The study area covers approximately 50,746 square kilometers of Rio Grande do Norte, a state where 155 of 167 municipalities fall within the semi-arid domain. Rainfall there is low, irregular, and highly variable from year to year, driven mainly by the seasonal migration of the Intertropical Convergence Zone, while the humid eastern coast receives moisture from Atlantic easterly disturbances and trade winds. This climatic gradient, superimposed on a dramatic geological contrast between ancient crystalline basement and younger sedimentary basins, produces a mosaic of hyperxerophilous and hypoxerophilous Caatinga, Atlantic Forest transitions, mangroves, dunes, and salt flats that has long frustrated automated mapping systems.

To capture that complexity, the researchers segmented the state into six analytical subunits by intersecting hydrographic basin subdivisions with geomorphological domains. The Apodi-Mossoró and Piranhas-Açu basins were each split into crystalline and sedimentary portions, and a cluster of smaller eastern basins was likewise divided, yielding AM Crystalline, AM Sedimentary, PA Crystalline, PA Sedimentary, A3 Crystalline, and A3 Sedimentary units. The logic is physical: sedimentary terrains carry deeper, flatter, more fertile soils suited to mechanized and irrigated agriculture, while crystalline landscapes are shallow, stony, and reddened by iron oxides, favoring open Caatinga and extensive livestock. Splitting the territory along these lines reduced spectral confusion between land use classes and bare soil, a persistent problem when a single classifier must handle an entire heterogeneous biome.

The classification engine itself was the random forest algorithm, an ensemble of decision trees trained by bootstrap aggregation, run on the Google Earth Engine cloud platform. Landsat-8 OLI imagery for 2023 was converted to surface reflectance and filtered for clouds and shadows, then enriched with an extraordinary stack of 153 predictor bands: six spectral bands, dozens of vegetation and water indices, tasseled cap brightness, greenness, and wetness transformations, fraction images from a linear spectral mixture model, land surface temperature derived from the thermal band, and estimated precipitation. The rainfall layer was itself a machine learning product, built from 198 rain gauges and modeled against elevation, slope, vegetation indices, temperature, humidity, and the De Martonne aridity index, averaged over 100 regression runs.

Training data came from 11,079 samples, combining manual photointerpretation of false-color Landsat mosaics, 59 field-collected GPS points, and 271 points adapted from MapBiomas. The classes followed the IBGE Land Use Classification System and MapBiomas hierarchy across five categories and thirteen level-II classes, from urban areas and aquaculture through croplands, pasture, forest and savanna vegetation, mangrove, herbaceous restinga, and hypersaline tidal flats, to water bodies and non-vegetated areas. Hyperparameter tuning was exhaustive: the number of trees was tested from 100 to 2,000 in ten repetitions per subunit, with out-of-bag error and overall accuracy guiding the choice, and the entire classification pipeline was run 100 times with different random seeds, consolidated by majority vote into a final map for each fragment.

The results were strikingly consistent. Accuracy metrics ranged from 78 to 88 percent and kappa values from 0.75 to 0.85 across the six subunits, with the crystalline portion of the Apodi-Mossoró basin posting the best performance at 0.88 overall accuracy and 0.85 kappa. Forest vegetation and the rivers, lakes, and ocean class achieved precision above 90 percent, and savanna vegetation exceeded 84 percent in crystalline areas. The trouble spots were revealing: salt marshes and herbaceous restinga in the Piranhas-Açu estuarine complex showed inconsistencies, aquaculture proved hard to separate from open water, and croplands were repeatedly confused with pasture, particularly where cassava and cashew grow on sandy soils with sparse cover, or where corn and sugarcane at the start of their cycles reflect light much like grazed land.

An uncertainty index, the normalized variety metric, mapped where the 100 iterations disagreed most. High uncertainty, above 35 percent, clustered in intensively managed croplands such as Serra do Mel and the irrigated fruit-growing hub of the Piranhas-Açu basin, in the heterogeneous urban fabric of Parnamirim and Macaíba, and in the desertification-prone Seridó microregion, where open savanna-steppe vegetation on shallow crystalline soils blurs into pasture. Dense natural vegetation on mountain slopes and massifs, by contrast, showed almost no disagreement, confirming that the model is most confident exactly where the landscape is most coherent.

Perhaps the most consequential finding lies in the variable importance rankings. Across every geomorphological compartment, the most influential predictors were not spectral indices but spatial position, longitude and latitude, altitude, and climatic conditions, namely precipitation and land surface temperature. This makes ecological sense: latitude governs solar input, altitude shapes local temperature and rainfall, and the amount and timing of rain drive the leafing and flowering cycles of seasonally dry vegetation. Surface temperature, meanwhile, separates irrigated fields and dense canopy from exposed soil and degraded pasture. The forest vegetation class, restricted mainly to rainier western and eastern highlands, was especially dependent on these topographic and climatic cues, while water bodies and urban areas remained best distinguished by spectral signatures.

The mapped landscape tells its own sobering story. Native vegetation, whether preserved, degraded, or recovering, still dominates at roughly 58 percent of the territory, about 3 million hectares, with savanna-steppe covering some 2.88 million hectares and forest formations a mere 37,740 hectares concentrated in the wetter ranges. But more than 41 percent of the biome in Rio Grande do Norte now bears anthropogenic uses: pastures span about 1.77 million hectares, croplands around 135,000 hectares, and a growing footprint of wind farms, solar installations, mining, oil extraction, aquaculture, and urban infrastructure reshapes coastal and upland scenery alike. The contrast between the irrigated fruit export economy of the sedimentary Açu-Mossoró hub and the rainfed subsistence farming of the crystalline Sertaneja Depression mirrors the underlying geology, a dichotomy the classifier captured through the sheer weight of longitude and elevation in its decision trees.

When the team compared their geomorphological approach against the traditional method of classifying whole river basins, overall accuracy was statistically similar, but the spatial errors told a different tale. Basin-wide classification inflated croplands in drainage lines, merged urban areas with other uses, and generalized exposed rock and dunes into broad anthropogenic categories, while the terrain-aware segmentation kept those distinctions intact. The authors caution that spatial coordinates can also let a model merely memorize training patterns rather than generalize, a known pitfall of spatially autocorrelated predictors. Still, their conclusion is clear: for seasonally dry tropical forests facing climate projections of rising heat and declining tree cover, mapping systems that understand relief, water, and position, not just reflected light, offer a genuinely better window on one of the world’s most threatened drylands.

Subject of Research: Machine learning land use and vegetation cover mapping in the Caatinga biome using morphometric and climatic variables

Article Title: Integrating morphometric and climatic variables into the mapping of land use and vegetation cover in the Caatinga biome

Article References: Guedes da Silva, A. D., Flor de Souza, S. F., Lucena, R. L., & Reis, J. S. (2026). Integrating morphometric and climatic variables into the mapping of land use and vegetation cover in the Caatinga biome. Environmental Monitoring and Assessment, 198(10), Article 1132. https://doi.org/10.1007/s10661-026-15949-z

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15949-z

Keywords: Caatinga, random forest, land use land cover, Google Earth Engine, remote sensing, semi-arid, Rio Grande do Norte, geomorphology, precipitation, machine learning, dry tropical forest, MapBiomas

The team, led by Romario Oliveira de Santana of the Chico Mendes Institute for Biodiversity Conservation, working with colleagues at the University of São Paulo and the Federal University of Acre, drew on gross primary productivity (GPP) estimates from the MODIS satellite instrument at 500-meter resolution and soil organic carbon (SOC) stocks from Brazil’s MapBiomas Soil initiative, which estimates carbon in the top 30 centimeters of soil at 30-meter resolution using machine learning trained on field observations. After harmonizing the datasets onto a common grid and classifying the landscape into six land use categories — forest, natural nonforest formation, agriculture, pasture, forestry plantations and mixed-use mosaics — the researchers analyzed annual records from 2001 through 2023.

The productivity side of the ledger behaved as expected. Natural forests recorded the highest average GPP at 2.89 kilograms of carbon per square meter per year, followed closely by commercial forestry plantations at 2.78. Agricultural fields and pastures trailed well behind, at roughly 2.0 kilograms per square meter per year. The Kruskal–Wallis test confirmed these differences were highly significant, and the pattern aligns with global findings that forest conversion reduces landscape-level photosynthetic capacity — one international analysis cited in the study estimated that historical agricultural expansion has cut global GPP by approximately 4.4 percent.

The soil carbon side of the ledger, however, defied intuition. The richest stocks, averaging 88.3 tonnes per hectare with enormous variability, occurred not under towering rainforest but in natural nonforest formations — the grasslands, savannas and open ecosystems that conservation assessments often overlook. Forests held intermediate stocks of 56.0 tonnes per hectare, while agriculture, pasture and plantations carried the lightest carbon loads. High aboveground productivity, in other words, did not translate into high belowground carbon storage.

Statistical modeling drove the point home. Simple regressions between GPP and SOC within each land use class explained very little variation — the strongest relationship, in agricultural areas, accounted for only about 12 percent of the variance, and in natural nonforest formations the relationship was statistically indistinguishable from zero. Multiple regression models incorporating soil clay content and land use explained roughly 38 percent of variation in each variable, underscoring that soil texture and management history, not current photosynthesis, largely govern where soil carbon accumulates.

Perhaps the study’s most methodologically significant twist came when the researchers confronted a problem that plagues landscape-scale ecology: spatial autocorrelation. Both GPP and SOC showed pronounced clustering, with Moran’s I averaging 0.578 for productivity and 0.704 for soil carbon. Conventional regression residuals remained strongly spatially structured, particularly for SOC. Fitting spatial error models year by year collapsed the residual autocorrelation to near zero and slashed the Akaike information criterion by thousands of points — but it also dissolved the GPP–SOC relationship. After properly accounting for spatial dependence, the effect of productivity on soil carbon was significant in only 3 of 23 years. Much of the apparent coupling detected by naive models was shared spatial structure, not ecology.

The temporal analysis added a further layer of divergence. Regional GPP showed no significant directional trend across the 23-year record, fluctuating year to year but remaining functionally stable — although natural forests stayed consistently productive while forestry plantations declined significantly. Estimated SOC, by contrast, crept upward at the regional scale until the end of the 2010s before dipping in recent years, and beneath that aggregate trend lay a patchwork: significant declines in agriculture, plantations and natural nonforest formations, a significant rise in pasture, and near-stability in forests. Productivity and soil carbon, the authors conclude, operate on different clocks — one responding within seasons to light, water and canopy structure, the other accumulating and eroding over decades through decomposition, mineral association and land management.

By classifying every pixel against regional median values for both variables, the study produced a map of four functional regimes that reads like a diagnosis of the corridor’s ecological health. Ecological hotspots — areas combining high productivity and high soil carbon — clustered overwhelmingly in forest ecosystems, which accounted for 66.9 percent of their extent. The low-functionality regime, where both variables fall short, was dominated by pasturelands, which made up 82.2 percent of its area and formed the principal landscape matrix across inland sections. Two intermediate regimes — carbon-dominated and productivity-dominated — filled the gaps, each with its own characteristic land use signature. The association between regimes and land use was, statistically speaking, about as strong as associations get.

The implications ripple outward from southern Bahia. For conservation planners, the message is that satellite greenness alone cannot identify a landscape’s most valuable carbon reservoirs; open natural ecosystems with modest tree cover may be quietly safeguarding enormous belowground stocks and should not be dismissed as degraded land. For restoration strategists, the pasture-dominated low-functionality zones emerge as prime candidates for improved management and ecological recovery, though the authors caution that site-specific assessments must guide priorities. And for anyone banking on tropical forests to offset emissions, the study offers a sobering technical reminder: planting something green does not guarantee that carbon will stay in the ground. Carbon persistence depends on organo-mineral stabilization, aggregate protection and the slow choreography of soil chemistry — processes that no satellite can see, but that this integrated spatial framework now brings into focus.

Subject of Research: Spatial relationships between gross primary productivity and soil organic carbon across land use systems in the Atlantic Forest of Brazil

Article Title: Environmental monitoring of soil carbon and ecosystem productivity across land use systems in the Atlantic Forest of Brazil

Article References: de Santana, R. O., Trindade, L. R. S. L. C., & Delgado, R. C. (2026). Environmental monitoring of soil carbon and ecosystem productivity across land use systems in the Atlantic Forest of Brazil. Environmental Monitoring and Assessment, 198(10), Article 1100. https://doi.org/10.1007/s10661-026-15952-4

Image Credits: AI Generated

DOI: 10.1007/s10661-026-15952-4

Keywords: Atlantic Forest, soil organic carbon, gross primary productivity, land use change, remote sensing, MODIS, MapBiomas, spatial autocorrelation, carbon cycle, tropical ecology, ecological restoration, pasture degradation