Evapotranspiration (ET) is a critical process within the hydrological cycle, susceptible to shifts due to changes in land use. In tropical forest regions, widespread transformations often result in mosaic patterns of land-use types. Our goal was to explore the importance of vegetation structure, topography, meteorology and soil for the spatial variability of ET in a tropical mosaic landscape. We used a random forest machine learning technique for spatial data, employing forward feature selection and cross-validation to prevent overfitting. Our study region is situated in north-eastern Madagascar and is mainly composed of forest fragments, vanilla agroforests, rice fields and fallow land of shifting cultivation. We used a combination of open-source data products derived from various satellite experiments. Daily ET data were retrieved from the ECOsystem Spaceborne Thermal Radiometer Experiment on Space Station (ECOSTRESS). Forest structure predictors from GEDI and PROBA-V, meteorological data from ERA5, topography from JAXA and soil data from ISRIC were obtained. The variables included in the L3 algorithm to calculate ECOSTRESS ET daily data were not included in the study to prevent bias in the models. The models achieved high accuracy for the spatial prediction of ET (R2) of 0.76 and 0.82 for different days. Besides other biophysical variables, leaf area index, tree cover and tree height were important variables in predicting ET. Our findings thereby underscore the crucial role of forest structure on ET even in complex structured tropical mosaic landscapes.

The Montgomery equation, which assumes a proportional relationship between the tepal area and the product of the tepal length and width, is validated using data drawn from four Magnolia species. An important metric of floral non-reproductive size is individual petal or tepal area (A). The Montgomery equation (ME) estimates A by assuming a proportional relationship between A and the product of petal or tepal length (L) and width (W), i.e., A ∝ LW , whereas the power-law equation (PLE) assumes the allometric relationship A ∝( LW)^_1 1 . If W/L has a small variation, four relationships are expected to hold true, i.e., A ∝ L^2 , A ∝ L^_2 , A ∝ W^2 , and A ∝ W^_3 , where α1, α2, and α3 are scaling exponents to be estimated. To assess the validity of these six formulae, 2031 the petal-like tepals of 250 flowers from four Magnolia species were measured. The root-mean-square error (RMSE) was used to determine the goodness of fit of each equation, and the percentage error (PE) was used to compare any two equations with the same predicator, i.e., LW, L and W. The ME was validated for calculating A at the species level and for the pooled data given that three of the four species had < 0.05 RMSEs and one had a < 0.07 RMSE. However, the PLE was more robust than the ME at the species level. For the pooled data, the ME and PLE had a negligible difference in RMSE values. These results show that the ME is a valid and non-destructive tool for measuring A for the Magnolia species examined in this study and likely holds true across other more diverse species.

Oil palms (Elaeis guineensis Jacq.) are increasingly cultivated throughout the humid tropics and are reported to have high transpiration rates. A potential contribution of stem water storage to transpiration has been discussed in previous studies. We assessed the water-use characteristics of oil palms at different horizontal and vertical positions in the plant by using three sap flux techniques, i.e., thermal dissipation probes, the heat ratio method and heat field deformation sensors. In a radial profile of the stem, sap flux densities were low at the outer margin, increased to 2.5 cm under the bark and remained relatively high to the innermost measured depth at 7.5 cm. In a vertical profile of the stem and with further sensors in leaf petioles, we found only small time lags in sap flux densities. Time lags along the flow path are often used for analyzing the contribution of water storage to transpiration. Thus, the small observed time differences in our study would leave only little room for the contribution of water storage to transpiration. However, water storage might still contribute to transpiration in ways that are not detected by time lag analysis. Such mechanisms may be explored in future studies.

Trees are major regulators of the forest water balance. To stay vital, water loss via transpiration has to be compensated by root water uptake. However, it is often not clear where exactly the trees get their water from. This has become particularly relevant in central Europe, as forests are facing more frequent and intense droughts with climate change. It is known that water uptake strategies are species-specific, influenced by environmental conditions and, potentially, neighboring species. Yet, knowledge about species-specific patterns of water uptake depth and how it is affected by tree-species mixture is scarce. Stable water isotopes present a valuable tool to elucidate these belowground processes.For our study, we selected mixtures of European beech (Fagus sylvatica), the dominant broadleaved tree species in central Europe, with native, but drought-prone Norway spruce (Picea abies), and non-native, but supposedly more drought-resistant Douglas fir (Pseudotsuga menziesii), as well as the respective pure stands. We aimed to uncover the effect of (1) species identity, (2) species mixture and (3) environmental conditions on water uptake depth. To achieve this, we conducted two sampling campaigns in climatically contrasting years: a natural abundance campaign in relatively wet 2021, covering 20 plots on 4 sites, and a tracer experiment focused on European beech and Douglas fir on a subset of plots, including weekly sampling of 12 trees throughout the drought summer 2022. We found species-specific patterns of water uptake depth, where Norway spruce tended to use the greatest share of shallow water, followed by European beech and Douglas fir. Within species, the data indicated differences in water uptake depth between pure and mixed stands, however, we did not detect a spatial differentiation between co-occurring species. Dry conditions tended to shift water uptake to deeper layers, with beech responding stronger than Douglas fir.Our results corroborate that species-specific traits have to be considered when assessing forest water pathways, especially in mixed forests and under drought. Considering central Europe, our data supports the assumption that Douglas fir may be more drought resistant than Norway spruce by tapping deeper water sources. In mixture, both European beech and Douglas fir seem to exploit similar soil depths, while none of the species is limited to the drought-prone topsoil.

Forests in the Himalayan region are crucial for maintaining the region's ecological balance, conserving biodiversity, and supporting the livelihoods of local people. However, because of limited accessibility and an adverse climate, scientific studies on how forest functions in this region depend on ecological drivers are rare. We used a handheld mobile laser scanner to assess the forest structural complexity (FSC) in the Annapurna Conservation Area of Nepal and related this to its potential drivers, including forest disturbances. Based on stratified sampling, we selected 69 plots across a gradient of elevations and precipitations. Other factors that might influence FSC were obtained from forest inventory data, climatic databases, the Google Earth platform, and digital elevation models. Using simple linear regression and multiple regression analysis, we tested for the dependency of FSC, measured using the box dimension (Db), on influential predictor variables. Overall, explanatory variables strongly influenced FSC (adjusted R2 = 0.60, P < 0.001), with Db being affected by the number of trees, the maximum height of the forests, species diversity, north-facing aspect, soil pH, and forest disturbance. Surprisingly, climatic variables, precipitation, and temperature did not show any effect on FSC. The LiDAR-based approach to FSC used in our study enabled rapid assessment in hard-to-access regions. It can be used to inform effective management and conservation, for example, in monitoring development over time or for benchmarking.

Climate change affects life in tropical landscapes by increasing temperatures, among other impacts. In addition, land transformation from forest to other land cover types typically increases temperatures of surfaces exposed to direct solar radiation. In rural areas, however, many organisms including humans live sheltered below canopies, but little is known about below-canopy surface temperatures. We assessed canopy leaf area index (LAI) and below-canopy surface temperatures in a landscape of rural lowland Sumatra, Indonesia. In this region, land-use trajectories in the past decades led from forest to monocultural plantations. We found that LAI varied more than 10-fold, with the highest values observed in forest and low values in rubber and oil palm plantations. Below-canopy surface temperatures increased significantly with decreasing LAI. A decrease in LAI by one unit (m2 leaf area per m2 ground area) increased ground surface temperature by 0.95 °C (confidence interval: 1.07–0.83 °C) around noon. Likewise, the temperatures measured on peoples’ cap or hijab surfaces under the canopy was strongly affected. For a light-brown hijab, the average noon surface temperature increased by 4.0 °C from forest to oil palm. Thus, in addition to climate warming, land transformation and associated declines in canopy LAI significantly increased below-canopy surface temperatures. Strategies for mitigating such potentially adverse temperature effects may include the integration of more and densely foliated trees into landscapes.

In monoculture-dominated landscapes, recovering biodiversity is a priority, but effective restoration strategies have yet to be identified. In this study, we experimentally tested passive and active restoration strategies to recover taxonomic, phylogenetic, and functional diversity of woody plants within 52 tree islands established in an oil palm landscape. Large tree islands and higher initial planted diversity catalyzed diversity recovery, particularly functional diversity at the landscape level. At the local scale, results demonstrated that greater initial planting diversity begets greater diversity of native recruits, overcoming limitations of natural recruitment in highly modified landscapes. Establishing large and diverse tree islands is crucial for safeguarding rare, endemic, and forest-associated species in oil palm landscapes.

Transformation of forests into agricultural lands threatens biodiversity and ecosystem functions globally. In the biodiversity hotspot Madagascar, key ecosystem functions along with highly endemic flora and fauna are under threat. Comprehensive studies identifying determinants of biodiversity and ecosystem function losses are lacking, with no studies accounting for endemic species richness across multiple taxa. We assess how plot-scale stand structural attributes (basal area, leaf area index, diameter diversity, and tree species richness) and landscape-scale forest cover influence biodiversity (species richness of butterflies, ants, birds, reptiles, amphibians, herbaceous plants, and multidiversity) and ecosystem functions (soil organic carbon, predation rate, acoustic diversity, and multifunctionality) in the tropical mosaic landscape of northeastern Madagascar. Complexly structured stands with a larger basal area and/or a higher tree species richness promoted greater endemic multidiversity (R-2 = 0.70), soil organic carbon, and acoustic diversity. However, our models did not indicate significant explanatory variables for multi-diversity (R-2 = 0.25) and multifunctionality (R-2 = 0.43). Landscape-scale forest cover significantly favored endemic bird species richness. Our findings highlight that a simplified stand structure, resulting from land-use change, substantially reduces endemic biodiversity and ecosystem functions in this historically forested region. Conservation actions should aim at maintaining complex stand structure in forests and agroforests, while restoration interventions should focus on re-building such structures.

Generative models and their possible applications are almost limitless. But there are still problems that such models have. On one hand, the models are difficult to train. Stability in training, mode collapse or non convergence, together with the huge parameter space make it extremely costly and difficult to train and optimize generative models. The following paper proposes an optimization method limited to a few hyperparameters with grid-search and early stopping which selects the best hyperparameter combination based on the results obtained with the Universal Image Quality Index (UIQ) by creating a copy of the source image and comparing it with the generated target. The proposed method allows to directly measure the impact of hyperparameter tuning by comparing the achieved UIQ score against a baseline.

The expansion of the oil palm industry in Indonesia has improved livelihoods in rural communities, but comes at the cost of biodiversity and ecosystem degradation. Here, we investigated ways to balance ecological and economic outcomes of oil palm cultivation. We compared a wide range of production systems, including smallholder plantations, industrialized company estates, estates with improved agronomic management, and estates with native tree enrichment. Across all management types, we assessed multiple indicators of biodiversity, ecosystem functions, management, and landscape structure to identify factors that facilitate economic-ecological win-wins, using palm yields as measure of economic performance. Although, we found that yields in industrialized estates were, on average, twice as high as those in smallholder plantations, ecological indicators displayed substantial variability across systems, regardless of yield variations, highlighting potential for economic-ecological win-wins. Reducing management intensity (e.g., mechanical weeding instead of herbicide application) did not lower yields but improved ecological outcomes at moderate costs, making it a potential measure for balancing economic and ecological demands. Additionally, maintaining forest cover in the landscape generally enhanced local biodiversity and ecosystem functioning within plantations. Enriching plantations with native trees is also a promising strategy to increase ecological value without reducing productivity. Overall, we recommend closing yield gaps in smallholder cultivation through careful intensification, whereas conventional plantations could reduce management intensity without sacrificing yield. Our study highlights various pathways to reconcile the economics and ecology of palm oil production and identifies management practices for a more sustainable future of oil palm cultivation.

Decades of studies have demonstrated links between biodiversity and ecosystem functioning, yet the generality of the relationships and the underlying mechanisms remain unclear, especially for forest ecosystems. Using 11 tree-diversity experiments, we tested tree species richness-community productivity relationships and the role of arbuscular (AM) or ectomycorrhizal (ECM) fungal-associated tree species in these relationships. Tree species richness had a positive effect on community productivity across experiments, modified by the diversity of tree mycorrhizal associations. In communities with both AM and ECM trees, species richness showed positive effects on community productivity, which could have resulted from complementarity between AM and ECM trees. Moreover, both AM and ECM trees were more productive in mixed communities with both AM and ECM trees than in communities assembled by their own mycorrhizal type of trees. In communities containing only ECM trees, species richness had a significant positive effect on productivity, whereas species richness did not show any significant effects on productivity in communities containing only AM trees. Our study provides novel explanations for variations in diversity-productivity relationships by suggesting that tree-mycorrhiza interactions can shape productivity in mixed-species forest ecosystems.

Vegetation structural complexity has been identified as a vital factor for forest ecosystem function, stability, and resilience. However, agricultural land with much reduced structural complexity has largely replaced natural forests in the tropics. Therefore, restoring structural complexity in large-scale plantation monocultures by introducing agroforestry systems may counteract the loss of biodiversity and ecosystem functions. However, we still have limited knowledge of how the structural complexity of agroforests develops under different restoration treatments. We established a large-scale biodiversity enrichment experiment in a conventional monoculture oil palm plantation in Sumatra, Indonesia. In this experiment, agroforests were implemented by planting clusters of native trees ("tree islands") within the oil palm plantation, systematically varying initially planted tree richness ranging from 0-6 (0 corresponding to natural regeneration only) and plot area (25-1600 m2). We tested the effect of the experimental treatments on nine years of the development of local structural complexity using a stand structural complexity index (SSCI) derived from terrestrial laser scanning. We found that tree planting and natural regeneration treatments promoted structural complexity by creating denser and more complex vegetation structures. Plots with a tree planting treatment tended to show greater structural complexity than plots with natural regeneration only. However, during the study period, oil palms still dominated heights, and the temporal change in structural complexity among plots with or without a tree planting treatment did not differ. As for plot area, our results indicate that structural complexity looking from the center of plots did not necessarily depend on the area during the study period, and even small tree islands can increase local structural complexity in a monoculture oil palm plantation. Initially planted tree richness did not significantly affect the development of structural complexity. Nine years after establishment, not planted trees but regenerated woody plants strongly positively affected vegetation density and structural complexity. Our findings highlight that sustaining vegetation density below oil palm canopies is a key strategy to increase the structural complexity of oil palm landscapes.

Plant diversity effects on community productivity often increase over time. Whether the strengthening of diversity effects is caused by temporal shifts in species-level overyielding (i.e., higher species-level productivity in diverse communities compared with monocultures) remains unclear. Here, using data from 65 grassland and forest biodiversity experiments, we show that the temporal strength of diversity effects at the community scale is underpinned by temporal changes in the species that yield. These temporal trends of species-level overyielding are shaped by plant ecological strategies, which can be quantitatively delimited by functional traits. In grasslands, the temporal strengthening of biodiversity effects on community productivity was associated with increasing biomass overyielding of resource-conservative species increasing over time, and with overyielding of species characterized by fast resource acquisition either decreasing or increasing. In forests, temporal trends in species overyielding differ when considering above- versus belowground resource acquisition strategies. Overyielding in stem growth decreased for species with high light capture capacity but increased for those with high soil resource acquisition capacity. Our results imply that a diversity of species with different, and potentially complementary, ecological strategies is beneficial for maintaining community productivity over time in both grassland and forest ecosystems.

Microclimate and vegetation structure control evapotranspiration (ET) from land surfaces at stand and landscape scales. Tropical rainforests are among the most diverse and complex terrestrial ecosystems, harbouring vast plant and animal species throughout their dense multistory canopy. They contribute substantially to global precipitation through their high ET. However, there is little information about ET influences at very small spatial scales under given climatic conditions. In a tropical rainforest on Sumatra, we studied the relationship between pixel-level ET as derived from high-resolution (similar to 10 cm), near-surface thermography from an unmanned aerial vehicle (UAV) and canopy structure as derived from red-green-blue (RGB) image and three-dimensional (3D) point cloud analyses. The 16 derived potential predictors encompassed vegetation height, height variability, vegetation density and reflectance variables. Using regression models, several of the studied variables had a significant linear relationship with ET, but the explained variance was only marginal. However, applying a random forest algorithm including forward feature selection and target oriented cross validation explained substantial parts of the pixel-level variance in ET (R-2 = 0.56-0.65), thus indicating multiple non-linear relationships with interactions among predictor variables. Therein, green leaf index, leaf area density and vegetation height were often the most important variables for the prediction outcome, but their sequence varied among the four study plots. Overall, combining canopy structure variables derived from RGB photogrammetry explained relatively large parts of spatial ET variations. Our study thus indicates the large potential of combining UAV-based thermography and photogrammetry techniques with machine learning approaches to better understand ET but also suggests that more work remains to be done in explaining ET patterns at very small spatial scales.

Enhancing tree diversity may be important to fostering resilience to drought-related climate extremes. So far, little attention has been given to whether tree diversity can increase the survival of trees and reduce its variability in young forest plantations. We conducted an analysis of seedling and sapling survival from 34 globally distributed tree diversity experiments (363,167 trees, 168 species, 3744 plots, 7 biomes) to answer two questions: (1) Do drought and tree diversity alter the mean and variability in plot-level tree survival, with higher and less variable survival as diversity increases? and (2) Do species that survive poorly in monocultures survive better in mixtures and do specific functional traits explain monoculture survival? Tree species richness reduced variability in plot-level survival, while functional diversity (Rao's Q entropy) increased survival and also reduced its variability. Importantly, the reduction in survival variability became stronger as drought severity increased. We found that species with low survival in monocultures survived comparatively better in mixtures when under drought. Species survival in monoculture was positively associated with drought resistance (indicated by hydraulic traits such as turgor loss point), plant height and conservative resource-acquisition traits (e.g. low leaf nitrogen concentration and small leaf size). Synthesis. The findings highlight: (1) The effectiveness of tree diversity for decreasing the variability in seedling and sapling survival under drought; and (2) the importance of drought resistance and associated traits to explain altered tree species survival in response to tree diversity and drought. From an ecological perspective, we recommend mixing be considered to stabilize tree survival, particularly when functionally diverse forests with drought-resistant species also promote high survival of drought-sensitive species. Rising climate extremes, such as drought, can cause major uncertainty in the survival of young trees. Tree diversity can reduce survival variability and stabilize tree survival. Functionally diverse communities with drought-tolerant species can promote the survival of drought-sensitive species.image

Vegetation structural complexity is an important component of forest ecosystems, influencing biodiversity and functioning. Due to the heterogeneous distribution of vegetation elements, structural complexity underpins ecological dynamics, species composition, microclimate, and habitat diversity. Field measurements and Light Detection and Ranging (LiDAR) data, such as airborne (ALS) and terrestrial (TLS), can assess structural characteristics of forest and agroforestry systems at various spatial scales. This assessment is urgently needed for monitoring ecosystem restoration in degraded lands (e.g., in oil palm landscapes), where it is not well-known how structural measures derived from these different approaches relate to each other. Here, we compared the degree of correlation between individual and multivariate datasets of vegetation structural complexity metrics derived from ALS, TLS, and ground-based inventory approaches. The study was conducted in a 140 ha oil palm monoculture, enriched with 52 plots in the form of tree islands representing agroforestry systems of varying sizes and planted diversity levels in Sumatra, Indonesia. Our datasets comprised 25 ALS, five TLS, and nine ground-based inventory metrics. We studied correlations among metrics related to traditional stand summary, heterogeneity, and vertical and horizontal stand structure. We used principal component analysis for data dimensionality reduction, correlation analysis to quantify the strength of relationships between metrics, and Procrustes analysis to investigate the agreement between datasets. Significant correlations were found between ALS and TLS metrics for canopy density (r = 0.79) and maximum tree height (r = 0.58) and between ALS and ground-based inventory measures of stand heterogeneity and height diversity (r between 0.60 and −0.63). Further, we observed significant agreements between the ordinations of multivariate datasets (r = 0.56 for ALS − TLS; and r = 0.46 for ALS – ground-based inventory). Our findings underline the ability of ALS to capture structural complexity patterns, especially for canopy gap dynamics and vegetation height metrics, as captured by TLS, and for measures of heterogeneity and vertical structure as captured by ground-based inventories. Our study highlights the strength of each approach and underscores the potential of integrating ALS and TLS with ground-based inventories for a comprehensive characterization of vegetation structure in complex agroforestry systems, which can provide guidance for their management and support ecosystem restoration monitoring efforts.

Abstract Total leaf area per plant is an important measure of the photosynthetic capacity of an individual plant that together with plant density drives the canopy leaf area index, that is, the total leaf area per unit ground area. Because the total number of leaves per plant (or per shoot) varies among conspecifics and among mixed species communities, this variation can affect the total leaf area per plant and per canopy but has been little studied. Previous studies have shown a strong linear relationship between the total leaf area per plant (or per shoot) (AT) and the total number of leaves per plant (or per shoot) (NT) on a log–log scale for several growth forms. However, little is known whether such a scaling relationship also holds true for bamboos, which are a group of Poaceae plants with great ecological and economic importance in tropical, subtropical, and warm temperate regions. To test whether the scaling relationship holds true in bamboos, two dwarf bamboo species (Shibataea chinensis Nakai and Sasaella kongosanensis ‘Aureostriatus’) with a limited but large number of leaves per culm were examined. For the two species, the leaves from 480 and 500 culms, respectively, were sampled and AT was calculated by summing the areas of individual leaves per culm. Linear regression and correlation analyses reconfirmed that there was a significant log–log linear relationship between AT and NT for each species. For S. chinensis, the exponent of the AT versus NT scaling relationship was greater than unity, whereas that of S. kongosanensis ‘Aureostriatus’ was smaller than unity. The coefficient of variation in individual leaf area increased with increasing NT for each species. The data reconfirm that there is a strong positive power‐law relationship between AT and NT for each of the two species, which may reflect adaptations of plants in response to intra‐ and inter‐specific competition for light.

In tropical regions, shifting from forests and traditional agroforestry to intensive plantations generates conflicts between human welfare (farmers’ demands and societal needs) and environmental protection. Achieving sustainability in this transformation will inevitably involve trade-offs between multiple ecological and socioeconomic functions. To address these trade-offs, our study used a new methodological approach allowing the identification of transformation scenarios, including theoretical landscape compositions that satisfy multiple ecological functions (i.e., structural complexity, microclimatic conditions, organic carbon in plant biomass, soil organic carbon and nutrient leaching losses), and farmers needs (i.e., labor and input requirements, total income to land, and return to land and labor) while accounting for the uncertain provision of these functions and having an actual potential for adoption by farmers. We combined a robust, multi-objective optimization approach with an iterative search algorithm allowing the identification of ecological and socioeconomic functions that best explain current land-use decisions. The model then optimized the theoretical land-use composition that satisfied multiple ecological and socioeconomic functions. Between these ends, we simulated transformation scenarios reflecting the transition from current land-use composition towards a normative multifunctional optimum. These transformation scenarios involve increasing the number of optimized socioeconomic or ecological functions, leading to higher functional richness (i.e., number of functions). We applied this method to smallholder farms in the Jambi Province, Indonesia, where traditional rubber agroforestry, rubber plantations, and oil palm plantations are the main land-use systems. Given the currently practiced land-use systems, our study revealed short-term returns to land as the principal factor in explaining current land-use decisions. Fostering an alternative composition that satisfies additional socioeconomic functions would require minor changes (“low-hanging fruits”). However, satisfying even a single ecological indicator (e.g., reduction of nutrient leaching losses) would demand substantial changes in the current land-use composition (“moonshot”). This would inevitably lead to a profit decline, underscoring the need for incentives if the societal goal is to establish multifunctional agricultural landscapes. With many oil palm plantations nearing the end of their production cycles in the Jambi province, there is a unique window of opportunity to transform agricultural landscapes.

Transformation of forests into agricultural lands threatens biodiversity and ecosystem functions globally. In the biodiversity hotspot Madagascar, key ecosystem functions along with highly endemic flora and fauna are under threat. Comprehensive studies identifying proximate drivers are lacking, with no studies accounting for endemic species richness across multiple taxa. We assess how plot-scale stand structural attributes (basal area, leaf area index, diameter diversity, and tree species richness) and landscape-scale forest cover affect biodiversity (species richness of butterflies, ants, birds, amphibians, reptiles, herbaceous plants, and multidiversity) and ecosystem functions (soil organic carbon, predation rate, acoustic diversity index, and multifunctionality) in the tropical mosaic landscape of northeastern Madagascar. Across a prevalent land-use gradient in the region, we sampled five woody land-use types: old-growth forest, forest fragment, woody fallow, forest-derived and fallow-derived vanilla agroforests. Complexly structured stands with a larger basal area and higher tree species richness promoted greater endemic multidiversity and certain ecosystem functions. Landscape-scale forest cover significantly favored, particularly, endemic bird species richness. Our models explained 70% of the variance in endemic multidiversity, 26% in multidiversity, and 43% in multifunctionality. Significant effects directions of our predictors differed between taxa but were univocally positive for endemic species richness. Our findings highlight that a simplified stand structure, resulting from land-use change, substantially reduces endemic biodiversity and ecosystem functions in this historically forested region. Conservation actions should aim at maintaining complex stand structure in forests and agroforests, while restoration interventions should focus on re-building such structures.### Competing Interest StatementThe authors have declared no competing interest.