Long-term monitoring of grasslands is pivotal for ensuring continuity of many environmental services and for supporting food security and environmental modeling. Remote sensing provides an irreplaceable source of information for studying changes in grasslands. Specifically, Spectral Mixture Analysis (SMA) allows for quantification of physically meaningful ground cover fractions of grassland ecosystems (i.e., green vegetation, nonphotosynthetic vegetation, and soil), which is crucial for our understanding of change processes and their drivers. However, although popular due to straightforward implementation and low computational cost, ‘classical’ SMA relies on a single endmember definition for each targeted ground cover component, thus offering limited suitability and generalization capability for heterogeneous landscapes. Furthermore, the impact of irregular data density on SMA-based long-term trends in grassland ground cover has also not yet been critically addressed. We conducted a systematic assessment of i) the impact of data density on long-term trends in ground cover fractions in grasslands; and ii) the effect of endmember definition used in ‘classical’ SMA on pixel- and map-level trends of grassland ground cover fractions. We performed our study for 13 sites across European grasslands and derived the trends based on the Cumulative Endmember Fractions calculated from monthly composites. We compared three different data density scenarios, i.e., 1984–2021 Landsat data record as is, 1984–2021 Landsat data record with the monthly probability of data after 2014 adjusted to the pre-2014 levels, and the combined 1984–2021 Landsat and 2015–2021 Sentinel-2 datasets. For each site we ran SMA using a selection of sitespecific and generalized endmembers, and compared the pixel- and map-level trends. Our results indicated no significant impact of varying data density on the long-term trends from Cumulative Endmember Fractions in European grasslands. Conversely, the use of different endmember definitions led in some regions to significantly different pixel- and map-level long-term trends raising questions about the suitability of the ‘classical’ SMA for complex landscapes and large territories. Therefore, we caution against using the ‘classical’ SMA for remotesensing- based applications across broader scales or in heterogenous landscapes, particularly for trend analyses, as the results may lead to erroneous conclusions.
File: Lewinska-et-al.-2025-Impact-of-data-density-and-endmember-definitions.pdf
Context Approaches estimating landscape effects
on biodiversity frequently focus on a single extent,
finding one ‘optimal’ extent, or use narrow extents.
However, species perceive the environment in different
ways, select habitat hierarchically, and respond to
multiple selection pressures at extents that best predict
each pressure.
Objective We aimed to assess multi-scale relationships
between primary productivity and species
occurrences and abundances.
Methods We used a multi-scale approach, called
‘scalograms’, to assess landscape level effects of primary
productivity, in the form of Dynamic Habitat
Indices (DHIs) on the occurrences and abundances of
100 Argentinian forest bird species. We used average
DHI values within multiple extents (3 × 3 to 101 ×
101 pixels; 30 m resolution), and 11 ‘scalogram’ metrics
as environmental inputs in occurrence and abundance
models.
Results Average cumulative DHI values in extents
81 × 81 to 101 × 101 pixels (5.9 – 9.2 km2)
and maximum
cumulative DHI across extents were in the top
three predictors of species occurrences (included in
models for 41% and 18% of species, respectively).
Average cumulative DHI values in various extents
contributed ~ 1.6 times more predictive power to
occurrence models than expected. For species abundances,
average DHI values and scalogram measures
were in the top three predictors for < 2% of species
and contributed less model predictive power than
expected, regardless of DHI type (cumulative, minimum,
variation).
Conclusions Argentinian forest bird occurrences,
but not abundances, respond to high levels of primary
productivity at multiple, broad extents rather than a
single ‘optimal’ extent. Factors other than primary
productivity appear to be more important for predicting
abundance.
File: Olah-et-al_2025_Argentina_DHIs_scalograms_bird-occurence_abundance.pdf
Biodiversity science requires effective tools to predict patterns of species diversity at multiple temporal and
spatial scales. The Dynamic Habitat Indices (DHIs) are remotely sensed indices that summarize aboveground
vegetation productivity in a way that is ecologically relevant for biodiversity assessments. Existing global DHIs,
derived from MODIS at 1-km resolution, predict species richness at broad scales well, but that resolution is coarse
relative to the grain at which many species perceive their habitat. With the much finer spatial resolution of
Sentinel-2 and Landsat data, plus Landsat’s longer data record, it is possible to track potential changes of
vegetation and its impacts on biodiversity at a finer grain over longer periods. Here, our main goals were to
derive the DHIs from 10-m Sentinel-2, 30-m Landsat, and 250-m MODIS data for the conterminous US and
compare all DHIs at two spatial extents, and to evaluate the ability of these DHIs to predict bird species richness
in 25 National Ecological Observatory Network terrestrial sites. In addition, we derived the Landsat DHIs for
1991–2000 and investigated how they changed by 2011–2020. We found that the Sentinel-2, Landsat, and
MODIS DHIs were highly correlated when summarized by ecoregion (Spearman correlation ranging from 0.89 to
0.99), indicating good agreement between them and that we were able to overcome the lower temporal resolution
of Sentinel-2 and Landsat. Sentinel-2 and Landsat DHIs outperformed MODIS in modeling species richness
for all bird guilds, explaining up to 49% of variance of grassland affiliates in linear regression models.
Furthermore medium-resolution DHIs (10–30 m resolution) captured spatial heterogeneity much better than
MODIS DHIs. We observed considerable changes in Landsat DHIs from 1991–2000 to 2011–2020, such as
increased cumulative DHI along the West Coast, in mountain ranges, and in the South, but lower cumulative DHI
in the Midwest. Our newly derived DHIs for the conterminous US have great potential for use in biodiversity
science and conservation.
File: Razenkova-etal-2025-DHIs-from-Landsat-and-Sentinal2.pdf
Traditional approaches of forest classifications were based on tree species composition, but recently combine phenology
and climate to characterise functional (cyclic and seasonal greenness) rather than structural or compositional
components (phenoclusters). The objective was to compare the conservation value (capacity to support more native
biodiversity) and provision of ecosystem services (ES) in different phenocluster categories of Nothofagus antarctica
forests in Tierra del Fuego (Argentina). We used available models (ES, potential biodiversity) and ground-truth data of
145 stands, comparing phenocluster values using uni- and multivariate analyses. Conservation value and capacity to
supply ES significantly varied among phenocluster categories: (i) cultural, regulating, and provisioning ES and potential
biodiversity at landscape level, (ii) soil carbon and nitrogen, (iii) dominant height, crown cover, basal area, total volume,
and domestic animal stock, and (iv) understory plant richness and cover at stand level. These differences are linked to
the forest capacity to support more native biodiversity and ES. Besides, multivariate analyses supporting the split of
this forest type into four phenocluster subtypes (coast, highland, ecotone with other types, and degraded or secondary
forests). Our findings suggest the needs of specific management and conservation proposals, based on phenoclusters
rather than forest types defined by tree canopy-cover composition.
File: Martinez-Pastur-et-al.-2025_Conservation-value_Nothofagus-antarctica_based-on-phenoclusters.pdf
Land cover change substantially affects ecosystems and leaves long-lasting legacies. Unfortunately, land cover analyses typically begin in the mid-1980s, when 30-m Landsat data became available, missing major global changes that occurred in the 1960s and 1970s. We aimed to quantify long-term land cover changes in the Caucasus (240,000 km2) comparing the magnitude of Soviet-era (1965–1987) versus post-Soviet changes (1987–2015). We (a) mapped land cover based on 1965 Corona spy satellite imagery and (b) quantified long-term changes by comparing 1965 Corona with 1987 and 2015 Landsatbased classifications while accounting for the differences in sensors’ spatial and spectral resolutions. Our Corona-derived map accuracy was 74.4 ± 3.7%, and change accuracies were 66.0 ± 4.2% for 1965–1987 and 61.6 ± 2.8% for 1965–2015. Overall, 30% of the land changed during the Soviet era compared to 20% during the post-Soviet era, highlighting the importance of mapping those early changes. Change trajectories differed considerably during the Soviet era and thereafter. For example, forests were lost during the Soviet era (− 6%) but gained area post-1987 (+ 5%). Croplands were often lost (− 18%) due to grassland gains (+ 11%), which were continuous, but at different rates (4% versus 7%), whereas croplands were lost in both eras, especially post-1987 (3% versus 16%). There were stark differences among countries: Azerbaijan underwent post-Soviet cropland gains, while the Russian Caucasus and Georgia experienced forest gains. Our results highlight the feasibility and value of early spy satellite data for long-term land cover change analyses, particularly in regions with substantial land cover changes then.
File: s10113-025-02360-6.pdf
Avian diversity, a key indicator of ecosystem health, is closely related to canopy structure. Most avian diversity models are based on either optical remote sensing or airborne lidar data, but the latter is limited to small study areas. The launch of the Global Ecosystem Dynamics Investigation (GEDI) instrument in 2018 has opened new avenues for exploring the influence of vegetation structure on avian diversity. To examine how direct measurements of canopy structural characteristics explain bird diversity across North America, we analyzed 18 GEDI metrics from 2019 to 2022, along with corresponding Breeding Bird Survey (BBS) counts and AVONET morphological data, analyzing effects across broad regions and at varying spatial extents. We grouped 440 bird species into 20 ecological guilds under six guild categories and employed random forest algorithms to model avian diversity across eight spatial extents (1, 2, 3, 4, 5, 10, 20, and 39.2 km). The models predicted six diversity indices, including species richness (sRich), functional richness (fRich), evenness (fEve), dispersion (fDis), divergence (fDiv), and redundancy (fRed) across eight spatial extents. The best-predicted guilds varied for each diversity index. The most accurate models were sRich (pseudo-R2 = 0.71, RMSE = 4.28) and fRed (pseudo-R2 = 0.60, RMSE = 0.13) for forest specialists guilds; fRich (pseudo-R2 = 0.55, RMSE = 0.18) for urban guilds; fEve (pseudo-R2 = 0.28, RMSE = 0.08) for insectivore guilds; and fDiv (pseudo-R2 = 0.38, RMSE = 0.12) and fDis (pseudo-R2 = 0.53, RMSE = 0.87) for short distance migrants guilds. Our results highlight the critical role of canopy structure, including its horizontal and vertical distribution and variation, in predicting avian diversity, as measured by the mean number of detected modes (num_detectedmodes), the standard deviation of foliage height diversity (FHD), num_detectedmodes, canopy cover, and plant area index (PAI) across the spatial extents centered on BBS routes. Therefore, we recommend incorporating the GEDI metrics into avian diversity modeling and mapping across North America, thereby potentially enhancing bird habitat management and conservation efforts.
File: Xu-1-s2.0-S0034425724004723-main.pdf
Knowing where and when croplands have been abandoned or otherwise removed from cultivation is fundamental to evaluating future uses of these areas, e.g. as sites for ecological restoration, recultivation, bioenergy production, or other uses. However, large uncertainties remain about the location and time of cropland abandonment and how this process and the availability of associated lands vary spatially and temporally across the United States. Here, we present a nationwide, 30 m resolution map of croplands abandoned throughout the period of 1986–2018 for the conterminous United States (CONUS). We mapped the location and time of abandonment from annual cropland layers we created in Google Earth Engine from 30 m resolution Landsat imagery using an automated classification method and training data from the U.S. Department of Agriculture Cropland Data Layer. Our abandonment map has overall accuracies of 0.91 and 0.65 for the location and time of abandonment, respectively. From 1986 to 2018, 12.3 (±2.87) million hectares (Mha) of croplands were abandoned across CONUS, with areas of greatest change over the Ogallala Aquifer, the southern Mississippi Alluvial Plain, the Atlantic Coast, North Dakota, northern Montana, and eastern Washington state. The average annual nationwide abandoned area across our study period was 0.51 Mha per year. Annual abandonment peaked between 1997 and 1999 at a rate of 0.63 Mha year−1 , followed by a continuous decrease to 0.41 Mha year−1 in 2009–2011. Among the abandoned croplands, 53% (6.5 Mha) changed to grassland and pasture, 18.6% (2.28 Mha) to shrubland and forest, 8.4% (1.03 Mha) to wetlands, and 4.6% (0.56 Mha) to non-vegetated lands. Of the areas that we mapped as abandoned, 19.6% (2.41 Mha) were enrolled in the Conservation Reserve Program as of 2020. Our new map highlights the long-term dynamic nature of agricultural land use and its relation to various competitive pressures and land use policies in the United States.
File: Xie_2024_Environ._Res._Lett._19_044009.pdf
Mapping land cover in highly heterogeneous landscapes is challenging, and classifications have inherent limitations where the spatial resolution of remotely sensed data exceeds the size of small objects. For example, classifications based on 30-m Landsat data do not capture urban or other heterogeneous environments well. This limitation may be overcome by quantifying the subpixel fractions of different land cover types. However, the selection process and transferability of models designed for subpixel land cover mapping across biomes is yet challenging. We asked to what extent (a) locally trained models can be used for sub-pixel land cover fraction estimates in other biomes, and (b) training data from different regions can be combined into spatially generalized models to quantify fractions across global biomes. We applied machine learning regression-based fraction mapping to quantify land cover fractions of 18 regions in five biomes using Landsat data from 2022. We used spectral-temporal metrics to incorporate intra-annual temporal information and compared the performance of local, spatially transferred, and spatially generalized models. Local models performed best when applied to their respective sites (average mean absolute error, MAE, 9–18%), and also well when transferred to other sites within the same biome, but not consistently so for out-of-biome sites. However, spatially generalized models that combined input data from many sites worked very well when analyzing sites in many different biomes, and their MAE values were only slightly higher than those of the respective local models. A weighted training data selection approach, preferring training data with a lower spectral distance to the image data to be predicted, further enhanced the performance of generalized models. Our results suggest that spatially generalized regression-based fraction models can support multi-class sub-pixel fraction estimates based on medium resolution satellite images globally. Such products would have great value for environmental monitoring in heterogeneous environments and where land cover varies along spatial or temporal gradients.
File: Schug-1-s2.0-S0034425724002785-main.pdf
Remote sensing data are important for assessing ecological change, but their value is often restricted by their limited temporal coverage. Major historical events that affected the environment, such as those associated with colonial history, World War II, or the Green Revolution are not captured by modern remote sensing. In the present article, we highlight the potential of globally available black-and-white satellite photographs to expand ecological and conservation assessments back to the 1960s and to illuminate ecological concepts such as shifting baselines, time-lag responses, and legacy effects. This historical satellite photography can be used to monitor ecosystem extent and structure, species’ populations and habitats, and human pressures on the environment. Even though the data were declassified decades ago, their use in ecology and conservation remains limited. But recent advances in image processing and analysis can now unlock this research resource. We encourage the use of this opportunity to address important ecological and conservation questions.
File: biae002.pdf
More frequent and widespread large fires are occurring in the western United States (US), yet reliable methods for predicting these fires, particularly with extended lead times and a high spatial resolution, remain challenging. In this study, we proposed an interpretable and accurate hybrid machine learning (ML) model, that explicitly represented the controls of fuel flammability, fuel availability, and human suppression effects on fires. The model demonstrated notable accuracy with a F1‐score of 0.846 ± 0.012, surpassing processdriven fire danger indices and four commonly used ML models by up to 40% and 9%, respectively. More importantly, the ML model showed remarkably higher interpretability relative to other ML models. Specifically, by demystifying the “black box” of each ML model using the explainable AI techniques, we identified substantial structural differences across ML fire models, even among those with similar accuracy. The relationships between fires and their drivers, identified by our model, were aligned closer with established fire physical principles. The ML structural discrepancy led to diverse fire predictions and our model predictions exhibited greater consistency with actual fire occurrence. With the highly interpretable and accurate model, we revealed the strong compound effects from multiple climate variables related to evaporative demand, energy release component, temperature, and wind speed, on the dynamics of large fires and megafires in the western US. Our findings highlight the importance of assessing the structural integrity of models in addition to their accuracy. They also underscore the critical need to address the rise in compound climate extremes linked to large wildfires.
File: Earth-s-Future-2024-Li-Projecting-Large-Fires-in-the-Western-US-With-an-Interpretable-and-Accurate-Hybrid-Machine.pdf
