Accurate characterization of the role of the dry tropics in the global carbon cycle requires precise estimation of woody biomass changes due to ecological and anthropogenic change, including deforestation, forest degradation, regrowth, mortality and enhanced tree growth due to climate change. L-band Synthetic Aperture Radar (SAR) backscatter observations offer a reliable option to consistently map these processes as they are (i) available globally since 2007 (JAXA ALOS-1, ALOS-2 and ALOS-4), and (ii) sensitive to woody structure, such as aboveground biomass density (AGBD) up to similar to 100 t ha(-1). However, we lack multi-site empirical understanding of the scattering processes that determine the relationship between L-band SAR and woody vegetation structure in the dry tropics, and how this is mediated by soil properties. This study used observations from ground plots in Africa (n = 171), Australia (n = 6), and South America (n = 44) to understand the impact of vegetation structure and soil properties on spatially and temporally coincident fully-polarimetric L-band SAR data. Fully-polarimetric L-band SAR single-look complex data were converted to scattering mechanisms/parameters using van Zyl, Cloude-Pottier, and Freeman-Durden polarimetric decompositions to elucidate the physical mechanisms involved. Multivariate SAR-vegetation-soil relationships were analysed using a theory-informed structural equation modelling approach. The strongest positive effects on volume scattering come from stem density (stems ha(-1)) and mean stem biomass of trees, and soil water and sand content (standardized regression coefficients of 0.3, 0.1, 0.2 and 0.1, respectively). The only significant effect on surface scattering is from stem density (0.1). Significant effects on double bounce scattering are from stem density (0.3) and soil sand content (-0.2). Since AGBD is the product of stem density and mean stem biomass, this modelling framework points to a stronger effect from the number of trees rather than their size/biomass. Therefore, AGBD maps relying solely on radar intensity may not reflect significant changes when AGBD is increasing due to the growth of existing stems. Additionally, such maps might overestimate changes in AGBD when driven by the recruitment of new stems or loss of existing stems. Full-polarimetric observations allow the decomposition of the radar signal into volume scattering, surface scattering, and double bounce, enabling the inversion of structural equation models to retrieve both stem density and mean stem biomass. This provides a more comprehensive description of forest structure compared to retrieving only AGBD. As this approach depends on full-polarimetric data, its effectiveness is closely tied to the availability of such observations. Our findings underscore the value of recent and upcoming missions such as ALOS-4 PALSAR-3, BIOMASS and ROSE-L, and highlight the need to prioritise the acquisition of quad-pol SAR data to support future large-scale retrieval of vegetation structure attributes.

The main hurdle in validating global maps of forest above-ground biomass (AGB) is the lack of reference data, particularly for validating maps that represent different epochs and spatial resolutions. Here we introduce AGBref the first global AGB reference dataset derived from a collection of National Forest Inventories (NFI), permanent research plots and local AGB maps from airborne LiDAR. This dataset provides biomass estimates for multiple epochs (2005, 2010, 2015, 2020) at various spatial resolutions or supports (500 m, 1 km, 10 km, 25 km). More importantly, the dataset comes with uncertainty estimates stemming from plot measurement errors and the biomass variability at different supports (within-pixel variability). AGBref also comes with a series of quality flags that inform users about the characteristics of grid cells according on the attributes of the plots within them. We further examine how filtering based on quality flags influences the representativeness of AGBref globally and across major biomes.

Habitat fragmentation is a major threat to biodiversity, but existing literature largely ignores naturally patchy ecosystems in favor of forests, where deforestation creates spatially distinct fragments. Here, we use savannas to highlight the problems with applying forest fragmentation principles to spatially patchy ecosystems. Identifying fragmentation using landscape functionality, specifically connectivity, enables better understanding of ecosystem dynamics. Tools and concepts from connectivity research are well suited to identifying barriers other than vegetation structure contributing to fragmentation. Opportunities exist to improve fragmentation mapping by combining remote-sensing data with field measurements related to connectivity to empirically test whether landscapes are functionally fragmented. Advancements in deep learning and increasingly accessible data open many possibilities for comprehensive maps of fragmentation.

Accurate estimation of forest height plays a pivotal role in mapping carbon stocks from space. Spaceborne LiDARs give accurate spot estimates of forest canopy height, but sample only a tiny fraction of the landscape. The gaps must therefore be filled using other satellite remote sensing data. Although several studies have employed machine learning methods to produce wall-to-wall forest height maps, they have generally overlooked the distinct characteristics of various remote sensing data sources and have not fully exploited the potential benefits of multisource remote sensing integration. In this study, we propose a novel deep learning framework termed the multimodal attention remote sensing network (MARSNet) to extrapolate dominant heights derived from Global Ecosystem Dynamics Investigation (GEDI), using Sentinel-1 C-band Synthetic Aperture Radar (SAR) data, Advanced Land Observing Satellite-2 (ALOS-2) Phased Array type L-band Synthetic Aperture Radar-2 (PALSAR-2) data, and Sentinel-2 passive optical data. MARSNet comprises separate encoders for each remote sensing data modality to extract multi-scale features, and a shared decoder to fuse the features and estimate height. Using individual encoders for each remote sensing data source avoids interference across modalities and extracts distinct representations. To focus on the useful information from each dataset, we reduce the prevalent spatial and layer redundancies in each remote sensing data by incorporating the extended spatial and layer reconstruction convolution (ESLConv) modules in the encoders. MARSNet achieves good performance in estimating dominant height, with a R2 of 0.62 and RMSE of 2.82 m on test data, outperforming the widely used random forest (RF) approach which attained an R2 of 0.55 and RMSE of 3.05 m using the same layers. We demonstrate the efficacy of the MARSNet modules and the expansion of data sources for improving dominant height estimation through network ablation studies and data ablation studies. Finally, we apply the trained MARSNet model to generate wall-to-wall maps at 10 m resolution for Jilin province, China. Through independent validation using field measurements, MARSNet demonstrates an R2 of 0.54 and RMSE of 3.76 m, compared to 0.39 and 4.37 m for the RF baseline model. Additionally, MARSNet effectively mitigates the common tendency of RF models to overestimate in low height areas and underestimate in high canopy areas (low sensitivity). Our research demonstrates the effectiveness of a multimodal deep learning approach fusing GEDI with SAR and passive optical imagery for enhancing the accuracy of high-resolution dominant height estimation. This method shows promise for enabling accurate large-scale forest height mapping in areas where high-quality ground data are available, potentially revolutionizing our understanding of global forest structure and carbon stocks.

Accurate biome delineation is difficult where biomes occupy the same climatic space, as is the case for tropical dry forest and savanna. The resulting confusion limits our ability to understand and manage impacts of global change on these biomes. To address this, we developed an unsupervised, repeatable method to delineate biomes and their component functional ecosystems, based on landscape-level vegetation structure measured using remote sensing and an understanding of the ecology of the region. This approach contrasts with previous definitions, based on climate differences amongst savanna, dry forest and rain forest. Using the heterogeneous north-east Brazil, where several biomes interdigitate, as a case study, a hierarchical functional ecosystem classification is proposed that aligns with both the IUCN Global Ecosystem Typology (GET) and previous work. Based on fuzzy clustering of remotely sensed vegetation attributes, seven groups were found, identified as rain forest, cerrado (savanna) and five caatinga vegetation groups. These groups broadly align with the literature, for example, sedimentary and arboreal caatinga. These groups align with three ‘Ecosystem Functional Groups’ (EFGs) described by the IUCN GET and, additionally, suggest there is a new, fourth EFG in the region: non-pyric shrublands. Random Forest models showed soil pH was the most important environmental variable distinguishing these vegetation groups. These results suggest a remotely sensed structure-based approach is an effective method for operationalising the IUCN GET. North-East Brazil – where many EFGs are interdigitated – serves as a challenging case study and, therefore, we hope our approach will have generality for other regions globally. There are seven vegetation groups in northeast Brazil, including savanna, rain forest and five types of caatinga. Most of these vegetation groups align with the IUCN Global Ecosystem Typology 2.0, but non-pyric shrubland (caatinga) vegetation may represent a new Ecosystem Functional Group. Soil pH is the strongest determinant of vegetation distribution in northeast Brazil. Remote sensing can provide objective, spatially explicit information on vegetation types in the region, largely consistent with previous vegetation classifications. Accurate biome mapping is vital for management, as biomes differ in ecosystem function and consequently require different management.

Wood density is a critical control on tree biomass, so poor understanding of its spatial variation can lead to large and systematic errors in forest biomass estimates and carbon maps. The need to understand how and why wood density varies is especially critical in tropical America where forests have exceptional species diversity and spatial turnover in composition. As tree identity and forest composition are challenging to estimate remotely, ground surveys are essential to know the wood density of trees, whether measured directly or inferred from their identity. Here, we assemble an extensive dataset of variation in wood density across the most forested and tree-diverse continent, examine how it relates to spatial and environmental variables, and use these relationships to predict spatial variation in wood density over tropical and sub-tropical South America. Our analysis refines previously identified east-west Amazon gradients in wood density, improves them by revealing fine-scale variation, and extends predictions into Andean, dry, and Atlantic forests. The results halve biomass prediction errors compared to a naïve scenario with no knowledge of spatial variation in wood density. Our findings will help improve remote sensing-based estimates of aboveground biomass carbon stocks across tropical South America.

Land systems are the nexus of many global sustainability and justice challenges. Here we present eight guiding principles (P1-8) for improved land system policies following the heuristic stages of a policy cycle. The principles are as follows: embrace recognitional justice (P1), be politically strategic (P2), consider multiple policy goals (P3), address systemic issues (P4), take an integrative scope (P5), foster co-development (P6), adopt clear and monitorable targets (P7) and integrate diagnostic and adaptive capacities (P8). We then explore how well policies align with these principles in two globally relevant cases (land-based climate mitigation and biodiversity-friendly agriculture). In both cases, we find that when policies align poorly with the principles at the agenda-setting stage, there is further misalignment at the policy formulation stage. In the instances when recognitional justice is embraced at the onset, policies subsequently integrate more diverse goals and co-development, but they insufficiently consider political strategy and struggle to handle system complexity. Nonetheless, we identify promising policy mixes that provide benefits to multiple actors, integrate multiple goals, take an integrative scope and have strong monitoring and adaptation, aligning well with multiple principles. Further investigation of these principles could reveal promising policy pathways for land systems.

Zimbabwe has pledged to halt and reverse forest loss by 2030, which if accomplished may enhance the delivery of ecosystem services. Uncertainty over the extent of woodland cover change and the impact of land redistribution could impede progress. Through comparative analysis of communal and resettlement areas we investigated the patterns, causes and implications of land-cover change in Hwedza, Zimbabwe between 1990 and 2020. Land-cover classification of remotely sensed data reveals that Hwedza has transitioned from a trajectory of net woodland loss to net woodland gain. There is no evidence that resettlement increased deforestation compared to communal areas. Changes in off-farm income, smallholder tobacco farming, and reduced profitability of staple crops were perceived by interviewees to be important factors affecting woodland change. Due to the importance of woodland services such as fuelwood, our findings highlight the need to address the societal implications of policies aiming to reduce deforestation .

Trees can differ enormously in their crown architectural traits, such as the scaling relationships between tree height, crown width and stem diameter. Yet despite the importance of crown architecture in shaping the structure and function of terrestrial ecosystems, we lack a complete picture of what drives this incredible diversity in crown shapes. Using data from 374,888 globally distributed trees, we explore how climate, disturbance, competition, functional traits, and evolutionary history constrain the height and crown width scaling relationships of 1914 tree species. We find that variation in height-diameter scaling relationships is primarily controlled by water availability and light competition. Conversely, crown width is predominantly shaped by exposure to wind and fire, while also covarying with functional traits related to mechanical stability and photosynthesis. Additionally, we identify several plant lineages with highly distinctive stem and crown forms, such as the exceedingly slender dipterocarps of Southeast Asia, or the extremely wide crowns of legume trees in African savannas. Our study charts the global spectrum of tree crown architecture and pinpoints the processes that shape the 3D structure of woody ecosystems.

Gross primary productivity (GPP), the amount of carbon plants fixed by photosynthesis, is pivotal for understanding the global carbon cycle and ecosystem functioning. Process-based models built on the knowledge of ecological processes are susceptible to biases stemming from their assumptions and approximations. These limitations potentially result in considerable uncertainties in global GPP estimation, which may pose significant challenges to our net zero goals. This study presents UFLUX v2.0, a process-informed model that integrates state-of-the-art ecological knowledge and advanced machine learning (ML) technique to reduce uncertainties in GPP estimation by learning the biases between process-based models and eddy covariance (EC) measurements. In our findings, UFLUX v2.0 demonstrated a substantial improvement in model accuracy, achieving an R-2 of 0.79 with a reduced RMSE of 1.60 g & sdot; Cm(-2)d(-1), compared to the process-based model's R-2 of 0.51 and RMSE of 3.09 g & sdot; Cm(-2)d(-1). Our global GPP distribution analysis indicates that while UFLUX v2.0 and the process-based model achieved similar global total GPP (137.47 and 132.23 PgC, respectively), they exhibited large differences in spatial distribution, particularly in latitudinal gradients. These differences are very likely due to systematic biases in the process-based model and differing sensitivities to climate and environmental conditions. This study offers improved adaptability for GPP modeling across diverse ecosystems and further enhances our understanding of global carbon cycles and its responses to environmental changes.

Southern African woodlands (SAW) are the world's largest savanna, covering ∼ 3 M km2, but their carbon balance and its interactions with climate and disturbance are poorly understood. Here we address three issues that hinder regional efforts to address international climate agreements: producing a state-of-the-art C budget of the SAW region; diagnosing C cycle functional variation and interactions with climate and fire across SAW; and evaluating SAW C cycle representation in land surface models (LSMs). Using 1506 independent 0.5° pixel model calibrations, each constrained with local Earth observation time series of woody carbon stocks (Cwood) and leaf area, we produce a regional SAW C analysis (2006–2017). The regional net biome production is neutral, i.e. −0.08 Mg C ha−1 yr−1 (95 % uncertainty interval −1.67/1.66), with fire emissions contributing ∼ 0.88 Mg C ha−1 yr−1 (95 % uncertainty interval 0.36–2.51). Fire-related mortality driving fluxes from the total Cwood to dead organic matter likely exceeds both fire-related emissions from Cwood into the atmosphere and non-fire Cwood mortality. The emergent spatial variation in biogenic fluxes and C pools is strongly correlated with mean annual precipitation and burned area. However, there are multiple, potentially confounding, causal pathways through which variation in environmental drivers impacts the spatial distribution of C stocks and fluxes, which is mediated by spatial variations in functional parameters like allocation, wood lifespan, and fire resilience. More Cwood in wetter areas is caused by positive precipitation effects on net primary production and on parameters for wood lifespan but is damped by a negative effect with rising precipitation increasing fire-related mortality. Compared to this analysis, LSMs showed marked differences in spatial distributions and magnitudes of C stocks and fire emissions. The current generation of LSMs represents savanna as a single plant functional type, missing important spatial functional variations identified here. Patterns of biomass and C cycling across the region are the outcome of climate controls on production and vegetation–fire interactions which determine residence times, which is linked to spatial variations in key ecosystem functional characteristics.

Miombo woodlands are the world’s largest savanna, covering 2-3 M km2, and are the dominant land cover in the dry tropics of southern Africa. Here we quantify the dynamics of the miombo region carbon cycle, diagnosing stocks and fluxes and their interactions with climate and disturbance, and evaluate their representation in Trendy land surface models (LSMs). We produce a constrained multi-year analysis (2006-2017) using earth observation time series of total wood C (Cwood) and leaf area index to calibrate an intermediate complexity ecosystem model forced with observed climate, deforestation and burned area. Statistical analyses determine the relationships between carbon cycling, environmental and disturbance variables, and evaluate LSMs. The analysis suggests that the regional net biome production is neutral, 0.0 Mg C ha-1 yr-1 (95% Confidence Interval -1.7 - 1.6), with fire emissions contributing ~1.0 Mg C ha-1 yr-1 (95% CI 0.4-2.5). Spatial variation in biogenic fluxes and C pools is strongly correlated with mean annual precipitation. Burned area is also positively correlated with these pools and fluxes. Areas that are more frequently burned tend to have greater precipitation, and shorter residence time of Cwood. Fire-related mortality from Cwood to dead organic matter likely exceeds fire-related emissions from Cwood to atmosphere, and likely exceeds natural rates of Cwood mortality. LSMs match the biogenic fluxes of the analysis, but diverge on C stocks, timings of heterotrophic respiration and magnitude of fire emissions. The analysis suggests that climate, through precipitation, drives spatial variability in Cwood and GPP across the region. Fire disturbance is the major driver of losses from Cwood. Larger annual precipitation is correlated with both greater GPP and greater fire disturbance. These factors have opposing but unbalanced impacts on Cwood, but the precipitation-GPP effect dominates. Patterns of C cycling across the region are a complex outcome of climate controls on production, and vegetation-fire interactions.

1. The global movement for ecosystem restoration has gained momentum in response to the Bonn Challenge (2010) and the UN Decade on Ecosystem Restoration (UNDER, 2021-2030). While several science-based guidelines exist to aid in achieving successful restoration outcomes, significant variation remains in the outcomes of restoration projects. Some of this disparity can be attributed to unexpected responses of ecosystem components to planned interventions. 2. Given the complex nature of ecosystems, we propose that concepts from Complex Systems Science (CSS) that are linked to non-linearity, such as regime shifts, ecological resilience and ecological feedbacks, should be employed to help explain this variation in restoration outcomes from an ecological perspective. 3. Our framework, Explore Before You Restore, illustrates how these concepts impact restoration outcomes by influencing degradation and recovery trajectories. Additionally, we propose incorporating CSS concepts into the typical restoration project cycle through a CSS assessment phase and suggest that the need for such assessment is explicitly included in the guidelines to improve restoration outcomes. 4. To facilitate this inclusion and make it workable by practitioners, we describe indicators and methods available for restoration teams to answer key questions that should make up such CSS assessment. In doing so, we identify key outstanding science and policy tasks that are needed to further operationalize CSS assessment in restoration. 5. Synthesis and applications. By illustrating how key Complex Systems Science (CSS) concepts linked to non-linear threshold behaviour can impact restoration outcomes through influencing recovery trajectories, our framework Explore Before You Restore demonstrates the need to incorporate Complex Systems thinking in ecosystem restoration. We argue that inclusion of CSS assessment into restoration project cycles, and more broadly, into international restoration guidelines, may significantly improve restoration outcomes.

The increase in Earth observations from space in recent years supports improved quantification of carbon storage by terrestrial vegetation and fosters studies that relate satellite measurements to biomass retrieval algorithms. However, satellite observations are only indirectly related to the carbon stored by vegetation. While ground surveys provide biomass stock measurements to act as reference for training the models, they are sparsely distributed. Here, we addressed this problem by designing an algorithm that harnesses the interplay of satellite observations, modeling frameworks and field measurements, and generated global estimates of above-ground biomass (AGB) density that meet the requirements of the scientific community in terms of accuracy, spatial and temporal resolution. The design was adapted to the amount, type and spatial distribution of satellite data available around the year 2020. The retrieval algorithm estimated AGB annually by merging estimates derived from C- and L-band Synthetic Aperture Radar (SAR) backscatter observations with a Water Cloud type of model and does not rely on AGB reference data at the same spatial scale as the SAR data. This model is integrated with functions relating to forest structural variables that were trained on spaceborne LiDAR observations and sub-national AGB statistics. The yearly estimates of AGB were successively harmonized using a cost function that minimizes spurious fluctuations arising from the moderate-to-weak sensitivity of the SAR backscatter to AGB. The spatial distribution of the AGB estimates was correctly reproduced when the retrieval model was correctly set. Over-predictions occasionally occurred in the low AGB range (<50 Mg ha−1) and under-predictions in the high AGB range (>300 Mg ha−1). These errors were a consequence of sometimes too strong generalizations made within the modeling framework to allow reliable retrieval worldwide at the expense of accuracy. The precision of the estimates was mostly between 30% and 80% relative to the estimated value. While the framework is well founded, it could be improved by incorporating additional satellite observations that capture structural properties of vegetation (e.g., from SAR interferometry, low-frequency SAR, or high-resolution observations), a dense network of regularly monitored high-quality forest biomass reference sites, and spatially more detailed characterization of all model parameters estimates to better reflect regional differences.

As part of the REgional Carbon Cycle Assessment and Processes Phase 2 (RECCAP2) project, we developed a comprehensive African Greenhouse gases (GHG) budget for the period 2010-2019 and compared it to the budget over the 1985-2009 (RECCAP1) period. We considered bottom-up process-based models, data-driven remotely sensed products, and national GHG inventories in comparison with top-down atmospheric inversions, accounting also for lateral fluxes. We incorporated emission estimates derived from novel methodologies for termites, herbivores, and fire, which are particularly important in Africa. We further constrained global woody biomass change products with high-quality regional observations. During the RECCAP2 period, Africa’s carbon sink capacity is decreasing, with net ecosystem exchange switching from a small sink of −0.61 ± 0.58 PgCyr−1 in RECCAP1 to a small source in RECCAP2 at 0.162 (-1.793/2.633) PgCyr-1. Net CO2 emissions estimated from bottom-up approaches were 1.588 (-6.461/11.439) PgCO2yr-1, net CH4 were 78.453 (36.665/59.677) TgCH4yr-1) and net N2O were 1.81 (1.716/2.239) TgN2Oyr-1. Top-down atmospheric inversions showed similar trends. LUC emissions increased, representing one of the largest contributions at 1.746 (0.841/2.651) PgCO2eq yr-1 to the African GHG budget and almost similar to emissions from fossil fuels at 1.743 (1.531/1.956) PgCO2eq yr-1, which also increased from RECCAP1. Additionally, wildfire emissions decreased, while fuelwood burning increased. For most component fluxes, uncertainty is large, highlighting the need for increased efforts to address Africa-specific data gaps. However, for RECCAP2, we improved our overall understanding of many of the important components of the African GHG budget that will assist to inform climate policy and action.

Local access to "wild,"common-pool terrestrial and aquatic resources is being diminished by global resource demand and large-scale conservation interventions. Many theories suggest the well-being of wild harvesters can be supported through transitions to other livelihoods, improved infrastructure, and market access. However, new theories argue that such benefits may not always occur because they are context dependent and vary across dimensions of well-being. We test these theories by comparing how wild harvesting and other livelihoods have been associated with food security and life satisfaction in different contexts across similar to 10,800 households in the tropics. Wild harvests coincided with high well-being in remote, assetpoor, and less-transformed landscapes. Yet, overall, well-being increased with electrical infrastructure, proximity to cities, and household capitals. This provides large-scale confirmation of the context dependence of nature's contributions to people, and suggests a need to maintain local wild resource access while investing in equitable access to infrastructure, markets, and skills.

Seasonal foliage display (leaf phenology) is a key determinant of ecosystem function. Variation in land surface phenology, observed via space‐borne remote sensing, can be explained at broad spatial scales by climate, but we lack understanding of how vegetation structure and floristic diversity mediates these relationships. This lack of understanding hampers our ability to predict changes in phenology and therefore ecosystem function, in light of rapid ongoing shifts in biodiversity and ecosystem structure due to land use and climate change. We combined a network of 619 vegetation monitoring sites across seasonally dry tropical deciduous woodlands in Zambia with land surface phenology metrics to investigate the role of tree species diversity, composition and vegetation structure on patterns of land surface phenology, including the phenomenon of pre‐rain green‐up. Tree species diversity was associated with earlier pre‐rain green‐up, a longer growing season, and greater cumulative foliage production. Independent of diversity, proportional abundance of Detarioideae species (subfamily of Fabaceae) was associated with a longer growing season by facilitating earlier pre‐rain green‐up. Woodland stands with larger trees green up earlier, suggesting access to deep groundwater reserves. Senescence metrics showed variation among sites but were not well‐explained by precipitation, temperature, structure or diversity. Synthesis: Tree diversity, composition and structure are co‐determinants of seasonal patterns of foliage display in seasonally dry tropical deciduous woodlands, as measured via land surface phenology, at regional scale. Our study identifies both a niche complementarity effect whereby diverse woodlands exhibit longer growing seasons, as well as a mass‐ratio effect whereby detarioid species drive earlier pre‐rain green‐up, providing insights into the mechanisms underlying the biodiversity ecosystem function relationship in this biome. We stress the importance of considering biotic controls on ecosystem functioning in the next generation of earth‐system models predicting the response of communities to global change.

The highlands of southeast Angola are one of the world's largest intact formations of miombo woodland. Recent interest from conservation groups is increasing the possibility of a new protected area in this conflict-afflicted, remote region, contributing to the “30 × 30” target of the Global Biodiversity Framework. With the potential for a new protected area, it is important to quantify the extent and change of natural and anthropogenic land covers in the region, not least because of the close dependence of livelihoods on natural resources in the miombo. We developed a 1990–2020 land cover time series, analysing deforestation, canopy opening, canopy closure, and vegetation regrowth after disturbance. Regional woodland extent has remained roughly constant despite frequent transitions between dense and open woodlands. Canopy opening peaked post-civil war, potentially related to the resettlement of displaced people. Over 30 years, 61 % ± 2 % of canopy opening was offset by subsequent canopy closure, which peaked a decade after the war ended, indicating the resilience of miombo systems. A woodland resource-use frontier, consisting of deforestation and canopy opening, is evident in the north-west of the area, likely driven by urban demand for agricultural products, charcoal, timber and other wood-derived goods. A distinct “core” of dynamic woodland occupies 52 % of the study region, where there is no evidence that shifting cultivation and local livelihoods are a net cause of land cover change. We do not find evidence for extensive net woody encroachment, only 2 % of the study region is being encroached by woody vegetation. This canopy closure is associated with remoteness from anthropogenic pressures and biophysical drivers that facilitate woody vegetation growth. Policymakers and conservation managers can use these data to aid in locating and prioritising interventions to sustainably produce agricultural and wood fuel products to meet increasing urban demand. Additionally, supporting conditions for maintaining both biophysical processes and livelihoods in remote areas is crucial to achieving 30 × 30 equitably.

Millions of households globally rely on uncultivated ecosystems for their livelihoods. However, much of the understanding about the broader contribution of uncultivated ecosystems to human wellbeing is still based on a series of small-scale studies due to limited availability of large-scale datasets. We pooled together 11 comparable datasets comprising 232 settlements and 10,971 households in ten low-and middle-income countries, representing forest, savanna and coastal ecosystems to analyse how uncultivated nature contributes to multi-dimensional wellbeing and how benefits from nature are distributed between households. The resulting dataset integrates secondary data on rural livelihoods, multidimensional human wellbeing, household demographics, resource tenure and social-ecological context, primarily drawing on nine existing household surveys and their associated contextual information together with selected variables, such as travel time to cities, population density, local area GDP and land use and land cover from existing global datasets. This integrated dataset has been archived with ReShare (UK Data Service) and will be useful for further analyses on nature-wellbeing relationships on its own or in combination with similar datasets.