Quantifying the environmental and disturbance drivers of tropical forest vertical structure is critical for monitoring ecosystem carbon dynamics and the impacts of land use change on tropical forests. Using GEDI LiDAR footprints, collected between 2019 and 2022 across the Xingu River Basin, we developed a framework to identify and characterize coherent forest vertical architecture classes across the basin. We applied a hierarchical unsupervised analysis of GEDI relative height waveforms, revealing six distinct vertical architecture classes differing in canopy height, plant area index, and vertical biomass distribution. To link these classes to specific environmental factors, we modelled their probability of occurrence as a function of 30 years of land use and landcover history, landscape configuration metrics, climatic variability, and terrain attributes using Generalized Additive Models with a lagged structure form space and time varying predictors. Our results show GEDI waveform derived architectures capture structural responses to both local environmental conditions and longer-term disturbance regimes. Three classes represent tall, high biomass interior forests with subtle climatic and landscape differences, while three classes correspond to forests near edges, associated with floodplain environments or anthropogenic activities. Edge associated architectures exhibit reduced canopy height and vertical complexity, highlighting the influence of fragmentation. Models’ performance increase with the inclusion of a full lagged structure capturing near 50% of deviance. This modelling framework, underscore the potential of integrating GEDI observations with climatic and land use / land cover datasets to predict forest structural states in response to future environmental scenarios.

The Indochinese leopard (Panthera pardus delacouri) is one of the most endangered leopard subspecies, restricted to less than 4% of its historical range across Southeast Asia. With severe population declines driven by poaching, habitat loss and fragmentation, and prey depletion, effective recovery requires urgent, spatially-informed conservation action. This study presents the first comprehensive connectivity analysis for the Indochinese leopard across its extant and former range, aiming to identify core habitats, connectivity corridors, and opportunities for population restoration. We used resistant kernel modelling through the CoLa Decision Support System to evaluate landscape connectivity across 14 forest complexes in the extant range and 11 potential reintroduction areas, considering scenarios of extant, reintroduction, and combined range connectivity. Importance to connectivity was assessed based on kernel extent, movement density, and protected area overlap. Our results confirmed the Dawna–Tenasserim landscape, spanning parts of Thailand and Myanmar, and Peninsular Malaysia as the most critical strongholds to conserve. In the former range, our analysis revealed potential for large-scale metapopulation structural connectivity across Cambodia and Lao PDR, despite recent local extirpations. Realizing reintroduction opportunities will require complementary actions such as reducing poaching, restoring prey, and improving habitat management. In some complexes, intensive landscape restoration may be needed to reconnect major core areas with smaller reintroduction complexes that could still be critical for meta-population connectivity. Despite increasing human pressure in the region, substantial habitat and connectivity potential remain. Our framework provides a spatial foundation to prioritize actions for preventing extinction and guiding large carnivore recovery across fragmented tropical landscapes.

Remote sensing is an important tool for monitoring species habitat spatially and temporally. Species distribution models (SDM) often rely on remotely-sensed geospatial datasets to predict probability of occurrence and infer habitat preferences. Lidar measurements from the Global Ecosystem Dynamics Investigation (GEDI) are shedding light on three dimensional forest structure in regions of the world where this aspect of species habitat has previously been poorly quantified. Here we combine a large camera trap dataset of mammal species in Borneo and Sumatra with a diverse set of geospatial data to predict the probability of occurrence of 47 species. Multi-temporal GEDI predictors were created through fusion with Landsat time series, extending back to the year 2001. The availability of these GEDI-based forest structure predictors and other temporally-resolved predictor variables enabled temporal matching of species occurrences and hindcast predictions of species probability of occurrence at years 2001 and 2021. Our GEDI-Landsat fusion approach worked well for forest structure metrics related to canopy height (relative height of the 95th percentile of returned energy R2 = 0.62 and relative RMSE = 41%) but, not surprisingly, was less accurate for metrics related to interior canopy vegetation structure (e.g., plant area volume density from 0 to 5 m above the ground R2 = 0.05 and relative RMSE = 85%). For the SDM analyses, we tested several combinations of predictor sets and found that when considering a large pool of multiscale predictors, the exact composition, and whether GEDI Fusion predictors were included, didn’t have a large impact on generalized linear modeling (GLM) and Random Forest (RF) model performance. Adding GEDI Fusion predictors to a baseline set only meaningfully improved performance for some species (n = 4 for RF and n = 3 for GLM). However, when GEDI Fusion predictors were used in a smaller predictor set that is more suitable for hindcasting species probability of occurrence, more SDMs showed meaningful performance improvements relative to the baseline model (n = 9 for RF and n = 4 for GLM) and the relative importance of GEDI-based canopy structure predictors increased relative to when they were combined with the baseline predictor set. Moreover, as we examined predictor importance and partial dependence, the utility of GEDI Fusion predictors in hindcast models was evident in regards to ecological interpretability. We produced a catalog of probability of occurrence maps for all 47 mammals species at 90 m spatial resolution for years 2001 and 2021, enabling subsequent ecological interpretation and conservation analyses.

Vegetation vertical structure refers to the 3D distribution of vegetation aboveground biomass. Vegetation vertical structure of tropical forests influences other ecological and environmental variables that are essential for the functioning of the ecosystems. Integrating over 5.9 million Globel Ecosystem Dynamics Investigation (GEDI) LiDAR (Light Detection and Ranging) footprints, multispectral, and synthetic aperture radar (SAR) imagery, we built five national maps at 25 m resolution of five forest structural metrics for Colombia, South America, for the year 2020. We mapped canopy height, the height of half the cumulative returned energy from GEDI (RH50), total canopy cover, foliage height diversity, and total plant area index. The resulting maps tended to have the highest errors in the Amazon and Andean regions. Total cover had the highest relative error. Interrelationship curves between forest structural metrics of GEDI footprints are maintained across mapped metrics, indicating that the predictive models preserve structural relationships observed in GEDI data. Due to the medium-high spatial resolution and national coverage of the forest structural maps presented in this work, these maps will be useful for evaluating and mapping other ecological variables and conservation priorities in Colombia.

Understanding habitat selection is critical for the conservation of ungulate species. Our aim was to (1) quantify herd-specific habitat selection for American pronghorn (Antilocapra americana) in the southwestern United States and (2) produce a habitat suitability map that can aid in the prioritization of management actions. We used GPS telemetry locations for individual pronghorn from 2007-2013 representing six herds and remotely sensed habitat covariates to model habitat selection. To determine the effect of each habitat covariate on habitat selection, we fit integrated step selection functions (iSSFs) to the data for each pronghorn herd using a mixed-effects modeling framework. We included random effects of individual pronghorn to account for intra-specific variability in selection. We used the coefficient values from iSSFs to produce a habitat suitability map averaged across the six herds and then evaluated the predictions against data from independently tracked pronghorn herds. Our findings indicated that while there is between-herd variability in pronghorn habitat selection in northern Arizona, there were also some common relationships. All herds selected for areas with a greater proportion of grassland and shrubland, however the magnitude of that selection varied between herds. Most herds also selected areas with low topographic diversity and a lower proportion of developed land with different magnitudes of response between herds. The responses to these two covariates seem to be related to function responses to local limiting factors. The average habitat suitability map indicated large swaths of unsuitable area separating some herds with large areas of habitat in the northeast. Our results demonstrate the importance of investigating herd-level variation in habitat selection analyses for herd-forming species, meaning managers can make decisions that are tailored to a herd's unique situation but also contextualize those decisions within conservation efforts across the landscape.

Unsustainable development continues to fragment natural landscapes and wildlife populations, contributing to declining global biodiversity. Advances in computation have enabled ever more sophisticated assessment of development and conservation impacts on functional landscape connectivity. However, accessibility of these advances to non-expert users has lagged. Here we present Connecting Landscapes (CoLa), an integration and expansion of existing software applications that model functional connectivity, population dynamics, and genetic exchange across landscapes. CoLa is a cloud-based or locally installable conservation decision-support system (DSS) that enables user-friendly assessment of the trade-offs between development and conservation in a data-driven framework. We describe the origins of the CoLa DSS, its functionality, and present two case studies at different spatial scales illustrating its use. We expect the CoLa DSS will be particularly useful to decision makers attempting to reconcile economic growth and biodiversity conservation, supporting the transition to conservation-led development needed to stem the ongoing loss of biodiversity.

SDG 15, part of the UN's 2030 Agenda, focuses on "Life on Land". Target 15.5 aims to address habitat degradation, halt biodiversity decline, and protect species from extinction by 2020. Linked to the Red List Index (15.1.1) and its sub-indicator (15.5.1.2), we integrated a methodology to support endangered species and ecoregions in the Life on Land-NASA Project for Peru, Ecuador, and Colombia. This complements the Red List metric, harmonizing reporting systems and facilitating comprehensive SDG 15 reporting. These countries selected the Spectacled bear and the Paramo ecoregion as pilot species and ecoregion to establish a method for forecasting their habitat suitability for SDG 15 reporting (2019-2022). We used the SDM R package to model habitat suitability with 1,192 occurrence records and 1,000 pseudo-absences for the Spectacled bear, and 50,147 occurrence records and 10,000 pseudo-absences for the Paramo. The modeling incorporated 19 Bioclim variables, elevation, and Human Footprint for a baseline (1970-2000) and three RCPs for 2050. The habitat suitability modeling indicated a decline in the spectacled bear's habitat for two RCPs in the high occurrence category. The Paramo ecoregion showed a decrease in all RCPs in the high occurrence probability. Random Forest outperformed other models within the SDM. For the spectacled bear, altitude was crucial for current conditions, while bio6 (Min Temperature of Coldest Month) was significant for RCPs. Elevation was the most important variable for the Paramos in both current conditions and RCPs 2050.

Climate change is a global concern, and its impact on environmental variables such as temperature and annual precipitation is unknown spatially in the desert, andes, and rainforest ecoregions of Peru, Ecuador, and Colombia. In this study, we conducted a general review of climate drivers for South America (SA) and explored climate data using the GCM compareR package (General Circulation Models) and average ensembles for temperature and precipitation. Our results showed that all GCMs demonstrated increases in the annual mean temperature (BIO1) and in the mean temperature of the driest quarter (BIO9) for Peru, Ecuador, and Colombia for 2050 in three RCPs (2.6, 4.5, and 8.5). Also, most of the GCMs showed increases in the annual precipitation (BIO12) and the precipitation in the driest quarter (BIO17). We conducted non-parametric tests (Kruskal-Wallis Test) to assess if the medians of temperature and precipitation in the three ecoregions are equal for both the baseline and the climate change scenarios. We rejected the null hypothesis that the medians are equal for both temperatures and precipitation in the baseline vs. 2050 RCPs (2.6, 4.5, and 8.5). A spatial analysis was conducted to visualize the variations in temperature and precipitation between the RCPs versus the baseline, and the spatial variation at the country or ecoregion level can be observed. The annual mean temperature (°C) or annual precipitation (mm) divided by its standard deviation for each ecoregion (M metric) was analyzed to see how much the average temperature or the annual precipitation is relatively large compared to the variability or dispersion of temperatures or precipitation respectively; the average temperature and the annual precipitation for the baseline and the three RCPs are relatively large and associated with the variability or dispersion of their temperatures in the Napo moist forest compared to the other ecoregions. Our study provides important insights into the potential impacts of climate change on these ecosystems. Prospects in the Napo moist forest ecoregion, where significant changes in temperature and humidity have already occurred, and new species have invaded or evolved in the western Amazon rainforest, are particularly highlighted and reflected in terms of risk mitigation, ecosystem restoration, surveillance, and monitoring.

Abstract Recently classified as a unique species by the IUCN, African forest elephants (Loxodonta cyclotis) are critically endangered due to severe poaching. With limited knowledge about their ecological role due to the dense tropical forests they inhabit in central Africa, it is unclear how the Afrotropics are influenced by elephants. Although their role as seed dispersers is well known, they may also drive large‐scale processes that determine forest structure through the creation of elephant trails and browsing the understory, allowing larger, carbon‐dense trees to succeed. Multiple scales of lidar were collected by NASA in Lopé National Park, Gabon from 2015 to 2022. Utilizing two airborne lidar datasets in an African forest elephant stronghold, detailed canopy structural information was used in conjunction with elephant trail data to determine how forest structure varies on and off trails. Forest along elephant trails displayed different structural characteristics than forested areas off trails, with lower canopy height, canopy cover, and different vertical distribution of plant density. Less plant area density was found on trails at 1 m in height, while more vegetation was found at 12 m, compared to off trail locations. Trails in forest areas with previous logging history had lower plant area in the top of the canopy. Forest elephants can be considered as “logging light” ecosystem engineers, affecting canopy structure through browsing and movement. Both airborne lidar scales were able to capture elephant impact along trails, with the high‐resolution discrete return lidar performing higher than waveform lidar.

Species distribution modeling (SDM) is a fundamental tool in theoretical and applied ecology. However, relatively little is known about the performance of different approaches for scale optimization, model selection, and algorithmic prediction in the context of nonlinear, multiscale and interactive relationships between environmental variables and species occurrence. Modelers often struggle to optimize a tradeoff between ecological relevance, model robustness, complexity, and overfitting. In this paper, we investigated several methods designed to optimize spatial scale and variable selection in SDMs, in each case evaluating model fitness, parsimony and predictive performance. We used a simulation approach to produce a large pool of alternative underlying habitat relationships that reflect a broad range of realistic habitat associations. We also compared several different modeling algorithms, including logistic regression with a generalized linear model (GLM), Lasso and Elastic-Net Regularized GLMs (GLMNet), and random forest (RF), as well as alternative variable and scale selection methods. We found that GLM methods employing all-subsets dredge routines for variable selection were consistently the best predictors based on all criteria of our model performance assessment and across all attributes of the simulated underlying relationship, including nonlinearity and interaction. We had expected machine learning approaches, such as random forest, to perform better in these more complex forms of species-environment relationships. GLM using dredge variable selection was also the method that included the fewest spurious covariates and included the most correct predictors as a proportion of all predictors. We found that univariate scaling was the most robust method of variable and scale selection, along with Minimal Redundancy Maximal Relevancy (MRMR) which performed equivalently. The simulation experiment presented here provides a robust assessment of simulated multi-species distribution model performance, complexity and fidelity. By simulating a large range of potential habitat relationships with varying spatial scale, effect sizes, linearity, and interactions, we comprehensively evaluated model performance across gradients of complexity of the underlying relationships and violations of classical statistical assumptions. This study provides a valuable assessment and a broader example of the power and utility of controlled simulation experiments in habitat relationships and other ecological spatial predictive modeling.

Earth’s ecosystems are characterized by numerous gradients related to the distribution of environmental conditions and resources. Niche theory predicts that animals will evolve traits to exploit changing resource availability and environmental conditions across these gradients. Much work has been done examining how animal traits like body mass and diet change across gradients from regional to global scales. Environmental and resource gradients in the vertical dimension tend to exhibit strong changes over relatively short distances due to the influence of elevation and vegetation. Vegetation structure may be an especially important vertical axis as it contributes to strong gradients in micro- climate, food resources, and predation risk. To investigate interrelationships between the vertical niche and its presumed drivers, we use functional traits, phylogenies, and predation risk to predict the vertical foraging niche for 4,828 mammals and 9,437 birds globally. To provide biogeographic context to the predictive analysis, we use species ranges to map geographic distributions of the vertical foraging niche and relationships between the niche and its presumed drivers. Linking trait databases with species range maps revealed distinct global distributions of vertical foraging niches for mammals and birds. The most important predictors of these niches varied by taxon but there were several systematic relationships. Diet, body mass, and phylogeny were strong predictors of vertical foraging niche across mammal and bird species. Percent fruit in diet exhibited progressively more positive relationships with higher canopy foraging positions. Predation pressure was relatively unimportant in predicting most vertical foraging niches for birds and mammals but displayed a positive trend with arboreal foraging. Geographic hotspots for the importance of fruit in both mammal and bird diets included the Andes-Amazon transition zone, the Amazon Basin, and New Guinea. Our results provide support for the theory of resource driven vertical niche partitioning but also reveal that vertical niches are strongly associated with phylogeny, suggesting niche conservatism in numerous mammal and bird families. Geographic patterns in variable importance values suggest multiple mechanisms behind spatial structure in eco- evolutionary relationships, including latitudinal gradients in vegetation structure and composition, historical patterns of island isolation (in Southeast Asia), and the influence of habitat heterogeneity driven by tectonic processes (in South America). ### Competing Interest Statement The authors have declared no competing interest.

Background Vegetation structure is increasingly recognized as a key variable to explain ecosystems states and dynamics. New Remote Sensing tools are available to complement labor intensive field investigations and consider the global biogeography of this parameter. Objectives We propose to model the processes explaining the interaction between vegetation structure and animal community assembly globally, while requiring minimal computing power, based on the most fundamentals assumptions. Methods We integrate spaceborne (GEDI: Global Ecosystem Dynamics Investigation) and ground based (TLS: Terrestrial Laser Scanning) Lidar data in the Madingley general ecosystem model. We compare the outcome of this integration to previous version and to the TetraDensity estimate of animal biomass and Elton traits database for arboreality. Results Animal biomass density simulated by Madingley is closer to global estimates when integrating vegetation structure. The strength of this effect increases with higher cohort body mass and varies with local environmental conditions and stochastic processes. Simulated proportion of arboreality across cohorts is consistently higher than observations. This is consistent with the divergence of biases between model and database. Conclusions Our results concur with our hypotheses about the role of vegetation structure on animal community assembly, as it reduces total animal biomass abundance. However, assessing the accuracy of its relative weight is challenging. While we have global products about arboreality and animal biomass density, they represent modern day ecosystem state, including anthropogenic activity, while Madingley simulates potential ecosystem optimum. Therefore, we call for further research in this field and for challenging modelling attempts to compare with.

Tropical Dry Forest (TDF) is one of the most threatened terrestrial environments in the neotropics because of high rates of conversion to agriculture. Despite this high degree of threat, many regions lack detailed maps on the actual extent and boundaries of their TDFs, which is fundamental information for their conservation. We developed a methodological framework to map TDF at 10 m in regions where they grade into other forest types. The approach uses climate variables, altitude, soil properties, and remote sensing data (multispectral and SAR Synthetic Aperture Radar-SAR imagery) to predict TDF in the Department of Caldas, Colombia. Accuracy of the resulting map was confirmed with field observations (Overall Accuracy = 0.88, Kappa = 0.73, Area Under the Receiver Operating Characteristics Curve = 0.87, and True Skill Statistic = 0.73). We estimated - 54 km2 of TDF, which represents 9.2 % of its potential area of occurrence in the study area. TDF gradually transitions to Tropical Moist Forest (TMF) from two main low altitude river valleys to higher altitudes in the Andes Mountains. With modification, our methods could be applied for mapping TDF in other tropical regions.

The United Nations recently agreed to major expansions of global protected areas (PAs) to slow biodiversity declines 1 .But while reserves often reduce habitat loss, their efficacy at preserving animal diversity is unclear, as is their influence on biodiversity in surrounding unprotected areas 2-5 .Unregulated hunting can empty PAs of larger animals 6 , illegal tree felling can degrade habitat quality 7 , and parks can simply displace disturbances such as logging and hunting to unprotected areas of the landscape ('leakage') 8 .Alternatively, well-functioning PAs could enhance animal diversity within reserves as well as in nearby unprotected sites ('spillover') 9 .Here we test if PAs across mega-diverse Southeast Asia contribute to vertebrate conservation inside and outside their boundaries.Reserves increased all facets of bird diversity.Large reserves also had substantially enhanced mammal diversity in the adjacent unprotected landscape.Rather than PAs generating leakage that deteriorated ecological conditions elsewhere, our results are consistent with PAs inducing spillover that benefits biodiversity in surrounding areas.These findings support the 2030 United Nations goals of achieving 30% PA coverage by demonstrating that protected areas are associated with higher vertebrate diversity both inside their boundaries and in the broader landscape.

Structurally intact native forests free from major human pressures are vitally important habitats for the persistence of forest biodiversity. However, the extent of such high-integrity forest habitats remaining for biodiversity is unknown. Here, we quantify the amount of high-integrity tropical rainforests, as a fraction of total forest cover, within the geographic ranges of 16,396 species of terrestrial vertebrates worldwide. We found up to 90% of the humid tropical ranges of forest-dependent vertebrates was encompassed by forest cover. Concerningly, however, merely 25% of these remaining rainforests are of high integrity. Forest-dependent species that are threatened and declining and species with small geographic ranges have disproportionately low proportions of high-integrity forest habitat left. Our work brings much needed attention to the poor quality of much of the forest estate remaining for biodiversity across the humid tropics. The targeted preservation of the world’s remaining high-integrity tropical rainforests that are currently unprotected is a critical conservation priority that may help alleviate the biodiversity crisis in these hyperdiverse and irreplaceable ecosystems. Enhanced efforts worldwide to preserve tropical rainforest integrity are essential to meet the targets of the Convention on Biological Diversity’s 2022 Kunming-Montreal Global Biodiversity Framework which aims to achieve near zero loss of high biodiversity importance areas (including ecosystems of high integrity) by 2030.

Global Ecosystem Dynamics Investigation (GEDI) is a relatively new technology for global forest research, acquiring LiDAR measurements of vertical vegetation structure across Earth’s tropical, sub-tropical, and temperate forests. Previous GEDI validation efforts have largely focused on top of canopy accuracy, and findings vary by geographic region and forest type. Despite this, many applications utilize measurements of vertical vegetation distribution from the lower canopy, with a wide diversity of uses for GEDI data appearing in the literature. Given the variability in data requirements across research applications and ecosystems, and the regional variability in GEDI data quality, it is imperative to understand GEDI error to draw strong inferences. Here, we quantify the accuracy of GEDI relative height metrics through canopy layers for the Brazilian Amazon. To assess the accuracy of on-orbit GEDI L2A relative height metrics, we utilize the GEDI waveform simulator to compare detailed airborne laser scanning (ALS) data from the Sustainable Landscapes Brazil project to GEDI data collected by the International Space Station. We also assess the impacts of data filtering based on biophysical and GEDI sensor conditions and geolocation correction on GEDI error metrics (RMSE, MAE, and Bias) through canopy levels. GEDI data accuracy attenuates through the lower percentiles in the relative height (RH) curve. While top of canopy (RH98) measurements have relatively high accuracy (R2 = 0.76, RMSE = 5.33 m), the accuracy of data decreases lower in the canopy (RH50: R2 = 0.54, RMSE = 5.59 m). While simulated geolocation correction yielded marginal improvements, this decrease in accuracy remained constant despite all error reduction measures. Some error rates for the Amazon are double those reported in studies from other regions. These findings have broad implications for the application of GEDI data, especially in studies where forest understory measurements are particularly challenging to acquire (e.g., dense tropical forests) and where understory accuracy is highly important.

Increased environmental threats require proper monitoring of animal communities to understand where and when changes occur. Ecoacoustic tools that quantify natural acoustic environments use a combination of biophony (animal sound) and geophony (wind, rain, and other natural phenomena) to represent the natural soundscape and, in comparison to anthropophony (technological human sound) can highlight valuable landscapes to both human and animal communities. However, recording these sounds requires intensive deployment of recording devices and storage and interpretation of large amounts of data, resulting in large data gaps across the landscape and periods in which recordings are absent. Interpolating ecoacoustic metrics like biophony, geophony, anthropophony, and acoustic indices can bridge these gaps in observations and provide insight across larger spatial extents and during periods of interest. Here, we use seven ecoacoustic metrics and acoustically-derived bird species richness across a heterogeneous landscape composed of densely urbanized, suburban, rural, protected, and recently burned lands in Sonoma County, California, U.S.A., to explore spatiotemporal patterns in ecoacoustic measurements. Predictive models of ecoacoustic metrics driven by land-use/land-cover, remotely-sensed vegetation structure, anthropogenic impact, climate, geomorphology, and phenology variables capture landscape and daily differences in ecoacoustic patterns with varying performance (avg. R ^2 = 0.38 ± 0.11) depending on metric and period-of-day and provide interpretable patterns in sound related to human activity, weather phenomena, and animal activity. We also offer a case study on the use of the data-driven prediction of biophony to capture changes in soniferous species activity before (1–2 years prior) and after (1–2 years post) wildfires in our study area and find that biophony may depict the reorganization of acoustic communities following wildfires. This is demonstrated by an upward trend in activity 1–2 years post-wildfire, particularly in more severely burned areas. Overall, we provide evidence of the importance of climate, spaceborne-lidar-derived forest structure, and phenological time series characteristics when modeling ecoacoustic metrics to upscale site observations and map ecoacoustic biodiversity in areas without prior acoustic data collection. Resulting maps can identify areas of attention where changes in animal communities occur at the edge of human and natural disturbances.

Intact native forests under negligible large-scale human pressures (i.e., high-integrity forests) are critical for biodiversity conservation. However, high-integrity forests are declining worldwide due to deforestation and forest degradation. Recognizing the importance of high-integrity ecosystems (including forests), the Kunming-Montreal Global Biodiversity Framework (GBF) has directly included the maintenance and restoration of ecosystem integrity, in addition to ecosystem extent, in its goals and targets. Yet, the headline indicators identified to help nations monitor forest ecosystems and their integrity can currently track changes only in 1) forest cover or extent, and 2) the risk of ecosystem collapse using the IUCN Red List of Ecosystems (RLE). These headline indicators are unlikely to facilitate the monitoring of forest integrity for two reasons. First, focusing on forest cover not only misses the impacts of anthropogenic degradation on forests but can also fail to detect the effect of positive management actions in enhancing forest integrity. Second, the risk of ecosystem collapse as measured by the ordinal RLE index (from Least Concern to Critically Endangered) makes it unlikely that changes to the continuum of forest integrity over space and time would be reported by nations. Importantly, forest ecosystems in many biodiverse African and Asian nations remain unassessed with the RLE. As such, many nations will likely resort to monitoring forest cover alone and therefore inadequately report progress against forest integrity goals and targets. We concur that monitoring changes in forest cover and the risk of ecosystem collapse are indeed vital aspects of conservation monitoring. Yet, they are insufficient for the specific purpose of tracking progress against crucial ecosystem integrity components of the GBF’s goals. We discuss the pitfalls of merely monitoring forest cover, a likely outcome with the current headline indicators. Augmenting forest cover monitoring with indicators that capture change in absolute area along the continuum of forest integrity would help monitor progress toward achieving area-based targets related to both integrity and extent of global forests.

The UN 2030 Agenda for Sustainable Development Goal 15, termed Life on Land, is monitored by indicators and sub-indicators that largely deal with forest extent. In countries with structurally complex and species-rich forests, indicators and sub-indicators of forest quality are also needed to effectively monitor and sustain ecological integrity. The goal of the paper is to demonstrate the use of complementary sub-indicators of forest quality for SDG15 reporting and conservation planning. Our objective is to apply these sub-indicators within Colombia, Ecuador, and Peru and evaluate spatial patterns and trends over time as a basis for revealing how the results complement the official indicators of forest extent and forest extent in protected areas in informing conservation. The sub-indicators of forest quality quantify naturalness, riparian forest, forest structure and integrity, forest fragmentation, and forest connectivity. We quantified change during 2000–2021 in these metrics and highlighted insights gained from the complementary sub-indicators of forest quality relative to the official sub-indicators based on forest extent,Forests covered about 60–70% of the forested ecoregions in each country in 2000 and this proportion declined in all three countries by approximately 4% by 2021. Only a subset of the forested area was of high forest quality. Natural forests represented about 40% of forests in Colombian and Ecuador in 2000 and 50% in Peru. Those proportions declined: by 6.3% in Colombia, 6.5% in Ecuador, and 3.4% in Peru. Even less of the forested area was Core Forest in 2013; less than 28% among countries. During 2013–2021, the proportion of forest that was Core decreased by 2.3% in Colombia, 4.5% in Ecuador, and 6.7% in Peru. Connected Forests were about 17–22% of forests among the countries in 2013 and declined 10.4% in Colombia, 1.6% in Ecuador, and 3.8% in Peru by 2021. Forests high in forest structure were 10–18% of forests in 2012 among the countries and increased by 1.1–2% by 2021. Forests of high integrity were 7–13% of forests in 2012 and increased by1.4–2% by 2021. Riparian forests represented less than about 7–9% among the countries and declined by 0.6–1.3% by 2021. Thus, the area of highly quality forest across the countries was substantially less than full forest extent and high-quality forest declined at a higher rate than forest extent during 2000–2021. Forest structure and integrity did increase slightly over this time period.Our results for trends in forest naturalness, riparian association, within stand structure, fragmentation, and connectivity demonstrate how consideration of forest quality provides a much stronger basis for evaluating success in meeting SDG15 targets than consideration of forest extent alone.