How has the advance of agribusiness reshaped land and water appropriation in the Upper Paranapanema region, considering changes in land use, the expansion of irrigation, and the land tenure structure?

Cartographic data on land use and central-pivot irrigation in the Upper Paranapanema Basin across different periods between the 1980s and the 2010s, produced by the Center for Studies in Spatial Ecology and Sustainable Development (Núcleo de Estudos em Ecologia Espacial e Desenvolvimento Sustentável - NEEDS).

Data on changes in the number of family farming establishments in the Upper Paranapanema region between 2006 and 2017, obtained from the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística - IBGE).

Census data from the LU-PA Project (Census Survey of Agricultural Production Units) of the São Paulo State Secretariat of Agriculture and Supply and the Institute of Agricultural Economics.

Analysis of cartographic data to assess changes in land use and the adoption of central-pivot irrigation in the Upper Paranapanema Basin.

Production of maps to represent changes in the number of family farming establishments in the region.

Integration and spatial analysis of Gini indices of land distribution calculated for the 44 municipalities in the region. 

Land use: Between 1987 and 2017, land use changed markedly, with agricultural expansion over areas previously dominated by pasture, which were strongly reduced.

Expansion of irrigation: The area irrigated by central pivots increased significantly between 1985 and 2017.

Family farming: The number of family farming establishments declined in 31 of the 44 municipalities between 2006 and 2017.

Land concentration: Most municipalities showed medium to high levels of land concentration, with an intensification of this process in about 75% of municipalities between 1995 and 2017. 

Spatial scale: Upper Paranapanema Basin, with analysis at the municipal level.

Temporal scale: Long-term change from the 1980s to 2022, capturing the transition from minimal irrigation to widespread adoption.

Do resources, capabilities, and/or perceptions of farmers influence forest cover within private rural properties and through which pathways do those associations occur? 

Quantitative research involving 257 farmers and their properties.

Coordinates of the rural properties evaluated in the research

Data from the CAR (Rural Environmental Registry) of the National Rural Environmental Registry System

Land use and forest cover data derived from geospatial images from the São Paulo Environmental System (DataGEO) and Melo (2024)

To estimate the association between variables, multiple linear regressions were performed, to compare variables within models through the standardized coefficients, and between models by the Adjusted R² and Akaike’s Information Criterion (AIC). 

Small and non-significant effects of testing only positive and negative perceptions of farmers on forest cover, but in combination with farmers’ resources and capabilities, models could explain up to 36% of variation on forest cover. 

The capability “slope”, as a proxy for production suitability, and the resource “farm size”, as a proxy for available capital, were always highly positively associated with forest cover in the models excluding and including farmers’ perceptions.

Spatial scale: Upper Paranapanema Basin, with analysis at the municipal level.

Temporal scale: Long-term change from the 1980s to 2022, capturing the transition from minimal irrigation to widespread adoption.

What are the determining factors in the change in forest area in various land use sectors (agriculture, forestry and livestock) in the Atlantic Forest of the state of São Paulo?

Literature review to identify potential drivers and enabling factors of reforestation at the landscape level.

Data on potential drivers and enabling factors were obtained from municipal-level datasets and statistical data from the Brazilian Institute of Geography and Statistics (IBGE) and the Atlas of Human Development in Brazil.

Land-use data from MapBiomas for the Atlantic Forest between 1990 and 2020, with a spatial resolution of 30 m.

Classification of potential drivers and enabling factors of reforestation at the landscape level, including:

•  Production factors related to human, economic, and natural attributes;
• Socioeconomic conditions;
• Agricultural conditions;
• Forest conditions.

Identification of annual forest dynamics, including land-use transitions to and from forest (deforestation and reforestation, respectively).

Multiple linear regression models to identify which combinations of drivers and enabling factors are statistically associated with cumulative deforestation and reforestation, and to assess whether these drivers and enabling factors differ for reforestation occurring on former croplands (agriculture), pastures (livestock), or planted forests (silviculture).

• Reforestation is more likely to occur in heterogeneous landscapes with mixed agriculture and low-intensity land use (pastures), and less likely in areas dominated by commercial crop production.
• Reforestation on mixed agricultural lands, which accounts for 77% of reforestation events, is strongly linked to the trajectory of forest policy.
• Reforestation on pastures is more common in municipalities with lower gross domestic product (GDP) and higher illiteracy rates and is also associated with legal compliance, being higher in areas with larger deficits of Permanent Preservation Areas (APPs) and Legal Reserves.

Spatial scale: Broad analysis covering 369 municipalities in the Atlantic Forest region of the state of São Paulo. 

Temporal scale: Contemporary habitat-loss patterns based on recent land-cover datasets.

How much forest remains in the Atlantic Forest, and how is this forest spatially distributed?

Forest cover data derived from an Atlantic Forest vegetation map produced by SOS Mata Atlântica and INPE.

The map was generated through visual interpretation of TM/Landsat-5 and CCD/CBERS-2 satellite images from 2005.

The analysis focused on five landscape configuration metrics: fragment size, edge area, connectivity, isolation, and distance to protected areas.

• Remaining forest cover represents approximately 11% to 16% of the original Atlantic Forest.
• More than 80% of forest fragments are smaller than 50 hectares.
• Almost half of the remaining forest is less than 100 m distant from edge.
• The average distance between forest fragments is approximately 1,440 meters.
• Nature reserves protect only 9% of the remaining forest and 1% of the original forest. 

Spatial scale: Brazilian Atlantic Forest

Temporal scale: Year 2005

What is the current state of knowledge on Atlantic Forest biodiversity?

How can the Atlantic Forest be subdivided into biogeographical sub-regions?

What is the relationship between forest distribution and topographic characteristics?

What are the protection, restoration, and ecosystem service potentials for the Atlantic Forest?

Published research from various research groups over recent decades

Biogeographical data on Atlantic Forest sub-regions

Historical and current forest distribution data

Altitudinal and geomorphometric datasets

Bioclimatic information from WorldClim and elevation map

Terrain aspect parameter derived from SRTM data

Literature review and synthesis of biodiversity knowledge

Biogeographical analysis to propose new sub-regional subdivisions

Spatial analysis relating forest distribution to altitude and geomorphometry

Assessment of protection efforts, restoration initiatives, and ecosystem services

Spatial: Entire Atlantic Forest biome (Brazil, extending into Paraguay and Argentina)

Temporal: Historical (original distribution) to present day, with synthesis of research from recent decades

How do landscape amount and configuration affect the richness and abundance of understory bird species in fragmented landscapes? Do these effects vary along a forest cover gradient and according to species’ sensitivity to fragmentation?

Understory bird species richness and abundance data were obtained using mist-net sampling in 53 Atlantic Forest fragments in southeastern Brazil.

SPOT-5 satellite images from 2005 with 10-m resolution and aerial photographs from 2000 with 5-m resolution were used to classify land cover and land use.

Selection of landscapes representing variation in total forest cover, classified as low (10%), intermediate (30%), and high (50%).

Development of statistical models to investigate the effects of fragment area and landscape connectivity on species richness and abundance.

Development of models with different combinations of area and connectivity variables to assess whether richness and abundance responses to fragment area and connectivity varied among landscapes (i.e., across different amounts of forest cover).

Model selection based on the Akaike Information Criterion (AIC).

Species richness was significantly higher in landscapes with 50% forest cover, exceeding that in 10% and 30% forested landscapes by more than 25%.

Highly fragmentation-sensitive species were 3 - 4 times more abundant in the 50% forest cover landscape than in the 10% and 30% landscapes, whereas less sensitive species were richer in landscapes with lower forest cover.

The relative importance of fragment size and connectivity changed along the forest cover gradient.

Fragment size was strongly associated with bird species richness and abundance in landscapes with 10% and 50% forest cover, whereas connectivity played a relatively greater role in the 30% forest cover landscape.

Responses to fragment area and connectivity varied with species’ sensitivity to habitat conversion and fragmentation, indicating the absence of a single fragmentation threshold and highlighting context- and group-dependent responses.

Spatial scale: Atlantic Forest of southeastern Brazil.

Temporal scale: Data were collected between 2001 and 2006.

Do bird species’ responses to habitat changes vary according to their distribution range and local environmental suitability?

Biogeographical and ecomorphological traits of birds obtained from published databases.

Abundance and occurrence data for 81 bird species from the ATLANTIC BIRDS database, based on mist-net captures conducted between 1990 and 2017.

Environmental variables (climate, topography, soil, and vegetation) used for species distribution modeling.

Geographic distribution maps of Atlantic Forest birds, used to estimate the distance to the edges of species’ distribution ranges.

Landscape structure data (forest cover, fragmentation, and connectivity) derived from global maps based on Landsat imagery (30 m resolution).

Historical forest cover data from MapBiomas (1990 – 2015).

Environmental suitability of species was estimated using species distribution modeling.

The distance of populations to the edges of species’ distribution ranges was calculated using geographic distribution maps.

Landscape structure around sampling sites was characterized using metrics of forest cover, fragmentation, and connectivity.

Bird population responses to habitat changes were analyzed using mixed regression models.

Species were classified according to their sensitivity patterns to forest loss, fragmentation, and connectivity across their distribution ranges.

The influence of species traits on these sensitivity patterns was assessed using principal component analysis (PCA) and PERMANOVA

The study identified four population sensitivity patterns to habitat loss, fragmentation, and connectivity across species’ distribution ranges and environmental suitability gradients: no effect of geographical range or environmental suitability (21% of the species), higher sensitivity at the geographical edges or low suitability (14%), lower sensitivity in core or better areas (11%) or both extre mesresponding similarly (11%), with species showing no response to landscape changes (43%).

Sensitivity patterns were not explained by species traits.

Spatial scale: Broad, multi-regional analyses across species’ distribution ranges, including transition zones such as Atlantic Forest–Cerrado interfaces

Temporal scale: Contemporary habitat-loss patterns based on recent land-cover datasets

What is the environmental liability of rural properties in Southwestern São Paulo registered in the Rural Environmental Registry (CAR), considering property size?

What is the area of environmental liability that is not yet registered in the CAR?

What proportion of Permanent Preservation Areas (APPs) must be restored according to property size?

Vector cartographic data on the boundaries of rural properties obtained from the SICAR (National Rural Environmental Registry) online portal.

Maps of hydrography, land use, and Permanent Preservation Areas (APPs) obtained from the Brazilian Foundation for Sustainable Development, with 2013 as the reference year.

Classification of rural properties according to property size, based on the number of fiscal modules (MF).

Use of cartographic data to identify consolidated Permanent Preservation Areas (APPs) and to calculate the total areas of environmental liability and APPs.

Exclusive consideration of APPs associated with the margins of water bodies, based on property size classification and legal requirements for the restoration of environmental liabilities in consolidated areas.

Creation of buffers around water bodies and springs, with widths defined according to property size, resulting in four buffer categories. These buffers were intersected with the boundaries of the corresponding properties and with land-use maps to identify areas of environmental liability and areas with native vegetation in environmental compliance.

Development of three analytical scenarios:

Scenario 1: Assessment of 100% of areas with declared CAR and identification of the total area to be restored.

Scenario 2: Combination of Scenario 1 with areas without declared CAR, assuming that all such areas correspond to properties smaller than one fiscal module, with a minimum APP restoration requirement (5 m).

Scenario 3: Combination of Scenario 1 with areas without declared CAR, assuming that all such areas correspond to properties larger than 10 fiscal modules, with a maximum APP restoration requirement (30 m).

Large properties account for 62% of the total analyzed area registered in the CAR, while medium-sized properties represent 15%, small properties 17%, and micro-properties only 6% of the total area.

Southwestern São Paulo contains approximately 155,065 hectares of Permanent Preservation Areas (APPs), of which about 47% are degraded. The results of the modeled scenarios vary depending on the assumptions regarding the size of properties without CAR registration; however, in all scenarios, more than 40% of APP areas require restoration.

Large properties hold the largest share of APP-related environmental liabilities. However, although small properties (less than two fiscal modules) have a smaller absolute area of APP to be restored, they show a proportionally higher amount of degraded areas.

Spatial scale: Rural properties registered in the Rural Environmental Registry (CAR) in 15 municipalities of Southwestern São Paulo.

Temporal scale: Current scenario based on 2020 data.

The article aims to document the rediscovery of Phrynomedusa appendiculata from a breeding population in the Atlantic Plateau forests of the state of São Paulo.

Records obtained during an amphibian survey conducted in different habitats within a protected Atlantic Forest area in southern São Paulo state.

Morphological, acoustic, and molecular data based on a specimen collected in the field.

Morphological variation was assessed using standard morphometric measurements and compared with published data and museum specimens.

Phylogenetic relationships were inferred using mitochondrial and nuclear DNA sequences generated for the species and combined with sequences from public databases (GenBank).

Most new data were consistent with the variation previously reported for the species, with subtle differences interpreted as intraspecific variation.

This record allowed an update of the species’ geographic distribution and provided the first inference of its phylogenetic relationships based on molecular data.

The resulting phylogeny corroborated the generic placement and evolutionary distinctiveness of Phrynomedusa appendiculata, identifying it as sister to the clade P. marginata + P. dryade.

How is climate change expected to alter rainfall patterns (amount, seasonality, and variability) in the Alto Paranapanema Basin, and what are the implications for hydrological conditions in the basin?

Historical rainfall records from meteorological stations within and around the Alto Paranapanema Basin

Downscaled climate model outputs representing future climate scenarios

Basin-scale spatial data used to aggregate and analyse rainfall patterns across the catchment

Statistical analysis of historical rainfall time series to establish baseline seasonal patterns

Application of climate change projections to assess future rainfall behaviour

Comparison of historical and projected periods to evaluate:

• changes in wet-season rainfall totals
• shifts in rainfall seasonality
• changes in interannual variability

Basin-wide aggregation of results to characterise system-level rainfall responses

Wet-season rainfall shows a projected reduction of up to ~40% relative to historical averages under climate change scenarios.

Rainfall patterns become more irregular, with increased variability between years.

The reduction is seasonally concentrated, affecting periods critical for water recharge, agriculture, and ecosystem functioning.

Spatial scale: Alto Paranapanema Basin (UGRHI-14), São Paulo State

Temporal scale: 

• Historical baseline: multi-decadal rainfall records (20th–early 21st century)
• Future projections: mid- to late-21st century climate scenarios (scenario-based projections rather than single deterministic forecasts)

How have changes in land use and land cover affected sediment production and sediment export in the Upper Paranapanema Basin over recent decades?

What is the relationship between agricultural expansion, erosion susceptibility, and sediment dynamics in the basin?

To what extent can riparian restoration and soil conservation practices reduce sediment production and export under different management scenarios?

Historical land use and land cover maps for the Upper Paranapanema Basin for 1987 and 2017

Spatial data on topography, soils, and erosion susceptibility

Hydrological and sediment modelling inputs representing basin conditions under different management scenarios

Statistical analysis of historical rainfall time series to establish baseline seasonal patterns

Application of climate change projections to assess future rainfall behaviour

Comparison of historical and projected periods to evaluate:

• changes in wet-season rainfall totals
• shifts in rainfall seasonality
• changes in interannual variability

Basin-wide aggregation of results to characterise system-level rainfall responses

Sediment production increased from 38.86 t ha⁻¹ year⁻¹ in 1987 to 91.80 t ha⁻¹ year⁻¹ in 2017, an increase of 136.2 percent

Sediment export increased from approximately 1.37 t ha⁻¹ year⁻¹ in 1987 to 4.1 t ha⁻¹ year⁻¹ in 2017

Restoration of PPAs alone reduced average sediment production to 42.69 t ha⁻¹ year⁻¹, a reduction of more than 50 percent

Conservation practices alone reduced sediment production to 11.42 t ha⁻¹ year⁻¹, a reduction of 87.6 percent compared to current conditions

Combined PPA restoration and conservation practices reduced sediment production to 10.81 t/ha/year

Sediment export under PPA restoration decreased to 1.66 t/ha/year, a reduction of approximately 59.5 percent

Combined PPA restoration and conservation practices reduced sediment export to 0.378 t/ha/year, corresponding to a reduction of approximately 90.8 percent

Spatial scale: Upper Paranapanema River Basin

Temporal scale: Land use and sediment dynamics assessed for the period between 1987 and 2017, with scenario simulations representing alternative management futures