , green icons). C. paniculatum is also predicted to move upslope in the next century, showing a mean elevational increase of 85m by 2100 (Fig. 7 ). This directional divergence, with species boundaries moving in different cardinal directions at significantly different rates, suggests potential disruption of their current spatial relationship under future climate scenarios, particularly under high emission pathways and later time periods. Host-pathogen elevational relationship Elevation responses show some consistency between host and pathogen (Fig. 7 ), with both species generally shifting upward in elevation at different rates and thresholds for the focal tree. C. paniculatum shows more pronounced elevational shifts (maximum vertical shift height of 253.6 m during 2071–2100 with SSP3) than the pathogen (up to 124.0 m for the SSP5 scenario). These different rates of elevational migration could temporarily alter disease dynamics along elevation gradients, although given their geographic distributions, the predicted probability of presence of the pathogen exceeds the overall elevational range of C. paniculatum. Geographic range overlap The predicted distribution of L. calophylli dominates the eastern forest corridor, maintaining stable geographic distribution into the future (Figs. 8– 9 ). In contrast, the host, C. paniculatum , has a very limited predicted distribution across Madagascar, persisting in patches in the north and isolated locations on the eastern humid belt under future climate and land use scenarios (Fig. 9 ). Distribution and overlap changes demonstrate the increasingly limited distribution of C. paniculatum along the northern and central eastern zones, while for the wilt L. calophylli , a pattern of expanding distribution along Madagascar's humid forest corridor. Their areas of overlap are concentrated along the central-eastern forests, particularly in the north (Fig. 9 ). Across all climate and land use pathways, there is a clear geographic stability of L. calophylli to 2100, with the low-emission fossil fuelled pathway showing the most extensive geographic distribution of the pathogen by the late century. Their shared overlap appears to reduce over time, both by shared pathway intensity and future year range, likely due to the overall contraction in the range of C. paniculatum (Fig. 8). The overlap change patterns suggest a southward shift in overlap areas over time. Standardised range shift metrics (habitat exposure and predicted area of occupancy) show C. paniculatum to be much more vulnerable to future change drivers than the wilt pathogen L. calophylli. Both show increasing vulnerability as the intensity of future climate scenario increases from 1 (low emissions) to 5 (high emissions), but C. paniculatum shows steeper declines and more variability of results across scenarios (Table 1 ). C. paniculatum exhibits high habitat exposure values (0.759–0.830), particularly under higher emission scenarios, indicating widespread habitat transformation to new areas by 2100. Climate exposure is lower for the wilt pathogen L. calophylli , and lowest (0.083) under the low-emission pathway (SSP1) but increases substantially under higher emission scenarios (0.298–0.315). The difference in habitat exposure between species suggests substantially greater climate sensitivity for the host tree. Potential area of occupancy (pAOO) changes show on average, that C. paniculatum is forecasted to lose up to 67% of its current area by 2100, compared to L. calophylli , which is a much lower predicted loss at up to 31% (Table 1 ). DISCUSSION Environmental change expands invasion opportunity in Madagascar This study has three key findings that together clarify how global environmental change can reshape host-pathogen dynamics and invasion processes in tropical biodiversity hotspots. First, climate change consistently maintains or expands climatically suitable habitat for the novel pathogen L. calophylli across much of Madagascar, even under high-emissions scenarios. Second, the endemic host C. paniculatum is projected to undergo consistent range contraction across all climate scenarios, compounded by fragmentation and dispersal limitation from deforestation, resulting in a small number of geographically isolated refugia by late century. Third, these contrasting responses generate a pronounced spatial and climatic decoupling between host and pathogen distributions, such that pathogen suitability increasingly exceeds and spatially envelops the shrinking host range. Similar host-pathogen mismatches have been documented historically in temperate regions following introduction of invasive pathogens under changing environmental conditions, most notably the extirpation of American chestnut ( Castanea dentata ) by the introduced fungus Cryphonectria parasitica (Anagnostakis, 1987, Santini et al., 2013) and the widespread collapse of elm populations following invasion of Dutch elm disease ( Ophiostoma spp ) (Karnosky, 1979, Santini et al., 2013). In those invasions, pathogen persistence and spread far exceeded the climatic and demographic resilience of their primary hosts, restructuring forest composition for decades to centuries (Santini et al., 2013). Our projections suggest that L. calophylli may follow a comparable trajectory in Madagascar, persisting across landscapes even as the endemic host becomes increasingly rare and spatially fragmented. The projected stability and breadth of the climatic niche of L. calophylli indicate that the pathogen is unlikely to be constrained by future climate change in the same way its host is. By the end of the century, the pathogen occupies orders of magnitude more climatically suitable area than C. paniculatum under high-emissions scenarios, suggesting that it may persist independently of this host. This pattern is consistent with invasion dynamics observed in other forest pathogens, where multi-host strategies allow pathogens to maintain populations in a reservoir host even as susceptible species decline (Desprez-Loustau et al., 2007). It is possible that regardless of whether C. paniculatum becomes locally extirpated across its range, L. calophylli may remain in local forests, increasing the likelihood of spill over into other native species or reinfection of remnant populations. By maintaining the area of climatically suitable habitat for L. calophylli, climate change effectively increases the invasion opportunity available to this novel pathogen. Here, invasion opportunity refers to the total extent of environments in which establishment and long-term persistence are climatically plausible. As suitable landscape increases, the probability that repeated introduction events result in successful establishment also increases, even if the likelihood of establishment per individual introduction remains unchanged. This distinction is critical for forest pathogens, where propagule pressure is typically chronic rather than episodic (Garbelotto and Pautasso, 2012, Lovett et al., 2016). For instance, continued movement of timber or contaminated equipment can generate multiple introduction events over time, and a larger suitable area increases the chance that at least some of these introductions occur in environments conducive to persistence (Santini et al., 2013). Alternatively, if climate change reduces suitable habitat for L. calophylli, repeated introductions would more often fail because arrival events would disproportionately occur in environmentally mismatched locations (Meentemeyer et al., 2011). However, our projections suggest the opposite trajectory. By expanding or maintaining climatic suitability across large parts of the island, environmental change weakens climatic barriers to establishment and makes invasion outcomes increasingly dependent on propagule pressure rather than environmental mismatch. In this way, climate change increases invasion opportunity and increases the likelihood that repeated introductions translate into establishment (Elad and Pertot, 2014), causing early mortality to already climatically weakened hosts like C. paniculatum . Importantly, the asymmetry of future habitat suitability between host and pathogen highlighted in this study is supported by multiple, independent diagnostics rather than being an artefact of model transfer or threshold choice. The MESS analysis (Online Resource 4, Figure S14 ) indicated that future projections for L. calophylli are largely made within environmental space represented by the training data, reducing the likelihood that the observed broad pathogen suitability arises from extrapolation into novel climates. Additionally, uncertainty mapping shows persistent areas of high model agreement for the pathogen across Madagascar’s humid forests. Together, these diagnostics strengthen confidence that the projected divergence reflects a real difference in (pathogen) climatic tolerance and (host) vulnerability, rather than modelling artefacts. Notwithstanding, the predicted distributions represent fully realised climatic suitability rather than realised occupancy and therefore cannot be interpreted as direct predictions of current spread or disease extent. Management implications and future conservation directions The contrasting responses the wilt pathogen L. calophylli to the endemic host tree C. paniculatum depict a troubling dimension to potential climate change impacts that extends beyond species range shifts. While the host faces clear range contractions, the pathogen shows steady tolerance to changes in the climate over time, with minor projected range losses across the same scenarios and time frames. This disparity creates an ecological mismatch that has concerning implications for forest health and conservation planning. In the current-day projected distribution of L. calophylli , the geographical range of presence spans a broader climate niche, extending much further longitudinally, with reducing but visible presence stretching from sea-level at the eastern coastline all the way westward, through the humid into the sub-humid region, and even into drier savanna land towards the centre of the country. This broader distribution suggests the pathogen can persist under climate conditions that far exceed the physiological limits of C. paniculatum . L. calophylli maintains suitable habitat in most of Madagascar’s central spine and eastern escarpment, covering most of the humid zone, and parts of the eastern sub-humid zone. This suggests that the pathogen could survive as a generalist and not require C. paniculatum specifically as a host for survival, where it may utilise other species as a ‘reservoir’ in areas where their ranges to do intersect (Grünwald et al., 2008, Rizzo and Garbelotto, 2003). In addition, the wilt pathogen L. calophylli causing mortality of Calophyllum paniculatum in Madagascar and Calophyllum inophyllum on nearby tropical islands is still an unknown to the conservation community. In the face of becoming widespread before official identification, our study provides a timely assessment of the potential impacts of L. calophylli on a focal endemic tree species in a tropical biodiversity hotspot. We provide suggestions for how predictive ecological modelling can support wider efforts to identify, monitor, and limit its spread by recommending that future work focuses on the following research questions and actions: Host range and other targets Research question: Do we know with certainty, that L. calophylli does not also infect or impact other endemic species or genera in Madagascar? Action : Collaborate with local partners to conduct systematic field sampling of multiple species across sites within the core probability of presence distributions of L. calophylli . Action : Collaborate with pathologists to officially identify the pathogen. Phase of spread Research question : What phase of spread is the pathogen in? Action : random sampling of species and sites within and outside of the probability of presence distribution of L. calophylli . Action : Multi-criteria analysis (MCA) to map the possible spread start point and dispersal patterns, using anthropogenic and natural spread routes: ferry ports, roads, known logging areas, built-up areas, hiking paths, prevailing wind direction and river catchments. Application of global tools Research question : How can IPBES tools and suggestions for invasive species management (IPBES, 2023) be adapted to the case of L. calophylli in the context of tropical, island biodiversity? Action : Assess applicability of IPBES frameworks (IPBES, 2023) for prevention, early detection and rapid response in Madagascar’s ecological and policy context. CONCLUSION We highlight that the range dynamics of a non-native, incipient pathogen can become a bigger threat to endemic island species under climate change. We identified and compared the current and future predicted geographic distributions of the wilt pathogen L. calophylli in relation to those of host C. paniculatum , and quantified predicted range changes and potential future spatial overlap. We showed the already sparse, currently suitable habitat of C. paniculatum may reduce further to isolated fragments in the central and north sub-humid ecoregions, as a direct result of combined climate change and deforestation by the end of this century. On the other hand, the climate resilience shown by the fungal wilt pathogen L. calophylli , across all measured climate projections, without a strong reliance on forested areas for survival could further put endemic vulnerable host species at further risk. These findings should alert the conservation community to explore the impacts of multiple global and regional change drivers simultaneously for other vulnerable and endangered species across the tropics, to better understand current future threats to species’ survival, and to limit further biodiversity loss across terrestrial biodiversity hotspots globally. Declarations Competing Interests The authors have no relevant financial or non-financial interests to disclose. Funding The author(s) received no financial support for the research, authorship, and/or publication of this article. Author Contribution All authors contributed to the study conception and design. Material preparation and analysis were performed by E.L.U, supported by A.R and R.A and K.A.B. The first draft of the manuscript was written by E.L.U, supported by K.A.B, A.R and R.A. All authors reviewed this and provided comments on previous versions of the manuscript. All authors read and approved the final manuscript. Acknowledgement The authors would like to thank Professor Patricia Chapple Wright and team at the Stony Brook University Institute for the Conservation of Tropical Environments (ICTE) Centre ValBio for open data sharing on current information pertaining to the wilt pathogen spread in Madagascar. Data Availability All data used in this study are derived from publicly available repositories and referenced in the manuscript accordingly:1. Global portals: [GBIF](https:/www.gbif.org) , [CHELSA](https:/www.chelsa-climate.org) , [ForestAtRisk](https:/forestatrisk.cirad.fr)2. Publicly available survey data from published peer-reviewed articles:1. Ramananjato, V., Razafindratsima, Onja. 2021. Data from: Structure of microhabitats used by Microcebus rufus across a heterogeneous landscape [Dataset]. [https://doi.org/10.5061/dryad.2280gb5rs](https:/doi.org/10.5061/dryad.2280gb5rs)2. Armstrong, A. H., Shugart, H. H. & Fatoyinbo, T. E. 2011. Characterization of Community Composition and Forest Structure in a Madagascar Lowland Rainforest. Tropical Conservation Science, 4, 428-444. [https://doi.org/10.1177/194008291100400406](https:/doi.org/10.1177/194008291100400406)3. Wright, P. C., Otero Jimenez, B., Rakotonirina, P., Andriananoely, D. H., Shea, A., Ratalata, B. & Razafimahaimodison, J. C. 2020. The Progressive Spread of the Vascular Wilt Like Pathogen of Calophyllum Detected in Ranomafana National Park, Madagascar. Frontiers in Forests and Global Change. [https://doi.org/10.3389/ffgc.2020.00091](https:/doi.org/10.3389/ffgc.2020.00091) .All custom code used for all preparation and analyses can be found at: [https://github.com/anon-017/anon-fungal-host-sdm](https:/github.com/anon-017/anon-fungal-host-sdm) . References Aguayo, J., Elegbede, F., Husson, C., Saintonge, F.-X. & Marçais, B. 2014. Modeling climate impact on an emerging disease, the Phytophthora alni-induced alder decline. Global Change Biology , 20, 3209–3221 https://doi.org/10.1111/gcb.12601 . Allouche, O., Tsoar, A. & Kadmon, R. 2006. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). Journal of Applied Ecology , 43, 1223–1232 https://doi.org/10.1111/j.1365-2664.2006.01214.x . Anagnostakis, S. L. 1987. Chestnut Blight: The Classical Problem of an Introduced Pathogen. Mycologia , 79, 23–37 10.1080/00275514.1987.12025367 . Araújo, M. B. & New, M. 2007. Ensemble forecasting of species distributions. Trends in Ecology & Evolution , 22, 42–47 https://doi.org/10.1016/j.tree.2006.09.010 . Araújo, M. B., Pearson, R. G. & Rahbek, C. 2005. Equilibrium of Species' Distributions with Climate. Ecography , 28, 693–695 Armstrong, A., Fischer, R., Huth, A., Shugart, H. & Fatoyinbo, T. 2018. Simulating Forest Dynamics of Lowland Rainforests in Eastern Madagascar. Forests [Online], 9. Augspurger, C. K. & Wilkinson, H. T. 2007. Host Specificity of Pathogenic Pythium Species: Implications for Tree Species Diversity. Biotropica , 39, 702–708 https://doi.org/10.1111/j.1744-7429.2007.00326.x . Barbet-Massin, M., Jiguet, F., Albert, C. H. & Thuiller, W. 2012. Selecting pseudo-absences for species distribution models: how, where and how many? Methods in Ecology and Evolution , 3, 327–338 https://doi.org/10.1111/j.2041-210X.2011.00172.x . Bebber, D. P., Ramotowski, M. a. T. & Gurr, S. J. 2013. Crop pests and pathogens move polewards in a warming world. Nature Climate Change , 3, 985–988 10.1038/nclimate1990 . Bellard, C., Rysman, J.-F., Leroy, B., Claud, C. & Mace, G. M. 2017. A global picture of biological invasion threat on islands. Nature Ecology & Evolution , 1, 1862–1869 10.1038/s41559-017-0365-6 . Boughorbel, S., Jarray, F. & El-Anbari, M. 2017. Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric. PLOS ONE , 12, e0177678 https://doi.org/10.1371/journal.pone.0177678 . Boyd, I. L., Freer-Smith, P. H., Gilligan, C. A. & Godfray, H. C. J. 2013. The Consequence of Tree Pests and Diseases for Ecosystem Services. Science , 342, 1235773 10.1126/science.1235773 . Brown, K. A., Yesuf, Gabriel., Carvalho, Fabio 2022. Assessing the quality of remaining forests on Madagascar. In : GOODMAN, S. M. (ed.) The new natural history of Madagascar . Princeton, NJ, USA: Princeton University Press. Burgess, T. I., Scott, J. K., Mcdougall, K. L., Stukely, M. J. C., Crane, C., Dunstan, W. A., Brigg, F., Andjic, V., White, D., Rudman, T., Arentz, F., Ota, N. & Hardy, G. E. S. J. 2017. Current and projected global distribution of Phytophthora cinnamomi, one of the world's worst plant pathogens. Global Change Biology , 23, 1661–1674 https://doi.org/10.1111/gcb.13492 . Chamberlain S, B. V., Mcglinn D, Oldoni D, Desmet P, Geffert L, Ram K. 2017. rgbif: Interface to the Global Biodiversity Information Facility API. R package version 3.8.0. https://CRAN.R-project.org/package=rgbif [Accessed] Chamberlain, S. A. & Boettiger, C. 2017. R Python, and Ruby clients for GBIF species occurrence data. PeerJ Preprints. Choe, H., Thorne, J. H., Hijmans, R., Kim, J., Kwon, H. & Seo, C. 2017. Meta-corridor solutions for climate-vulnerable plant species groups in South Korea. Journal of Applied Ecology , 54, 1742–1754 https://doi.org/10.1111/1365-2664.12865 . Dawson, W., Moser, D., Van Kleunen, M., Kreft, H., Pergl, J., Pyšek, P., Weigelt, P., Winter, M., Lenzner, B., Blackburn, T. M., Dyer, E. E., Cassey, P., Scrivens, S. L., Economo, E. P., Guénard, B., Capinha, C., Seebens, H., García-Díaz, P., Nentwig, W., García-Berthou, E., Casal, C., Mandrak, N. E., Fuller, P., Meyer, C. & Essl, F. 2017. Global hotspots and correlates of alien species richness across taxonomic groups. Nature Ecology & Evolution , 1, 0186 10.1038/s41559-017-0186 . De Moraes, L. F. D., Assumpção, J. M., Luchiari, C. & Pereira, T. S. 2006. Native tree species planting for restoration at Poço das Antas Biological Reserve, Rio de Janeiro, Brazil. Rodriguesia , https://doi.org/10.1590/2175-7860200657307 . Desprez-Loustau, M.-L., Robin, C., Buee, M., Courtecuisse, R., Garbaye, J., Suffert, F., Sache, I. & Rizzo, D. M. 2007. The fungal dimension of biological invasions. Trends in ecology & evolution , 22, 472–480 https://doi.org/10.1016/j.tree.2007.04.005 . 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 (Copenhagen) , 36, 27–46 https://doi.org/10.1111/j.1600-0587.2012.07348.x . Elad, Y. & Pertot, I. 2014. Climate Change Impacts on Plant Pathogens and Plant Diseases. Journal of Crop Improvement , 28, 99–139 10.1080/15427528.2014.865412 . Elith, J., Kearney, M. & Phillips, S. 2010. The art of modelling range-shifting species. Methods in Ecology and Evolution , 1, 330–342 https://doi.org/10.1111/j.2041-210X.2010.00036.x . Engelbrecht, B. M. J., Comita, L. S., Condit, R., Kursar, T. A., Tyree, M. T., Turner, B. L. & Hubbell, S. P. 2007. Drought sensitivity shapes species distribution patterns in tropical forests. Nature , 447, 80–82 https://doi.org/10.1038/nature05747 . Escobar, L. E. & Craft, M. E. 2016. Advances and Limitations of Disease Biogeography Using Ecological Niche Modeling. Frontiers in Microbiology , Volume 7–2016 Esquivel-Muelbert, A., Baker, T. R., Dexter, K. G., Lewis, S. L., Ter Steege, H., Lopez-Gonzalez, G., Monteagudo Mendoza, A., Brienen, R., Feldpausch, T. R., Pitman, N., Alonso, A., Van Der Heijden, G., Peña-Claros, M., Ahuite, M., Alexiaides, M., Álvarez Dávila, E., Murakami, A. A., Arroyo, L., Aulestia, M., Balslev, H., Barroso, J., Boot, R., Cano, A., Chama Moscoso, V., Comiskey, J. A., Cornejo, F., Dallmeier, F., Daly, D. C., Dávila, N., Duivenvoorden, J. F., Duque Montoya, A. J., Erwin, T., Di Fiore, A., Fredericksen, T., Fuentes, A., García-Villacorta, R., Gonzales, T., Guevara Andino, J. E., Honorio Coronado, E. N., Huamantupa-Chuquimaco, I., Eliana Maria Jiménez, R., Killeen, T. J., Malhi, Y., Mendoza, C., Mogollón, H., Jørgensen, P. M., Montero, J. C., Mostacedo, B., Nauray, W., Neill, D., Vargas, P. N., Palacios, S., Palacios Cuenca, W., Pallqui Camacho, N. C., Peacock, J., Phillips, J. F., Pickavance, G., Quesada, C. A., Ramírez-Angulo, H., Restrepo, Z., Reynel Rodriguez, C., Paredes, M. R., Peñuela-Mora, M. C., Sierra, R., Silveira, M., Stevenson, P., Stropp, J., Terborgh, J., Tirado, M., Toledo, M., Torres-Lezama, A., Umaña, M. N., Urrego, L. E., Vasquez Martinez, R., Gamarra, L. V., Vela, C. I. A., Vilanova Torre, E., Vos, V., Von Hildebrand, P., Vriesendorp, C., Wang, O., Young, K. R., Zartman, C. E. & Phillips, O. L. 2017. Seasonal drought limits tree species across the Neotropics. Ecography , 40, 618–629 https://doi.org/10.1111/ecog.01904 . Fonte, S. J. & Schowalter, T. D. 2004. Decomposition of Greenfall vs. Senescent Foliage in a Tropical Forest Ecosystem in Puerto Rico. Biotropica , 36, 474–482 https://doi.org/10.1111/j.1744-7429.2004.tb00343.x . Garbelotto, M. & Pautasso, M. 2012. Impacts of exotic forest pathogens on Mediterranean ecosystems: four case studies. European Journal of Plant Pathology , 133, 101–116 10.1007/s10658-011-9928-6 . Garrett, K. A., Dendy, S. P., Frank, E. E., Rouse, M. N. & Travers, S. E. 2006. Climate Change Effects on Plant Disease: Genomes to Ecosystems. Annual Review of Phytopathology , 44, 489–509 https://doi.org/10.1146/annurev.phyto.44.070505.143420 . Gbif. 2025a. Occurrence Download. https://doi.org/10.15468/dl.284amw Gbif. 2025b. Occurrence Download. https://doi.org/10.15468/dl.dj8pmv Gbif. 2025c. Occurrence Download. https://doi.org/10.15468/dl.7kd9hb Goberville, E., Hautekèete, N.-C., Kirby, R. R., Piquot, Y., Luczak, C. & Beaugrand, G. 2016. Climate change and the ash dieback crisis. Scientific Reports , 6, 35303 10.1038/srep35303 . Goodman, S. M. 2023. Updated estimates of biotic diversity and endemism for Madagascar—revisited after 20 years. Oryx , 57, 588–600 https://doi.org/10.1017/S0030605322001284 . Grünwald, N. J., Goss Em Fau - Press, C. M. & Press, C. M. 2008. Phytophthora ramorum: a pathogen with a remarkably wide host range causing sudden oak death on oaks and ramorum blight on woody ornamentals. Molecular Plant Pathology , 9, 729–740 https://doi.org/10.1111/j.1364-3703.2008.00500.x . Harper, G. J., Steininger, M. K., Tucker, C. J., Juhn, D. & Hawkins, F. 2007. Fifty years of deforestation and forest fragmentation in Madagascar. Environmental Conservation , 34, 325–333 https://doi.org/10.1017/S0376892907004262 . Hijmans, R. 2025. terra: Spatial Data Analysis. R package version 1.8.21. https://CRAN.R-project.org/package=terra [Accessed] Ipbes. 2023. Summary for Policymakers of the Thematic Assessment Report on Invasive Alien Species and their Control of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services [Online]. Bonn, Germany: Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) secretariat. [Accessed 2025]. Iucn. 2019. The IUCN Red List of Threatened Species [Online]. Available: www.iucnredlist.org [Accessed]. Iucn 2022. Guidelines for Using IUCN Red List Categories and Criteria. IUCN. Iucn. 2023. The IUCN Red List of Threatened Species. Karger, D. N., Conrad, O., Böhner, J., Kawohl, T., Kreft, H., Soria-Auza, R. W., Zimmermann, N. E., Linder, H. P. & Kessler, M. 2017. Climatologies at high resolution for the earth’s land surface areas. Scientific Data , 4, 170122 https://doi.org/10.1038/sdata.2017.122 . Karger, D. N., Conrad, O., Böhner, J., Kawohl, T., Kreft, H., Soria-Auza, R. W., Zimmermann, N. E., Linder, H. P. & Kessler, M. 2021. Climatologies at high resolution for the earth’s land surface areas. https://doi.org/10.16904/envidat.228 Karnosky, D. F. 1979. Dutch Elm Disease: A Review of the History, Environmental Implications, Control, and Research Needs. Environmental Conservation , 6, 311–322 Krasting, J. P., John, J. G., Blanton, C., Mchugh, C., Nikonov, S., Radhakrishnan, A., Rand, K., Zadeh, N. T., Balaji, V., Durachta, J., Dupuis, C., Menzel, R., Robinson, T., Underwood, S., Vahlenkamp, H., Dunne, K. A., Gauthier, P. P. G., Ginoux, P., Griffies, S. M., Hallberg, R., Harrison, M., Hurlin, W., Malyshev, S., Naik, V., Paulot, F., Paynter, D. J., Ploshay, J., Reichl, B. G., Schwarzkopf, D. M., Seman, C. J., Silvers, L., Wyman, B., Zeng, Y., Adcroft, A., Dunne, J. P., Dussin, R., Guo, H., He, J., Held, I. M., Horowitz, L. W., Lin, P., Milly, P. C. D., Shevliakova, E., Stock, C., Winton, M., Wittenberg, A. T., Xie, Y. & Zhao, M. 2018. NOAA-GFDL GFDL-ESM4 model output prepared for CMIP6 CMIP. https://doi.org/10.22033/ESGF/CMIP6.1407 Li, X.-D., Chen, Y., Zhang, C.-L., Wang, J., Song, X.-J., Zhang, X.-R., Zhu, Z.-H. & Liu, G. 2024. Assessing the climatic niche changes and global invasion risk of Solanum elaeagnifolium in relation to human activities. Science of The Total Environment , 954, 176723 https://doi.org/10.1016/j.scitotenv.2024.176723 . Li, X. & Wang, Y. 2013. Applying various algorithms for species distribution modelling. Integrative Zoology , 8, 124–135 https://doi.org/10.1111/1749-4877.12000 . Liu, C., White, M. & Newell, G. 2013. Selecting thresholds for the prediction of species occurrence with presence-only data. Journal of Biogeography , 40, 778–789 https://doi.org/10.1111/jbi.12058 . Lovett, G. M., Weiss, M., Liebhold, A. M., Holmes, T. P., Leung, B., Lambert, K. F., Orwig, D. A., Campbell, F. T., Rosenthal, J., Mccullough, D. G., Wildova, R., Ayres, M. P., Canham, C. D., Foster, D. R., Ladeau, S. L. & Weldy, T. 2016. Nonnative forest insects and pathogens in the United States: Impacts and policy options. Ecological Applications , 26, 1437–1455 https://doi.org/10.1890/15-1176 . Mcdowell, N. G., Allen, C. D., Anderson-Teixeira, K., Aukema, B. H., Bond-Lamberty, B., Chini, L., Clark, J. S., Dietze, M., Grossiord, C., Hanbury-Brown, A., Hurtt, G. C., Jackson, R. B., Johnson, D. J., Kueppers, L., Lichstein, J. W., Ogle, K., Poulter, B., Pugh, T. a. M., Seidl, R., Turner, M. G., Uriarte, M., Walker, A. P. & Xu, C. 2020. Pervasive shifts in forest dynamics in a changing world. Science , 368, eaaz9463 https://doi.org/10.1126/science.aaz9463 . Meentemeyer, R. K., Cunniffe, N. J., Cook, A. R., Filipe, J. a. N., Hunter, R. D., Rizzo, D. M. & Gilligan, C. A. 2011. Epidemiological modeling of invasion in heterogeneous landscapes: spread of sudden oak death in California (1990–2030). Ecosphere , 2, art17 https://doi.org/10.1890/ES10-00192.1 . Metz, M. R., Frangioso, K. M., Meentemeyer, R. K. & Rizzo, D. M. 2011. Interacting disturbances: wildfire severity affected by stage of forest disease invasion. Ecological Applications , 21, 313–320 https://doi.org/10.1890/10-0419.1 . Morellato, L. P. C., Alberton, B., Alvarado, S. T., Borges, B., Buisson, E., Camargo, M. G. G., Cancian, L. F., Carstensen, D. W., Escobar, D. F. E., Leite, P. T. P., Mendoza, I., Rocha, N. M. W. B., Soares, N. C., Silva, T. S. F., Staggemeier, V. G., Streher, A. S., Vargas, B. C. & Peres, C. A. 2016. Linking plant phenology to conservation biology. Biological Conservation , 195, 60–72 https://doi.org/10.1016/j.biocon.2015.12.033 . Naimi, B. & Araújo, M. B. 2016. sdm: a reproducible and extensible R platform for species distribution modelling. Ecography , 39, 368–375 https://doi.org/10.1111/ecog.01881 . Owens, H. L., Campbell, L. P., Dornak, L. L., Saupe, E. E., Barve, N., Soberón, J., Ingenloff, K., Lira-Noriega, A., Hensz, C. M., Myers, C. E. & Peterson, A. T. 2013. Constraints on interpretation of ecological niche models by limited environmental ranges on calibration areas. Ecological Modelling , 263, 10–18 https://doi.org/10.1016/j.ecolmodel.2013.04.011 . Park Williams, A., Allen, C. D., Macalady, A. K., Griffin, D., Woodhouse, C. A., Meko, D. M., Swetnam, T. W., Rauscher, S. A., Seager, R., Grissino-Mayer, H. D., Dean, J. S., Cook, E. R., Gangodagamage, C., Cai, M. & Mcdowell, N. G. 2013. Temperature as a potent driver of regional forest drought stress and tree mortality. Nature Climate Change , 3, 292–297 https://doi.org/10.1038/nclimate1693 . Peterson, A. T., Soberón, J., Pearson, R. G., Anderson, R. P., Martínez-Meyer, E., Nakamura, M. & Araújo, M. B. 2012. Ecological Niches and Geographic Distributions , Princeton University Press. Ramananjato, V. & Razafindratsima, O. H. 2021. Structure of Microhabitats Used by Microcebus rufus Across a Heterogeneous Landscape. International Journal of Primatology , 42, 682–700 https://doi.org/10.1007/s10764-021-00224-4 . Ramananjato, V., Razafindratsima, Onja. 2021. Data from: Structure of microhabitats used by Microcebus rufus across a heterogeneous landscape [Dataset]. https://doi.org/10.5061/dryad.2280gb5rs Razafindratsima, O. H., Jones, T. A. & Dunham, A. E. 2014. Patterns of movement and seed dispersal by three lemur species. American Journal of Primatology , 76, 84–96 https://doi.org/10.1002/ajp.22199 . Reyer, C. P. O., Brouwers, N., Rammig, A., Brook, B. W., Epila, J., Grant, R. F., Holmgren, M., Langerwisch, F., Leuzinger, S., Lucht, W., Medlyn, B., Pfeifer, M., Steinkamp, J., Vanderwel, M. C., Verbeeck, H. & Villela, D. M. 2015. Forest resilience and tipping points at different spatio-temporal scales: approaches and challenges. Journal of Ecology , 103, 5–15 https://doi.org/10.1111/1365-2745.12337 . Rizzo, D. M. & Garbelotto, M. 2003. Sudden oak death: endangering California and Oregon forest ecosystems. Frontiers in Ecology and the Environment , 1, 197–204 https://doi.org/10.1890/1540-9295(2003 )001[0197:SODECA]2.0.CO;2. Rovero, F. & Ahumada, J. 2017. The Tropical Ecology, Assessment and Monitoring (TEAM) Network: An early warning system for tropical rain forests. Science of The Total Environment , 574, 914–923 https://doi.org/10.1016/j.scitotenv.2016.09.146 . Santini, A., Ghelardini, L., De Pace, C., Desprez-Loustau, M. L., Capretti, P., Chandelier, A., Cech, T., Chira, D., Diamandis, S., Gaitniekis, T., Hantula, J., Holdenrieder, O., Jankovsky, L., Jung, T., Jurc, D., Kirisits, T., Kunca, A., Lygis, V., Malecka, M., Marcais, B., Schmitz, S., Schumacher, J., Solheim, H., Solla, A., Szabò, I., Tsopelas, P., Vannini, A., Vettraino, A. M., Webber, J., Woodward, S. & Stenlid, J. 2013. Biogeographical patterns and determinants of invasion by forest pathogens in Europe. New Phytologist , 197, 238–250 https://doi.org/10.1111/j.1469-8137.2012.04364.x . Slik, J. W. F., Paoli, G., Mcguire, K., Amaral, I., Barroso, J., Bastian, M., Blanc, L., Bongers, F., Boundja, P., Clark, C., Collins, M., Dauby, G., Ding, Y., Doucet, J.-L., Eler, E., Ferreira, L., Forshed, O., Fredriksson, G., Gillet, J.-F., Harris, D., Leal, M., Laumonier, Y., Malhi, Y., Mansor, A., Martin, E., Miyamoto, K., Araujo-Murakami, A., Nagamasu, H., Nilus, R., Nurtjahya, E., Oliveira, Á., Onrizal, O., Parada-Gutierrez, A., Permana, A., Poorter, L., Poulsen, J., Ramirez-Angulo, H., Reitsma, J., Rovero, F., Rozak, A., Sheil, D., Silva-Espejo, J., Silveira, M., Spironelo, W., Ter Steege, H., Stevart, T., Navarro-Aguilar, G. E., Sunderland, T., Suzuki, E., Tang, J., Theilade, I., Van Der Heijden, G., Van Valkenburg, J., Van Do, T., Vilanova, E., Vos, V., Wich, S., Wöll, H., Yoneda, T., Zang, R., Zhang, M.-G. & Zweifel, N. 2013. Large trees drive forest aboveground biomass variation in moist lowland forests across the tropics. Global Ecology and Biogeography , 22, 1261–1271 https://doi.org/10.1111/geb.12092 . Soberón, J., And A. Townsend Peterson 2005. Interpretation of Models of Fundamental Ecological Niches and Species’ Distributional Areas. Biodiversity Informatics , http://dx.doi.org/10.17161/bi.v2i0.4 . Sopniewski, J., Scheele, B. C. & Cardillo, M. 2022. Predicting the distribution of Australian frogs and their overlap with Batrachochytrium dendrobatidis under climate change. Diversity and Distributions , 28, 1255–1268 https://doi.org/10.1111/ddi.13533 . Stevens, P. F. 1980. A Revision Of The Old World Species Of Calophyllum (Guttiferae). Journal of the Arnold Arboretum , 61, 117–424 Sturrock, R. N., Frankel, S. J., Brown, A. V., Hennon, P. E., Kliejunas, J. T., Lewis, K. J., Worrall, J. J. & Woods, A. J. 2011. Climate change and forest diseases. Plant Pathology , 60, 133–149 https://doi.org/10.1111/j.1365-3059.2010.02406.x . Thompson, S. E., Levin, S. & Rodriguez-Iturbe, I. 2014. Rainfall and temperatures changes have confounding impacts on hytophthora cinnamomi occurrence risk in the southwestern USA under climate change scenarios. Global Change Biology , 20, 1299–1312 https://doi.org/10.1111/gcb.12463 . Thuiller, W., Guéguen, M., Renaud, J., Karger, D. N. & Zimmermann, N. E. 2019. Uncertainty in ensembles of global biodiversity scenarios. Nature Communications , 10, 1446 https://doi.org/10.1038/s41467-019-09519-w . Vieilledent, G. 2021. forestatrisk: a Python package for modelling and forecasting deforestation in the tropics. Journal of Open Source Software , 6, 2975 https://doi.org/10.21105/joss.02975 . Vieilledent, G., Grinand, C., Rakotomalala, F. A., Ranaivosoa, R., Rakotoarijaona, J.-R., Allnutt, T. F. & Achard, F. 2018. Combining global tree cover loss data with historical national forest cover maps to look at six decades of deforestation and forest fragmentation in Madagascar. Biological Conservation , 222, 189–197 https://doi.org/10.1016/j.biocon.2018.04.008 . Vieilledent, G., Grinand, C. & Vaudry, R. 2013. Forecasting deforestation and carbon emissions in tropical developing countries facing demographic expansion: a case study in Madagascar. Ecology and Evolution , 3, 1702–1716 https://doi.org/10.1002/ece3.550 . Visser, M. D., Muller-Landau, H. C., Schnitzer, S. A., De Kroon, H., Jongejans, E. & Wright, S. J. 2018. A host–parasite model explains variation in liana infestation among co-occurring tree species. Journal of Ecology , 106, 2435–2445 https://doi.org/10.1111/1365-2745.12997 . Wainhouse, D., Murphy, S., Greig, B., Webber, J. & Vielle, M. 1998. The role of the bark beetle Cryphalus trypanus in the transmission of the vascular wilt pathogen of takamaka (Calophyllum inophyllum) in the Seychelles. Forest Ecology and Management , 108, 193–199 https://doi.org/10.1016/S0378-1127(98)00234-5 . Webber, J. F., Jacobs, K. & Wingfield, M. J. 1999. A re-examination of the vascular wilt pathogen of takamaka (Calophyllum inophyllum). Mycological Research , 103, 1588–1592 https://doi.org/10.1017/S0953756299001021 . Wiehe, P. O. 1949. Wilt of Calophyllum inophyllom L. var. tacamaha (Willd.) REV caused by Haplographium calophylli sp. nov. in Mauritius. Williams, J. N., Seo, C., Thorne, J., Nelson, J. K., Erwin, S., O’brien, J. M. & Schwartz, M. W. 2009. Using species distribution models to predict new occurrences for rare plants. Diversity and Distributions , 15, 565–576 https://doi.org/10.1111/j.1472-4642.2009.00567.x . Wright, P. C., Otero Jimenez, B., Rakotonirina, P., Andriananoely, D. H., Shea, A., Ratalata, B. & Razafimahaimodison, J. C. 2020. The Progressive Spread of the Vascular Wilt Like Pathogen of Calophyllum Detected in Ranomafana National Park, Madagascar. Frontiers in Forests and Global Change , https://doi.org/10.3389/ffgc.2020.00091 . Zurell, D., Elith, J. & Schröder, B. 2012. Predicting to new environments: tools for visualizing model behaviour and impacts on mapped distributions. Diversity and Distributions , 18, 628–634 https://doi.org/10.1111/j.1472-4642.2012.00887.x . Zurell, D., Franklin, J., König, C., Bouchet, P. J., Dormann, C. F., Elith, J., Fandos, G., Feng, X., Guillera-Arroita, G., Guisan, A., Lahoz-Monfort, J. J., Leitão, P. J., Park, D. S., Peterson, A. T., Rapacciuolo, G., Schmatz, D. R., Schröder, B., Serra-Diaz, J. M., Thuiller, W., Yates, K. L., Zimmermann, N. E. & Merow, C. 2020. A standard protocol for reporting species distribution models. Ecography , 43, 1261–1277 https://doi.org/10.1111/ecog.04960 . Tables Table 1 . Standardised range shift metrics - habitat exposure index and potential area of occupancy (pAOO) change for each species and emissions scenario compared between current projections and year 2100. Model Climate pathway* Habitat exposure index ** IUCN pAOO % change ** Wilt pathogen L. calophylli SSP1 0.083 -7.936 SSP3 0.298 -28.986 SSP5 0.315 -30.656 Host C. paniculatum SSP1 0.759 -45.173 SSP3 0.786 -64.047 SSP5 0.830 -67.485 *SSP1 = low-emissions, sustainable socio-economic scenario; SSP3 = high emissions, regional rivalry scenario; SSP5 = high emissions, fossil-fuelled scenario. **Range shift metrics (habitat exposure and pAOO) were computed after converting continuous probability of presence to binary presence-absence predictions. We used species specific binary optimisation thresholds: L. calophylli models were optimised for the minimum distance to ROC curve. Host plant model optimised for maximizing the sum of sensitivity and specificity (TSS). Additional Declarations No competing interests reported. Supplementary Files OnlineResources.docx Cite Share Download PDF Status: Published Journal Publication published 02 May, 2026 Read the published version in Landscape Ecology → Version 1 posted Editorial decision: Revision requested 02 Mar, 2026 Reviews received at journal 02 Mar, 2026 Reviews received at journal 24 Feb, 2026 Reviewers agreed at journal 10 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers agreed at journal 06 Feb, 2026 Reviews received at journal 06 Feb, 2026 Reviewers agreed at journal 03 Feb, 2026 Reviewers agreed at journal 03 Feb, 2026 Reviewers invited by journal 03 Feb, 2026 Editor assigned by journal 31 Jan, 2026 Submission checks completed at journal 31 Jan, 2026 First submitted to journal 30 Jan, 2026 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-8741587","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":585296057,"identity":"2f65b8b0-1cf4-4d2e-b5be-f2a67e07d1ae","order_by":0,"name":"Emma L Underwood","email":"data:image/png;base64,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","orcid":"","institution":"Kingston University","correspondingAuthor":true,"prefix":"","firstName":"Emma","middleName":"L","lastName":"Underwood","suffix":""},{"id":585296058,"identity":"f5c5dea7-e217-442b-a3f0-cbcf67d2685d","order_by":1,"name":"Kerry A Brown","email":"","orcid":"","institution":"Kingston University","correspondingAuthor":false,"prefix":"","firstName":"Kerry","middleName":"A","lastName":"Brown","suffix":""},{"id":585296059,"identity":"9ad74021-ae0b-4f2a-b471-a2d76c1f3721","order_by":2,"name":"Anna Rönnfeldt","email":"","orcid":"","institution":"University of Potsdam","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Rönnfeldt","suffix":""},{"id":585296062,"identity":"380cd0e1-9c5d-4a6d-bcc2-bd9a62ccb8a8","order_by":3,"name":"Mark Mulligan","email":"","orcid":"","institution":"King's College London","correspondingAuthor":false,"prefix":"","firstName":"Mark","middleName":"","lastName":"Mulligan","suffix":""},{"id":585296063,"identity":"ebb2f3ed-1f38-4ba5-88cd-a977def84fcc","order_by":4,"name":"Nigel Walford","email":"","orcid":"","institution":"Kingston University","correspondingAuthor":false,"prefix":"","firstName":"Nigel","middleName":"","lastName":"Walford","suffix":""},{"id":585296064,"identity":"73312522-c36e-4a2a-a833-35e5ee5956a9","order_by":5,"name":"Rebekka Allgayer","email":"","orcid":"","institution":"University of Aberdeen","correspondingAuthor":false,"prefix":"","firstName":"Rebekka","middleName":"","lastName":"Allgayer","suffix":""}],"badges":[],"createdAt":"2026-01-30 13:23:44","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8741587/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8741587/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10980-026-02360-9","type":"published","date":"2026-05-02T15:57:40+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":101957353,"identity":"730c8fc4-4ec5-4e88-aed2-8b9b03b01636","added_by":"auto","created_at":"2026-02-05 11:58:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":214877,"visible":true,"origin":"","legend":"\u003cp\u003eMap of Africa showing the training extent (A) for the fungal wilt pathogen \u003cem\u003eL. calophylli \u003c/em\u003eSDM at 1 km spatial resolution. Madagascar (B) is the prediction extent where both the fungal wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e and host tree distributions were mapped in time and space. \u003cem\u003eVerticillium\u003c/em\u003eand \u003cem\u003eLeptographium\u003c/em\u003e occurrences were downloaded and cleaned from GBIF (2025a) to represent the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e (n = 34), and field surveys bolstered further GBIF occurrences of the host tree (n = 62) \u003cem\u003eC. paniculatum\u003c/em\u003e (Armstrong et al., 2018, Gbif, 2025b, Ramananjato, 2021, Ramananjato and Razafindratsima, 2021, Rovero and Ahumada, 2017). Ranomafana National Park, shown by clustered presence of \u003cem\u003eC. paniculatum\u003c/em\u003e in the south-eastern humid forest was the first recorded sighting of \u003cem\u003eL. calophylli\u003c/em\u003e (Wright et al., 2020). Pseudoabsences were randomly generated for the wilt pathogen which excluded a set distance from pathogen (black circles) and \u003cem\u003eCryphalus\u003c/em\u003e bark beetle dispersal vector occurrences (triangles) (Barbet-Massin et al., 2012, Gbif, 2025c). Seychelles (grey square) and Mauritius (black square) are also identified as nearby islands of known pathogen spread sightings (Webber et al., 1999; Wiehe, 1949).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/f10ebdb9aac3f01fcf02d2e6.png"},{"id":101957328,"identity":"37f2c723-8aaf-4459-80c6-506ecf760478","added_by":"auto","created_at":"2026-02-05 11:57:44","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":241355,"visible":true,"origin":"","legend":"\u003cp\u003eCurrent projected binary presence-absence of (a) fungal wilt of \u003cem\u003eL. calophylli\u003c/em\u003e and (b) \u003cem\u003eC. paniculatum\u003c/em\u003e with associated confidence categories. Confidence was determined through inversing and thresholding each ensemble model’s entropy layer). Green areas on the maps represent high confidence of species presence (where more than 75% of model outputs agree), and purple areas represent lower confidence presences (less than 25% model outputs agree). With the same method, grey areas on the maps represent high confidence absences, and red shows low confidence absences. For the host tree, mapping prediction uncertainty was calculated using the 80th percentile (as 75th led to zero).\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/16aba6b8a09d72158bec6f03.png"},{"id":101957329,"identity":"ab43a170-291d-4892-89e1-97aa041f1614","added_by":"auto","created_at":"2026-02-05 11:57:45","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":585762,"visible":true,"origin":"","legend":"\u003cp\u003eFuture predictions of wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e binary presence-absence across climate scenarios (columns: SSP126, SSP370 and SSP585) and time periods (rows: 2011-2040, 2041-2070, and 2071-2100). Data was converted to binary presence-absence using the ‘minimum distance to ROC curve’ optimisation method in the \u003cem\u003esdm\u003c/em\u003e R package. Each binary prediction visualises high and low confidence for presences and absences, using an uncertainty threshold (high confidence where 75% models in ensemble are in agreement). Green areas on the maps represent high confidence species presence, and purple areas represent low confidence presences. Grey areas on the maps represent high confidence absences, and red shows low confidence absences.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/a750fc4106aa0b6b0a1ffec4.png"},{"id":101957340,"identity":"22953727-0ec4-4271-8627-e23015dbaf44","added_by":"auto","created_at":"2026-02-05 11:57:57","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":452733,"visible":true,"origin":"","legend":"\u003cp\u003eFuture predictions of host \u003cem\u003eC. paniculatum\u003c/em\u003e binary presence-absence across climate scenarios (columns: SSP126, SSP370 and SSP585) and time periods (rows: 2011-2040, 2041-2070, and 2071-2100). Data was converted to binary presence-absence using the ‘maximising sensitivity and specificity’ optimisation method in the \u003cem\u003esdm\u003c/em\u003e R package. Each binary prediction visualises high and low confidence for presences and absences, using an uncertainty threshold (high confidence result is where 80% models in ensemble are in agreement). Green areas on the maps represent high confidence species presence, and purple areas represent low confidence presences. Grey areas on the maps represent high confidence absences, and red shows low confidence absences.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/9882d99342a3cd96ba43ad19.png"},{"id":101957441,"identity":"4dc3ec84-a730-40e8-bd04-b34089bb91f2","added_by":"auto","created_at":"2026-02-05 11:58:17","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":102001,"visible":true,"origin":"","legend":"\u003cp\u003eCumulative range change comparison including predicted percentage contraction (orange – percentage of current range becoming unsuitable in under future climates), stability (light yellow – proportion of current range that is maintained) and expansion (blue – proportion of future range that is newly suitable) for \u003cem\u003eL. calophylli\u003c/em\u003e and \u003cem\u003eC. paniculatum\u003c/em\u003eby climate scenario (SSP126, SSP370, SSP585) and time frame (2011-2040, 2041-2070, 2071-2100). Percentages are higher than 100% as the bars are stacked and proportional change to both current (contraction and stability) and future (expansion).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/5e702fa7024fd95e42319e61.png"},{"id":101957338,"identity":"3a610f4c-5d8e-491a-a3eb-f4312640ce6a","added_by":"auto","created_at":"2026-02-05 11:57:55","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":176316,"visible":true,"origin":"","legend":"\u003cp\u003eCentroid shift direction (cardinal and ordinal) and magnitude (km) shown for wilt pathogen (\u003cem\u003eL. calophylli\u003c/em\u003e – purple) and host (\u003cem\u003eC. paniculatum\u003c/em\u003e – green, \u003cem\u003eL. calophylli\u003c/em\u003e) across three future time frames and future climate scenarios (low-emissions (SSP1) – circle, medium scenario of high-emissions regional rivalry (SSP3) – triangle, high emissions fossil fuelled scenario (SSP5) – square.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/1365347fb4359f6d0a831cdc.png"},{"id":101957358,"identity":"b7e945f7-a4fc-4ee6-aef8-304430dc27b8","added_by":"auto","created_at":"2026-02-05 11:58:04","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":58277,"visible":true,"origin":"","legend":"\u003cp\u003ePredicted elevational range shifts by species, year and scenario between 2010 and 2100. The top row shows wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e (purple), converted to binary presence using the ‘minimum distance to ROC curve’ optimisation threshold. The bottom row shows \u003cem\u003eC. paniculatum\u003c/em\u003e elevational range changes (green) using the ‘maximising the sum of sensitivity and specificity’ threshold. Climate scenarios represent low-emissions, sustainable pathway (SSP1), high-emissions regional rivalry (SSP3) and high-emissions, fossil-fuelled pathway (SSP5).\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/90d4384c9c46095805a7a8b4.png"},{"id":101957359,"identity":"fba0fc8e-736a-4510-b10e-dc249dbc0a7f","added_by":"auto","created_at":"2026-02-05 11:58:04","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":317050,"visible":true,"origin":"","legend":"\u003cp\u003eFuture predicted distributions of \u003cem\u003eL. calophylli\u003c/em\u003e (purple) and \u003cem\u003eC. paniculatum\u003c/em\u003e(green), and their combined overlap (black) for each future year-range (columns) and climate scenario (rows). These distributions have been created using a binary threshold optimisation, maximising Sum of Sensitivity and Specificity for \u003cem\u003eC. paniculatum\u003c/em\u003e and the Minimum Distance to ROC Curve for \u003cem\u003eL. calophylli \u003c/em\u003e(Online Resource 4, \u003cstrong\u003eTable S4\u003c/strong\u003e).\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/b5b54bd906cc21e69a7b798e.png"},{"id":101957354,"identity":"2b92154e-35df-476a-9716-bb6582247d11","added_by":"auto","created_at":"2026-02-05 11:58:02","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":367244,"visible":true,"origin":"","legend":"\u003cp\u003eFuture predicted cumulative spatial overlap \u003cstrong\u003egain\u003c/strong\u003e (blue) and \u003cstrong\u003eloss\u003c/strong\u003e(orange) between the fungal wilt pathogen (\u003cem\u003eL. calophylli\u003c/em\u003e) and host tree (\u003cem\u003eC. paniculatum)\u003c/em\u003e and for each future year-range (columns) and future climate scenario (rows). The underlying predicted distributions in which the overlap was calculated were created using a binary threshold optimisation, Maximising the Sum of Sensitivity and Specificity (opt 2) for \u003cem\u003eC. paniculatum,\u003c/em\u003e and the Minimum Distance to ROC Curve (opt 4) for \u003cem\u003eL. calophylli \u003c/em\u003e(Online Resource 4, \u003cstrong\u003eTable S4\u003c/strong\u003e). Each overlapping cell change is measured between the current projection and the respective future year.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/7dffac8817f7d38d58e38f53.png"},{"id":108495642,"identity":"4d860ec5-eb31-4be5-9e8f-8977d60cefd1","added_by":"auto","created_at":"2026-05-05 10:10:10","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2882779,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/008357a2-4cae-4b83-a574-aa5c46a2d753.pdf"},{"id":101957348,"identity":"11b92a2c-6800-4a31-b758-fa08714df844","added_by":"auto","created_at":"2026-02-05 11:58:00","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":5920662,"visible":true,"origin":"","legend":"","description":"","filename":"OnlineResources.docx","url":"https://assets-eu.researchsquare.com/files/rs-8741587/v1/47824e54e891c3692301a81a.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Climate change facilitates fungal pathogen expansion while driving endemic host range contractions in a tropical biodiversity hotspot","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eEmerging plant pathogens are increasingly recognised as a major driver of biodiversity loss (IPBES, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In tropical island hotspots, where host species are endemic, often range-restricted, ecologically specialised and already stressed by habitat loss and climate change (Brown, 2022), pathogen invasion can have profound consequences. In such systems, pathogen-driven mortality of dominant or functionally unique trees can rapidly alter community structure and ecosystem function, potentially triggering local extinctions and cascading ecological consequences (Boyd et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). For instance, such infections can truncate seed-shadow and recruitment by removing parent trees (Augspurger and Wilkinson, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), reduce fruit and nectar availability that in-turn depress frugivores and pollinators (Razafindratsima et al., \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2014\u003c/span\u003e), diminish above-ground carbon stocks (Slik et al., \u003cspan citationid=\"CR65\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), and alter litter inputs (Fonte and Schowalter, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). In combination, these impacts can shift community composition toward disturbance-tolerant states (e.g., liana-dominant, (Visser et al., \u003cspan citationid=\"CR75\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) and heighten fire susceptibility at forest margins (Metz et al., \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). These changes are difficult to reverse once thresholds are crossed (Reyer et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Consequently, as global change intensifies, understanding how plant-pathogen dynamics may progress temporally and spatially becomes central to predicting and mitigating biodiversity loss.\u003c/p\u003e \u003cp\u003eClimate change influences host-pathogen interactions through changes in temperature and moisture regimes that govern pathogen sporulation, infection efficiency and survival (Elad and Pertot, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2014\u003c/span\u003e, Garrett et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2006\u003c/span\u003e), as well as host stress and defence (Desprez-Loustau et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Sturrock et al., \u003cspan citationid=\"CR69\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Depending on the pathogen-host system, warming, altered precipitation and seasonality patterns may facilitate pathogens to track hosts and expand into previously unoccupied refugia (Bebber et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Conversely, some pathogens might encounter environmental thresholds that limit their survival and spread, thereby creating spatial refugia for host species. The development of robust disease risk models that forecast pathogen distribution under altered climatic conditions is therefore crucial, not only for predicting disease outbreaks but also for informing effective conservation strategies and management practices in vulnerable ecosystems.\u003c/p\u003e \u003cp\u003eMany conservation strategies in invasion ecology rely on correlative species distribution models (SDMs) to predict how species\u0026rsquo; ranges may shift in response to environmental change. These SDMs are widely used to identify climate refugia and prioritise conservation strategies (Li and Wang, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2013\u003c/span\u003e, Peterson et al., \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). However, most SDMs focus solely on abiotic variables and assume species respond independently to environmental gradients. This assumption breaks down when biotic factors, such as host-pathogen dynamics, and human-led dispersal shape a species\u0026rsquo; realised niche, e.g., taking consideration of the \u0026lsquo;BAM\u0026rsquo; framework (Escobar and Craft, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Sober\u0026oacute;n, 2005). For host species facing pathogen pressure, projected range shifts may not reflect future viability if pathogens are able to track or outpace hosts. Conversely, if pathogen ranges contract or shift differently than their hosts, it may result in spatial refugia for the host species, potentially increasing their resilience. These outcomes have fundamentally different implications for extinction risk, yet few SDM-based conservation studies consider such spatial overlap explicitly (Goberville et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, Sopniewski et al., \u003cspan citationid=\"CR67\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCritically, this omission may lead to over-optimistic or misleading conservation assessments. Yet the methodological and data challenges of doing so are considerable, especially for under-studied pathogens in poorly sampled regions. These knowledge gaps are particularly stark in tropical island ecosystems, where high rates of endemism intersect with high vulnerability to invasive species, habitat loss and climate change (Bellard et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Dawson et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Madagascar is emblematic of this convergence. The island harbours exceptional biodiversity, including over 12,000 plant species, of which 83% are endemic (Goodman, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Yet this richness coexists with accelerating environmental degradation: deforestation, driven by slash-and-burn agriculture and fuelwood harvesting, has already reduced the country\u0026rsquo;s natural forest cover to less than 10% of its original extent (Harper et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Vieilledent et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Concurrently, climate models forecast substantial shifts in temperature and precipitation patterns across the island (Brown, 2022; Goodman, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). These stressors interact with increasing biotic pressures, including the incipient introduction of pathogens not previously known to occur in these systems. However, the extent to which disease dynamics will interact with climate and land-use change to reshape species distributions in Madagascar remains largely unknown.\u003c/p\u003e \u003cp\u003eOne illustrative example is the recently described vascular wilt disease affecting \u003cem\u003eCalophyllum paniculatum\u003c/em\u003e, a native tree on Madagascar. The fungal pathogen thought to be \u003cem\u003eLeptographium calophylli\u003c/em\u003e (formerly \u003cem\u003eVerticillium\u003c/em\u003e) has been observed causing mortality in \u003cem\u003eC. paniculatum\u003c/em\u003e populations in Ranomafana National Park since 2016 (Wright et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Although \u003cem\u003eLeptographium\u003c/em\u003e species are globally distributed and known to cause disease in other woody hosts, their ecology and epidemiology in Madagascar\u0026rsquo;s forests remain poorly characterised. The origins of this outbreak are uncertain, which underscores the urgency of proactive monitoring and spatial forecasting under future environmental change.\u003c/p\u003e \u003cp\u003eTo address these ecological knowledge gaps, we use ensemble SDMs to model current and future suitability for both the vascular wilt pathogen (\u003cem\u003eLeptographium calophylli\u003c/em\u003e) and a susceptible endemic host (\u003cem\u003eCalophyllum paniculatum\u003c/em\u003e) across Madagascar under three climate and deforestation scenarios, at three future time slices. The overarching aim of this study is to identify persistent risk hotspots and candidate refugia to inform conservation prioritisation and providing a template for data-poor tropical ecosystems. For the final research objective, we will map host-pathogen spatial coupling and identify mortality and risk hotspots by deriving overlap, divergence and elevational range metrics to classify regions of intensified exposure and potential refugia. One plausible outcome is partial decoupling with persistent overlap, for example, if pathogen suitability remains relatively stable while the host contracts or shifts differently in space or elevation, yielding refugia alongside enduring co-occurrence and sustained mortality risk. The alternatives are equally plausible: (i) tight coupling if both taxa track similar directional and elevational shifts, maintaining high overlap; (ii) risk attenuation if pathogen suitability contracts or shifts away faster than the host, enlarging refugia; or (iii) risk escalation if pathogen suitability expands or shifts faster than the host, increasing exposure.\u003c/p\u003e"},{"header":"MATERIALS AND METHODS","content":"\u003cp\u003eWe chose to accompany this methods section with Overview, Data, Model, Assessment and Prediction (ODMAP) protocols and full code for each species modelled (Zurell et al., \u003cspan citationid=\"CR82\" class=\"CitationRef\"\u003e2020\u003c/span\u003e) in the name of openness and transparency (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/anon-017/anon-fungal-host-sdm\u003c/span\u003e\u003cspan address=\"https://github.com/anon-017/anon-fungal-host-sdm\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). These protocols summarise all parameter decisions and inputs made to build and predict future range shifts of both the focal tree and its wilt pathogen.\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eStudy site\u003c/h2\u003e \u003cp\u003eMadagascar (and Indian Ocean Islands) is one of 36 global biodiversity hotspots, combining exceptional endemism with extensive habitat loss (Mittermeier et al., 2011). The island lies \u003cem\u003eca.\u003c/em\u003e 90 km east of mainland Africa (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent A) and has lost much of its natural forest cover to deforestation and degradation, primarily through shifting agriculture and fuelwood extraction (Brown, 2022, Goodman, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2023\u003c/span\u003e, Harper et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Vieilledent et al., \u003cspan citationid=\"CR74\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eBecause the focal non-native fungal pathogen \u0026ndash; \u003cem\u003eLeptographium calophylli\u003c/em\u003e \u0026ndash; has not yet been systematically recorded in Madagascar but has a documented history on neighbouring Indian Ocean islands and mainland Africa (Webber et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1999\u003c/span\u003e; Wiehe, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1949\u003c/span\u003e), we treated continental Africa as the putative source region and Madagascar as the potential invasion front. We therefore defined two raster extents at 1-km resolution (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e): (i) a training extent covering mainland Africa (extent A) used to model the pathogen\u0026rsquo;s climatic niche based on African occurrences and (ii) a projection extent covering Madagascar (extent B) where we projected both pathogen and host suitability through time. Raster templates for both extents were prepared using Global Administrative Areas (GADM.org) data, including only mainland extents and excluding coastal islands to avoid artefacts along shorelines. Ranomafana National Park (extent B) was highlighted as the location where wilt symptoms on \u003cem\u003eCalophyllum paniculatum\u003c/em\u003e were first reported and monitored (Wright et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), providing an independent reference for evaluating projected invasion risk.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudy species\u003c/h3\u003e\n\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003eEndemic host plant\u003c/h2\u003e \u003cp\u003e \u003cem\u003eCalophyllum\u003c/em\u003e is a genus of evergreen tropical trees in the family Calophyllaceae (formerly \u0026ldquo;Clusiaceae\u0026rdquo;), comprising 87 recognised species globally (IUCN, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Many species in the genus show negative population trends, with 29% assessed as decreasing and 57% are data deficient (IUCN, \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). \u003cem\u003eCalophyllum paniculatum (\u003c/em\u003ehereafter \u003cem\u003eC. paniculatum)\u003c/em\u003e is endemic to mid-level elevation humid and sub-humid ecoregions of Madagascar, where it forms part of the canopy in evergreen rainforest. It is currently categorised as vulnerable on the IUCN Red List (IUCN, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2019\u003c/span\u003e), and recent reports of high adult mortality associated with the vascular wilt (Wright et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), raise concern that climate change, land-use change and pathogen invasion may interact to accelerate its decline.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eFungal wilt pathogen\u003c/h3\u003e\n\u003cp\u003eA vascular wilt disease has been causing increasing adult mortality in \u003cem\u003eC. paniculatum\u003c/em\u003e in Ranomafana National Park since 2016 and is attributed to a fungal pathogen in the genus \u003cem\u003eLeptographium\u003c/em\u003e, most likely \u003cem\u003eLeptographium calophylli\u003c/em\u003e (\u003cem\u003eL. calophylli\u003c/em\u003e) (Wright et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). The fungal wilt \u003cem\u003eL. calophylli\u003c/em\u003e has long been recognised as a serious pathogen of \u003cem\u003eCalophyllum\u003c/em\u003e trees in other tropical and subtropical moist broadleaf biomes in the Indian Ocean islands of Seychelles and Mauritius (Wainhouse et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Webber et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1999\u003c/span\u003e, Wiehe, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1949\u003c/span\u003e). The external symptoms include rapid crown wilting and canopy dieback, while internally the disease is characterised by dark brown vascular streaking visible in the xylem tissue (Wiehe, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1949\u003c/span\u003e, Wright et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eDispersal vector\u003c/h3\u003e\n\u003cp\u003eAlthough aerial dispersal may only contribute locally, long distance spread of \u003cem\u003eL. calophylli\u003c/em\u003e across islands separated by hundreds of kilometres is likely to have been facilitated by human movement of infected timber. An ecologically important vector for the dispersal within forests is the bark beetle, \u003cem\u003eCryphalus\u003c/em\u003e Erichson, 1836. This genus of beetle range in size from 0.8\u0026ndash;3 mm in length and inhabit tree bark where fungal spores attach to their exoskeleton and are transported to new hosts (Wainhouse et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e1998\u003c/span\u003e). At least 25 \u003cem\u003eCryphalus\u003c/em\u003e species occur on the African Continent (GBIF, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025c\u003c/span\u003e) and \u003cem\u003eCryphalus trypanus\u003c/em\u003e has been identified as the principal vector of the \u003cem\u003eCalophyllum\u003c/em\u003e wilt pathogen in Seychelles and Mauritius (Wainhouse et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e1998\u003c/span\u003e).\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eOccurrence records\u003c/h2\u003e \u003cp\u003eTo our knowledge, there are no officially reported geolocated occurrences of the fungal wilt \u003cem\u003eL calophylli\u003c/em\u003e in Madagascar in any global biodiversity database, aside from surveys from Ranomafana National Park (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent B; Wright et al. \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Due to recent taxonomic and historical uncertainty surrounding the fungal pathogen\u0026rsquo;s identity (Wainhouse et al., \u003cspan citationid=\"CR76\" class=\"CitationRef\"\u003e1998\u003c/span\u003e, Webber et al., \u003cspan citationid=\"CR77\" class=\"CitationRef\"\u003e1999\u003c/span\u003e, Wiehe, \u003cspan citationid=\"CR78\" class=\"CitationRef\"\u003e1949\u003c/span\u003e), we broadened our query to the wilt genera \u003cem\u003eVerticillium\u003c/em\u003e Nees, 1816 and \u003cem\u003eLeptographium\u003c/em\u003e Lagerb. \u0026amp; Melin. All records for these genera restricted to mainland Africa (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent A) were downloaded from Global Biodiversity Information Facility (GBIF, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2025a\u003c/span\u003e). We removed records with missing, inconsistent, or obviously incorrect geographic coordinates and excluded records with zero counts. The initial dataset comprised \u003cem\u003eLeptographium\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;10) occurrences from southern Africa and \u003cem\u003eVerticillium\u003c/em\u003e (n\u0026thinsp;=\u0026thinsp;86) occurrences from northern, western, and eastern Africa. After thinning spatial duplicates within 1 km grid cells, n\u0026thinsp;=\u0026thinsp;34 unique fungal wilt occurrences remained. For simplicity, we refer to these the genera-level occurrences as \u003cem\u003eL. calophylli\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eFor the endemic host plant \u003cem\u003eC. paniculatum\u003c/em\u003e, we again, collated all records (n\u0026thinsp;=\u0026thinsp;65) with accepted coordinates (GBIF, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2025b\u003c/span\u003e) and supplemented these with field observations from multiple surveys in Madagascar, including the Tropical Ecological Assessment and Monitoring (TEAM) network (Rovero and Ahumada, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), Centre ValBio (CVB) wilt surveys (Wright et al., \u003cspan citationid=\"CR80\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), a microhabitat study (Ramananjato, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Ramananjato and Razafindratsima, \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2021\u003c/span\u003e) and a soil survey in which the host was recorded as present (Armstrong et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). To minimise temporal duplication, we retained one record per plot for a common survey year (2018) within each dataset where appropriate. This yielded a final set of 62 unique \u003cem\u003eC. paniculatum\u003c/em\u003e occurrences for terrestrial Madagascar (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent B).\u003c/p\u003e \u003cp\u003eTo accompany the collated species presence points within the ensemble modelling framework, pseudoabsences and background points were generated then combined with the presence data for the wilt and host. The fungal wilt pseudoabsences were constructed using the \u0026lsquo;2\u0026deg; far\u0026rsquo; approach (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent A (Barbet-Massin et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2012\u003c/span\u003e)), where following Li et al. (\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), we also combined presence records for the bark beetle vector \u003cem\u003eCryphalus\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, extent A) obtained from GBIF and grouped with the fungal wilt occurrences (GBIF, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2025c\u003c/span\u003e). Host background points were randomly generated across the whole of Madagascar. For more information on these approaches, see Online Resource 3 \u0026ndash; Methods.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eEnvironmental information\u003c/h3\u003e\n\u003cp\u003eTo characterise contemporary and future macroclimate, we extracted bioclimatic variables from \u003cem\u003eCHELSA\u003c/em\u003e v2.1 (Karger et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Karger et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), at each occurrence location. These represent the 19 standard temperature and precipitation predictors at 30 arc-seconds (~\u0026thinsp;1 km at the equator) spatial resolution. These variables capture seasonal variability and extremes to constrain tropical tree distributions and fungal pathogen dynamics. We extracted climate projections from the CMIP6 GFDL-ESM4 general circulation model under three Shared Socioeconomic Pathway scenarios: SSP1-RCP2.6, SSP3-RCP7, and SSP5-RCP8.5 (Krasting et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2018\u003c/span\u003e), spanning low, medium and high emissions futures and thus bracketing plausible climates that could mediate both host range shifts and climate-facilitated pathogen invasion. We assembled these variables across four time periods: 1981\u0026ndash;2010 (current baseline), 2011\u0026ndash;2040, 2041\u0026ndash;2070 and 2071\u0026ndash;2100. In addition, we included forest cover projections from the \u003cem\u003eForestAtRisk\u003c/em\u003e model for Madagascar (Vieilledent, \u003cspan citationid=\"CR72\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, Vieilledent et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). \u003cem\u003eForestAtRisk\u003c/em\u003e provides 30 m resolution maps of historical forest cover and projected deforestation under a \u0026ldquo;business as usual\u0026rdquo; scenario with baseline data for 2010\u0026ndash;2020 and projections for 2040, 2070, and 2100 aligned with our climate time slices. These projections suggest that, under continued deforestation, humid forest in Madagascar could be almost entirely lost by 2100, with large uncertainty in the exact timing of collapse but consistent trends of decline across scenarios (Vieilledent et al., \u003cspan citationid=\"CR73\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). These combined data represented temporally matched, dynamic forest cover layers that were used as a proxy for intact humid and sub-humid forest structure relevant to host \u003cem\u003eC. paniculatum\u003c/em\u003e establishment and to microclimatic buffering of the wilt pathogen and its bark beetle vector.\u003c/p\u003e\n\u003ch3\u003eSpecies distribution modelling (SDM)\u003c/h3\u003e\n\u003cp\u003eWe used ensemble species distribution models (SDMs) to relate species occurrences to climatic and land-cover variables and to project habitat suitability under future scenarios (Ara\u0026uacute;jo and New, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Our framework models an emerging fungal wilt (\u003cem\u003eL. calophylli\u003c/em\u003e) and its endemic host tree \u003cem\u003eC. paniculatum\u003c/em\u003e to: (i) characterise their current environmental niches, (ii) forecast potential invasion and range shifts under combinations of climate change and deforestation and (iii) quantify changing spatial overlap between pathogen and host. In the context of Madagascar\u0026rsquo;s highlighted endemic flora, these models provide an effective way to explore how climate-mediated pathogen invasion could interact with ongoing habitat loss to reshape host distributions and extinction risk.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003ePredictor variable selection\u003c/h2\u003e \u003cp\u003eBecause both the pathogen and host were represented by relatively few occurrences, we restricted SDM complexity and minimised multicollinearity among predictors to avoid overfitting (Dormann et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Initially, we considered the 19 \u003cem\u003eCHELSA\u003c/em\u003e bioclimatic variables and \u003cem\u003eForestAtRisk\u003c/em\u003e forest cover and quantified collinearity separately for the pathogen (African training extent) and host (Madagascar) datasets using Spearman\u0026rsquo;s rank correlation matrices and Variance Inflation Factors (VIF). Highly correlated pairs (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left|r\\right|\\ge\\:0.7\\)\u003c/span\u003e\u003c/span\u003e) were identified and one variable from each pair was removed based on lower VIF, yielding a pool of non-collinear predictors for each species (Online Resource 4, \u003cb\u003eFigures S2-3\u003c/b\u003e). To keep model complexity, proportionate to sample size, we limited the number of climate predictors to approximately one variable per 8\u0026ndash;10 presence records. For the fungal pathogen (\u003cem\u003eL. calophylli\u003c/em\u003e) this resulted in a maximum of four climate predictors, while for the host plant (\u003cem\u003eC. paniculatum\u003c/em\u003e) we allowed for eight variables, including the \u003cem\u003eForestAtRisk\u003c/em\u003e forest cover predictor. These were ecologically interpretable and formed the basis for SDM fitting and subsequent ensemble model predictions.\u003c/p\u003e \u003cp\u003eFor the fungal wilt pathogen, the final non-collinear climate predictors were: (i) precipitation of the driest month, (ii) precipitation seasonality, (iii) mean temperature of the coldest quarter, and (iv) precipitation of the coldest quarter. Together, these variables captured dry-season moisture stress (Desprez-Loustau et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), intra-annual rainfall variability (Thompson et al., \u003cspan citationid=\"CR70\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) and cool season thermal (Aguayo et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2014\u003c/span\u003e) and moisture regimes (Burgess et al., \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) that are known to constrain survival, overwintering and infection windows of vascular wilt fungi. For the endemic host tree, the final predictors comprised: (i) temperature seasonality, (ii) mean temperature of the driest quarter, (iii) annual temperature range, (iv) precipitation of the wettest month, (v) precipitation of the coldest quarter, and (vi) forest cover. These variables summarise thermal stability and extremes (Engelbrecht et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Morellato et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2016\u003c/span\u003e, Park Williams et al., \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), water availability during both wet and cool seasons (Esquivel-Muelbert et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2017\u003c/span\u003e, Mcdowell et al., \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), and the presence of closed-canopy forest, consistent with \u003cem\u003eC. paniculatum\u003c/em\u003e as a humid rainforest tree whose distribution is shaped by relatively stable tropical climates and the persistence of intact forest structure (De Moraes et al., \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, Stevens, \u003cspan citationid=\"CR68\" class=\"CitationRef\"\u003e1980\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eEnsemble model fitting and validation\u003c/h2\u003e \u003cp\u003eWe used a weighted ensemble approach to reduce bias from individual algorithms and better capture uncertainties in predictions across climate and deforestation scenarios (Ara\u0026uacute;jo and New, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2007\u003c/span\u003e, Thuiller et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). Ensemble models were built using the \u003cem\u003esdm\u003c/em\u003e R package (Naimi and Ara\u0026uacute;jo, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). For each species, we fitted three algorithms that perform well with limited occurrence data: Boosted Regression Tree (BRT), Random Forest (RF) and MaxEnt (Williams et al., \u003cspan citationid=\"CR79\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). Models were replicated ten times for each algorithm using bootstrap resampling, producing 30 models \u003cem\u003eper\u003c/em\u003e species in the final model. For each ensemble, to account for the much larger number of background data relative to presences, we applied a weighting ratio such that each pseudo-absence/background point contributed one-tenth of the weight of a presence, making the total weighted contribution of presences and pseudo-absences/background points approximately equal in each model replicate. We used five-fold cross-validation, with an 80:20 training-testing split in each fold (Ara\u0026uacute;jo et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). Two- and three-dimensional partial response curves were generated for all predictors and suitability predictions were averaged across replicates and algorithms to obtain ensemble mean predictions for subsequent performance evaluation and projection (Online Resource 4, \u003cb\u003eFigures S11-12\u003c/b\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eEnsemble model performance evaluation\u003c/h2\u003e \u003cp\u003eWe assessed the performance of each individual model replicate using threshold-independent discrimination metrics: Area under the receiver operating curve (AUC), True Skill Statistic (TSS) and the standard deviation of both for each cross-validation fold (Allouche et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e, Liu et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The top trained, ensembles were inspected for ecologically plausible response curves and spatial distribution patterns before being retained for future projections (Online Resource 4, \u003cb\u003eFigures S4-5\u003c/b\u003e). For the final wilt ensemble, we chose Matthews Correlation Coefficient (MCC) as the primary evaluation metric due to its robustness in novel environmental conditions and effectiveness in handling class imbalance (Boughorbel et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). For the host tree \u003cem\u003eC. paniculatum\u003c/em\u003e, the final ensemble used AUC-weighted averaging, further optimised with respect to maximising TSS to balance omission and commission errors.\u003c/p\u003e \u003cp\u003eFor each species, we quantified variable importance for the top ensemble model by permuting predictors one at a time and measuring the resulting loss in predictive performance (Dormann et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In the permutation approach, values of a given predictor were randomly shuffled across sites while all other predictors were kept unchanged, so the greater the reduction in model performance, the higher the inferred importance of that predictor (Online Resource 4, \u003cb\u003eFigures S9-10\u003c/b\u003e). Importance scores were averaged across algorithms within the top ensemble and rescaled to facilitate comparison among variables within each species (Naimi and Ara\u0026uacute;jo, \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eThresholding to create binary presence/absence\u003c/h2\u003e \u003cp\u003eTo analyse future range shift and host-pathogen overlap, we converted continuous probability of presence maps (Online Resource 4, \u003cb\u003eFigures S7-8\u003c/b\u003e) to binary presence-absence maps for each species, climate scenario and time period. We used the \u003cem\u003ethreshold\u003c/em\u003e function in the \u003cem\u003esdm\u003c/em\u003e package to identify an optimal binary cut-off for each ensemble, weighting towards the best-performing algorithms and replicates. For the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e, we used the minimum distance to ROC curve (\"minROCdist\") optimiser to improve transferability between Africa and Madagascar and reduce overprediction in novel climates (Owens et al., \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). For the host tree \u003cem\u003eC. paniculatum\u003c/em\u003e, modelled within its native range, we optimised the weighted ensemble towards the maximum sum of sensitivity and specificity (\"max(se\u0026thinsp;+\u0026thinsp;sp)\", or TSS) to balance omission and commission error rates (Liu et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). The binary maps were used to calculate distributional changes using area, centroid, altitudinal changes, as well as habitat exposure, and the predicted area of occupancy (pAOO), following Choe et al. (\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) and the IUCN (\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). See Online Resource 3, \u003cb\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eMapping novel environments and uncertainty\u003c/h2\u003e \u003cp\u003eWe performed a Multivariate Environmental Similarity Surface (MESS) analysis on the top ensemble for each species to assess the similarity between current and future climatic conditions (Elith et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2010\u003c/span\u003e, Zurell et al., \u003cspan citationid=\"CR81\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Given the model transfer from Africa to Madagascar, this was particularly helpful to analyse any differences in conditions between training and testing (geographic disparity) as well as changes in environmental conditions over time. In addition, we evaluated the spatial uncertainty of predictions, using high and low confidence presence and absence categories, allowing for spatially explicit interpretation of model outputs mapped across Madagascar. For this, we used a percentile-based approach to threshold the uncertainty (entropy) layer created for each ensemble, using 75% and 25% as thresholds for high and low confidence, respectively (Thuiller et al., \u003cspan citationid=\"CR71\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). For high confidence predictions, we applied a threshold at the point where \u0026gt;\u0026thinsp;75% of models were in agreement for each pixel (the inverse of the entropy layer). The same was applied for low confidence predictions, where a threshold was chosen where \u0026lt;\u0026thinsp;25% of models agreed. We wrote custom code to make this process dynamic, testing 5% increments to account for any cases of high agreement across model algorithms and replicates in case the default resulted in a zero-value threshold. This ensured the uncertainty measurements were specific to each species and ensemble model.\u003c/p\u003e \u003cp\u003ePreliminary environmental data processing was carried out in QGIS (3.40 LTR). All further data preparation, modelling and analysis was performed using R Statistical Software (v4.3.2 (2023-10-31 ucrt); (R Core Team, 2023)) with packages \u003cem\u003ergbif\u003c/em\u003e (Chamberlain S, 2017; Chamberlain and Boettiger, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2017\u003c/span\u003e) and \u003cem\u003eterra\u003c/em\u003e (Hijmans, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) for spatial data processing of pseudoabsences and background points. Further methods information is found in Online Resources 1 \u0026amp; 2 (ODMAP protocols).\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003eProjections of host pathogen distributions under current climate\u003c/h2\u003e \u003cp\u003eWhen visualising current projections for the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea), the pathogen appears to be suitable across a larger and overlapping geographic extent than host \u003cem\u003eC. paniculatum\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). There are high confidence areas of suitable habitat across the entire north-south humid belt of the island, and high confidence absences to the west, largely encompassing the entire distribution of C. p\u003cem\u003eaniculatum\u003c/em\u003e, although the host distribution is sparser within the same humid and sub-humid belts. There are some small patches of uncertainty (low confidence absence \u0026ndash; red) within around the outside of the projected distribution for the wilt, and in the south and west which does not overlap with the host (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). The host presence distribution is small and isolated, with low confidence absences projected along western coastal fringes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eSpatial-temporal patterns of future range dynamics\u003c/h2\u003e \u003cp\u003eOverall, ensemble SDMs for both species project changes in the geographic distributions of suitable habitat of the fungal wilt \u003cem\u003eL. calophylli\u003c/em\u003e, and endemic host, \u003cem\u003eC. paniculatum\u003c/em\u003e across all climate scenarios examined. These projections, however, reveal markedly different trajectories for the pathogen and its host, suggesting potential disruption of their current spatial relationship by the end of the century. The wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e, demonstrates considerably greater resilience under future climates, with projected range contractions of only 7.7\u0026ndash;31.6% across all scenarios and timeframes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, Online Resource 4, \u003cb\u003eTable S5\u003c/b\u003e). Even under the most extreme climate scenario (SSP5) by 2071\u0026ndash;2100, the pathogen is projected to retain 68.5% of its current range.\u003c/p\u003e \u003cp\u003eIn contrast to the wilt pathogen, \u003cem\u003eC. paniculatum\u003c/em\u003e is projected to experience severe range shifting, with an overall net contraction across all scenarios, except for the first period (2011\u0026ndash;2040 for the low-emission pathway only) (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e). Moderate reductions of 22.5\u0026ndash;41.1% under the low emissions pathway become progressively more severe under higher emission scenarios, with the most extreme reductions of 65.9% by the end of the century observed under the fossil-fuelled shared socioeconomic pathway (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Online Resource 4, \u003cb\u003eTable S5\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eMeasuring range differences along a temporal gradient reveals distinctly different trajectories of range stability, contraction and expansion. The fungal pathogen shows consistent range stability across scenarios and time periods, with slight expansion into new range by 2040 under high emissions scenario, and even minor expansion by end of century under low emissions scenario (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Meanwhile, a progressive intensification of range contractions across all scenarios (except the low-emissions pathway) are predicted for the host \u003cem\u003eC. paniculatum\u003c/em\u003e, with the most dramatic changes occurring in the latter half of the century, equating to more than double range losses compared to the wilt (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Towards the end of the century (2071\u0026ndash;2100), the most dramatic changes occur, particularly for \u003cem\u003eC. paniculatum\u003c/em\u003e, which predicts the highest contractions of original range area, along with the lowest levels of stability and expansion compared to earlier in the century. Equally, the pathogen - although still experiencing some range contractions during this period (8.3\u0026ndash;31.5%) - maintains relatively high range stability (68.5\u0026ndash;91.7%) (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e, Online Resource 4, \u003cb\u003eTable S5\u003c/b\u003e).\u003c/p\u003e \u003cp\u003e \u003cem\u003eL. calophylli\u003c/em\u003e exhibits conservative spatial shifts with a consistent directional pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, purple icons). These directional shift results reveal a potential spatial decoupling in the geographic responses of host and pathogen. For \u003cem\u003eC. paniculatum\u003c/em\u003e, predominant shifts were observed in southward and eastward directions across most scenarios, with substantial centroid shifts ranging from 29.7-111.2 km (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e, green icons). \u003cem\u003eC. paniculatum\u003c/em\u003e is also predicted to move upslope in the next century, showing a mean elevational increase of 85m by 2100 (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). This directional divergence, with species boundaries moving in different cardinal directions at significantly different rates, suggests potential disruption of their current spatial relationship under future climate scenarios, particularly under high emission pathways and later time periods.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eHost-pathogen elevational relationship\u003c/h2\u003e \u003cp\u003eElevation responses show some consistency between host and pathogen (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e), with both species generally shifting upward in elevation at different rates and thresholds for the focal tree. \u003cem\u003eC. paniculatum\u003c/em\u003e shows more pronounced elevational shifts (maximum vertical shift height of 253.6 m during 2071\u0026ndash;2100 with SSP3) than the pathogen (up to 124.0 m for the SSP5 scenario). These different rates of elevational migration could temporarily alter disease dynamics along elevation gradients, although given their geographic distributions, the predicted probability of presence of the pathogen exceeds the overall elevational range of \u003cem\u003eC. paniculatum.\u003c/em\u003e\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eGeographic range overlap\u003c/h2\u003e \u003cp\u003eThe predicted distribution of \u003cem\u003eL. calophylli\u003c/em\u003e dominates the eastern forest corridor, maintaining stable geographic distribution into the future (Figs.\u0026nbsp;8\u0026ndash;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e9\u003c/span\u003e). In contrast, the host, \u003cem\u003eC. paniculatum\u003c/em\u003e, has a very limited predicted distribution across Madagascar, persisting in patches in the north and isolated locations on the eastern humid belt under future climate and land use scenarios (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Distribution and overlap changes demonstrate the increasingly limited distribution of \u003cem\u003eC. paniculatum\u003c/em\u003e along the northern and central eastern zones, while for the wilt \u003cem\u003eL. calophylli\u003c/em\u003e, a pattern of expanding distribution along Madagascar's humid forest corridor. Their areas of overlap are concentrated along the central-eastern forests, particularly in the north (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e9\u003c/span\u003e). Across all climate and land use pathways, there is a clear geographic stability of \u003cem\u003eL. calophylli\u003c/em\u003e to 2100, with the low-emission fossil fuelled pathway showing the most extensive geographic distribution of the pathogen by the late century. Their shared overlap appears to reduce over time, both by shared pathway intensity and future year range, likely due to the overall contraction in the range of \u003cem\u003eC. paniculatum\u003c/em\u003e (Fig.\u0026nbsp;8). The overlap change patterns suggest a southward shift in overlap areas over time.\u003c/p\u003e \u003cp\u003eStandardised range shift metrics (habitat exposure and predicted area of occupancy) show \u003cem\u003eC. paniculatum\u003c/em\u003e to be much more vulnerable to future change drivers than the wilt pathogen \u003cem\u003eL. calophylli.\u003c/em\u003e Both show increasing vulnerability as the intensity of future climate scenario increases from 1 (low emissions) to 5 (high emissions), but \u003cem\u003eC. paniculatum\u003c/em\u003e shows steeper declines and more variability of results across scenarios (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). \u003cem\u003eC. paniculatum\u003c/em\u003e exhibits high habitat exposure values (0.759\u0026ndash;0.830), particularly under higher emission scenarios, indicating widespread habitat transformation to new areas by 2100. Climate exposure is lower for the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e, and lowest (0.083) under the low-emission pathway (SSP1) but increases substantially under higher emission scenarios (0.298\u0026ndash;0.315). The difference in habitat exposure between species suggests substantially greater climate sensitivity for the host tree. Potential area of occupancy (pAOO) changes show on average, that \u003cem\u003eC. paniculatum\u003c/em\u003e is forecasted to lose up to 67% of its current area by 2100, compared to \u003cem\u003eL. calophylli\u003c/em\u003e, which is a much lower predicted loss at up to 31% (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003ch2\u003eEnvironmental change expands invasion opportunity in Madagascar\u003c/h2\u003e\n\u003cp\u003eThis study has three key findings that together clarify how global environmental change can reshape host-pathogen dynamics and invasion processes in tropical biodiversity hotspots. First, climate change consistently maintains or expands climatically suitable habitat for the novel pathogen \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003eacross much of Madagascar, even under high-emissions scenarios. Second, the endemic host \u003cem\u003eC. paniculatum\u003c/em\u003e is projected to undergo consistent range contraction across all climate scenarios, compounded by fragmentation and dispersal limitation from deforestation, resulting in a small number of geographically isolated refugia by late century. Third, these contrasting responses generate a pronounced spatial and climatic decoupling between host and pathogen distributions, such that pathogen suitability increasingly exceeds and spatially envelops the shrinking host range. Similar host-pathogen mismatches have been documented historically in temperate regions following introduction of invasive pathogens under changing environmental conditions, most notably the extirpation of American chestnut (\u003cem\u003eCastanea dentata\u003c/em\u003e) by the introduced fungus \u003cem\u003eCryphonectria parasitica\u003c/em\u003e (Anagnostakis, 1987, Santini et al., 2013) and the widespread collapse of elm populations following invasion of Dutch elm disease (\u003cem\u003eOphiostoma spp\u003c/em\u003e) (Karnosky, 1979, Santini et al., 2013). In those invasions, pathogen persistence and spread far exceeded the climatic and demographic resilience of their primary hosts, restructuring forest composition for decades to centuries (Santini et al., 2013). Our projections suggest that \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003emay follow a comparable trajectory in Madagascar, persisting across landscapes even as the endemic host becomes increasingly rare and spatially fragmented.\u003c/p\u003e\n\u003cp\u003eThe projected stability and breadth of the climatic niche of \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003eindicate that the pathogen is unlikely to be constrained by future climate change in the same way its host is. By the end of the century, the pathogen occupies orders of magnitude more climatically suitable area than \u003cem\u003eC. paniculatum\u003c/em\u003e under high-emissions scenarios, suggesting that it may persist independently of this host. This pattern is consistent with invasion dynamics observed in other forest pathogens, where multi-host strategies allow pathogens to maintain populations in a reservoir host even as susceptible species decline (Desprez-Loustau et al., 2007). It is possible that regardless of whether \u003cem\u003eC. paniculatum\u0026nbsp;\u003c/em\u003ebecomes locally extirpated across its range, \u003cem\u003eL. calophylli\u003c/em\u003e may remain in local forests, increasing the likelihood of spill over into other native species or reinfection of remnant populations.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBy maintaining the area of climatically suitable habitat for \u003cem\u003eL. calophylli,\u003c/em\u003e climate change effectively increases the invasion opportunity available to this novel pathogen. Here, \u003cem\u003einvasion opportunity\u003c/em\u003e refers to the total extent of environments in which establishment and long-term persistence are climatically plausible. As suitable landscape increases, the probability that repeated introduction events result in successful establishment also increases, even if the likelihood of establishment per individual introduction remains unchanged. This distinction is critical for forest pathogens, where propagule pressure is typically chronic rather than episodic (Garbelotto and Pautasso, 2012, Lovett et al., 2016). For instance, continued movement of timber or contaminated equipment can generate multiple introduction events over time, and a larger suitable area increases the chance that at least some of these introductions occur in environments conducive to persistence (Santini et al., 2013). Alternatively, if climate change reduces suitable habitat for \u003cem\u003eL. calophylli,\u003c/em\u003e repeated introductions would more often fail because arrival events would disproportionately occur in environmentally mismatched locations (Meentemeyer et al., 2011). However, our projections suggest the opposite trajectory. By expanding or maintaining climatic suitability across large parts of the island, environmental change weakens climatic barriers to establishment and makes invasion outcomes increasingly dependent on propagule pressure rather than environmental mismatch. In this way, climate change increases invasion opportunity and increases the likelihood that repeated introductions translate into establishment (Elad and Pertot, 2014), causing early mortality to already climatically weakened hosts like \u003cem\u003eC. paniculatum\u003c/em\u003e.\u003c/p\u003e\n\u003cp\u003eImportantly, the asymmetry of future habitat suitability between host and pathogen highlighted in this study is supported by multiple, independent diagnostics rather than being an artefact of model transfer or threshold choice. The MESS analysis (Online Resource 4, \u003cstrong\u003eFigure S14\u003c/strong\u003e) indicated that future projections for \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003eare largely made within environmental space represented by the training data, reducing the likelihood that the observed broad pathogen suitability arises from extrapolation into novel climates. Additionally, uncertainty mapping shows persistent areas of high model agreement for the pathogen across Madagascar’s humid forests. Together, these diagnostics strengthen confidence that the projected divergence reflects a real difference in (pathogen) climatic tolerance and (host) vulnerability, rather than modelling artefacts. Notwithstanding, the predicted distributions represent fully realised climatic suitability rather than realised occupancy and therefore cannot be interpreted as direct predictions of current spread or disease extent.\u003c/p\u003e\n\u003ch2\u003eManagement implications and future conservation directions\u003c/h2\u003e\n\u003cp\u003eThe contrasting responses the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e to the endemic host tree \u003cem\u003eC. paniculatum\u0026nbsp;\u003c/em\u003edepict a troubling dimension to potential climate change impacts that extends beyond species range shifts. While the host faces clear range contractions, the pathogen shows steady tolerance to changes in the climate over time, with minor projected range losses across the same scenarios and time frames. This disparity creates an ecological mismatch that has concerning implications for forest health and conservation planning. In the current-day projected distribution of \u003cem\u003eL. calophylli\u003c/em\u003e, the geographical range of presence spans a broader climate niche, extending much further longitudinally, with reducing but visible presence stretching from sea-level at the eastern coastline all the way westward, through the humid into the sub-humid region, and even into drier savanna land towards the centre of the country. This broader distribution suggests the pathogen can persist under climate conditions that far exceed the physiological limits of \u003cem\u003eC. paniculatum\u003c/em\u003e. \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003emaintains suitable habitat in most of Madagascar’s central spine and eastern escarpment, covering most of the humid zone, and parts of the eastern sub-humid zone. This suggests that the pathogen could survive as a generalist and not require \u003cem\u003eC. paniculatum\u0026nbsp;\u003c/em\u003especifically as a host for survival, where it may utilise other species as a ‘reservoir’ in areas where their ranges to do intersect (Grünwald et al., 2008, Rizzo and Garbelotto, 2003).\u003c/p\u003e\n\u003cp\u003eIn addition, the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e causing mortality of \u003cem\u003eCalophyllum paniculatum\u0026nbsp;\u003c/em\u003ein Madagascar and \u003cem\u003eCalophyllum inophyllum\u0026nbsp;\u003c/em\u003eon nearby tropical islands is still an unknown to the conservation community. In the face of becoming widespread before official identification, our study provides a timely assessment of the potential impacts of \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003eon a focal endemic tree species in a tropical biodiversity hotspot. We provide suggestions for how predictive ecological modelling can support wider efforts to identify, monitor, and limit its spread by recommending that future work focuses on the following research questions and actions:\u003c/p\u003e\n\u003col\u003e\n \u003cli\u003eHost range and other targets\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003eResearch question:\u003c/strong\u003e Do we know with certainty, that \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003edoes not also infect or impact other endemic species or genera in Madagascar?\u0026nbsp;\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eAction\u003c/strong\u003e: Collaborate with local partners to conduct systematic field sampling of multiple species across sites within the core probability of presence distributions of \u003cem\u003eL. calophylli\u003c/em\u003e.\u0026nbsp;\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eAction\u003c/strong\u003e: Collaborate with pathologists to officially identify the pathogen.\u003c/li\u003e\n\u003c/ul\u003e\n\u003col start=\"2\"\u003e\n \u003cli\u003ePhase of spread\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003eResearch question\u003c/strong\u003e: What phase of spread is the pathogen in?\u0026nbsp;\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eAction\u003c/strong\u003e: random sampling of species and sites within and \u003cem\u003eoutside\u0026nbsp;\u003c/em\u003eof the probability of presence distribution of \u003cem\u003eL. calophylli\u003c/em\u003e.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eAction\u003c/strong\u003e: Multi-criteria analysis (MCA) to map the possible spread start point and dispersal patterns, using anthropogenic and natural spread routes: ferry ports, roads, known logging areas, built-up areas, hiking paths, prevailing wind direction and river catchments.\u003c/li\u003e\n\u003c/ul\u003e\n\u003col start=\"3\"\u003e\n \u003cli\u003eApplication of global tools\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003e\u003cstrong\u003eResearch question\u003c/strong\u003e: How can IPBES tools and suggestions for invasive species management (IPBES, 2023) be adapted to the case of \u003cem\u003eL. calophylli\u0026nbsp;\u003c/em\u003ein the context of tropical, island biodiversity?\u0026nbsp;\u003c/p\u003e\n\u003cul\u003e\n \u003cli\u003e\u003cstrong\u003eAction\u003c/strong\u003e: Assess applicability of IPBES frameworks (IPBES, 2023) for prevention, early detection and rapid response in Madagascar’s ecological and policy context.\u003c/li\u003e\n\u003c/ul\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eWe highlight that the range dynamics of a non-native, incipient pathogen can become a bigger threat to endemic island species under climate change. We identified and compared the current and future predicted geographic distributions of the wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e in relation to those of host \u003cem\u003eC. paniculatum\u003c/em\u003e, and quantified predicted range changes and potential future spatial overlap. We showed the already sparse, currently suitable habitat of \u003cem\u003eC. paniculatum\u003c/em\u003e may reduce further to isolated fragments in the central and north sub-humid ecoregions, as a direct result of combined climate change and deforestation by the end of this century. On the other hand, the climate resilience shown by the fungal wilt pathogen \u003cem\u003eL. calophylli\u003c/em\u003e, across all measured climate projections, without a strong reliance on forested areas for survival could further put endemic vulnerable host species at further risk. These findings should alert the conservation community to explore the impacts of multiple global and regional change drivers simultaneously for other vulnerable and endangered species across the tropics, to better understand current future threats to species\u0026rsquo; survival, and to limit further biodiversity loss across terrestrial biodiversity hotspots globally.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003ch2\u003eCompeting Interests\u003c/h2\u003e \u003cp\u003eThe authors have no relevant financial or non-financial interests to disclose.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eFunding\u003c/h2\u003e \u003cp\u003eThe author(s) received no financial support for the research, authorship, and/or publication of this article.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAll authors contributed to the study conception and design. Material preparation and analysis were performed by E.L.U, supported by A.R and R.A and K.A.B. The first draft of the manuscript was written by E.L.U, supported by K.A.B, A.R and R.A. All authors reviewed this and provided comments on previous versions of the manuscript. All authors read and approved the final manuscript.\u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e\u003cp\u003eThe authors would like to thank Professor Patricia Chapple Wright and team at the Stony Brook University Institute for the Conservation of Tropical Environments (ICTE) Centre ValBio for open data sharing on current information pertaining to the wilt pathogen spread in Madagascar.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003eAll data used in this study are derived from publicly available repositories and referenced in the manuscript accordingly:1. Global portals: [GBIF](https:/www.gbif.org) , [CHELSA](https:/www.chelsa-climate.org) , [ForestAtRisk](https:/forestatrisk.cirad.fr)2. Publicly available survey data from published peer-reviewed articles:1. Ramananjato, V., Razafindratsima, Onja. 2021. Data from: Structure of microhabitats used by Microcebus rufus across a heterogeneous landscape [Dataset]. [https://doi.org/10.5061/dryad.2280gb5rs](https:/doi.org/10.5061/dryad.2280gb5rs)2. Armstrong, A. H., Shugart, H. H. \u0026amp; Fatoyinbo, T. E. 2011. Characterization of Community Composition and Forest Structure in a Madagascar Lowland Rainforest. Tropical Conservation Science, 4, 428-444. [https://doi.org/10.1177/194008291100400406](https:/doi.org/10.1177/194008291100400406)3. Wright, P. C., Otero Jimenez, B., Rakotonirina, P., Andriananoely, D. H., Shea, A., Ratalata, B. \u0026amp; Razafimahaimodison, J. C. 2020. The Progressive Spread of the Vascular Wilt Like Pathogen of Calophyllum Detected in Ranomafana National Park, Madagascar. Frontiers in Forests and Global Change. [https://doi.org/10.3389/ffgc.2020.00091](https:/doi.org/10.3389/ffgc.2020.00091) .All custom code used for all preparation and analyses can be found at: [https://github.com/anon-017/anon-fungal-host-sdm](https:/github.com/anon-017/anon-fungal-host-sdm) .\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAguayo, J., Elegbede, F., Husson, C., Saintonge, F.-X. \u0026amp; Mar\u0026ccedil;ais, B. 2014. Modeling climate impact on an emerging disease, the Phytophthora alni-induced alder decline. \u003cem\u003eGlobal Change Biology\u003c/em\u003e, 20, 3209\u0026ndash;3221 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.12601\u003c/span\u003e\u003cspan address=\"10.1111/gcb.12601\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllouche, O., Tsoar, A. \u0026amp; Kadmon, R. 2006. Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). \u003cem\u003eJournal of Applied Ecology\u003c/em\u003e, 43, 1223\u0026ndash;1232 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-2664.2006.01214.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-2664.2006.01214.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnagnostakis, S. L. 1987. Chestnut Blight: The Classical Problem of an Introduced Pathogen. \u003cem\u003eMycologia\u003c/em\u003e, 79, 23\u0026ndash;37 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/00275514.1987.12025367\u003c/span\u003e\u003cspan address=\"10.1080/00275514.1987.12025367\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAra\u0026uacute;jo, M. B. \u0026amp; New, M. 2007. Ensemble forecasting of species distributions. \u003cem\u003eTrends in Ecology \u0026amp; Evolution\u003c/em\u003e, 22, 42\u0026ndash;47 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tree.2006.09.010\u003c/span\u003e\u003cspan address=\"10.1016/j.tree.2006.09.010\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAra\u0026uacute;jo, M. B., Pearson, R. G. \u0026amp; Rahbek, C. 2005. Equilibrium of Species' Distributions with Climate. \u003cem\u003eEcography\u003c/em\u003e, 28, 693\u0026ndash;695\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eArmstrong, A., Fischer, R., Huth, A., Shugart, H. \u0026amp; Fatoyinbo, T. 2018. Simulating Forest Dynamics of Lowland Rainforests in Eastern Madagascar. \u003cem\u003eForests\u003c/em\u003e [Online], 9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAugspurger, C. K. \u0026amp; Wilkinson, H. T. 2007. Host Specificity of Pathogenic Pythium Species: Implications for Tree Species Diversity. \u003cem\u003eBiotropica\u003c/em\u003e, 39, 702\u0026ndash;708 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1744-7429.2007.00326.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1744-7429.2007.00326.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarbet-Massin, M., Jiguet, F., Albert, C. H. \u0026amp; Thuiller, W. 2012. Selecting pseudo-absences for species distribution models: how, where and how many? \u003cem\u003eMethods in Ecology and Evolution\u003c/em\u003e, 3, 327\u0026ndash;338 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.2041-210X.2011.00172.x\u003c/span\u003e\u003cspan address=\"10.1111/j.2041-210X.2011.00172.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBebber, D. P., Ramotowski, M. a. T. \u0026amp; Gurr, S. J. 2013. Crop pests and pathogens move polewards in a warming world. \u003cem\u003eNature Climate Change\u003c/em\u003e, 3, 985\u0026ndash;988 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/nclimate1990\u003c/span\u003e\u003cspan address=\"10.1038/nclimate1990\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBellard, C., Rysman, J.-F., Leroy, B., Claud, C. \u0026amp; Mace, G. M. 2017. A global picture of biological invasion threat on islands. \u003cem\u003eNature Ecology \u0026amp; Evolution\u003c/em\u003e, 1, 1862\u0026ndash;1869 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41559-017-0365-6\u003c/span\u003e\u003cspan address=\"10.1038/s41559-017-0365-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoughorbel, S., Jarray, F. \u0026amp; El-Anbari, M. 2017. Optimal classifier for imbalanced data using Matthews Correlation Coefficient metric. \u003cem\u003ePLOS ONE\u003c/em\u003e, 12, e0177678 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1371/journal.pone.0177678\u003c/span\u003e\u003cspan address=\"10.1371/journal.pone.0177678\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoyd, I. L., Freer-Smith, P. H., Gilligan, C. A. \u0026amp; Godfray, H. C. J. 2013. The Consequence of Tree Pests and Diseases for Ecosystem Services. \u003cem\u003eScience\u003c/em\u003e, 342, 1235773 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1126/science.1235773\u003c/span\u003e\u003cspan address=\"10.1126/science.1235773\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrown, K. A., Yesuf, Gabriel., Carvalho, Fabio 2022. Assessing the quality of remaining forests on Madagascar. \u003cem\u003eIn\u003c/em\u003e: GOODMAN, S. M. (ed.) \u003cem\u003eThe new natural history of Madagascar\u003c/em\u003e. Princeton, NJ, USA: Princeton University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurgess, T. I., Scott, J. K., Mcdougall, K. L., Stukely, M. J. C., Crane, C., Dunstan, W. A., Brigg, F., Andjic, V., White, D., Rudman, T., Arentz, F., Ota, N. \u0026amp; Hardy, G. E. S. J. 2017. Current and projected global distribution of Phytophthora cinnamomi, one of the world's worst plant pathogens. \u003cem\u003eGlobal Change Biology\u003c/em\u003e, 23, 1661\u0026ndash;1674 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.13492\u003c/span\u003e\u003cspan address=\"10.1111/gcb.13492\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChamberlain S, B. V., Mcglinn D, Oldoni D, Desmet P, Geffert L, Ram K. 2017. rgbif: Interface to the Global Biodiversity Information Facility API. R package version 3.8.0. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://CRAN.R-project.org/package=rgbif\u003c/span\u003e\u003cspan address=\"https://CRAN.R-project.org/package=rgbif\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [Accessed]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChamberlain, S. A. \u0026amp; Boettiger, C. 2017. R Python, and Ruby clients for GBIF species occurrence data. PeerJ Preprints.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoe, H., Thorne, J. H., Hijmans, R., Kim, J., Kwon, H. \u0026amp; Seo, C. 2017. Meta-corridor solutions for climate-vulnerable plant species groups in South Korea. \u003cem\u003eJournal of Applied Ecology\u003c/em\u003e, 54, 1742\u0026ndash;1754 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2664.12865\u003c/span\u003e\u003cspan address=\"10.1111/1365-2664.12865\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDawson, W., Moser, D., Van Kleunen, M., Kreft, H., Pergl, J., Pyšek, P., Weigelt, P., Winter, M., Lenzner, B., Blackburn, T. M., Dyer, E. E., Cassey, P., Scrivens, S. L., Economo, E. P., Gu\u0026eacute;nard, B., Capinha, C., Seebens, H., Garc\u0026iacute;a-D\u0026iacute;az, P., Nentwig, W., Garc\u0026iacute;a-Berthou, E., Casal, C., Mandrak, N. E., Fuller, P., Meyer, C. \u0026amp; Essl, F. 2017. Global hotspots and correlates of alien species richness across taxonomic groups. \u003cem\u003eNature Ecology \u0026amp; Evolution\u003c/em\u003e, 1, 0186 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/s41559-017-0186\u003c/span\u003e\u003cspan address=\"10.1038/s41559-017-0186\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Moraes, L. F. D., Assump\u0026ccedil;\u0026atilde;o, J. M., Luchiari, C. \u0026amp; Pereira, T. S. 2006. Native tree species planting for restoration at Po\u0026ccedil;o das Antas Biological Reserve, Rio de Janeiro, Brazil. \u003cem\u003eRodriguesia\u003c/em\u003e, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1590/2175-7860200657307\u003c/span\u003e\u003cspan address=\"10.1590/2175-7860200657307\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDesprez-Loustau, M.-L., Robin, C., Buee, M., Courtecuisse, R., Garbaye, J., Suffert, F., Sache, I. \u0026amp; Rizzo, D. M. 2007. The fungal dimension of biological invasions. \u003cem\u003eTrends in ecology \u0026amp; evolution\u003c/em\u003e, 22, 472\u0026ndash;480 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tree.2007.04.005\u003c/span\u003e\u003cspan address=\"10.1016/j.tree.2007.04.005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\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. \u003cem\u003eEcography (Copenhagen)\u003c/em\u003e, 36, 27\u0026ndash;46 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1600-0587.2012.07348.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1600-0587.2012.07348.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElad, Y. \u0026amp; Pertot, I. 2014. Climate Change Impacts on Plant Pathogens and Plant Diseases. \u003cem\u003eJournal of Crop Improvement\u003c/em\u003e, 28, 99\u0026ndash;139 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1080/15427528.2014.865412\u003c/span\u003e\u003cspan address=\"10.1080/15427528.2014.865412\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eElith, J., Kearney, M. \u0026amp; Phillips, S. 2010. The art of modelling range-shifting species. \u003cem\u003eMethods in Ecology and Evolution\u003c/em\u003e, 1, 330\u0026ndash;342 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.2041-210X.2010.00036.x\u003c/span\u003e\u003cspan address=\"10.1111/j.2041-210X.2010.00036.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEngelbrecht, B. M. J., Comita, L. S., Condit, R., Kursar, T. A., Tyree, M. T., Turner, B. L. \u0026amp; Hubbell, S. P. 2007. Drought sensitivity shapes species distribution patterns in tropical forests. \u003cem\u003eNature\u003c/em\u003e, 447, 80\u0026ndash;82 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nature05747\u003c/span\u003e\u003cspan address=\"10.1038/nature05747\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEscobar, L. E. \u0026amp; Craft, M. E. 2016. Advances and Limitations of Disease Biogeography Using Ecological Niche Modeling. \u003cem\u003eFrontiers in Microbiology\u003c/em\u003e, Volume 7\u0026ndash;2016\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEsquivel-Muelbert, A., Baker, T. R., Dexter, K. G., Lewis, S. L., Ter Steege, H., Lopez-Gonzalez, G., Monteagudo Mendoza, A., Brienen, R., Feldpausch, T. R., Pitman, N., Alonso, A., Van Der Heijden, G., Pe\u0026ntilde;a-Claros, M., Ahuite, M., Alexiaides, M., \u0026Aacute;lvarez D\u0026aacute;vila, E., Murakami, A. A., Arroyo, L., Aulestia, M., Balslev, H., Barroso, J., Boot, R., Cano, A., Chama Moscoso, V., Comiskey, J. A., Cornejo, F., Dallmeier, F., Daly, D. C., D\u0026aacute;vila, N., Duivenvoorden, J. F., Duque Montoya, A. J., Erwin, T., Di Fiore, A., Fredericksen, T., Fuentes, A., Garc\u0026iacute;a-Villacorta, R., Gonzales, T., Guevara Andino, J. E., Honorio Coronado, E. N., Huamantupa-Chuquimaco, I., Eliana Maria Jim\u0026eacute;nez, R., Killeen, T. J., Malhi, Y., Mendoza, C., Mogoll\u0026oacute;n, H., J\u0026oslash;rgensen, P. M., Montero, J. C., Mostacedo, B., Nauray, W., Neill, D., Vargas, P. N., Palacios, S., Palacios Cuenca, W., Pallqui Camacho, N. C., Peacock, J., Phillips, J. F., Pickavance, G., Quesada, C. A., Ram\u0026iacute;rez-Angulo, H., Restrepo, Z., Reynel Rodriguez, C., Paredes, M. R., Pe\u0026ntilde;uela-Mora, M. C., Sierra, R., Silveira, M., Stevenson, P., Stropp, J., Terborgh, J., Tirado, M., Toledo, M., Torres-Lezama, A., Uma\u0026ntilde;a, M. N., Urrego, L. E., Vasquez Martinez, R., Gamarra, L. V., Vela, C. I. A., Vilanova Torre, E., Vos, V., Von Hildebrand, P., Vriesendorp, C., Wang, O., Young, K. R., Zartman, C. E. \u0026amp; Phillips, O. L. 2017. Seasonal drought limits tree species across the Neotropics. \u003cem\u003eEcography\u003c/em\u003e, 40, 618\u0026ndash;629 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ecog.01904\u003c/span\u003e\u003cspan address=\"10.1111/ecog.01904\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFonte, S. J. \u0026amp; Schowalter, T. D. 2004. Decomposition of Greenfall vs. Senescent Foliage in a Tropical Forest Ecosystem in Puerto Rico. \u003cem\u003eBiotropica\u003c/em\u003e, 36, 474\u0026ndash;482 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1744-7429.2004.tb00343.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1744-7429.2004.tb00343.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarbelotto, M. \u0026amp; Pautasso, M. 2012. Impacts of exotic forest pathogens on Mediterranean ecosystems: four case studies. \u003cem\u003eEuropean Journal of Plant Pathology\u003c/em\u003e, 133, 101\u0026ndash;116 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1007/s10658-011-9928-6\u003c/span\u003e\u003cspan address=\"10.1007/s10658-011-9928-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGarrett, K. A., Dendy, S. P., Frank, E. E., Rouse, M. N. \u0026amp; Travers, S. E. 2006. Climate Change Effects on Plant Disease: Genomes to Ecosystems. \u003cem\u003eAnnual Review of Phytopathology\u003c/em\u003e, 44, 489\u0026ndash;509 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev.phyto.44.070505.143420\u003c/span\u003e\u003cspan address=\"10.1146/annurev.phyto.44.070505.143420\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGbif. 2025a. Occurrence Download. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15468/dl.284amw\u003c/span\u003e\u003cspan address=\"10.15468/dl.284amw\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGbif. 2025b. Occurrence Download. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15468/dl.dj8pmv\u003c/span\u003e\u003cspan address=\"10.15468/dl.dj8pmv\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGbif. 2025c. Occurrence Download. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.15468/dl.7kd9hb\u003c/span\u003e\u003cspan address=\"10.15468/dl.7kd9hb\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoberville, E., Hautek\u0026egrave;ete, N.-C., Kirby, R. R., Piquot, Y., Luczak, C. \u0026amp; Beaugrand, G. 2016. Climate change and the ash dieback crisis. \u003cem\u003eScientific Reports\u003c/em\u003e, 6, 35303 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e10.1038/srep35303\u003c/span\u003e\u003cspan address=\"10.1038/srep35303\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoodman, S. M. 2023. Updated estimates of biotic diversity and endemism for Madagascar\u0026mdash;revisited after 20 years. \u003cem\u003eOryx\u003c/em\u003e, 57, 588\u0026ndash;600 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0030605322001284\u003c/span\u003e\u003cspan address=\"10.1017/S0030605322001284\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGr\u0026uuml;nwald, N. J., Goss Em Fau - Press, C. M. \u0026amp; Press, C. M. 2008. Phytophthora ramorum: a pathogen with a remarkably wide host range causing sudden oak death on oaks and ramorum blight on woody ornamentals. \u003cem\u003eMolecular Plant Pathology\u003c/em\u003e, 9, 729\u0026ndash;740 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1364-3703.2008.00500.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1364-3703.2008.00500.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarper, G. J., Steininger, M. K., Tucker, C. J., Juhn, D. \u0026amp; Hawkins, F. 2007. Fifty years of deforestation and forest fragmentation in Madagascar. \u003cem\u003eEnvironmental Conservation\u003c/em\u003e, 34, 325\u0026ndash;333 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0376892907004262\u003c/span\u003e\u003cspan address=\"10.1017/S0376892907004262\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHijmans, R. 2025. terra: Spatial Data Analysis. R package version 1.8.21. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://CRAN.R-project.org/package=terra\u003c/span\u003e\u003cspan address=\"https://CRAN.R-project.org/package=terra\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e [Accessed]\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIpbes. 2023. \u003cem\u003eSummary for Policymakers of the Thematic Assessment Report on Invasive Alien Species and their Control of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services\u003c/em\u003e [Online]. Bonn, Germany: Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) secretariat. [Accessed 2025].\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIucn. 2019. \u003cem\u003eThe IUCN Red List of Threatened Species\u003c/em\u003e [Online]. Available: www.iucnredlist.org [Accessed].\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIucn 2022. Guidelines for Using IUCN Red List Categories and Criteria. IUCN.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIucn. 2023. The IUCN Red List of Threatened Species.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarger, D. N., Conrad, O., B\u0026ouml;hner, J., Kawohl, T., Kreft, H., Soria-Auza, R. W., Zimmermann, N. E., Linder, H. P. \u0026amp; Kessler, M. 2017. Climatologies at high resolution for the earth\u0026rsquo;s land surface areas. \u003cem\u003eScientific Data\u003c/em\u003e, 4, 170122 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/sdata.2017.122\u003c/span\u003e\u003cspan address=\"10.1038/sdata.2017.122\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarger, D. N., Conrad, O., B\u0026ouml;hner, J., Kawohl, T., Kreft, H., Soria-Auza, R. W., Zimmermann, N. E., Linder, H. P. \u0026amp; Kessler, M. 2021. Climatologies at high resolution for the earth\u0026rsquo;s land surface areas. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.16904/envidat.228\u003c/span\u003e\u003cspan address=\"10.16904/envidat.228\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarnosky, D. F. 1979. Dutch Elm Disease: A Review of the History, Environmental Implications, Control, and Research Needs. \u003cem\u003eEnvironmental Conservation\u003c/em\u003e, 6, 311\u0026ndash;322\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrasting, J. P., John, J. G., Blanton, C., Mchugh, C., Nikonov, S., Radhakrishnan, A., Rand, K., Zadeh, N. T., Balaji, V., Durachta, J., Dupuis, C., Menzel, R., Robinson, T., Underwood, S., Vahlenkamp, H., Dunne, K. A., Gauthier, P. P. G., Ginoux, P., Griffies, S. M., Hallberg, R., Harrison, M., Hurlin, W., Malyshev, S., Naik, V., Paulot, F., Paynter, D. J., Ploshay, J., Reichl, B. G., Schwarzkopf, D. M., Seman, C. J., Silvers, L., Wyman, B., Zeng, Y., Adcroft, A., Dunne, J. P., Dussin, R., Guo, H., He, J., Held, I. M., Horowitz, L. W., Lin, P., Milly, P. C. D., Shevliakova, E., Stock, C., Winton, M., Wittenberg, A. T., Xie, Y. \u0026amp; Zhao, M. 2018. NOAA-GFDL GFDL-ESM4 model output prepared for CMIP6 CMIP. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.22033/ESGF/CMIP6.1407\u003c/span\u003e\u003cspan address=\"10.22033/ESGF/CMIP6.1407\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, X.-D., Chen, Y., Zhang, C.-L., Wang, J., Song, X.-J., Zhang, X.-R., Zhu, Z.-H. \u0026amp; Liu, G. 2024. Assessing the climatic niche changes and global invasion risk of Solanum elaeagnifolium in relation to human activities. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, 954, 176723 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2024.176723\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2024.176723\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, X. \u0026amp; Wang, Y. 2013. Applying various algorithms for species distribution modelling. \u003cem\u003eIntegrative Zoology\u003c/em\u003e, 8, 124\u0026ndash;135 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1749-4877.12000\u003c/span\u003e\u003cspan address=\"10.1111/1749-4877.12000\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, C., White, M. \u0026amp; Newell, G. 2013. Selecting thresholds for the prediction of species occurrence with presence-only data. \u003cem\u003eJournal of Biogeography\u003c/em\u003e, 40, 778\u0026ndash;789 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/jbi.12058\u003c/span\u003e\u003cspan address=\"10.1111/jbi.12058\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLovett, G. M., Weiss, M., Liebhold, A. M., Holmes, T. P., Leung, B., Lambert, K. F., Orwig, D. A., Campbell, F. T., Rosenthal, J., Mccullough, D. G., Wildova, R., Ayres, M. P., Canham, C. D., Foster, D. R., Ladeau, S. L. \u0026amp; Weldy, T. 2016. Nonnative forest insects and pathogens in the United States: Impacts and policy options. \u003cem\u003eEcological Applications\u003c/em\u003e, 26, 1437\u0026ndash;1455 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/15-1176\u003c/span\u003e\u003cspan address=\"10.1890/15-1176\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcdowell, N. G., Allen, C. D., Anderson-Teixeira, K., Aukema, B. H., Bond-Lamberty, B., Chini, L., Clark, J. S., Dietze, M., Grossiord, C., Hanbury-Brown, A., Hurtt, G. C., Jackson, R. B., Johnson, D. J., Kueppers, L., Lichstein, J. W., Ogle, K., Poulter, B., Pugh, T. a. M., Seidl, R., Turner, M. G., Uriarte, M., Walker, A. P. \u0026amp; Xu, C. 2020. Pervasive shifts in forest dynamics in a changing world. \u003cem\u003eScience\u003c/em\u003e, 368, eaaz9463 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.aaz9463\u003c/span\u003e\u003cspan address=\"10.1126/science.aaz9463\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeentemeyer, R. K., Cunniffe, N. J., Cook, A. R., Filipe, J. a. N., Hunter, R. D., Rizzo, D. M. \u0026amp; Gilligan, C. A. 2011. Epidemiological modeling of invasion in heterogeneous landscapes: spread of sudden oak death in California (1990\u0026ndash;2030). \u003cem\u003eEcosphere\u003c/em\u003e, 2, art17 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/ES10-00192.1\u003c/span\u003e\u003cspan address=\"10.1890/ES10-00192.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMetz, M. R., Frangioso, K. M., Meentemeyer, R. K. \u0026amp; Rizzo, D. M. 2011. Interacting disturbances: wildfire severity affected by stage of forest disease invasion. \u003cem\u003eEcological Applications\u003c/em\u003e, 21, 313\u0026ndash;320 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/10-0419.1\u003c/span\u003e\u003cspan address=\"10.1890/10-0419.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorellato, L. P. C., Alberton, B., Alvarado, S. T., Borges, B., Buisson, E., Camargo, M. G. G., Cancian, L. F., Carstensen, D. W., Escobar, D. F. E., Leite, P. T. P., Mendoza, I., Rocha, N. M. W. B., Soares, N. C., Silva, T. S. F., Staggemeier, V. G., Streher, A. S., Vargas, B. C. \u0026amp; Peres, C. A. 2016. Linking plant phenology to conservation biology. \u003cem\u003eBiological Conservation\u003c/em\u003e, 195, 60\u0026ndash;72 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.biocon.2015.12.033\u003c/span\u003e\u003cspan address=\"10.1016/j.biocon.2015.12.033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNaimi, B. \u0026amp; Ara\u0026uacute;jo, M. B. 2016. sdm: a reproducible and extensible R platform for species distribution modelling. \u003cem\u003eEcography\u003c/em\u003e, 39, 368\u0026ndash;375 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ecog.01881\u003c/span\u003e\u003cspan address=\"10.1111/ecog.01881\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOwens, H. L., Campbell, L. P., Dornak, L. L., Saupe, E. E., Barve, N., Sober\u0026oacute;n, J., Ingenloff, K., Lira-Noriega, A., Hensz, C. M., Myers, C. E. \u0026amp; Peterson, A. T. 2013. Constraints on interpretation of ecological niche models by limited environmental ranges on calibration areas. \u003cem\u003eEcological Modelling\u003c/em\u003e, 263, 10\u0026ndash;18 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ecolmodel.2013.04.011\u003c/span\u003e\u003cspan address=\"10.1016/j.ecolmodel.2013.04.011\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark Williams, A., Allen, C. D., Macalady, A. K., Griffin, D., Woodhouse, C. A., Meko, D. M., Swetnam, T. W., Rauscher, S. A., Seager, R., Grissino-Mayer, H. D., Dean, J. S., Cook, E. R., Gangodagamage, C., Cai, M. \u0026amp; Mcdowell, N. G. 2013. Temperature as a potent driver of regional forest drought stress and tree mortality. \u003cem\u003eNature Climate Change\u003c/em\u003e, 3, 292\u0026ndash;297 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/nclimate1693\u003c/span\u003e\u003cspan address=\"10.1038/nclimate1693\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePeterson, A. T., Sober\u0026oacute;n, J., Pearson, R. G., Anderson, R. P., Mart\u0026iacute;nez-Meyer, E., Nakamura, M. \u0026amp; Ara\u0026uacute;jo, M. B. 2012. \u003cem\u003eEcological Niches and Geographic Distributions\u003c/em\u003e, Princeton University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamananjato, V. \u0026amp; Razafindratsima, O. H. 2021. Structure of Microhabitats Used by Microcebus rufus Across a Heterogeneous Landscape. \u003cem\u003eInternational Journal of Primatology\u003c/em\u003e, 42, 682\u0026ndash;700 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10764-021-00224-4\u003c/span\u003e\u003cspan address=\"10.1007/s10764-021-00224-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRamananjato, V., Razafindratsima, Onja. 2021. Data from: Structure of microhabitats used by Microcebus rufus across a heterogeneous landscape [Dataset]. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.5061/dryad.2280gb5rs\u003c/span\u003e\u003cspan address=\"10.5061/dryad.2280gb5rs\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRazafindratsima, O. H., Jones, T. A. \u0026amp; Dunham, A. E. 2014. Patterns of movement and seed dispersal by three lemur species. \u003cem\u003eAmerican Journal of Primatology\u003c/em\u003e, 76, 84\u0026ndash;96 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ajp.22199\u003c/span\u003e\u003cspan address=\"10.1002/ajp.22199\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eReyer, C. P. O., Brouwers, N., Rammig, A., Brook, B. W., Epila, J., Grant, R. F., Holmgren, M., Langerwisch, F., Leuzinger, S., Lucht, W., Medlyn, B., Pfeifer, M., Steinkamp, J., Vanderwel, M. C., Verbeeck, H. \u0026amp; Villela, D. M. 2015. Forest resilience and tipping points at different spatio-temporal scales: approaches and challenges. \u003cem\u003eJournal of Ecology\u003c/em\u003e, 103, 5\u0026ndash;15 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2745.12337\u003c/span\u003e\u003cspan address=\"10.1111/1365-2745.12337\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRizzo, D. M. \u0026amp; Garbelotto, M. 2003. Sudden oak death: endangering California and Oregon forest ecosystems. \u003cem\u003eFrontiers in Ecology and the Environment\u003c/em\u003e, 1, 197\u0026ndash;204 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1890/1540-9295(2003\u003c/span\u003e\u003cspan address=\"10.1890/1540-9295(2003\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e)001[0197:SODECA]2.0.CO;2.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRovero, F. \u0026amp; Ahumada, J. 2017. The Tropical Ecology, Assessment and Monitoring (TEAM) Network: An early warning system for tropical rain forests. \u003cem\u003eScience of The Total Environment\u003c/em\u003e, 574, 914\u0026ndash;923 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.scitotenv.2016.09.146\u003c/span\u003e\u003cspan address=\"10.1016/j.scitotenv.2016.09.146\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantini, A., Ghelardini, L., De Pace, C., Desprez-Loustau, M. L., Capretti, P., Chandelier, A., Cech, T., Chira, D., Diamandis, S., Gaitniekis, T., Hantula, J., Holdenrieder, O., Jankovsky, L., Jung, T., Jurc, D., Kirisits, T., Kunca, A., Lygis, V., Malecka, M., Marcais, B., Schmitz, S., Schumacher, J., Solheim, H., Solla, A., Szab\u0026ograve;, I., Tsopelas, P., Vannini, A., Vettraino, A. M., Webber, J., Woodward, S. \u0026amp; Stenlid, J. 2013. Biogeographical patterns and determinants of invasion by forest pathogens in Europe. \u003cem\u003eNew Phytologist\u003c/em\u003e, 197, 238\u0026ndash;250 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1469-8137.2012.04364.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1469-8137.2012.04364.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlik, J. W. F., Paoli, G., Mcguire, K., Amaral, I., Barroso, J., Bastian, M., Blanc, L., Bongers, F., Boundja, P., Clark, C., Collins, M., Dauby, G., Ding, Y., Doucet, J.-L., Eler, E., Ferreira, L., Forshed, O., Fredriksson, G., Gillet, J.-F., Harris, D., Leal, M., Laumonier, Y., Malhi, Y., Mansor, A., Martin, E., Miyamoto, K., Araujo-Murakami, A., Nagamasu, H., Nilus, R., Nurtjahya, E., Oliveira, \u0026Aacute;., Onrizal, O., Parada-Gutierrez, A., Permana, A., Poorter, L., Poulsen, J., Ramirez-Angulo, H., Reitsma, J., Rovero, F., Rozak, A., Sheil, D., Silva-Espejo, J., Silveira, M., Spironelo, W., Ter Steege, H., Stevart, T., Navarro-Aguilar, G. E., Sunderland, T., Suzuki, E., Tang, J., Theilade, I., Van Der Heijden, G., Van Valkenburg, J., Van Do, T., Vilanova, E., Vos, V., Wich, S., W\u0026ouml;ll, H., Yoneda, T., Zang, R., Zhang, M.-G. \u0026amp; Zweifel, N. 2013. Large trees drive forest aboveground biomass variation in moist lowland forests across the tropics. \u003cem\u003eGlobal Ecology and Biogeography\u003c/em\u003e, 22, 1261\u0026ndash;1271 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/geb.12092\u003c/span\u003e\u003cspan address=\"10.1111/geb.12092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSober\u0026oacute;n, J., And A. Townsend Peterson 2005. Interpretation of Models of Fundamental Ecological Niches and Species\u0026rsquo; Distributional Areas. \u003cem\u003eBiodiversity Informatics\u003c/em\u003e, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://dx.doi.org/10.17161/bi.v2i0.4\u003c/span\u003e\u003cspan address=\"10.17161/bi.v2i0.4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSopniewski, J., Scheele, B. C. \u0026amp; Cardillo, M. 2022. Predicting the distribution of Australian frogs and their overlap with Batrachochytrium dendrobatidis under climate change. \u003cem\u003eDiversity and Distributions\u003c/em\u003e, 28, 1255\u0026ndash;1268 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ddi.13533\u003c/span\u003e\u003cspan address=\"10.1111/ddi.13533\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStevens, P. F. 1980. A Revision Of The Old World Species Of Calophyllum (Guttiferae). \u003cem\u003eJournal of the Arnold Arboretum\u003c/em\u003e, 61, 117\u0026ndash;424\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSturrock, R. N., Frankel, S. J., Brown, A. V., Hennon, P. E., Kliejunas, J. T., Lewis, K. J., Worrall, J. J. \u0026amp; Woods, A. J. 2011. Climate change and forest diseases. \u003cem\u003ePlant Pathology\u003c/em\u003e, 60, 133\u0026ndash;149 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1365-3059.2010.02406.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1365-3059.2010.02406.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThompson, S. E., Levin, S. \u0026amp; Rodriguez-Iturbe, I. 2014. Rainfall and temperatures changes have confounding impacts on hytophthora cinnamomi occurrence risk in the southwestern USA under climate change scenarios. \u003cem\u003eGlobal Change Biology\u003c/em\u003e, 20, 1299\u0026ndash;1312 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/gcb.12463\u003c/span\u003e\u003cspan address=\"10.1111/gcb.12463\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThuiller, W., Gu\u0026eacute;guen, M., Renaud, J., Karger, D. N. \u0026amp; Zimmermann, N. E. 2019. Uncertainty in ensembles of global biodiversity scenarios. \u003cem\u003eNature Communications\u003c/em\u003e, 10, 1446 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41467-019-09519-w\u003c/span\u003e\u003cspan address=\"10.1038/s41467-019-09519-w\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVieilledent, G. 2021. forestatrisk: a Python package for modelling and forecasting deforestation in the tropics. \u003cem\u003eJournal of Open Source Software\u003c/em\u003e, 6, 2975 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.21105/joss.02975\u003c/span\u003e\u003cspan address=\"10.21105/joss.02975\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVieilledent, G., Grinand, C., Rakotomalala, F. A., Ranaivosoa, R., Rakotoarijaona, J.-R., Allnutt, T. F. \u0026amp; Achard, F. 2018. Combining global tree cover loss data with historical national forest cover maps to look at six decades of deforestation and forest fragmentation in Madagascar. \u003cem\u003eBiological Conservation\u003c/em\u003e, 222, 189\u0026ndash;197 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.biocon.2018.04.008\u003c/span\u003e\u003cspan address=\"10.1016/j.biocon.2018.04.008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVieilledent, G., Grinand, C. \u0026amp; Vaudry, R. 2013. Forecasting deforestation and carbon emissions in tropical developing countries facing demographic expansion: a case study in Madagascar. \u003cem\u003eEcology and Evolution\u003c/em\u003e, 3, 1702\u0026ndash;1716 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/ece3.550\u003c/span\u003e\u003cspan address=\"10.1002/ece3.550\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVisser, M. D., Muller-Landau, H. C., Schnitzer, S. A., De Kroon, H., Jongejans, E. \u0026amp; Wright, S. J. 2018. A host\u0026ndash;parasite model explains variation in liana infestation among co-occurring tree species. \u003cem\u003eJournal of Ecology\u003c/em\u003e, 106, 2435\u0026ndash;2445 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1365-2745.12997\u003c/span\u003e\u003cspan address=\"10.1111/1365-2745.12997\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWainhouse, D., Murphy, S., Greig, B., Webber, J. \u0026amp; Vielle, M. 1998. The role of the bark beetle Cryphalus trypanus in the transmission of the vascular wilt pathogen of takamaka (Calophyllum inophyllum) in the Seychelles. \u003cem\u003eForest Ecology and Management\u003c/em\u003e, 108, 193\u0026ndash;199 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0378-1127(98)00234-5\u003c/span\u003e\u003cspan address=\"10.1016/S0378-1127(98)00234-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWebber, J. F., Jacobs, K. \u0026amp; Wingfield, M. J. 1999. A re-examination of the vascular wilt pathogen of takamaka (Calophyllum inophyllum). \u003cem\u003eMycological Research\u003c/em\u003e, 103, 1588\u0026ndash;1592 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0953756299001021\u003c/span\u003e\u003cspan address=\"10.1017/S0953756299001021\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWiehe, P. O. 1949. Wilt of Calophyllum inophyllom L. var. tacamaha (Willd.) REV caused by Haplographium calophylli sp. nov. in Mauritius.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWilliams, J. N., Seo, C., Thorne, J., Nelson, J. K., Erwin, S., O\u0026rsquo;brien, J. M. \u0026amp; Schwartz, M. W. 2009. Using species distribution models to predict new occurrences for rare plants. \u003cem\u003eDiversity and Distributions\u003c/em\u003e, 15, 565\u0026ndash;576 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1472-4642.2009.00567.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1472-4642.2009.00567.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWright, P. C., Otero Jimenez, B., Rakotonirina, P., Andriananoely, D. H., Shea, A., Ratalata, B. \u0026amp; Razafimahaimodison, J. C. 2020. The Progressive Spread of the Vascular Wilt Like Pathogen of Calophyllum Detected in Ranomafana National Park, Madagascar. \u003cem\u003eFrontiers in Forests and Global Change\u003c/em\u003e, \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3389/ffgc.2020.00091\u003c/span\u003e\u003cspan address=\"10.3389/ffgc.2020.00091\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZurell, D., Elith, J. \u0026amp; Schr\u0026ouml;der, B. 2012. Predicting to new environments: tools for visualizing model behaviour and impacts on mapped distributions. \u003cem\u003eDiversity and Distributions\u003c/em\u003e, 18, 628\u0026ndash;634 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1472-4642.2012.00887.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1472-4642.2012.00887.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZurell, D., Franklin, J., K\u0026ouml;nig, C., Bouchet, P. J., Dormann, C. F., Elith, J., Fandos, G., Feng, X., Guillera-Arroita, G., Guisan, A., Lahoz-Monfort, J. J., Leit\u0026atilde;o, P. J., Park, D. S., Peterson, A. T., Rapacciuolo, G., Schmatz, D. R., Schr\u0026ouml;der, B., Serra-Diaz, J. M., Thuiller, W., Yates, K. L., Zimmermann, N. E. \u0026amp; Merow, C. 2020. A standard protocol for reporting species distribution models. \u003cem\u003eEcography\u003c/em\u003e, 43, 1261\u0026ndash;1277 \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ecog.04960\u003c/span\u003e\u003cspan address=\"10.1111/ecog.04960\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003e1\u003c/strong\u003e\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;Standardised range shift metrics - habitat exposure index and potential area of occupancy (pAOO) change for each species and emissions scenario compared between current projections and year 2100.\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" align=\"left\" width=\"100%\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eModel\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eClimate pathway*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eHabitat exposure index\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eIUCN pAOO % change\u003csup\u003e**\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eWilt pathogen\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eL. calophylli\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.083\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-7.936\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.298\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-28.986\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.315\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-30.656\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"3\"\u003e\n \u003cp\u003eHost\u003c/p\u003e\n \u003cp\u003e\u003cem\u003eC. paniculatum\u003c/em\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.759\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-45.173\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.786\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-64.047\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSSP5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.830\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e-67.485\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e*SSP1 = low-emissions, sustainable socio-economic scenario; SSP3 = high emissions, regional rivalry scenario; SSP5 = high emissions, fossil-fuelled scenario.\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e**Range shift metrics (habitat exposure and pAOO) were computed after converting continuous probability of presence to binary presence-absence predictions. We used species specific binary optimisation thresholds: \u003cem\u003eL. calophylli\u003c/em\u003e models were optimised for the minimum distance to ROC curve. Host plant model optimised for maximizing the sum of sensitivity and specificity (TSS).\u003c/sup\u003e\u003c/p\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":"climate change, fungal pathogen, tropics, endemism, biodiversity hotspot, sdm","lastPublishedDoi":"10.21203/rs.3.rs-8741587/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8741587/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eContext.\u003c/p\u003e \u003cp\u003eNon-native plant pathogens are reshaping ecosystems globally, yet their spread and potential spatial distributions under future climate and land-use change remains underexplored, particularly in biodiversity hotspots with vulnerable hosts. In Madagascar, a vascular wilt pathogen (\u003cem\u003eLeptographium calophylli\u003c/em\u003e) has been increasingly observed infecting native forest trees raising conservation concerns.\u003c/p\u003e \u003cp\u003eObjectives.\u003c/p\u003e \u003cp\u003eWe aimed to model future distributions of both a vulnerable host tree \u003cem\u003eCalophyllum paniculatum\u003c/em\u003e and incipient wilt pathogen \u003cem\u003eLeptographium calophylli\u003c/em\u003e under climate and land cover change scenarios over the next 80 years to assess overlap, divergence, and implications for extinction risk.\u003c/p\u003e \u003cp\u003eMethods.\u003c/p\u003e \u003cp\u003eUsing ensemble SDMs across Madagascar, we forecasted potential distributional ranges for both pathogen and host using the Global Climate Model (GFDL-ESM4) under three scenarios (SSP1\u0026ndash;2.6, SSP3\u0026ndash;7.0, SSP5\u0026ndash;8.5) and time periods (2011\u0026ndash;2040, 2041\u0026ndash;2070, 2071\u0026ndash;2100). We measured future range shifts using a habitat exposure index and potential area of occupancy.\u003c/p\u003e \u003cp\u003eResults.\u003c/p\u003e \u003cp\u003eThe pathogen is predicted to retain 68.5% of the current projected distribution by 2100, with expansion into previously uninhabited regions. \u003cem\u003eC. paniculatum\u003c/em\u003e is forecast to experience range contraction (65.9% loss by 2100) with persistent distributional overlap predicted across all scenarios.\u003c/p\u003e \u003cp\u003eConclusion.\u003c/p\u003e \u003cp\u003eWe demonstrate that future climatic conditions may facilitate fungal pathogen expansion, while simultaneously exposing vulnerable hosts within their native, endemic range to infection. The asymmetric dynamic of host range losses could intensify biodiversity loss in insular ecosystems, particularly when compounded by deforestation. This study highlights the importance of considering disease threats in biodiversity forecasts and conservation strategies, particularly in tropical systems facing rapid environmental change.\u003c/p\u003e","manuscriptTitle":"Climate change facilitates fungal pathogen expansion while driving endemic host range contractions in a tropical biodiversity hotspot","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-05 11:55:10","doi":"10.21203/rs.3.rs-8741587/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-02T16:19:05+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-02T15:18:46+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-24T22:03:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"51087380118954421510803875332096816339","date":"2026-02-10T15:20:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"123322907193593572671963923434451907289","date":"2026-02-09T17:29:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"167475217362914974276164289948924300921","date":"2026-02-09T15:57:33+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"150693229632096324598057471925561118943","date":"2026-02-09T07:33:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"240309690043651959589117598843658413572","date":"2026-02-06T14:25:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-06T11:17:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"44338150659938897468411945052144758254","date":"2026-02-03T21:49:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210778322426931755153820906856104696619","date":"2026-02-03T17:30:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-03T16:37:50+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-01-31T06:25:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-01-31T06:22:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"Landscape Ecology","date":"2026-01-30T13:06:39+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":"b6f000ef-93d4-432f-a152-9bb772942b97","owner":[],"postedDate":"February 5th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-05-05T09:59:23+00:00","versionOfRecord":{"articleIdentity":"rs-8741587","link":"https://doi.org/10.1007/s10980-026-02360-9","journal":{"identity":"landscape-ecology","isVorOnly":false,"title":"Landscape Ecology"},"publishedOn":"2026-05-02 15:57:40","publishedOnDateReadable":"May 2nd, 2026"},"versionCreatedAt":"2026-02-05 11:55:10","video":"","vorDoi":"10.1007/s10980-026-02360-9","vorDoiUrl":"https://doi.org/10.1007/s10980-026-02360-9","workflowStages":[]},"version":"v1","identity":"rs-8741587","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8741587","identity":"rs-8741587","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}