Acquisitive traits improve seedling survival with drought at the edges of a fragmented tropical forest

Acquisitive traits improve seedling survival with drought at the edges of a fragmented tropical forest | 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 Acquisitive traits improve seedling survival with drought at the edges of a fragmented tropical forest Peddiraju Bandaru, Lakshmipriya Cannanbilla, Ashish Nambiar, Rishanth Kuruvankunath Ravi, and 6 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6845805/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 05 Dec, 2025 Read the published version in Plant Ecology → Version 1 posted 11 You are reading this latest preprint version Abstract Climate change and habitat fragmentation threaten biodiversity, but their interactive effects remain poorly understood. In closed-canopy forests, altered rainfall patterns may induce drought conditions that are exacerbated at forest edges due to warmer, drier microclimates. Plant responses to water limitation can be mediated by functional traits related to resource acquisition and stress tolerance. We examined how drought and edge conditions jointly affect seedling survival, and whether species’ responses are explained by their traits. In a human-modified forest in the central Western Ghats, India, we transplanted ~ 1-year-old seedlings in a factorial combination of habitat (forest edge vs. interior) and drought (throughfall exclusion vs. control). We monitored survival through one year and estimated drought response (survival in drought relative to control), which was related to six traits. Throughfall exclusion reduced soil moisture more at edges, particularly during dry months. At the edge, three species showed significantly lower survival under drought, whereas survival in the interior did not differ with water treatment. Acquisitive traits (high leaf area, low stem specific density, low leaf dry matter content, and low leaf mass per area) improved survival with drought at edges. Trait-mediated responses were not evident in the interior, likely due to buffered microclimates. Multi-trait combinations were better predictors of drought response than individual traits, suggesting trait coordination. Our results suggest that droughts may favour acquisitive species at forest edges, potentially altering community composition, which has implications for management and restoration of fragmented forests in a changing climate. tropical forests forest fragmentation seedling survival trait-environment interactions resource acquisition Figures Figure 1 Figure 2 Figure 3 1 Introduction Climate change and habitat fragmentation are among the most pressing threats to global biodiversity (Haddad et al. 2015 ; Weiskopf et al. 2020 ). Rising temperature and altered precipitation patterns are projected to result in droughts, which can increase plant mortality (Duffy et al. 2015 ; Allen et al. 2017 ). Even in moist biomes, plant species performance relates to water availability (Engelbrecht et al. 2007 ; Comita and Engelbrecht 2009 ; Esquivel-Muelbert et al. 2019 ; Krishnadas et al. 2021 ), and drought will affect species to differing extents. The impact of drought on plant performance may be modulated by other global change factors, an example being habitat fragmentation. Forest fragmentation alters microclimatic conditions closer to forest edges, where higher temperatures, increased vapour pressure deficit, and drier soils can exacerbate the effects of drought (De Frenne et al. 2015 , 2019 ; Zellweger et al. 2020 ), but there are few empirical tests of this possibility. While drought can harm plants at all life stages, seedlings are highly vulnerable to water deficits due to their shallow root systems, limited stored resources, and light-limited understory conditions for growth and survival (Comita and Engelbrecht, 2014 ; Engelbrecht et al., 2007 ). Seedling establishment and community composition are affected by edge effects in forest fragments (Krishnadas and Comita 2018 ; Krishnadas et al. 2019 , 2020 ; Krishnadas 2023 ), to which drought may contribute. Since the seedling bank is critical for forest regeneration, understanding how drought interacts with fragmentation to shape species performance is essential to predict future forest composition (Engelbrecht and Kursar 2003 ; Poorter and Markesteijn 2008 ; Markesteijn and Poorter 2009 ; O’Brien et al. 2015 ). Plants cope with environmental stress through functional traits that relate to resource acquisition and stress tolerance (Krishnadas et al. 2025 ). Synthesis of plant functional traits across the climatic zones has shown that tree species with higher wood density survived better during drought in the tropics (Phillips et al., 2009 ; Van Nieuwstadt and Sheil, 2005 ; O’Brien et al., 2017a ). Similar trait-based responses to drought were seen in forests in the Mediterranean (Martínez-Vilalta et al. 2010 ) and temperate regions (Martinez-Meier et al. 2008 ; Nardini et al. 2013 ). A regional assessment of species distributions in peninsular India found that species with higher LMA and lower SSD increased with greater seasonal water deficit, suggesting a fitness advantage for these traits in drier conditions (Krishnadas et al. 2021 ). Less is known about how traits influence drought responses at earlier life stages. Greenhouse studies suggest that indicate that xylem structure and stomatal traits influence drought response of seedlings, larger xylem conduits reduced growth and photosynthesis, and smaller stomata decreased survival rates in drought relative to well-watered controls (Jhaveri et al. 2024 ). Other experiments seedlings respond to water deficit by shifting biomass allocation towards roots at the expense of leaves (Sunny et al. 2025 ). While greenhouse experiments provide valuable insights into seedling responses to drought, they do not capture the complexity of drought effects in natural forest conditions, such as competition, water lift, or variable soil resource availability (Comita and Engelbrecht 2014 ). To understand how drought and edge effects together shape seedling performance, we used throughfall exclusion to simulate drought at the edges and interiors of forest fragments in a human-modified landscape in the Western Ghats biodiversity hotspot in southern India. The region has experienced significant deforestation, with a 0.20% annual loss in forest cover, decreasing patch size, and increasing edge density from 1975 to 2005 (Reddy et al. 2013 ), which provides an ideal setting to examine the interaction between drought and forest fragmentation. Specifically, we asked: Does rainfall exclusion decrease soil moisture availability and properties by a greater extent at forest edges than interiors? Does the effect of drought on the survival of forest seedlings vary between the forest edge vs. interior? Do plant traits mediate the drought response of seedlings, and does this vary between the forest edge and interior? We hypothesised that drought-induced reduction in soil moisture would be more pronounced at forest edges, possibly due to greater solar radiation and wind exposure. We expected seedling survival to decrease with drought on average, and more so at edges due to harsher microclimatic conditions. Finally, we predicted that traits associated with resource-conservative strategies, such as thicker leaves and denser stems and roots, would show smaller declines in performance under drought, and that the influence of traits would be more prominent at forest edges. 2 Methods 2.1 Study site: This experiment was conducted in a 30 km 2 human-modified forest landscape in the central Western Ghats (12°56’N, 75°39’E) located in the Hassan district of Karnataka state (Krishnadas et al. 2018 ). The landscape comprises tropical humid forests and receives a mean annual rainfall of ca. 5000 mm, most of which falls during the monsoon season from July through October, with a pronounced dry season from December through May. Most seedling recruitment occurs during or just after monsoon rains, but seedlings have to survive the dry season to persist. Forest fragments make up ca. 60% of the study landscape, with the remainder being tea plantations, human settlements, and montane grasslands. The soils are primarily clayey Alfisols with good drainage, originating from a gneissic base. 2.2 Experimental design We conducted this study on saplings of 16 native tree species (Table S1 ). Species were chosen according to availability and germination of sufficient seeds and a sufficient degree of shade-tolerance for seedling survival in the understory. Seeds were sown in plastic grow-bags and emergent seedlings cultivated for 9–12 months after germination, then transplanted into the field between January and February in 2021. Each sapling was tagged with a unique ID and planted within a 2m X 2m plot in the pairwise combination of control and drought treatments at the forest edge and interior, replicated across 13 locations (hereafter blocks). Drought treatment was simulated using the throughfall exclusion technique by covering the assigned plot with polycarbonate sheeting to prevent rain from falling onto seedlings and the underlying soil layers. This method has been successfully used in multiple drought studies across the world and shown to reduce soil moisture without compromising light availability (Engelbrecht and Kursar 2003 ; O’Brien et al. 2017b ). The throughfall exclosure was maintained from March 2021 to February 2022. In each treatment block at both forest habitats, soil moisture was measured every month using a hand-held volumetric soil moisture sensor (HS2-20-HS2 CSA, Hydrosense II). Every month, the status of tagged saplings was recorded as alive/dead, new leaves/fallen leaves, and herbivory presence/absence was noted. At the end of the experiment, soil samples from the top 20 cm were collected from the corners of each plot to analyse soil physical (pH, and Electrical conductivity) and chemical (Organic Carbon, Available Cu, Mn, Fe, Zn, K, and P) properties. 2.4 Trait data We followed protocols recommended by (Pérez-Harguindeguy et al. 2013 ) to quantify functional traits. Three to five alive individuals from both the forest edge and interior of each species were harvested, brought back to the field station on the same day, and saturated overnight by immersing the petiole, root, and branch in a container filled with water. Water-saturated leaves were weighed to determine fresh weight, scanned with a desktop scanner for quantifying leaf area (LA), and then oven–dried at 70°C for 72 hours to determine dry weight. Leaf mass per unit area (LMA) was quantified as the ratio of dry weight to area, and leaf dry matter content (LDMC) as the ratio of dry weight to saturated fresh weight. A portion of the stem, main root and fine root was taken and used the water displacement method to estimate the volume, followed by oven-drying at 70°C for 72 hours to determine dry weight. Stem-specific density (SSD) was estimated as the ratio of dry weight to volume. Main root specific density (MRSD) and fine root specific density (FRSD) were estimated as the dry weight of the main root and fine root to their volumes. 2.6 Statistical analysis For question 1, linear mixed effects models with beta error distribution were used to model soil moisture percentage in relation to an interaction between habitat and treatment. Plot ID and month were included as random intercepts. Random slopes were modelled for each month to capture temporal variability and dependence of responses. In Wilkinson-Roger’s notation, the model is expressed as: Soil moisture ~ Forest habitat * Treatment + (1 | Plot id) + (1 + Forest habitat: Treatment | Month) Similarly, we checked whether rainfall exclusion altered the physical and chemical properties of the soil by the end of the experiment. We used a linear mixed effects model using a Gaussian error distribution to relate soil properties to an interaction between forest habitat and treatment, with plot ID included as a random intercept. Soil property ~ Forest habitat * Treatment + (1 | Plot id) For question 2, we assessed the survival of individual seedlings using a generalised linear mixed-effects model with a binomial error distribution. Survival ~ Forest habitat * Treatment + (1 | Forest habitat * Treatment | Species) Fixed effects tested whether survival response to drought varied between forest edge and interior habitat, while random slope for the interaction of effect and treatment and a random intercept for species was included to account for species-specific differences in response to habitat and treatment. Inference for random slopes of species between control and drought were made using the overlap of CIs in one treatment factor with the estimate of another treatment factor (Cumming et al. 2007 ). For question 3, we quantified the drought response of each species as the proportion of individuals surviving in the drought treatment relative to the control at the end of the experiment across all locations (Engelbrecht and Kursar 2003 ). $$\:Drought\:response\:=Number\:surviving\:in\:drought\:/Number\:surviving\:in\:control$$ To test if traits mediate drought response, first, we did a principal component analysis (PCA) on traits to obtain composite phenotypes defined by trait combinations based on leaf mass per area, leaf dry matter content, leaf area, stem specific density, main and fine root specific density. To test whether the composite phenotype defined by PCA axes and individual plant traits correlates with drought response, and if this relation differed between forest habitats, we used linear mixed-effects models with gamma error distribution. Gamma errors were suitable as the responses varied from 0–2.5. Drought response ~ Forest habitat * Trait + (1 + Forest habitat | Species) Species identity was included as a random intercept, accommodating additional species-specific variation in drought response that is not related to the traits we measured. We also studied this question with the binomial survival data as a function of traits, treatment and forest habitat with three-way interaction. We also tested if the individual survival was mediated by trait interaction with forest habitat and treatment using generalised linear mixed-effects models (GLMMs) with a binomial error distribution and a logit link function, as survival was measured as a binary outcome (alive or dead). The fixed effects included plant trait values, forest habitat (edge vs. interior), and drought treatment (control vs. drought). Species identity was included as a random effect, with random slopes for forest habitat and drought treatment to account for species-specific responses. The model was specified as: Survival ~ Trait × Forest habitat × Drought Treatment + (1 + Forest habitat + Treatment | Species) All data management and analysis were conducted using the R programming language version 4.3.1 (R Core Team 2023 ). Mixed effects models were implemented using glmmTMB (Brooks et al. 2017 ), and visualised using ggplot2 (Wickham et al. 2007 ), FactoMineR (Lê et al. 2008 ), and sjPlot (Lüdecke 2013 ). 3 Results 3.1 Soil moisture, physical and chemical properties The lowest soil moisture (VWC) was observed in the month of January with values of 5.02% ± 0.52, 7.21% ± 0.6, 7.56% ± 0.56, and 8.47% ± 0.78 in the forest edge drought, forest edge control, forest interior drought and forest interior control respectively (Table S2, Fig. 1 ). The highest soil moisture was observed in the month of August for the control treatment at both forest habitats, whereas drought treatments at the forest edge and interior experienced highest soil moisture in the month of September and July respectively. The highest soil moisture values (VWC) observed were 16.02% ± 1.48, 32.95% ± 0.59, 20.17% ± 1.45, and 34.82% ± 0.93 in the edge-drought, edge-control, interior-drought and interior-control, respectively. As expected, throughfall exclusion significantly reduced soil moisture over a year, with a stronger effect at the forest edge than in the interior. At the edge, soil moisture content halved under drought treatment, decreasing from 20% (95% CI: 15–26%) in control plots to 10% (8–13%) in drought plots. In contrast, in the forest interior, soil moisture declined by approximately 32%, from 22% (17–29%) in control to 15% (12–19%) under drought (Fig. S1 ). These results indicate that the impact of rainfall exclusion on soil moisture availability is more pronounced at forest edges compared to interiors. Monthly values across the factor types showed that moisture decreased in the throughfall treatment for all the months except March in the forest edge and March and January in the forest interior (Table S3, Fig. 1 ). Principal Component Analysis (PCA) of soil physical and chemical properties showed that the first two principal components together explained 51.0% of the total variance, with Dim1 and Dim2 accounting for 29.3% and 21.7%, respectively (Fig. S2). Dim1 was primarily associated with organic carbon (Org_C), electrical conductivity (E_cond), pH, and available potassium (Available_K), while Dim2 was mainly influenced by micronutrients such as available manganese (Available_Mn), zinc (Available_Zn), and iron (Available_Fe). Available copper (Available_Cu) and phosphorus (Available_P) contributed moderately to both dimensions but were more aligned with Dim1. However, our analysis found no significant differences in soil physical (pH and electrical conductivity) and chemical (organic carbon, and available Cu, Mn, Fe, Zn, K, and P) properties between the treatments of both forest edge and interior (Table S4). 3.2 Seedling survival The predicted survival probability of seedlings was 80%, 73%, 73% and 75% in edge control, edge drought, interior control and interior drought, respectively. Seedling survival probability decreased by 8.75% between control and drought in the forest edge, but was not statistically significant (Fig. S3, Table S5). However, species-specific random slopes showed substantial variation in drought responses, with three species having significantly lower survival under drought conditions at the forest edge (Fig. 2 , Table S6): Artocarpus heterophyllus (ARTHET), Artocarpus hirsutus (ARTHIR), and Calophyllum apetalum (CALAPE) . Survival differences between control and drought were less pronounced in the forest interior, with no species showing strong treatment-specific declines. 3.3 Traits and drought response 3.3.1 Principal Component Analysis of Traits The first two axes of the Principal Component Analysis (PCA) explained 65% (Fig. S4) of the total variation in traits for the 16 species (PC1: 46.4%, PC2: 18.6%). PC1 was associated with Leaf Mass per Area (LMA), Stem Specific Density (SSD), and Leaf Dry Matter Content (LDMC), while PC2 represented variation in Fine Root Specific Density (FRSD) and Leaf Area (LA). The PCA biplot showed that LMA, SSD, and LDMC were positively correlated and LA negatively correlated, indicating a resource acquisitive vs conservative dimension. FRSD was orthogonal to this dimension. 3.3.2 Trait mediation of drought response at forest edge vs. interior: Results were largely consistent between analyses of trait-mediated individual survival and survival ratios. We chose to present survival ratios as they capture the aggregate species-level response in relation to mean trait values. Outputs from models analysing individual survival are available in the Supplementary Information (Table S7, Fig. S5). In the trait-mediated survival ratio results (Table 1 ) that follow, coefficients from the gamma regression and their 95% confidence intervals are presented on the exponentiated scale, where values less than 1 indicate a negative relationship and values greater than 1 indicate a positive relationship. Table 1 Trait-mediated drought response of species. Drought response, quantified as the number of seedlings surviving in drought relative to control, was modelled as an interaction of plant functional traits and forest habitat (Edge and Interior) using generalized linear mixed effect model using gamma family and log link. Table contains exponentiated estimates, CIs written in parenthesis. CI values range having 1 means non-significant, range greater than one represents positive significant and range less than 1 represents significant negative correlation. Significant relationships with alpha ≤ 0.05 were written in bold and significant relationships with alpha ≤ 0.1 were written in bold and italics. Trait Intercept Edge Trait Edge*Trait PC1 1.04 (0.93–1.17) 0.94 (0.83–1.06) 1.01 (0.94–1.09) 0.88 (0.82–0.95) PC2 1.04 (0.93–1.17) 0.94 (0.83–1.06) 1.02 (0.91–1.14) 0.89 (0.79–1.00) LMA 1.11 (0.79–1.56) 1.46 (0.89–2.38) 0.91 (0.53–1.55) 0.49 (0.23–1.04) LDMC 1.02 (0.66–1.57) 1.74 (0.95–3.22) 1.09 (0.29–4.05) 0.14 (0.02–0.90) LA 1.15 (0.83–1.59) 0.63 (0.41–0.98) 1.00 (1.00–1.00) 1.00 (1.00–1.01) SSD 0.87 (0.61–1.25) 1.92 (1.23–2.98) 1.44 (0.72–2.88) 0.24 (0.10–0.55) MRSD 1.01 (0.66–1.56) 1.23 (0.63–2.41) 1.09 (0.39–3.06) 0.51 (0.10–2.57) FRSD 0.97 (0.78–1.22) 1.17 (0.86–1.58) 1.20 (0.63–2.26) 0.48 (0.20–1.15) Multi-trait phenotypes (principal component axes) showed that at the edge, drought response had a significant negative relationship with PC1 (β = 0.88, 95% CI: 0.82–0.95, p = 0.001) and PC2 (β = 0.89, 95% CI: 0.79–1.00, p = 0.04). In the interior, neither PC axis influenced drought response. Individual traits showed patterns consistent with the PC axes (Fig. 3 ). At the forest edge, drought response showed negative trends with LMA (β = 0.49, 95% CI: 0.23–1.04, p = 0.06), LDMC (β = 0.14, 95% CI: 0.02–0.90, p = 0.03), and SSD (β = 0.24, 95% CI: 0.10–0.55, p = 0.001). Thus, higher LMA, LDMC, and SSD, corresponding to resource-conservative strategies, were associated with greater detriment due to drought at the forest edge. At the edge, LA was associated with positive drought response (β = 1.00, 95% CI: 1.00–1.01, p = 0.06), i.e., larger leaf area improved survival in drought relative to control. None of these traits explained drought response in the interior. Root traits (MRSD and FRSD) did not influence drought response, suggesting that the effects of PC2 were driven primarily by LA. 4 Discussion Both forest edge and interior habitats experienced declines in soil moisture with drought treatments compared to controls, but the decrease was more pronounced at the forest edge. Soil moisture at the edge halved from 20–10% volumetric water content (VWC), while the interior saw a smaller decrease of 32%, from 22% VWC to 15% VWC during the course of the experiment. This hints at forest edges exacerbating drought stress compared to forest interiors, likely due to their increased exposure to environmental extremes such as higher light and temperature. Of course, the degree of soil moisture decrease with complete throughfall exclusion does not reflect real outcomes of diminished rainfall and only serves as a qualitative indicator of the edge-interior variation in drought. Real-time monitoring of soil moisture and microclimate will reveal the extent to which interannual variation in climate alters drought conditions at forest edges vs interiors. Interestingly, the strongest impacts of simulated drought were observed during the monsoon months, when soil moisture levels are typically highest under natural conditions. These findings suggest that drier-than-usual monsoons in this system can alter the spatial availability of soil moisture. This has implications for seed germination and seedling establishment since most tree species depend on the wetter months for their regeneration in the humid forests of the Western Ghats. Decreases in moisture may also alter biotic interactions during regeneration, such as microbially-mediated dynamics, which deserves further study (Dudenhöffer et al., 2018; Milici et al., 2025). Response to drought may also depend on seedling neighbourhoods, and drought may alter the relative importance of intra- vs inter-specific competition (O’Brien et al. 2017c ; Lebrija-Trejos et al. 2023 ). Drought did not affect mean seedling survival even at the forest edges, where we expected prominent declines in survival, even with the soil moisture dropping as low as 10%. One potential explanation is the ability of seedlings to adjust physiologically or modify resource allocation to cope with temporary moisture stress (Sunny et al. 2025 ). Also, in field conditions, the effects of reduced soil moisture may be alleviated by other factors such as plasticity in root allocation, hydraulic lift by larger trees, or changes in the microbiome, e.g., mycorrhizae that may help seedlings to withstand low soil moisture, which were not directly measured in this study. Alternatively, the duration of the drought treatment may not have been sufficient to drive mortality, and effects may emerge only over longer time frames of multiple dry seasons. Seedling performance could have been driven by drought-induced changes in soil physical properties (e.g., pH and electrical conductivity) or chemical properties (such as organic carbon and available nutrients). However, these soil properties did not vary between drought and well-watered conditions at the end of the experiment. At forest edges, characterised by greater exposure to light, temperature fluctuations, and lower soil moisture retention, drought appears to favour species with resource-acquisitive traits. Seedling drought response was higher in species with traits linked to drought avoidance strategies. These included low specific stem density (SSD), low leaf dry matter content (LDMC), larger leaf area, and higher fine root surface density (FRSD). Such traits are commonly associated with a fast resource acquisition strategy, which may allow plants to take advantage of brief periods of increased soil moisture following rainfall (Grime 2006 ; Comas et al. 2013 ). High FRSD may improve a plant's ability to absorb water efficiently during short windows of availability, as found in edge environments. The consistency in the drought response mediated by PC1 and PC2 suggests coordination of above- and below-ground strategies to deal with water use. Acquisitive traits in general may relate to drought avoidance, and this should be assessed from water-use physiology. In contrast, traits typically linked to drought tolerance, such as higher tissue density or lower root surface area that reduce water loss and/or promote conservative water use, were not associated with better drought response at the edge. This suggests that in environments where water availability fluctuates widely, avoiding drought through rapid uptake may be more beneficial than tolerating extended dry periods (Chaves et al. 2003 ; McDowell et al. 2008 ). Trait effects on drought response were weaker in the forest interior, indicating that trait-environment interactions were stronger at the more stressful edge conditions. Our inferences are, however, based on the average trait values of species: intraspecific traits variation or physiological responses may help improve individual predictions of individual performance in response to drought ( Premugh et al. In revision) . Nonetheless, the trait-based drought response of seedlings was consistent with landscape-scale findings in the Western Ghats, where species with lower stem-specific density (SSD) were more likely to occur in areas with greater seasonal water deficit (Krishnadas et al., 2021 ). With ongoing fragmentation of tropical forests, edge effects may further favour the regeneration of light-wooded, low LDMC species, which are known to increase at forest edges (Silva Da Costa et al. 2020; Zuñe-da-Silva et al. 2022 ). In contrast to edges, within the interior, traits did not play a prominent role in mediating drought responses. Microclimate buffering by the intact canopy may mitigate drought effects in the interior, as indicated in most species showing similar numbers surviving in drought vs control. 5 Conclusion Our study demonstrates that forest edges experience greater soil moisture declines under drought, leading to stronger trait-mediated filtering of seedling survival compared to forest interiors. Seedlings with resource-acquisitive traits were more resilient to drought at forest edges, whereas traits played no role in the buffered microclimate of the interior. These patterns suggest that drier future conditions could favour acquisitive species at edges, potentially slowing succession and altering long-term species composition. Given ongoing forest fragmentation and restoration efforts across tropical landscapes like the Western Ghats, our findings underscore the need to incorporate microhabitat variation and functional traits into restoration planning. Specifically, restoration strategies may need to prioritise drought-resilient species at forest edges to ensure successful regeneration, and resource-conservative species should be planted in cooler microclimates to ensure better outcomes. Long-term monitoring across life stages and quantifying physiological traits relevant to light and water use will help predict restoration outcomes under increasingly variable climates. As fragmentation and climate change continue to reshape tropical ecosystems, trait-based frameworks offer a valuable tool for conservation and restoration (Laughlin et al. 2017). Declarations Conflict of interest Authors declare no conflict of interest Funding information National Geographic Society Author Contribution Peddiraju Bandaru: Data curation (equal); formal analysis (equal); investigation (equal); validation (equal); visualisation (equal); writing – original draft (equal); writing – review and editing (equal). Lakshmipriya Cannanbilla, Ashish Nambiar, Rishanth Kuruvankunath Ravi, Ashok Kumar Mani, Sarvanan, Sharath Prakash, Ranjana Gauri Muniraja, and Malavika Kamath: Experimental setup (equal), investigation (equal), Data curation (equal). Meghna Krishnadas: Conceptualization(equal); data curation (equal); formal analysis (equal); funding acquisition (lead); experimental setup (equal); investigation (equal); methodology (equal); project administration (lead); resources (lead); supervision (lead); validation(lead); visualization (equal); writing – original draft (equal); writing– review and editing (equal) Acknowledgement We thank Kadamane Estate Company for permission to conduct this research on their property. We thank Rishiddh Jhaveri for help with initial data analysis and code. Data Availability Data and code can be found at https://github.com/Peddiraju14/Kadumane-NG. Please contact Peddiraju Bandaru ( [email protected] ) with any issues. References Allen K, Dupuy JM, Gei MG, et al (2017) Will seasonally dry tropical forests be sensitive or resistant to future changes in rainfall regimes? Environ Res Lett 12:023001. https://doi.org/10.1088/1748-9326/aa5968 Brooks M, Bolker B, Kristensen K, et al (2017) glmmTMB: Generalized Linear Mixed Models using Template Model Builder. 1.1.9 Chaves MM, Maroco JP, Pereira JS (2003) Understanding plant responses to drought — from genes to the whole plant. Funct Plant Biol 30:239. https://doi.org/10.1071/FP02076 Comas LH, Becker SR, Cruz VMV, et al (2013) Root traits contributing to plant productivity under drought. Front Plant Sci 4:. https://doi.org/10.3389/fpls.2013.00442 Comita LS, Engelbrecht BMJ (2009) Seasonal and spatial variation in water availability drive habitat associations in a tropical forest. Ecology 90:2755–2765. https://doi.org/10.1890/08-1482.1 Comita LS, Engelbrecht BMJ (2014) Drought as a driver of tropical tree species regeneration dynamics and distribution patterns. In: Coomes DA, Burslem DFRP, Simonson WD (eds) Forests and Global Change, 1st edn. Cambridge University Press, pp 261–308 Cumming G, Fidler F, Vaux DL (2007) Error bars in experimental biology. J Cell Biol 177:7–11. https://doi.org/10.1083/jcb.200611141 De Frenne P, Rodríguez-Sánchez F, De Schrijver A, et al (2015) Light accelerates plant responses to warming. Nat Plants 1:15110. https://doi.org/10.1038/nplants.2015.110 De Frenne P, Zellweger F, Rodríguez-Sánchez F, et al (2019) Global buffering of temperatures under forest canopies. Nat Ecol Evol 3:744–749. https://doi.org/10.1038/s41559-019-0842-1 Duffy PB, Brando P, Asner GP, Field CB (2015) Projections of future meteorological drought and wet periods in the Amazon. Proc Natl Acad Sci 112:13172–13177. https://doi.org/10.1073/pnas.1421010112 Engelbrecht BMJ, Comita LS, Condit R, et al (2007) Drought sensitivity shapes species distribution patterns in tropical forests. Nature 447:80–82. https://doi.org/10.1038/nature05747 Engelbrecht BMJ, Kursar TA (2003) Comparative drought-resistance of seedlings of 28 species of co-occurring tropical woody plants. Oecologia 136:383–393. https://doi.org/10.1007/s00442-003-1290-8 Esquivel‐Muelbert A, Baker TR, Dexter KG, et al (2019) Compositional response of Amazon forests to climate change. Glob Change Biol 25:39–56. https://doi.org/10.1111/gcb.14413 Grime JP (2006) Plant strategies, vegetation processes, and ecosystem properties, 2. ed. Wiley, Chichester Haddad NM, Brudvig LA, Clobert J, et al (2015) Habitat fragmentation and its lasting impact on Earth’s ecosystems. Sci Adv 1:e1500052. https://doi.org/10.1126/sciadv.1500052 Jhaveri R, Cannanbilla L, Bhat KSA, et al (2024) Anatomical traits explain drought response of seedlings from wet tropical forests. Ecol Evol 14:e70155. https://doi.org/10.1002/ece3.70155 Krishnadas M (2023) Rainfall affects interactions between plant neighbours. Nature 615:39–40. https://doi.org/10.1038/d41586-023-00267-y Krishnadas M, Agarwal K, Comita LS (2020) Edge effects alter the role of fungi and insects in mediating functional composition and diversity of seedling recruits in a fragmented tropical forest. Ann Bot 126:1181–1191. https://doi.org/10.1093/aob/mcaa138 Krishnadas M, Bagchi R, Sridhara S, Comita LS (2018) Weaker plant-enemy interactions decrease tree seedling diversity with edge-effects in a fragmented tropical forest. Nat Commun 9:4523. https://doi.org/10.1038/s41467-018-06997-2 Krishnadas M, Comita LS (2018) Influence of soil pathogens on early regeneration success of tropical trees varies between forest edge and interior. Oecologia 186:259–268. https://doi.org/10.1007/s00442-017-4006-1 Krishnadas M, Kumar AN, Comita LS (2019) Edge effects reduce α‐diversity but not β‐diversity during community assembly in a human‐modified tropical forest. Ecol Appl 29:e01996. https://doi.org/10.1002/eap.1996 Krishnadas M, Peddiraju B, Vadigi S, Nambiar A (2025) Unraveling the relationship between environment and plant functional traits. In: Plant Functional Traits. Elsevier, pp 59–86 Krishnadas M, Sankaran M, Page N, et al (2021) Seasonal drought regulates species distributions and assembly of tree communities across a tropical wet forest region. Glob Ecol Biogeogr 30:1847–1862. https://doi.org/10.1111/geb.13350 Lê S, Josse J, Husson F (2008) FactoMineR : An R Package for Multivariate Analysis. J Stat Softw 25:. https://doi.org/10.18637/jss.v025.i01 Lebrija-Trejos E, Hernández A, Wright SJ (2023) Effects of moisture and density-dependent interactions on tropical tree diversity. Nature 615:100–104. https://doi.org/10.1038/s41586-023-05717-1 Lüdecke D (2013) sjPlot: Data Visualization for Statistics in Social Science. 2.8.17 Markesteijn L, Poorter L (2009) Seedling root morphology and biomass allocation of 62 tropical tree species in relation to drought‐ and shade‐tolerance. J Ecol 97:311–325. https://doi.org/10.1111/j.1365-2745.2008.01466.x Martinez-Meier A, Sanchez L, Pastorino M, et al (2008) What is hot in tree rings? The wood density of surviving Douglas-firs to the 2003 drought and heat wave. For Ecol Manag 256:837–843. https://doi.org/10.1016/j.foreco.2008.05.041 Martínez‐Vilalta J, Mencuccini M, Vayreda J, Retana J (2010) Interspecific variation in functional traits, not climatic differences among species ranges, determines demographic rates across 44 temperate and Mediterranean tree species. J Ecol 98:1462–1475. https://doi.org/10.1111/j.1365-2745.2010.01718.x McDowell N, Pockman WT, Allen CD, et al (2008) Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytol 178:719–739. https://doi.org/10.1111/j.1469-8137.2008.02436.x Nardini A, Battistuzzo M, Savi T (2013) Shoot desiccation and hydraulic failure in temperate woody angiosperms during an extreme summer drought. New Phytol 200:322–329. https://doi.org/10.1111/nph.12288 O’Brien MJ, Burslem DFRP, Caduff A, et al (2015) Contrasting nonstructural carbohydrate dynamics of tropical tree seedlings under water deficit and variability. New Phytol 205:1083–1094. https://doi.org/10.1111/nph.13134 O’Brien MJ, Engelbrecht BMJ, Joswig J, et al (2017a) A synthesis of tree functional traits related to drought‐induced mortality in forests across climatic zones. J Appl Ecol 54:1669–1686. https://doi.org/10.1111/1365-2664.12874 O’Brien MJ, Reynolds G, Ong R, Hector A (2017b) Resistance of tropical seedlings to drought is mediated by neighbourhood diversity. Nat Ecol Evol 1:1643–1648. https://doi.org/10.1038/s41559-017-0326-0 O’Brien MJ, Reynolds G, Ong R, Hector A (2017c) Resistance of tropical seedlings to drought is mediated by neighbourhood diversity. Nat Ecol Evol 1:1643–1648. https://doi.org/10.1038/s41559-017-0326-0 Pérez-Harguindeguy N, Díaz S, Garnier E, et al (2013) New handbook for standardised measurement of plant functional traits worldwide. Aust J Bot 61:167. https://doi.org/10.1071/BT12225 Phillips OL, Aragão LEOC, Lewis SL, et al (2009) Drought Sensitivity of the Amazon Rainforest. Science 323:1344–1347. https://doi.org/10.1126/science.1164033 Poorter L, Markesteijn L (2008) Seedling Traits Determine Drought Tolerance of Tropical Tree Species. Biotropica 40:321–331. https://doi.org/10.1111/j.1744-7429.2007.00380.x R Core Team (2023) R: A Language and Environment for Statistical Computing. Vienna, Austria Reddy CS, Sreelekshmi S, Jha CS, Dadhwal VK (2013) National assessment of forest fragmentation in India: Landscape indices as measures of the effects of fragmentation and forest cover change. Ecol Eng 60:453–464. https://doi.org/10.1016/j.ecoleng.2013.09.064 Silva Da Costa W, Da Cunha M, José F. Pena Rodrigues P, et al (2020) Intraspecific variation in functional wood anatomy of tropical trees caused by effects of forest edge. For Ecol Manag 473:118305. https://doi.org/10.1016/j.foreco.2020.118305 Sunny R, Guha A, Jezeera A, et al (2025) Responses to water limitation are independent of light for saplings of a seasonally dry tropical forest. Biotropica 57:e13404. https://doi.org/10.1111/btp.13404 Van Nieuwstadt MGL, Sheil D (2005) Drought, fire and tree survival in a Borneo rain forest, East Kalimantan, Indonesia. J Ecol 93:191–201. https://doi.org/10.1111/j.1365-2745.2004.00954.x Weiskopf SR, Rubenstein MA, Crozier LG, et al (2020) Climate change effects on biodiversity, ecosystems, ecosystem services, and natural resource management in the United States. Sci Total Environ Wickham H, Chang W, Henry L, et al (2007) ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics. 3.5.1 Zellweger F, De Frenne P, Lenoir J, et al (2020) Forest microclimate dynamics drive plant responses to warming. Science 368:772–775. https://doi.org/10.1126/science.aba6880 Zuñe-da-Silva F, Rodrigues PJFP, Rojas-Idrogo C, et al (2022) EDGE INFLUENCE OVER FUNCTIONAL TREE TRAITS IN AN ATLANTIC FOREST REMNANT. Rev Árvore 46:e4603. https://doi.org/10.1590/1806-908820220000003 Additional Declarations No competing interests reported. Supplementary Files PlantecologyMSSI.docx Cite Share Download PDF Status: Published Journal Publication published 05 Dec, 2025 Read the published version in Plant Ecology → Version 1 posted Editorial decision: Revision requested 29 Jul, 2025 Reviews received at journal 25 Jul, 2025 Reviews received at journal 23 Jul, 2025 Reviews received at journal 11 Jul, 2025 Reviewers agreed at journal 01 Jul, 2025 Reviewers agreed at journal 01 Jul, 2025 Reviewers agreed at journal 25 Jun, 2025 Reviewers invited by journal 12 Jun, 2025 Editor assigned by journal 10 Jun, 2025 Submission checks completed at journal 09 Jun, 2025 First submitted to journal 08 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6845805","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470688030,"identity":"85cd84c5-b622-47dc-bd5b-62f3cc047b1f","order_by":0,"name":"Peddiraju Bandaru","email":"","orcid":"","institution":"National Centre for Biological Science","correspondingAuthor":false,"prefix":"","firstName":"Peddiraju","middleName":"","lastName":"Bandaru","suffix":""},{"id":470688031,"identity":"c4425eaf-29fb-4a5f-b0b2-f76f31a9c889","order_by":1,"name":"Lakshmipriya Cannanbilla","email":"","orcid":"","institution":"University of Bayreuth","correspondingAuthor":false,"prefix":"","firstName":"Lakshmipriya","middleName":"","lastName":"Cannanbilla","suffix":""},{"id":470688032,"identity":"8252d871-a578-4112-bd43-be64a3ce5117","order_by":2,"name":"Ashish Nambiar","email":"","orcid":"","institution":"Indiana University","correspondingAuthor":false,"prefix":"","firstName":"Ashish","middleName":"","lastName":"Nambiar","suffix":""},{"id":470688033,"identity":"2b421de2-5a87-4b50-be6c-4d54b8526945","order_by":3,"name":"Rishanth Kuruvankunath Ravi","email":"","orcid":"","institution":"Kadumane estate","correspondingAuthor":false,"prefix":"","firstName":"Rishanth","middleName":"Kuruvankunath","lastName":"Ravi","suffix":""},{"id":470688034,"identity":"935bf03f-1959-4f5f-ad9b-cd0d8962f889","order_by":4,"name":"Ashok Kumar Mani","email":"","orcid":"","institution":"Kadumane estate","correspondingAuthor":false,"prefix":"","firstName":"Ashok","middleName":"Kumar","lastName":"Mani","suffix":""},{"id":470688035,"identity":"746eed05-01b0-4532-b1f3-b8a621cb4028","order_by":5,"name":"Saravanan Karuna Moorthy","email":"","orcid":"","institution":"Pondicherry University","correspondingAuthor":false,"prefix":"","firstName":"Saravanan","middleName":"Karuna","lastName":"Moorthy","suffix":""},{"id":470688036,"identity":"8b76e76d-3205-4898-bd60-37654587083d","order_by":6,"name":"Sharath Prakash","email":"","orcid":"","institution":"Kodachadri Integrated Development Society","correspondingAuthor":false,"prefix":"","firstName":"Sharath","middleName":"","lastName":"Prakash","suffix":""},{"id":470688037,"identity":"d6bfe270-03bb-4adb-ba4d-c9de7ab8cdb6","order_by":7,"name":"Ranjana Gauri Muniraja","email":"","orcid":"","institution":"Rheinische-Friedrich-Wilhelms University of Bonn","correspondingAuthor":false,"prefix":"","firstName":"Ranjana","middleName":"Gauri","lastName":"Muniraja","suffix":""},{"id":470688038,"identity":"306aa6d5-e700-4566-b426-02de4ef42052","order_by":8,"name":"Malavika Kamath","email":"","orcid":"","institution":"Ashoka Trust for Research in Ecology and the Environment","correspondingAuthor":false,"prefix":"","firstName":"Malavika","middleName":"","lastName":"Kamath","suffix":""},{"id":470688039,"identity":"f66ce126-1c5e-4a1d-b1e4-a346e8fc2a44","order_by":9,"name":"Meghna Krishnadas","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAz0lEQVRIiWNgGAWjYHACNiC2YWCQQOISoyWNdC2HEVoIAt329muPC2rO5/HPbn/AXFFRx8An3YBfi9mZM+XGM47dLpa4c8aA8cyZwwxsMgcIaLmRkybNw3Y7cYNEDgNjY9sBBjaJBAJa7r8Bavl3Dqgl/QFj4786IrTcYD8mzdt2AKglwYCxsYGZCC1nctikefuSE2cA/XKw4dhhHsJajh9/Js3zzS6xf3b7w4cNNXVy8jMIaGFg4DGAMw+AuITUAwH7AyIUjYJRMApGwYgGAFoqQR0aB44qAAAAAElFTkSuQmCC","orcid":"","institution":"National Centre for Biological Science","correspondingAuthor":true,"prefix":"","firstName":"Meghna","middleName":"","lastName":"Krishnadas","suffix":""}],"badges":[],"createdAt":"2025-06-08 06:23:21","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6845805/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6845805/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11258-025-01577-z","type":"published","date":"2025-12-05T15:58:16+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":84707617,"identity":"73223f0b-57ea-438c-b862-eecfc414408b","added_by":"auto","created_at":"2025-06-16 12:44:03","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":251180,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eSoil moisture recorded across habitats (edge \u0026amp; interior) and treatments (control \u0026amp; drought).\u003c/strong\u003eSoil moisture percent was modelled using a generalized linear mixed-effects model with a beta error distribution. Plot identity was included as a random intercept, and treatment was modelled as a random slope within month to account for temporal variation in moisture responses. Upper and lower panel shows the soil moisture percentage variation in edge and interior respectively. Blue and red colour represents control and drought (throughfall exclusion) treatment respectively. * Indicates significant difference between treatments with alpha ≤ 0.05.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6845805/v1/790e4fe0e9ca244dc44f8cf3.jpeg"},{"id":84706844,"identity":"4fd17b56-b597-44f1-bfd0-ab31d2cde26a","added_by":"auto","created_at":"2025-06-16 12:36:03","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":176199,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePredicted survival of native tree seedlings across locations (edge and interior) and treatments (control and drought). \u003c/strong\u003eSurvival (binary: alive or dead) was modelled using a generalized linear mixed-effects model with a binomial error distribution. Species identity was included as a random intercept, and random slopes for treatment and location were modelled per species to account for species-specific responses. Left and right panels show the predicted survival probabilities across treatments for each species in edge and interior habitats, respectively. Blue and red colors represent control and drought (throughfall exclusion) treatments. Asterisks (*) indicate statistically significant differences between treatments (α ≤ 0.05).\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6845805/v1/673440ff8e6628b6fec6bd80.jpeg"},{"id":84706847,"identity":"47deb955-a86f-409a-83c9-79d6f1f77038","added_by":"auto","created_at":"2025-06-16 12:36:03","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":582787,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eRelationship between seedling drought response and functional traits at edge vs interior.\u003c/strong\u003eAggregate drought response, measured as the ratio of individuals surviving in drought relative to control conditions, was modelled as a function of composite and individual traits using linear mixed-effects models with a gamma error distribution. Forest habitat and trait values were included as fixed effects, and species identity was included as a random intercept and as random slopes for species responses to drought and forest habitat. Panels show drought response as a function of: (a) PC1 and (b) PC2 from a principal component analysis (PCA) of six traits; (c) leaf mass per area (LMA); (d) leaf dry matter content (LDMC); (e) leaf area; (f) stem specific density (SSD); (g) main root specific density (MRSD); and (h) fine root specific density (FRSD). Lines indicate model fits, with bold solid lines for significant relationships (\u003cem\u003ep\u003c/em\u003e\u0026lt; 0.05), bold dashed lines for marginal significance (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.1), and thin lines for non-significant trends. Green and yellow colour represents drought response at forest interior and edge respectively. Error bars represent 95% confidence intervals.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6845805/v1/db248b693bbaed90554eb747.png"},{"id":97724684,"identity":"2e741889-05e8-419b-979e-022203d95b03","added_by":"auto","created_at":"2025-12-08 16:13:20","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1854768,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6845805/v1/3bc54228-8894-4906-a8fc-5003c40d688b.pdf"},{"id":84706846,"identity":"e792fbcb-13a9-426e-a29e-78571f558856","added_by":"auto","created_at":"2025-06-16 12:36:03","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":646468,"visible":true,"origin":"","legend":"","description":"","filename":"PlantecologyMSSI.docx","url":"https://assets-eu.researchsquare.com/files/rs-6845805/v1/26ce15d5e7a5244240ccde6a.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Acquisitive traits improve seedling survival with drought at the edges of a fragmented tropical forest","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eClimate change and habitat fragmentation are among the most pressing threats to global biodiversity (Haddad et al. \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Weiskopf et al. \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Rising temperature and altered precipitation patterns are projected to result in droughts, which can increase plant mortality (Duffy et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Allen et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Even in moist biomes, plant species performance relates to water availability (Engelbrecht et al. \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Comita and Engelbrecht \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Esquivel-Muelbert et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Krishnadas et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e), and drought will affect species to differing extents. The impact of drought on plant performance may be modulated by other global change factors, an example being habitat fragmentation. Forest fragmentation alters microclimatic conditions closer to forest edges, where higher temperatures, increased vapour pressure deficit, and drier soils can exacerbate the effects of drought (De Frenne et al. \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2015\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Zellweger et al. \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2020\u003c/span\u003e), but there are few empirical tests of this possibility.\u003c/p\u003e \u003cp\u003eWhile drought can harm plants at all life stages, seedlings are highly vulnerable to water deficits due to their shallow root systems, limited stored resources, and light-limited understory conditions for growth and survival (Comita and Engelbrecht, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e; Engelbrecht et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Seedling establishment and community composition are affected by edge effects in forest fragments (Krishnadas and Comita \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Krishnadas et al. \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2019\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Krishnadas \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2023\u003c/span\u003e), to which drought may contribute. Since the seedling bank is critical for forest regeneration, understanding how drought interacts with fragmentation to shape species performance is essential to predict future forest composition (Engelbrecht and Kursar \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Poorter and Markesteijn \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Markesteijn and Poorter \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; O\u0026rsquo;Brien et al. \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePlants cope with environmental stress through functional traits that relate to resource acquisition and stress tolerance (Krishnadas et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Synthesis of plant functional traits across the climatic zones has shown that tree species with higher wood density survived better during drought in the tropics (Phillips et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2009\u003c/span\u003e; Van Nieuwstadt and Sheil, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; O\u0026rsquo;Brien et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2017a\u003c/span\u003e). Similar trait-based responses to drought were seen in forests in the Mediterranean (Mart\u0026iacute;nez-Vilalta et al. \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2010\u003c/span\u003e) and temperate regions (Martinez-Meier et al. \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Nardini et al. \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). A regional assessment of species distributions in peninsular India found that species with higher LMA and lower SSD increased with greater seasonal water deficit, suggesting a fitness advantage for these traits in drier conditions (Krishnadas et al. \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Less is known about how traits influence drought responses at earlier life stages.\u003c/p\u003e \u003cp\u003eGreenhouse studies suggest that indicate that xylem structure and stomatal traits influence drought response of seedlings, larger xylem conduits reduced growth and photosynthesis, and smaller stomata decreased survival rates in drought relative to well-watered controls (Jhaveri et al. \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Other experiments seedlings respond to water deficit by shifting biomass allocation towards roots at the expense of leaves (Sunny et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). While greenhouse experiments provide valuable insights into seedling responses to drought, they do not capture the complexity of drought effects in natural forest conditions, such as competition, water lift, or variable soil resource availability (Comita and Engelbrecht \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo understand how drought and edge effects together shape seedling performance, we used throughfall exclusion to simulate drought at the edges and interiors of forest fragments in a human-modified landscape in the Western Ghats biodiversity hotspot in southern India. The region has experienced significant deforestation, with a 0.20% annual loss in forest cover, decreasing patch size, and increasing edge density from 1975 to 2005 (Reddy et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), which provides an ideal setting to examine the interaction between drought and forest fragmentation. Specifically, we asked:\u003c/p\u003e \u003cp\u003e \u003col\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDoes rainfall exclusion decrease soil moisture availability and properties by a greater extent at forest edges than interiors?\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDoes the effect of drought on the survival of forest seedlings vary between the forest edge vs. interior?\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003cspan\u003e \u003cli\u003e \u003cp\u003eDo plant traits mediate the drought response of seedlings, and does this vary between the forest edge and interior?\u003c/p\u003e \u003c/li\u003e \u003c/span\u003e \u003c/ol\u003e \u003c/p\u003e \u003cp\u003eWe hypothesised that drought-induced reduction in soil moisture would be more pronounced at forest edges, possibly due to greater solar radiation and wind exposure. We expected seedling survival to decrease with drought on average, and more so at edges due to harsher microclimatic conditions. Finally, we predicted that traits associated with resource-conservative strategies, such as thicker leaves and denser stems and roots, would show smaller declines in performance under drought, and that the influence of traits would be more prominent at forest edges.\u003c/p\u003e"},{"header":"2 Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Study site:\u003c/h2\u003e \u003cp\u003eThis experiment was conducted in a 30 km\u003csup\u003e2\u003c/sup\u003e human-modified forest landscape in the central Western Ghats (12\u0026deg;56\u0026rsquo;N, 75\u0026deg;39\u0026rsquo;E) located in the Hassan district of Karnataka state (Krishnadas et al. \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). The landscape comprises tropical humid forests and receives a mean annual rainfall of \u003cem\u003eca.\u003c/em\u003e 5000 mm, most of which falls during the monsoon season from July through October, with a pronounced dry season from December through May. Most seedling recruitment occurs during or just after monsoon rains, but seedlings have to survive the dry season to persist. Forest fragments make up \u003cem\u003eca.\u003c/em\u003e 60% of the study landscape, with the remainder being tea plantations, human settlements, and montane grasslands. The soils are primarily clayey Alfisols with good drainage, originating from a gneissic base.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Experimental design\u003c/h2\u003e \u003cp\u003eWe conducted this study on saplings of 16 native tree species (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Species were chosen according to availability and germination of sufficient seeds and a sufficient degree of shade-tolerance for seedling survival in the understory. Seeds were sown in plastic grow-bags and emergent seedlings cultivated for 9\u0026ndash;12 months after germination, then transplanted into the field between January and February in 2021. Each sapling was tagged with a unique ID and planted within a 2m X 2m plot in the pairwise combination of control and drought treatments at the forest edge and interior, replicated across 13 locations (hereafter blocks). Drought treatment was simulated using the throughfall exclusion technique by covering the assigned plot with polycarbonate sheeting to prevent rain from falling onto seedlings and the underlying soil layers. This method has been successfully used in multiple drought studies across the world and shown to reduce soil moisture without compromising light availability (Engelbrecht and Kursar \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; O\u0026rsquo;Brien et al. \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2017b\u003c/span\u003e). The throughfall exclosure was maintained from March 2021 to February 2022. In each treatment block at both forest habitats, soil moisture was measured every month using a hand-held volumetric soil moisture sensor (HS2-20-HS2 CSA, Hydrosense II). Every month, the status of tagged saplings was recorded as alive/dead, new leaves/fallen leaves, and herbivory presence/absence was noted. At the end of the experiment, soil samples from the top 20 cm were collected from the corners of each plot to analyse soil physical (pH, and Electrical conductivity) and chemical (Organic Carbon, Available Cu, Mn, Fe, Zn, K, and P) properties.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Trait data\u003c/h2\u003e \u003cp\u003eWe followed protocols recommended by (P\u0026eacute;rez-Harguindeguy et al. \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) to quantify functional traits. Three to five alive individuals from both the forest edge and interior of each species were harvested, brought back to the field station on the same day, and saturated overnight by immersing the petiole, root, and branch in a container filled with water. Water-saturated leaves were weighed to determine fresh weight, scanned with a desktop scanner for quantifying leaf area (LA), and then oven\u0026ndash;dried at 70\u0026deg;C for 72 hours to determine dry weight. Leaf mass per unit area (LMA) was quantified as the ratio of dry weight to area, and leaf dry matter content (LDMC) as the ratio of dry weight to saturated fresh weight. A portion of the stem, main root and fine root was taken and used the water displacement method to estimate the volume, followed by oven-drying at 70\u0026deg;C for 72 hours to determine dry weight. Stem-specific density (SSD) was estimated as the ratio of dry weight to volume. Main root specific density (MRSD) and fine root specific density (FRSD) were estimated as the dry weight of the main root and fine root to their volumes.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Statistical analysis\u003c/h2\u003e \u003cp\u003eFor question 1, linear mixed effects models with beta error distribution were used to model soil moisture percentage in relation to an interaction between habitat and treatment. Plot ID and month were included as random intercepts. Random slopes were modelled for each month to capture temporal variability and dependence of responses. In Wilkinson-Roger\u0026rsquo;s notation, the model is expressed as:\u003c/p\u003e \u003cp\u003eSoil moisture\u0026thinsp;~\u0026thinsp;Forest habitat * Treatment + (1 | Plot id) + (1\u0026thinsp;+\u0026thinsp;Forest habitat: Treatment | Month)\u003c/p\u003e \u003cp\u003eSimilarly, we checked whether rainfall exclusion altered the physical and chemical properties of the soil by the end of the experiment. We used a linear mixed effects model using a Gaussian error distribution to relate soil properties to an interaction between forest habitat and treatment, with plot ID included as a random intercept.\u003c/p\u003e \u003cp\u003eSoil property\u0026thinsp;~\u0026thinsp;Forest habitat * Treatment + (1 | Plot id)\u003c/p\u003e \u003cp\u003eFor question 2, we assessed the survival of individual seedlings using a generalised linear mixed-effects model with a binomial error distribution.\u003c/p\u003e \u003cp\u003eSurvival\u0026thinsp;~\u0026thinsp;Forest habitat * Treatment + (1 | Forest habitat * Treatment | Species)\u003c/p\u003e \u003cp\u003eFixed effects tested whether survival response to drought varied between forest edge and interior habitat, while random slope for the interaction of effect and treatment and a random intercept for species was included to account for species-specific differences in response to habitat and treatment. Inference for random slopes of species between control and drought were made using the overlap of CIs in one treatment factor with the estimate of another treatment factor (Cumming et al. \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2007\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor question 3, we quantified the drought response of each species as the proportion of individuals surviving in the drought treatment relative to the control at the end of the experiment across all locations (Engelbrecht and Kursar \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:Drought\\:response\\:=Number\\:surviving\\:in\\:drought\\:/Number\\:surviving\\:in\\:control$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eTo test if traits mediate drought response, first, we did a principal component analysis (PCA) on traits to obtain composite phenotypes defined by trait combinations based on leaf mass per area, leaf dry matter content, leaf area, stem specific density, main and fine root specific density. To test whether the composite phenotype defined by PCA axes and individual plant traits correlates with drought response, and if this relation differed between forest habitats, we used linear mixed-effects models with gamma error distribution. Gamma errors were suitable as the responses varied from 0\u0026ndash;2.5.\u003c/p\u003e \u003cp\u003eDrought response\u0026thinsp;~\u0026thinsp;Forest habitat * Trait + (1\u0026thinsp;+\u0026thinsp;Forest habitat | Species)\u003c/p\u003e \u003cp\u003eSpecies identity was included as a random intercept, accommodating additional species-specific variation in drought response that is not related to the traits we measured. We also studied this question with the binomial survival data as a function of traits, treatment and forest habitat with three-way interaction. We also tested if the individual survival was mediated by trait interaction with forest habitat and treatment using generalised linear mixed-effects models (GLMMs) with a binomial error distribution and a logit link function, as survival was measured as a binary outcome (alive or dead). The fixed effects included plant trait values, forest habitat (edge vs. interior), and drought treatment (control vs. drought). Species identity was included as a random effect, with random slopes for forest habitat and drought treatment to account for species-specific responses. The model was specified as:\u003c/p\u003e \u003cp\u003eSurvival\u0026thinsp;~\u0026thinsp;Trait \u0026times; Forest habitat \u0026times; Drought Treatment + (1\u0026thinsp;+\u0026thinsp;Forest habitat\u0026thinsp;+\u0026thinsp;Treatment | Species)\u003c/p\u003e \u003cp\u003eAll data management and analysis were conducted using the R programming language version 4.3.1 (R Core Team \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Mixed effects models were implemented using glmmTMB (Brooks et al. \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2017\u003c/span\u003e), and visualised using ggplot2 (Wickham et al. \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2007\u003c/span\u003e), FactoMineR (L\u0026ecirc; et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2008\u003c/span\u003e), and sjPlot (L\u0026uuml;decke \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Results","content":"\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Soil moisture, physical and chemical properties\u003c/h2\u003e \u003cp\u003eThe lowest soil moisture (VWC) was observed in the month of January with values of 5.02% \u0026plusmn; 0.52, 7.21% \u0026plusmn; 0.6, 7.56% \u0026plusmn; 0.56, and 8.47% \u0026plusmn; 0.78 in the forest edge drought, forest edge control, forest interior drought and forest interior control respectively (Table S2, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The highest soil moisture was observed in the month of August for the control treatment at both forest habitats, whereas drought treatments at the forest edge and interior experienced highest soil moisture in the month of September and July respectively. The highest soil moisture values (VWC) observed were 16.02% \u0026plusmn; 1.48, 32.95% \u0026plusmn; 0.59, 20.17% \u0026plusmn; 1.45, and 34.82% \u0026plusmn; 0.93 in the edge-drought, edge-control, interior-drought and interior-control, respectively. As expected, throughfall exclusion significantly reduced soil moisture over a year, with a stronger effect at the forest edge than in the interior. At the edge, soil moisture content halved under drought treatment, decreasing from 20% (95% CI: 15\u0026ndash;26%) in control plots to 10% (8\u0026ndash;13%) in drought plots. In contrast, in the forest interior, soil moisture declined by approximately 32%, from 22% (17\u0026ndash;29%) in control to 15% (12\u0026ndash;19%) under drought (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). These results indicate that the impact of rainfall exclusion on soil moisture availability is more pronounced at forest edges compared to interiors. Monthly values across the factor types showed that moisture decreased in the throughfall treatment for all the months except March in the forest edge and March and January in the forest interior (Table S3, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003ePrincipal Component Analysis (PCA) of soil physical and chemical properties showed that the first two principal components together explained 51.0% of the total variance, with Dim1 and Dim2 accounting for 29.3% and 21.7%, respectively (Fig. S2). Dim1 was primarily associated with organic carbon (Org_C), electrical conductivity (E_cond), pH, and available potassium (Available_K), while Dim2 was mainly influenced by micronutrients such as available manganese (Available_Mn), zinc (Available_Zn), and iron (Available_Fe). Available copper (Available_Cu) and phosphorus (Available_P) contributed moderately to both dimensions but were more aligned with Dim1. However, our analysis found no significant differences in soil physical (pH and electrical conductivity) and chemical (organic carbon, and available Cu, Mn, Fe, Zn, K, and P) properties between the treatments of both forest edge and interior (Table S4).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Seedling survival\u003c/h2\u003e \u003cp\u003eThe predicted survival probability of seedlings was 80%, 73%, 73% and 75% in edge control, edge drought, interior control and interior drought, respectively. Seedling survival probability decreased by 8.75% between control and drought in the forest edge, but was not statistically significant (Fig. S3, Table S5). However, species-specific random slopes showed substantial variation in drought responses, with three species having significantly lower survival under drought conditions at the forest edge (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e2\u003c/span\u003e, Table S6): \u003cem\u003eArtocarpus heterophyllus (ARTHET), Artocarpus hirsutus (ARTHIR), and Calophyllum apetalum (CALAPE)\u003c/em\u003e. Survival differences between control and drought were less pronounced in the forest interior, with no species showing strong treatment-specific declines.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Traits and drought response\u003c/h2\u003e \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e \u003ch2\u003e3.3.1 Principal Component Analysis of Traits\u003c/h2\u003e \u003cp\u003eThe first two axes of the Principal Component Analysis (PCA) explained 65% (Fig. S4) of the total variation in traits for the 16 species (PC1: 46.4%, PC2: 18.6%). PC1 was associated with Leaf Mass per Area (LMA), Stem Specific Density (SSD), and Leaf Dry Matter Content (LDMC), while PC2 represented variation in Fine Root Specific Density (FRSD) and Leaf Area (LA). The PCA biplot showed that LMA, SSD, and LDMC were positively correlated and LA negatively correlated, indicating a resource acquisitive vs conservative dimension. FRSD was orthogonal to this dimension.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e \u003ch2\u003e3.3.2 Trait mediation of drought response at forest edge vs. interior:\u003c/h2\u003e \u003cp\u003eResults were largely consistent between analyses of trait-mediated individual survival and survival ratios. We chose to present survival ratios as they capture the aggregate species-level response in relation to mean trait values. Outputs from models analysing individual survival are available in the Supplementary Information (Table S7, Fig. S5). In the trait-mediated survival ratio results (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) that follow, coefficients from the gamma regression and their 95% confidence intervals are presented on the exponentiated scale, where values less than 1 indicate a negative relationship and values greater than 1 indicate a positive relationship.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003e\u003cb\u003eTrait-mediated drought response of species.\u003c/b\u003e Drought response, quantified as the number of seedlings surviving in drought relative to control, was modelled as an interaction of plant functional traits and forest habitat (Edge and Interior) using generalized linear mixed effect model using gamma family and log link. Table contains exponentiated estimates, CIs written in parenthesis. CI values range having 1 means non-significant, range greater than one represents positive significant and range less than 1 represents significant negative correlation. Significant relationships with alpha\u0026thinsp;\u0026le;\u0026thinsp;0.05 were written in bold and significant relationships with alpha\u0026thinsp;\u0026le;\u0026thinsp;0.1 were written in bold and italics.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTrait\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIntercept\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEdge\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTrait\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eEdge*Trait\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePC1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.04 (0.93\u0026ndash;1.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.94 (0.83\u0026ndash;1.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.01 (0.94\u0026ndash;1.09)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.88 (0.82\u0026ndash;0.95)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePC2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.04 (0.93\u0026ndash;1.17)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.94 (0.83\u0026ndash;1.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.02 (0.91\u0026ndash;1.14)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.89 (0.79\u0026ndash;1.00)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLMA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.11 (0.79\u0026ndash;1.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.46 (0.89\u0026ndash;2.38)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e0.91 (0.53\u0026ndash;1.55)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.49 (0.23\u0026ndash;1.04)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLDMC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.02 (0.66\u0026ndash;1.57)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.74 (0.95\u0026ndash;3.22)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09 (0.29\u0026ndash;4.05)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.14 (0.02\u0026ndash;0.90)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLA\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.15 (0.83\u0026ndash;1.59)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e0.63 (0.41\u0026ndash;0.98)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.00 (1.00\u0026ndash;1.00)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e1.00 (1.00\u0026ndash;1.01)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSSD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.87 (0.61\u0026ndash;1.25)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u003cb\u003e1.92 (1.23\u0026ndash;2.98)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.44 (0.72\u0026ndash;2.88)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.24 (0.10\u0026ndash;0.55)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMRSD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.01 (0.66\u0026ndash;1.56)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.23 (0.63\u0026ndash;2.41)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.09 (0.39\u0026ndash;3.06)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.51 (0.10\u0026ndash;2.57)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFRSD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e0.97 (0.78\u0026ndash;1.22)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.17 (0.86\u0026ndash;1.58)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e1.20 (0.63\u0026ndash;2.26)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u003cb\u003e0.48 (0.20\u0026ndash;1.15)\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eMulti-trait phenotypes (principal component axes) showed that at the edge, drought response had a significant negative relationship with PC1 (β\u0026thinsp;=\u0026thinsp;0.88, 95% CI: 0.82\u0026ndash;0.95, p\u0026thinsp;=\u0026thinsp;0.001) and PC2 (β\u0026thinsp;=\u0026thinsp;0.89, 95% CI: 0.79\u0026ndash;1.00, p\u0026thinsp;=\u0026thinsp;0.04). In the interior, neither PC axis influenced drought response. Individual traits showed patterns consistent with the PC axes (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e3\u003c/span\u003e). At the forest edge, drought response showed negative trends with LMA (β\u0026thinsp;=\u0026thinsp;0.49, 95% CI: 0.23\u0026ndash;1.04, p\u0026thinsp;=\u0026thinsp;0.06), LDMC (β\u0026thinsp;=\u0026thinsp;0.14, 95% CI: 0.02\u0026ndash;0.90, p\u0026thinsp;=\u0026thinsp;0.03), and SSD (β\u0026thinsp;=\u0026thinsp;0.24, 95% CI: 0.10\u0026ndash;0.55, p\u0026thinsp;=\u0026thinsp;0.001). Thus, higher LMA, LDMC, and SSD, corresponding to resource-conservative strategies, were associated with greater detriment due to drought at the forest edge. At the edge, LA was associated with positive drought response (β\u0026thinsp;=\u0026thinsp;1.00, 95% CI: 1.00\u0026ndash;1.01, p\u0026thinsp;=\u0026thinsp;0.06), i.e., larger leaf area improved survival in drought relative to control. None of these traits explained drought response in the interior. Root traits (MRSD and FRSD) did not influence drought response, suggesting that the effects of PC2 were driven primarily by LA.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eBoth forest edge and interior habitats experienced declines in soil moisture with drought treatments compared to controls, but the decrease was more pronounced at the forest edge. Soil moisture at the edge halved from 20\u0026ndash;10% volumetric water content (VWC), while the interior saw a smaller decrease of 32%, from 22% VWC to 15% VWC during the course of the experiment. This hints at forest edges exacerbating drought stress compared to forest interiors, likely due to their increased exposure to environmental extremes such as higher light and temperature. Of course, the degree of soil moisture decrease with complete throughfall exclusion does not reflect real outcomes of diminished rainfall and only serves as a qualitative indicator of the edge-interior variation in drought. Real-time monitoring of soil moisture and microclimate will reveal the extent to which interannual variation in climate alters drought conditions at forest edges vs interiors.\u003c/p\u003e \u003cp\u003eInterestingly, the strongest impacts of simulated drought were observed during the monsoon months, when soil moisture levels are typically highest under natural conditions. These findings suggest that drier-than-usual monsoons in this system can alter the spatial availability of soil moisture. This has implications for seed germination and seedling establishment since most tree species depend on the wetter months for their regeneration in the humid forests of the Western Ghats. Decreases in moisture may also alter biotic interactions during regeneration, such as microbially-mediated dynamics, which deserves further study (Dudenh\u0026ouml;ffer et al., 2018; Milici et al., 2025). Response to drought may also depend on seedling neighbourhoods, and drought may alter the relative importance of intra- vs inter-specific competition (O\u0026rsquo;Brien et al. \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2017c\u003c/span\u003e; Lebrija-Trejos et al. \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Drought did not affect mean seedling survival even at the forest edges, where we expected prominent declines in survival, even with the soil moisture dropping as low as 10%. One potential explanation is the ability of seedlings to adjust physiologically or modify resource allocation to cope with temporary moisture stress (Sunny et al. \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). Also, in field conditions, the effects of reduced soil moisture may be alleviated by other factors such as plasticity in root allocation, hydraulic lift by larger trees, or changes in the microbiome, e.g., mycorrhizae that may help seedlings to withstand low soil moisture, which were not directly measured in this study. Alternatively, the duration of the drought treatment may not have been sufficient to drive mortality, and effects may emerge only over longer time frames of multiple dry seasons. Seedling performance could have been driven by drought-induced changes in soil physical properties (e.g., pH and electrical conductivity) or chemical properties (such as organic carbon and available nutrients). However, these soil properties did not vary between drought and well-watered conditions at the end of the experiment.\u003c/p\u003e \u003cp\u003eAt forest edges, characterised by greater exposure to light, temperature fluctuations, and lower soil moisture retention, drought appears to favour species with resource-acquisitive traits. Seedling drought response was higher in species with traits linked to drought avoidance strategies. These included low specific stem density (SSD), low leaf dry matter content (LDMC), larger leaf area, and higher fine root surface density (FRSD). Such traits are commonly associated with a fast resource acquisition strategy, which may allow plants to take advantage of brief periods of increased soil moisture following rainfall (Grime \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Comas et al. \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). High FRSD may improve a plant's ability to absorb water efficiently during short windows of availability, as found in edge environments. The consistency in the drought response mediated by PC1 and PC2 suggests coordination of above- and below-ground strategies to deal with water use. Acquisitive traits in general may relate to drought avoidance, and this should be assessed from water-use physiology. In contrast, traits typically linked to drought tolerance, such as higher tissue density or lower root surface area that reduce water loss and/or promote conservative water use, were not associated with better drought response at the edge. This suggests that in environments where water availability fluctuates widely, avoiding drought through rapid uptake may be more beneficial than tolerating extended dry periods (Chaves et al. \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; McDowell et al. \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTrait effects on drought response were weaker in the forest interior, indicating that trait-environment interactions were stronger at the more stressful edge conditions. Our inferences are, however, based on the average trait values of species: intraspecific traits variation or physiological responses may help improve individual predictions of individual performance in response to drought (\u003cem\u003ePremugh et al. In revision)\u003c/em\u003e. Nonetheless, the trait-based drought response of seedlings was consistent with landscape-scale findings in the Western Ghats, where species with lower stem-specific density (SSD) were more likely to occur in areas with greater seasonal water deficit (Krishnadas et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). With ongoing fragmentation of tropical forests, edge effects may further favour the regeneration of light-wooded, low LDMC species, which are known to increase at forest edges (Silva Da Costa et al. 2020; Zu\u0026ntilde;e-da-Silva et al. \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). In contrast to edges, within the interior, traits did not play a prominent role in mediating drought responses. Microclimate buffering by the intact canopy may mitigate drought effects in the interior, as indicated in most species showing similar numbers surviving in drought vs control.\u003c/p\u003e"},{"header":"5 Conclusion","content":"\u003cp\u003eOur study demonstrates that forest edges experience greater soil moisture declines under drought, leading to stronger trait-mediated filtering of seedling survival compared to forest interiors. Seedlings with resource-acquisitive traits were more resilient to drought at forest edges, whereas traits played no role in the buffered microclimate of the interior. These patterns suggest that drier future conditions could favour acquisitive species at edges, potentially slowing succession and altering long-term species composition. Given ongoing forest fragmentation and restoration efforts across tropical landscapes like the Western Ghats, our findings underscore the need to incorporate microhabitat variation and functional traits into restoration planning. Specifically, restoration strategies may need to prioritise drought-resilient species at forest edges to ensure successful regeneration, and resource-conservative species should be planted in cooler microclimates to ensure better outcomes. Long-term monitoring across life stages and quantifying physiological traits relevant to light and water use will help predict restoration outcomes under increasingly variable climates. As fragmentation and climate change continue to reshape tropical ecosystems, trait-based frameworks offer a valuable tool for conservation and restoration (Laughlin et al. 2017).\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eConflict of interest\u003c/h2\u003e\n\u003cp\u003eAuthors declare no conflict of interest\u003c/p\u003e\n\u003ch2\u003eFunding information\u003c/h2\u003e\n\u003cp\u003eNational Geographic Society\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003ePeddiraju Bandaru: Data curation (equal); formal analysis (equal); investigation (equal); validation (equal); visualisation (equal); writing \u0026ndash; original draft (equal); writing \u0026ndash; review and editing (equal). Lakshmipriya Cannanbilla, Ashish Nambiar, Rishanth Kuruvankunath Ravi, Ashok Kumar Mani, Sarvanan, Sharath Prakash, Ranjana Gauri Muniraja, and Malavika Kamath: Experimental setup (equal), investigation (equal), Data curation (equal). Meghna Krishnadas: Conceptualization(equal); data curation (equal); formal analysis (equal); funding acquisition (lead); experimental setup (equal); investigation (equal); methodology (equal); project administration (lead); resources (lead); supervision (lead); validation(lead); visualization (equal); writing \u0026ndash; original draft (equal); writing\u0026ndash; review and editing (equal)\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eWe thank Kadamane Estate Company for permission to conduct this research on their property. We thank Rishiddh Jhaveri for help with initial data analysis and code.\u003c/p\u003e\n\u003ch2\u003eData Availability\u003c/h2\u003e\n\u003cp\u003eData and code can be found at https://github.com/Peddiraju14/Kadumane-NG. Please contact Peddiraju Bandaru ([email protected]) with any issues.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n \u003cli\u003eAllen K, Dupuy JM, Gei MG, et al (2017) Will seasonally dry tropical forests be sensitive or resistant to future changes in rainfall regimes? Environ Res Lett 12:023001. https://doi.org/10.1088/1748-9326/aa5968\u003c/li\u003e\n \u003cli\u003eBrooks M, Bolker B, Kristensen K, et al (2017) glmmTMB: Generalized Linear Mixed Models using Template Model Builder. 1.1.9\u003c/li\u003e\n \u003cli\u003eChaves MM, Maroco JP, Pereira JS (2003) Understanding plant responses to drought \u0026mdash; from genes to the whole plant. Funct Plant Biol 30:239. https://doi.org/10.1071/FP02076\u003c/li\u003e\n \u003cli\u003eComas LH, Becker SR, Cruz VMV, et al (2013) Root traits contributing to plant productivity under drought. Front Plant Sci 4:. https://doi.org/10.3389/fpls.2013.00442\u003c/li\u003e\n \u003cli\u003eComita LS, Engelbrecht BMJ (2009) Seasonal and spatial variation in water availability drive habitat associations in a tropical forest. Ecology 90:2755\u0026ndash;2765. https://doi.org/10.1890/08-1482.1\u003c/li\u003e\n \u003cli\u003eComita LS, Engelbrecht BMJ (2014) Drought as a driver of tropical tree species regeneration dynamics and distribution patterns. In: Coomes DA, Burslem DFRP, Simonson WD (eds) Forests and Global Change, 1st edn. Cambridge University Press, pp 261\u0026ndash;308\u003c/li\u003e\n \u003cli\u003eCumming G, Fidler F, Vaux DL (2007) Error bars in experimental biology. J Cell Biol 177:7\u0026ndash;11. https://doi.org/10.1083/jcb.200611141\u003c/li\u003e\n \u003cli\u003eDe Frenne P, Rodr\u0026iacute;guez-S\u0026aacute;nchez F, De Schrijver A, et al (2015) Light accelerates plant responses to warming. Nat Plants 1:15110. https://doi.org/10.1038/nplants.2015.110\u003c/li\u003e\n \u003cli\u003eDe Frenne P, Zellweger F, Rodr\u0026iacute;guez-S\u0026aacute;nchez F, et al (2019) Global buffering of temperatures under forest canopies. Nat Ecol Evol 3:744\u0026ndash;749. https://doi.org/10.1038/s41559-019-0842-1\u003c/li\u003e\n \u003cli\u003eDuffy PB, Brando P, Asner GP, Field CB (2015) Projections of future meteorological drought and wet periods in the Amazon. Proc Natl Acad Sci 112:13172\u0026ndash;13177. https://doi.org/10.1073/pnas.1421010112\u003c/li\u003e\n \u003cli\u003eEngelbrecht BMJ, Comita LS, Condit R, et al (2007) Drought sensitivity shapes species distribution patterns in tropical forests. Nature 447:80\u0026ndash;82. https://doi.org/10.1038/nature05747\u003c/li\u003e\n \u003cli\u003eEngelbrecht BMJ, Kursar TA (2003) Comparative drought-resistance of seedlings of 28 species of co-occurring tropical woody plants. Oecologia 136:383\u0026ndash;393. https://doi.org/10.1007/s00442-003-1290-8\u003c/li\u003e\n \u003cli\u003eEsquivel‐Muelbert A, Baker TR, Dexter KG, et al (2019) Compositional response of Amazon forests to climate change. Glob Change Biol 25:39\u0026ndash;56. https://doi.org/10.1111/gcb.14413\u003c/li\u003e\n \u003cli\u003eGrime JP (2006) Plant strategies, vegetation processes, and ecosystem properties, 2. ed. Wiley, Chichester\u003c/li\u003e\n \u003cli\u003eHaddad NM, Brudvig LA, Clobert J, et al (2015) Habitat fragmentation and its lasting impact on Earth\u0026rsquo;s ecosystems. Sci Adv 1:e1500052. https://doi.org/10.1126/sciadv.1500052\u003c/li\u003e\n \u003cli\u003eJhaveri R, Cannanbilla L, Bhat KSA, et al (2024) Anatomical traits explain drought response of seedlings from wet tropical forests. Ecol Evol 14:e70155. https://doi.org/10.1002/ece3.70155\u003c/li\u003e\n \u003cli\u003eKrishnadas M (2023) Rainfall affects interactions between plant neighbours. Nature 615:39\u0026ndash;40. https://doi.org/10.1038/d41586-023-00267-y\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Agarwal K, Comita LS (2020) Edge effects alter the role of fungi and insects in mediating functional composition and diversity of seedling recruits in a fragmented tropical forest. Ann Bot 126:1181\u0026ndash;1191. https://doi.org/10.1093/aob/mcaa138\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Bagchi R, Sridhara S, Comita LS (2018) Weaker plant-enemy interactions decrease tree seedling diversity with edge-effects in a fragmented tropical forest. Nat Commun 9:4523. https://doi.org/10.1038/s41467-018-06997-2\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Comita LS (2018) Influence of soil pathogens on early regeneration success of tropical trees varies between forest edge and interior. Oecologia 186:259\u0026ndash;268. https://doi.org/10.1007/s00442-017-4006-1\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Kumar AN, Comita LS (2019) Edge effects reduce \u0026alpha;‐diversity but not \u0026beta;‐diversity during community assembly in a human‐modified tropical forest. Ecol Appl 29:e01996. https://doi.org/10.1002/eap.1996\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Peddiraju B, Vadigi S, Nambiar A (2025) Unraveling the relationship between environment and plant functional traits. In: Plant Functional Traits. Elsevier, pp 59\u0026ndash;86\u003c/li\u003e\n \u003cli\u003eKrishnadas M, Sankaran M, Page N, et al (2021) Seasonal drought regulates species distributions and assembly of tree communities across a tropical wet forest region. Glob Ecol Biogeogr 30:1847\u0026ndash;1862. https://doi.org/10.1111/geb.13350\u003c/li\u003e\n \u003cli\u003eL\u0026ecirc; S, Josse J, Husson F (2008) \u003cstrong\u003eFactoMineR\u003c/strong\u003e : An \u003cem\u003eR\u003c/em\u003e Package for Multivariate Analysis. J Stat Softw 25:. https://doi.org/10.18637/jss.v025.i01\u003c/li\u003e\n \u003cli\u003eLebrija-Trejos E, Hern\u0026aacute;ndez A, Wright SJ (2023) Effects of moisture and density-dependent interactions on tropical tree diversity. Nature 615:100\u0026ndash;104. https://doi.org/10.1038/s41586-023-05717-1\u003c/li\u003e\n \u003cli\u003eL\u0026uuml;decke D (2013) sjPlot: Data Visualization for Statistics in Social Science. 2.8.17\u003c/li\u003e\n \u003cli\u003eMarkesteijn L, Poorter L (2009) Seedling root morphology and biomass allocation of 62 tropical tree species in relation to drought‐ and shade‐tolerance. J Ecol 97:311\u0026ndash;325. https://doi.org/10.1111/j.1365-2745.2008.01466.x\u003c/li\u003e\n \u003cli\u003eMartinez-Meier A, Sanchez L, Pastorino M, et al (2008) What is hot in tree rings? The wood density of surviving Douglas-firs to the 2003 drought and heat wave. For Ecol Manag 256:837\u0026ndash;843. https://doi.org/10.1016/j.foreco.2008.05.041\u003c/li\u003e\n \u003cli\u003eMart\u0026iacute;nez‐Vilalta J, Mencuccini M, Vayreda J, Retana J (2010) Interspecific variation in functional traits, not climatic differences among species ranges, determines demographic rates across 44 temperate and Mediterranean tree species. J Ecol 98:1462\u0026ndash;1475. https://doi.org/10.1111/j.1365-2745.2010.01718.x\u003c/li\u003e\n \u003cli\u003eMcDowell N, Pockman WT, Allen CD, et al (2008) Mechanisms of plant survival and mortality during drought: why do some plants survive while others succumb to drought? New Phytol 178:719\u0026ndash;739. https://doi.org/10.1111/j.1469-8137.2008.02436.x\u003c/li\u003e\n \u003cli\u003eNardini A, Battistuzzo M, Savi T (2013) Shoot desiccation and hydraulic failure in temperate woody angiosperms during an extreme summer drought. New Phytol 200:322\u0026ndash;329. https://doi.org/10.1111/nph.12288\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Brien MJ, Burslem DFRP, Caduff A, et al (2015) Contrasting nonstructural carbohydrate dynamics of tropical tree seedlings under water deficit and variability. New Phytol 205:1083\u0026ndash;1094. https://doi.org/10.1111/nph.13134\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Brien MJ, Engelbrecht BMJ, Joswig J, et al (2017a) A synthesis of tree functional traits related to drought‐induced mortality in forests across climatic zones. J Appl Ecol 54:1669\u0026ndash;1686. https://doi.org/10.1111/1365-2664.12874\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Brien MJ, Reynolds G, Ong R, Hector A (2017b) Resistance of tropical seedlings to drought is mediated by neighbourhood diversity. Nat Ecol Evol 1:1643\u0026ndash;1648. https://doi.org/10.1038/s41559-017-0326-0\u003c/li\u003e\n \u003cli\u003eO\u0026rsquo;Brien MJ, Reynolds G, Ong R, Hector A (2017c) Resistance of tropical seedlings to drought is mediated by neighbourhood diversity. Nat Ecol Evol 1:1643\u0026ndash;1648. https://doi.org/10.1038/s41559-017-0326-0\u003c/li\u003e\n \u003cli\u003eP\u0026eacute;rez-Harguindeguy N, D\u0026iacute;az S, Garnier E, et al (2013) New handbook for standardised measurement of plant functional traits worldwide. Aust J Bot 61:167. https://doi.org/10.1071/BT12225\u003c/li\u003e\n \u003cli\u003ePhillips OL, Arag\u0026atilde;o LEOC, Lewis SL, et al (2009) Drought Sensitivity of the Amazon Rainforest. Science 323:1344\u0026ndash;1347. https://doi.org/10.1126/science.1164033\u003c/li\u003e\n \u003cli\u003ePoorter L, Markesteijn L (2008) Seedling Traits Determine Drought Tolerance of Tropical Tree Species. Biotropica 40:321\u0026ndash;331. https://doi.org/10.1111/j.1744-7429.2007.00380.x\u003c/li\u003e\n \u003cli\u003eR Core Team (2023) R: A Language and Environment for Statistical Computing. Vienna, Austria\u003c/li\u003e\n \u003cli\u003eReddy CS, Sreelekshmi S, Jha CS, Dadhwal VK (2013) National assessment of forest fragmentation in India: Landscape indices as measures of the effects of fragmentation and forest cover change. Ecol Eng 60:453\u0026ndash;464. https://doi.org/10.1016/j.ecoleng.2013.09.064\u003c/li\u003e\n \u003cli\u003eSilva Da Costa W, Da Cunha M, Jos\u0026eacute; F. Pena Rodrigues P, et al (2020) Intraspecific variation in functional wood anatomy of tropical trees caused by effects of forest edge. For Ecol Manag 473:118305. https://doi.org/10.1016/j.foreco.2020.118305\u003c/li\u003e\n \u003cli\u003eSunny R, Guha A, Jezeera A, et al (2025) Responses to water limitation are independent of light for saplings of a seasonally dry tropical forest. Biotropica 57:e13404. https://doi.org/10.1111/btp.13404\u003c/li\u003e\n \u003cli\u003eVan Nieuwstadt MGL, Sheil D (2005) Drought, fire and tree survival in a Borneo rain forest, East Kalimantan, Indonesia. J Ecol 93:191\u0026ndash;201. https://doi.org/10.1111/j.1365-2745.2004.00954.x\u003c/li\u003e\n \u003cli\u003eWeiskopf SR, Rubenstein MA, Crozier LG, et al (2020) Climate change effects on biodiversity, ecosystems, ecosystem services, and natural resource management in the United States. Sci Total Environ\u003c/li\u003e\n \u003cli\u003eWickham H, Chang W, Henry L, et al (2007) ggplot2: Create Elegant Data Visualisations Using the Grammar of Graphics. 3.5.1\u003c/li\u003e\n \u003cli\u003eZellweger F, De Frenne P, Lenoir J, et al (2020) Forest microclimate dynamics drive plant responses to warming. Science 368:772\u0026ndash;775. https://doi.org/10.1126/science.aba6880\u003c/li\u003e\n \u003cli\u003eZu\u0026ntilde;e-da-Silva F, Rodrigues PJFP, Rojas-Idrogo C, et al (2022) EDGE INFLUENCE OVER FUNCTIONAL TREE TRAITS IN AN ATLANTIC FOREST REMNANT. Rev \u0026Aacute;rvore 46:e4603. https://doi.org/10.1590/1806-908820220000003\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"plant-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"vege","sideBox":"Learn more about [Plant Ecology](https://www.springer.com/journal/11258)","snPcode":"11258","submissionUrl":"https://submission.nature.com/new-submission/11258/3","title":"Plant Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"tropical forests, forest fragmentation, seedling survival, trait-environment interactions, resource acquisition","lastPublishedDoi":"10.21203/rs.3.rs-6845805/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6845805/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eClimate change and habitat fragmentation threaten biodiversity, but their interactive effects remain poorly understood. In closed-canopy forests, altered rainfall patterns may induce drought conditions that are exacerbated at forest edges due to warmer, drier microclimates. Plant responses to water limitation can be mediated by functional traits related to resource acquisition and stress tolerance. We examined how drought and edge conditions jointly affect seedling survival, and whether species\u0026rsquo; responses are explained by their traits. In a human-modified forest in the central Western Ghats, India, we transplanted\u0026thinsp;~\u0026thinsp;1-year-old seedlings in a factorial combination of habitat (forest edge vs. interior) and drought (throughfall exclusion vs. control). We monitored survival through one year and estimated drought response (survival in drought relative to control), which was related to six traits. Throughfall exclusion reduced soil moisture more at edges, particularly during dry months. At the edge, three species showed significantly lower survival under drought, whereas survival in the interior did not differ with water treatment. Acquisitive traits (high leaf area, low stem specific density, low leaf dry matter content, and low leaf mass per area) improved survival with drought at edges. Trait-mediated responses were not evident in the interior, likely due to buffered microclimates. Multi-trait combinations were better predictors of drought response than individual traits, suggesting trait coordination. Our results suggest that droughts may favour acquisitive species at forest edges, potentially altering community composition, which has implications for management and restoration of fragmented forests in a changing climate.\u003c/p\u003e","manuscriptTitle":"Acquisitive traits improve seedling survival with drought at the edges of a fragmented tropical forest","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-16 12:35:59","doi":"10.21203/rs.3.rs-6845805/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-07-29T16:51:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-25T19:29:23+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-23T20:06:12+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-11T09:34:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"338511843704668756662689858885743579201","date":"2025-07-01T16:35:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"245023499910424461050494610445913380674","date":"2025-07-01T16:19:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"253211165577612499700708709339741111812","date":"2025-06-26T03:13:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-12T15:31:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-10T15:30:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-09T04:05:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Plant Ecology","date":"2025-06-08T06:18:52+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"plant-ecology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"vege","sideBox":"Learn more about [Plant Ecology](https://www.springer.com/journal/11258)","snPcode":"11258","submissionUrl":"https://submission.nature.com/new-submission/11258/3","title":"Plant Ecology","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"6945c416-4460-4542-bb47-69de0d2cf724","owner":[],"postedDate":"June 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-12-08T16:10:33+00:00","versionOfRecord":{"articleIdentity":"rs-6845805","link":"https://doi.org/10.1007/s11258-025-01577-z","journal":{"identity":"plant-ecology","isVorOnly":false,"title":"Plant Ecology"},"publishedOn":"2025-12-05 15:58:16","publishedOnDateReadable":"December 5th, 2025"},"versionCreatedAt":"2025-06-16 12:35:59","video":"","vorDoi":"10.1007/s11258-025-01577-z","vorDoiUrl":"https://doi.org/10.1007/s11258-025-01577-z","workflowStages":[]},"version":"v1","identity":"rs-6845805","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6845805","identity":"rs-6845805","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}