Ученые создали карту вторичной растительности Амазонии для контроля восстановления лесов

Newswise — Secondary vegetation develops after primary forest is cleared or degraded and can restore carbon storage, biodiversity, hydrological regulation, and soil functions. However, forest recovery is uneven: patches differ in age, area, interior habitat, shape, isolation, persistence, and exposure to renewed clearing. Existing maps often show where regrowth occurs without indicating whether it is consolidated or likely to survive. This information gap complicates Brazil's commitment to restore 12 million hectares of native vegetation by 2030, where land ownership remains weak or undefined. Because of these challenges, deeper research is needed into the structure, connectivity, persistence, and land-tenure context of Amazonian secondary vegetation. Researchers from the National Institute for Space Research (INPE), the National Institute of Science and Technology in Synthesis of Amazonian Biodiversity (INCT SinBiAm), the Federal University of Rio Grande do Norte (UFRN), and the Federal University of São Carlos (UFSCar) published (DOI: 10.34133/remotesensing.1056) the study on June 10, 2026, in Journal of Remote Sensing. The work addresses a practical restoration challenge: policymakers require more than forest-cover maps to distinguish vulnerable, isolated regrowth from mature patches suitable for long-term protection across the region. The team converted satellite-derived land-cover records into a spatially explicit diagnosis of regeneration across the Amazon biome. Unlike conventional mapping that treats secondary vegetation (SV) as one category, the framework combines patch age, area, core area, boundary complexity, isolation, neighboring vegetation, and land tenure. It identified four profiles: Initial regeneration, Isolated small fragments, Consolidated regeneration, and Largest patch. Initial regeneration contained 56.60% of polygons, while Consolidated regeneration accounted for 38.13%. By distinguishing patches needing active intervention from those more likely to recover through protection, the classification provides a stronger evidence base for restoration planning and resource allocation at regional scale. The 2022 dataset included 689,288 SV polygons covering 16.25 million hectares. Their mean weighted age was 6.96 years, while a Weibull survival model estimated a half-life of approximately 7.14 years with an excellent fit (R² = 0.997). Small fragments dominated: 42.25% measured 2–5 hectares and 39.40% measured 5–20 hectares. Moreover, 81.14% had no core area beyond a 120-meter edge zone, indicating widespread exposure to edge effects. Connectivity was high: 69.80% directly touched primary vegetation (PV), and PV was the nearest vegetation type for 73.85% of patches. Land tenure revealed a governance concern, with 43.92% of SV occurring in areas without land registration. Consolidated regeneration averaged 10.29 years old, whereas isolated small fragments averaged only 7.11 hectares and lay about 535 meters from their nearest neighbors. Protected and traditional territories were more strongly associated with mature, large, interconnected regeneration. The Initial regeneration cluster covered 8.54 million hectares, or 52.57% of total SV area. “This study establishes a spatial baseline to guide policy design, environmental regularization, and restoration planning,” the research team wrote. The authors emphasized that effective protection should identify responsible actors, reconcile conservation with traditional land-use practices, and align restoration actions with the Brazilian National Vegetation Recovery Plan (PLANAVEG) during its implementation. The researchers analyzed TerraClass land-use and land-cover maps from 2008 to 2022, using the 10-meter-resolution 2022 product as the reference. They retained SV polygons of at least two hectares and calculated weighted age, area, core area, fractal dimension index, Euclidean nearest-neighbor distance, neighboring vegetation type, and overlap with nine land-tenure categories. A Weibull model estimated patch persistence, while principal component analysis (PCA) and K-means clustering grouped polygons into regeneration profiles. Nonparametric tests assessed differences among clusters. Statistical robustness was evaluated. Secondary-vegetation mapping polygons are already freely available through TerraBrasilis, while the qualification outputs are available in GeoPackage format and will be formally integrated into the platform’s next release. Agencies and conservation organizations could use the dataset to protect regeneration, reconnect isolated fragments, strengthen governance in unregistered areas, and monitor long-term persistence. With repeated satellite updates, the approach may help assess carbon and biodiversity recovery, improve restoration accountability, and guide tropical regions facing deforestation, fragmented land tenure, and limited conservation resources. ### References DOI Original Source URL https://doi.org/10.34133/remotesensing.1056 Funding information This research was funded by the National Council for Scientific and Technological Development (CNPq) grant number 422354/2023-6 (Monitoramento e avisos de mudanças de cobertura da terra nos Biomas Brasileiros — capacitação e semiautomatização do programa BiomasBR). About Journal of Remote Sensing The Journal of Remote Sensing, an online-only Open Access journal published in association with AIR-CAS, promotes the theory, science, and technology of remote sensing, as well as interdisciplinary research within earth and information science.