Deforestation Patterns Research Articles

Effectiveness of protected areas in reducing deforestation and forest fragmentation in Bangladesh

The establishment of protected areas has been considered a tool to battle deforestation in densely populated countries like Bangladesh. Since 2000, Bangladesh's Forest Department has declared about 41 protected areas. However, before more land is designated as a protected area, it is important to know how effective existing protected areas are in achieving conservation goals. Unfortunately, the determination of the conservation effectiveness of protected areas, such as their capacity to reduce deforestation and forest fragmentation over a considerable period at high temporal frequency (e.g., yearly), is still unavailable despite some known methods being available, for example, the System for the Integrated Assessment of Protected Areas. In this study, we processed and analyzed the Hansen dataset from 2000 to 2018 to produce yearly forest/non-forest maps of four protected areas in Bangladesh and used these maps, with a matching method, to estimate the effectiveness of protected area in reducing deforestation after controlling for potential hidden bias. We also analyzed the forest fragmentation scenario over the same time frame. The forest cover change results from 2010 to 2018 revealed a large-scale deforestation pattern in areas adjacent to the protected area boundary of Chunati Wildlife Sanctuary (CWS) and Baroiyadhala National Park–Hazarikhil Wildlife Sanctuary (together B-HWS). Using a propensity score matching (PSM) approach with a caliper of 0.25, we found that B-HWS was the best performing of the studied protected areas, and that 37% of forest pixels in B-HWS would have been deforested in 2018 if they had not been brought under protection in 2010. Similarly, the estimated avoided deforestation rates were approximately 21% and 4% for CWS and Dudpukuria–Dhopachari Wildlife Sanctuary, respectively. Despite an improvement in deforestation scenarios, during the period 2010–2018, for all protected areas, the forest fragmentation scenarios were exacerbated both inside their boundaries and in adjacent unprotected areas. Therefore, it remains questionable whether protected areas can ultimately maintain the integrity of conservation.

Forest Loss and Susceptible Area Prediction at Sefwi Wiawso District (SWD), Ghana

Forests provide immeasurable merits for the economies of most developing countries. Forests in developing countries experience harmful human-induced impacts such as unregulated removal of biodiversity and unsustainable land conversion. The Sefwi Wiawso District (SWD) located in Ghana, which includes portions of six protected forest reserves (FRs) such as Muro, Tano Suhien, Tano Suraw, Suhuma, Sui River, and Krokosua, is the subject of this study. The impacts of selected spatial variables on forest losses were examined using retrospective and predictive approaches. Past deforestation patterns were analyzed using classified Landsat 5 and 7 imagery from 1984 to 2017. Pixel areas in hectares (ha) from land use land cover (LULC) classifications were used to detect land cover classes that were vulnerable to potential loss. The study also carried out a simple forest prediction using the simple moving averages (SMA) forecasting model based on the past and present deforestation patterns from LULC classification. The results showed that 3587.49 hectares (ha) of protected forest cover was converted into agricultural lands and barelands. In addition, 2532.96 hectares (ha) was converted from close forest to nonforest land cover from 2000 to 2017, which is equivalent to a 16% reduction in close forest cover within the FRs in the SWD. This loss was also 11% higher than close forest areas between 2000 and 2010. SMA forecasting showed that from 2017 to 2024, 877.38 hectares (ha) of close forest resources will convert to open forest resources and other nonforest land cover. Subtle accessibility routes such as navigable rivers and unofficial roads are the key instigators of protected forest clearance in the Sefwi Wiawso Forest District (SWFD). The SWFD is surrounded by many communities and is susceptible to uncontrollable biodiversity removal due to lack of proper monitoring of agricultural practices, mining operations, fuelwood collection, and illegal hunting, which represents a means of livelihood for the forest fringe community dwellers. The research serves as a benchmark for similar studies in efforts to investigate, measure, and project land cover change in protected forest areas.

Semantic Unsupervised Change Detection of Natural Land Cover With Multitemporal Object-Based Analysis on SAR Images

Change detection is one of the most addressed topics in the remote sensing community. When performed on synthetic aperture radar images, the most critical issues are as follows: 1) the labeling of the identified changing patterns and 2) the scarce robustness of classic pixel-based approaches based on threshold segmentation of an appropriate change index, which tend to fail when multiple changes are present in the study area. In this work, a new methodology for unsupervised change detection in vegetation canopy is presented. It overcomes these limitations by exploiting multitemporal geographical object-based image analysis with the aim to make the intrinsic semantic of data emerge and direct the processing toward the identification of precise classes of changes through dictionary-based preclassification and fuzzy combination of class-specific information layers. The proposed methodology has been tested in ten different experiments covering agriculture and clear-cut deforestation applications. The results, validated against literature methods, highlighted the superiority of the proposed approach, which was quantitatively assessed in terms of standard classification quality parameters. On agriculture experiments, it allowed for an average increase in the detection accuracy of about 11% with respect to the best performing literature method, with an increment of the false alarm rate in the order of 0.5%. In case of deforestation, the registered detection accuracy was comparable to that achieved by the literature, while the most significant benefit was the reduction, of more than one-third, of the number of detected false deforestation patterns. Overall, the main characteristics of the proposed architecture are the robustness and the lack of any supervision, which makes it very well-suited for operational scenarios.

Multitemporal Analysis of Deforestation in Response to the Construction of the Tucuruí Dam

The expansion of hydroelectric dams that is planned, and under construction, in the Amazon basin is a proposal to generate “clean” energy, with the purposes of meeting the regional energy demand, and the insertion of Brazil into the international economic market. However, this type of megaproject can change the dynamics of natural ecosystems. In the present article, the spatiotemporal patterns of deforestation according to distance from the reservoir in the vicinity of the lake of Tucuruí, and within a radius of 30 km from it, are analyzed. A linear spectral mixture model of segmented Landsat-thematic mapper (TM), enhanced thematic mapper plus (ETM+), and operational land imager (OLI) images, and proximity analysis were used for the mapping of the land-cover classes in the vicinity of the artificial lake of Tucuruí. Likewise, landscape metrics were determined with the purpose of quantifying the reduction of primary forest, as a mechanism of loss of ecosystem services in the region. These methods were also used for the evaluation of the influence of the distance from the reservoir on the expansion of anthropogenic activities. This methodology was used for the scenarios of pre-inauguration, completion of phase I, beginning of construction phase II, full completion of the Tucuruí hydroelectric project, and the current scenario of the region. The results showed that the highest deforestation rate occurred in the first period of the analysis, due to the areas submerged by the reservoir and due to the anthropogenic disturbances, such as timber extraction, road construction, and the conversion of forests into large areas of agribusiness.

Brazil’s conservation reform and the reduction of deforestation in Amazonia

The Action Plan for the Prevention and Control of Deforestation in the Legal Amazon (PPCDAm), launched in 2004, promoted a drastic conservation reform in Brazil. To date, the plan has been through four distinct phases: PPCDAm-I (2004–2008), II (2009–2011), III (2012–2015), and IV (2016–2020). A substantial decline from historically high deforestation rates in the Amazon occurred from 2004 to 2012, part of which (particularly after 2007) can be attributed to the PPCDAm, and part to economic factors such as commodity prices and the currency exchange rates that affect the profitability of agricultural exports. After 2012 deforestation has trended upward despite control efforts. The PPCDAm evolved through its four phases, consequently changing the patterns of deforestation in Amazonia. Here, we review the conservation instruments and strategies promoted under the PPCDAm umbrella and, when possible, provide estimates of their impacts on deforestation based on rigorous, counterfactual evaluations. While the literature emphasizes some of the most well-known PPCDAm actions, e.g., improved satellite-based monitoring, the “blacklist” of municipalities, expansion of protected areas, land tenure regularization, credit access restrictions, and moratoria, we highlight a number of other PPCDAm activities that, along with the previous examples, are arguably also responsible for a significant share of the reduction of deforestation in the region.

Linking deforestation patterns to soil types: A multifractal approach

AbstractIn the three last decades, continental Ecuador biosphere reserves (CEBRs) have undergone important land cover transformations because of human‐managed systems such as agriculture, livestock, forestry, and urbanization. Thus, the native landscape structure has been altered, exhibiting a mosaic of patches with varying sizes and shapes. The resulting landscape could shape some patterns over time, reflecting the complex interactions between human activities and biotic‐abiotic factors, such as soil, climate, and water availability. Quantifying the landscape patterns and their dynamics is essential for the monitoring and assessment of the ecological consequences of land‐cover change in time. Based on the soil maps and raster land‐cover maps from 1990 to 2016 of two CEBRs, the spatial deforestation patterns were quantified through lacunarity and multifractal analysis. To conduct these analyses, the deforested areas were identified and overlapped to predominant soil types. The results showed that the deforestation process pattern and soil types interact through complex relations due to specific site characteristics. Thereby, both soil and land cover interacted in fractal structures. The patch area distribution of deforestation patterns linked to soil types reveals a power‐law relationship in which, as the area of the deforested patches increased, their frequency decreased. Also, this relation presents self‐similarity and multifractal behaviour over a wide range of scales from 46–2,674 pixels, with a pixel size of 30 m × 30 m. However, each CEBR showed varying degrees of heterogeneity and complexity, which were reflected through lacunarity and multifractal parameters. The dynamics of deforestation patterns linked to soil types can be spatially dependent on the initial patchiness state, the distance to the adjacent patches, and the suitability of the soil for agrarian activities. Altogether, these were the spatial underlying factors of the deforestation evolution in the two CEBRs.Highlights Deforestation patterns show spatial self‐similarity Deforestation patterns are linked to soil types Landscape ecology, lacunarity, and multifractal analyses are useful to understand deforestation patterns Initial patch location, distance to the nearest neighbour and soil suitability boost the deforestation expansion

Can de facto governance influence deforestation drivers in the Zambian Miombo?

Weak forest governance is posited as a key underlying driver of deforestation and forest degradation, but empirical evidence of this linkage is scarce. Many related studies capture the de jure (legal) conditions and miss out the de facto (implementation practices on the ground), particularly when considering the proximate drivers and other factors of deforestation. However, this is central for identifying the specifics of governance for curbing deforestation and forest degradation. We analyse the influence of de facto governance quality on deforestation, accounting for proximate drivers and other factors using stepwise regression. We further compare deforestation rates and drivers across different governance arrangements with differing institutions, tenure and forest access restrictions using Wilcoxon tests to derive conclusions for promising policy instruments that address deforestation. Data for the analysis were obtained through participatory mapping, focus group discussions and geographical information systems. To generate empirical evidence, 238,296 ha of land were mapped within 24 communities spanning three provinces, Copperbelt, North-Western and Eastern, in the Zambian Miombo. Regression results revealed that de facto governance quality has some effect but proximate drivers particularly charcoal production, crop agriculture and proximity to roads explain most of the deforestation patterns in the Zambian Miombo. Those drivers seem hardly affected by the weak governance processes. Since scores of governance quality were in general low and hardly varying, we conclude that in our case they were too weak to show effects on the proximate drivers. Only the governance indicator ‘local government capacity and effectiveness’ although still weak, was significantly linked to low deforestation rates. Comparative results further showed that restricted arrangements (state and traditionally restricted) exhibit lower deforestation than non-restricted arrangements (communal, forests with overlapping community claims, private and individual customary forests). But while crop agriculture was negligible, forest resource extraction was still substantial in restricted state forests, indicating a higher possibility for forest degradation instead. Although private and individual customary forests had higher tenure security, they showed higher deforestation rates than communal and state arrangements. This challenges the notion that tenure security alone guarantees successful forest conservation. Our results suggest that governance can only affect deforestation drivers positively above certain thresholds. This needs to be further complemented by specific measures such as sustainable production systems, incentives and alternative livelihoods to regulate the proximate and other underlying drivers of deforestation.

Rapidly Accelerating Deforestation in Cambodia’s Mekong River Basin: A Comparative Analysis of Spatial Patterns and Drivers

The Mekong River is a globally important river system, known for its unique flood pulse hydrology, ecological productivity, and biodiversity. Flooded forests provide critical terrestrial nutrient inputs and habitat to support aquatic species. However, the Mekong River is under threat from anthropogenic stressors, including deforestation from land cultivation and urbanization, and dam construction that inundates forests and encourages road development. This study investigated spatio-temporal patterns of deforestation in Cambodia and portions of neighboring Laos and Vietnam that form the Srepok–Sesan–Sekong watershed. A random forest model predicted tree cover change over a 25-year period (1993–2017) using the Landsat satellite archive. Then, a statistical predictive deforestation model was developed using annual-resolution predictors such as land-cover change, hydropower development, forest fragmentation, and socio-economic, topo-edaphic and climatic predictors. The results show that almost 19% of primary forest (nearly 24,000 km2) was lost, with more deforestation in floodplain (31%) than upland (18%) areas. Our results corroborate studies showing extremely high rates of deforestation in Cambodia. Given the rapidly accelerating deforestation rates, even in protected areas and community forests, influenced by a growing population and economy and extreme poverty, our study highlights landscape features indicating an increased risk of future deforestation, supporting a spatial framework for future conservation and mitigation efforts.

Detection of deforestation and analysis land conversion from high resolution satellite imageries in Bintulu District, Serawak, Malaysia

Primary forests are continuously threatened day to day by urbanisation and the conversion to agricultural plantation such as palm oil production and other land uses being one of the major reasons. In South-East Asia, the widespread of oil palm has boomed over the last two decades, resulting in the downfall of tropical forest land. This change has been particularly prevalent in Borneo with protected lands increasingly developed for palm oil and already deforested lands are being converted into industrial plantations. The primary concerns relating to this pattern of land use change are the short and long-term impacts of logging on our natural environments and ecosystems and how patterns of deforestation are contributing to global environment issues such as climate change. By detecting and mapping logging activities, forestry departments are better able to predict land cover changes in a particular region as a result of development and deforestation. In this study, land cover assessment based on remote sensing techniques was used to analyse changes in the Bintulu district, Borneo especially oil palm growth and its influence on the decline of forest areas between 2016 and 2018. High resolution satellite imageries (3 m spatial resolution) from PlanetScope were used due to the benefits in helping to differentiate several land cover classes over a higher spatial resolution. Results showed that a decline of primary forests in Bintulu is about 26.5% over the past 2 years. With that decline, comes a rise of oil palm growth by 17.6% just within 2 years. An increased in logged areas (36.1%) for conversion to other land-covers with a steady decline of other land cover classes by at least 20% each year was detected. The accuracy of results proved a reasonable accuracy with 90.0% confidence with satellite imagery. Thus, with the result a necessary step can be considered to monitor and prevent deforestation and various encroachment in the forest. With high-resolution satellite data, monitoring at local scales has become possible to resource managers as a way to create timely and reliable assessments.

Back to the people: The role of community-based responses in shaping landscape trajectories in Oaxaca, Mexico

Land use change results from top-down drivers, such as policies, trade, and migration. Land use change may also result from community-based responses. In Mexico, rural communities govern most of the country's forests. This study aimed to assess how socio-economic and biophysical factors affected the landscape trajectories of rural communities in southern Mexico. It also aimed at evaluating the role of communities in landscape change. Land use change of 63 rural communities was analyzed for the years 1987 and 2017. Four land uses were distinguished: forest, shrubland, agriculture, and bare soil. Five groups of communities were identified according to their socio-economic and biophysical factors. Two groups located in areas with high slopes and elevated marginalization index values showed deforestation patterns. Two other groups, consisting of more than half of the municipalities assessed, showed reforestation trends. The final group did not reveal major changes in land use. Two municipalities with reforestation trends were selected for an in-depth analysis of how community-based responses impacted natural resource management and conservation. Through local assemblies, the population voted for regulations that increased the forest area and reduced the bare soil. There was no evidence that these regulations affected croplands. These results show how a combination of socio-economic and biophysical factors can affect landscape change, but it also shows the often overlooked role of communities as a relevant bottom-up driver of change.

Spatiotemporal analysis of deforestation in the Chapare region of Bolivia using LANDSAT images

AbstractThe purpose of this study is to quantify (a) spatiotemporal deforestation patterns, and (b) the relationships between changes in the main land‐cover types in the Chapare region of Bolivia. We applied subspace classification methods to LANDSAT data from 1986, 1999, and 2018 and used grid cells at scales of 150, 300, 600, and 900 m to measure deforestation trajectories. The 150 m grids provided better detail to determine deforestation trajectories than coarser‐scale grid cells. Differences in grid‐cell scale did not influence the statistical trends in land‐cover changes significantly. Changes in forest area were negatively correlated with changes in cropland (r = −.44), grassland (r = −.34), swamp grassland (r = −.38), and regrowth (r = −.32) areas. Correlations between forest losses in cropland, grassland, and regrowth change analyses were weaker between 1999 and 2018 compared to 1986 to 1999. Forest cover declined from 6,635 km2 (1986) to 3,800 km2 (2018), and the deforestation rate increased from an annual average of 1.36% between 1986 and 1999 to 2.0% between 1999 and 2018. The key proximate drivers of forest clearance rates and patterns were increasing population, agricultural expansion, and road building. While coca is an economically important crop in Chapare, its direct and indirect effects on deforestation could not be determined unambiguously. It is probable that the expansion of agriculture will lead to further deforestation and forest fragmentation and, along with decreases in forest cover, further changes will take place between non‐forest categories.

Seeing the forest through many trees: Multi‐taxon patterns of phylogenetic diversity in the Atlantic Forest hotspot

AbstractAimWe combine phylogenetic and point locality data from selected lineages of the Atlantic Forest flora and fauna to compare spatial patterns of biodiversity sustained by the current configuration of forest remnants to a scenario of complete forest preservation. We then ask the question "how much biodiversity is likely lost, already"? Specifically, we assess how habitat loss likely impacted the climatic spaces occupied by the local species, the inferred composition of local communities and the spatial distribution of phylogenetic diversity and endemism.LocationAtlantic Forest, Brazil.MethodsUsing carefully curated point localities, phylogenetic data and parameterized models of species distributions, we generate maps of phylogenetic diversity, phylogenetic endemism and phylogenetic turnover for the entire Atlantic Forest. We map patterns of clade‐specific diversity under complete preservation of forest and then incorporate present‐day deforestation patterns to provide a more realistic scenario.ResultsInstead of a singular pattern, three different reoccurring syndromes described the flora and fauna of the Atlantic Forest. These patterns emerged irrespectively of clade age and life history traits. General turnover patterns were highly consistent with previous analyses of species composition and panbiogeographical studies. Deforestation has altered the availability of climatic spaces in the Atlantic Forest, its biological communities and the distribution of evolutionary lineages in space. However, approximately 60% of the pre‐Columbian climatic space persists in forest remnants, and today's biological communities are estimated to be 45% similar to pre‐deforestation times.Main conclusionsThe Atlantic Forest has been reduced to 8% of its once largely continuous range. However, the disproportionately large amounts of climate, community and lineage diversity that persist in remnants provide hope and support for conservation efforts that combine species occurrence and phylogenetic data. Inclusion of evolutionary thinking into strategic approaches to restoring Brazilian ecosystems could further conservation effectiveness by incorporating the adaptive potential of local assemblages in the face of further environmental shifts.

The impacts of cocaine-trafficking on conservation governance in Central America

This research is motivated by the compelling finding that the illicit cocaine trade is responsible for extensive patterns of deforestation in Central America. This pattern is most pronounced in the region's large protected areas. We wanted to know how cocaine trafficking affects conservation governance in Central America's protected areas, and whether deforestation is a result of impacts on governance. To answer this question, we interviewed conservation stakeholders from key institutions at various levels in three drug-trafficking hotspots: Peten, Guatemala, Northeastern Honduras, and the Osa Peninsula in Costa Rica. We found that, in order to establish and maintain drug transit operations, drug-trafficking organizations compete with and undermine conservation governance actors and institutions. Drug trafficking impacts conservation governance in three ways: 1) it undermines long standing conservation coalitions; 2) it fuels booms in extractive activities inside protected lands; and 3) it erodes the territorial control that conservation institutions exert, exploiting strict “fortress” conservation governance models. Participatory governance models that provide locals with strong expectations of land tenure and/or institutional support for local decision-making may offer resistance to the impacts on governance institutions that we documented.

Long-term changes in forest cover in a global biodiversity hotspot in southern Mozambique

Background: Deforestation is a complex and dynamic process of widespread concern in sub-Saharan Africa which is influenced by a range of social, economic and biophysical factors. Objectives: The aim of this study was to analyse patterns of deforestation and its potential drivers in the Licuáti Forest Reserve, a biodiversity hotspot in southern Mozambique between 1990 and 2016. Method: We performed image classification on Landsat imagery at six time steps and interviewed local community members to understand the spatial pattern and rate of forest cover loss over time. We also examined changes in the incidence of fire. Results: A substantial increase in the rate of deforestation since 1990 was detected in this vulnerable thicket vegetation. The probability of deforestation was significantly higher near the major roads, where houses are located. This suggests that the proximity of human settlements to the forest, and access to charcoal markets in urban areas, influenced the spatial pattern. Two key factors 1) charcoal production, and 2) the establishment of settlements and agricultural lands were identified as proximate causes of deforestation. In addition, fires associated with these two causes might amplify the loss of forests in the area. Conclusions: Complex interactions between the drivers of deforestation and socio-economic factors were suggested, as most of the charcoal produced in the region is transported to Maputo. Ongoing road improvements and infrastructural development in the region will likely accelerate the decline in forest cover in the future. This has implications for the biodiversity of the region as well as for the sustainability of local livelihoods, as they often depend on forest products for their daily uses.

Deforestation dynamics in Brazil's Amazonian settlements: Effects of land-tenure concentration

Brazil's Amazon deforestation is a major global and national environmental concern, and the ability to model and project both its course and the effect of different policy options depends on understanding how this process occurs at present and how it might change in the future. The present paper addresses one key factor in Amazon deforestation: land-tenure concentration in settlements. Brazil's policies for establishing and regulating settlement projects represent critical government decisions shaping the landscape in the 5 × 106 km2 Legal Amazonia region. We used remote-sensing data and information provided by the National Institute for Colonization and Agrarian Reform (INCRA) to evaluate the effect of land-tenure concentration in a settlement project (Projeto de Assentamento) located in a frontier area where cattle-ranching is expanding. We identified the actors and their deforestation patterns in the Matupi settlement in the southern part of Brazil's state of Amazonas. We spatially identified actors who concentrated “lots” (the parcels of land distributed to individual settlers) in 2011 and assessed whether the concentration was done by individual landholders or by “families” (where members merged their lots and the clearing was done together). Deforestation rates (1995–2011) were estimated for each type of actor and the trajectory of deforestation in the settlement (cumulative deforestation to 1994 and annual deforestation 1995–2016) was also analyzed. Concentrators occupied 28% (9653 ha) of the settlement and 29% of the lots (152 lots) analyzed; the numbers of lots concentrated ranged from two to ten. Concentrators of two lots and non-concentrators were the predominant actor types in the settlement. The mean annual clearing per landholding for concentrators of two lots (families: 4.1 ± 2.8 ha (mean ± SD); individuals: 5.1 ± 4.6 ha) was greater than for non-concentrators (1.7 ± 1.2 ha), despite their having similar patterns of small clearings. Concentrators of three or more lots had mean annual clearing per landholding between 6.2 ± 12.2 ha and 23.9 ± 38.7 ha and, the pattern of patches cleared per year >34 ha in area was predominant. The deforestation rate per lot was higher among concentrators as compared to non-concentrators, showing that lot concentration speeds deforestation. Analysis of deforestation patterns helps to better understand the process of lot concentration by spatially identifying the predominant patterns of each type of actor. The approach used in our study could assist authorities in identifying and monitoring land-tenure concentration in settlements. Agrarian-reform policymakers need to monitor this process, since it speeds deforestation in Amazonian settlement projects, as well as undermining the social objectives of the agrarian-reform program.

Urban and rural household energy consumption and deforestation patterns in Zaire province, Northern Angola: A landscape approach

Angola has been labelled a “petro-state”, since independence, due to its oil-wealth and the country's economic dependence on its revenues. Considering that oil availability and price could be a factor reducing deforestation and forest degradation through accelerating energy-transition, our objective was to study the energy consumption patterns and the role of woody biomass as a source of income of rural villagers and urban poor dwellers practicing agriculture, living in one of the oil-rich provinces. We used a “people and pixels” approach, combining interviews with households' heads with remote sensing analysis of land use/cover change. Our findings revealed that while kerosene is the main energy source for cooking and lighting in the city, in rural areas it is firewood and flashlights, respectively, and that the selling of firewood and/or charcoal is practiced by few households. Contrary to what has been reported about other Angolan regions and African countries, satellite imagery analysis showed that deforestation around Zaire province capital city and surrounding rural villages is comparatively small. Bearing in mind that the consumption of biomass for domestic household needs is reduced, other drivers of deforestation/forest degradation must be considered. Nonetheless, the high domestic use of kerosene contributes to the exhaustion of its limited global supplies and to increasing greenhouse gas emissions, disregarding the country's potential to produce renewable energy.

Evaluation of Deep Learning Techniques for Deforestation Detection in the Brazilian Amazon and Cerrado Biomes From Remote Sensing Imagery

Deforestation is one of the major threats to natural ecosystems. This process has a substantial contribution to climate change and biodiversity reduction. Therefore, the monitoring and early detection of deforestation is an essential process for preservation. Techniques based on satellite images are among the most attractive options for this application. However, many approaches involve some human intervention or are dependent on a manually selected threshold to identify regions that suffer deforestation. Motivated by this scenario, the present work evaluates Deep Learning-based strategies for automatic deforestation detection, namely, Early Fusion (EF), Siamese Network (SN), and Convolutional Support Vector Machine (CSVM) as well as Support Vector Machine (SVM), used as the baseline. The target areas are two regions with different deforestation patterns: the Amazon and Cerrado biomes in Brazil. The experiments used two co-registered Landsat 8 images acquired at different dates. The strategies based on Deep Learning achieved the best performance in our analysis in comparison with the baseline, with SN and EF superior to CSVM and SVM. In the same way, a reduction of the salt-and-pepper effect in the generated probabilistic change maps was noticed as the number of training samples increased. Finally, the work assesses how the methods can reduce the time invested in the visual inspection of deforested areas.

Identification of crucial stepping stone habitats for biodiversity conservation in northeastern Madagascar using remote sensing and comparative predictive modeling

Madagascar is a global biodiversity hotspot of conservation concern. The decline of natural forest habitats due to shifting cultivation has been one of the major land use changes during the last decades. We analyzed satellite images between 1990 and 2018 from northeastern Madagascar to evaluate the contribution of nine variables (e.g., topographic, demographic, forest protection) to explain past forest loss, predict future deforestation probabilities to define important areas that require further conservation attention. Forest cover declined by 21% since 1990 and the once continuous rain forest belt of the region is disrupted twice, in the center and at the southern limit of the study region. Status of forest protection and proximity to the forest edge were identified as most important predictors, but all variables contributed to explaining the observed pattern of deforestation. At least 20% of the 3136 villages in the area were established since 1990 at the expense of previously forested areas. This housing sprawl was mainly driven by accessibility, decreasing landscape connectivity. To conserve the unique biodiversity of the region, the expansion of protected forests and active reforestation measures are urgently needed. Sustainable land use planning and forest management integrating the needs of local land users and conservation priorities should be promoted. We see the highest potential for external stakeholders (e.g., national NGOs) to implement targeted interventions embedded in community-based approaches. Our land cover maps and predictive modeling highlight crucial areas that could act as stepping stone habitats for dispersing or retreating species and therefore important locations to intensify conservation measures.

Historical Aerial Surveys Map Long-Term Changes of Forest Cover and Structure in the Central Congo Basin

Given the impact of tropical forest disturbances on atmospheric carbon emissions, biodiversity, and ecosystem productivity, accurate long-term reporting of Land-Use and Land-Cover (LULC) change in the pre-satellite era (<1972) is an imperative. Here, we used a combination of historical (1958) aerial photography and contemporary remote sensing data to map long-term changes in the extent and structure of the tropical forest surrounding Yangambi (DR Congo) in the central Congo Basin. Our study leveraged structure-from-motion and a convolutional neural network-based LULC classifier, using synthetic landscape-based image augmentation to map historical forest cover across a large orthomosaic (~93,431 ha) geo-referenced to ~4.7 ± 4.3 m at submeter resolution. A comparison with contemporary LULC data showed a shift from previously highly regular industrial deforestation of large areas to discrete smallholder farming clearing, increasing landscape fragmentation and providing opportunties for substantial forest regrowth. We estimated aboveground carbon gains through reforestation to range from 811 to 1592 Gg C, partially offsetting historical deforestation (2416 Gg C), in our study area. Efforts to quantify long-term canopy texture changes and their link to aboveground carbon had limited to no success. Our analysis provides methods and insights into key spatial and temporal patterns of deforestation and reforestation at a multi-decadal scale, providing a historical context for past and ongoing forest research in the area.

Long-term changes in forest cover in a global biodiversity hotspot in southern Mozambique

Background: Deforestation is a complex and dynamic process of widespread concern in sub-Saharan Africa which is influenced by a range of social, economic and biophysical factors. Objectives: The aim of this study was to analyse patterns of deforestation and its potential drivers in the Licuáti Forest Reserve, a biodiversity hotspot in southern Mozambique between 1990 and 2016. Method: We performed image classification on Landsat imagery at six time steps and interviewed local community members to understand the spatial pattern and rate of forest cover loss over time. We also examined changes in the incidence of fire. Results: A substantial increase in the rate of deforestation since 1990 was detected in this vulnerable thicket vegetation. The probability of deforestation was significantly higher near the major roads, where houses are located. This suggests that the proximity of human settlements to the forest, and access to charcoal markets in urban areas, influenced the spatial pattern. Two key factors 1) charcoal production, and 2) the establishment of settlements and agricultural lands were identified as proximate causes of deforestation. In addition, fires associated with these two causes might amplify the loss of forests in the area. Conclusions: Complex interactions between the drivers of deforestation and socio-economic factors were suggested, as most of the charcoal produced in the region is transported to Maputo. Ongoing road improvements and infrastructural development in the region will likely accelerate the decline in forest cover in the future. This has implications for the biodiversity of the region as well as for the sustainability of local livelihoods, as they often depend on forest products for their daily uses.