Quantify Forest Cover Research Articles

Quantifying Forest Cover Loss during the COVID-19 Pandemic in the Lubumbashi Charcoal Production Basin (DR Congo) through Remote Sensing and Landscape Analysis

In the context of the Lubumbashi Charcoal Production Basin (LCPB), the socio-economic repercussions of the COVID-19 pandemic have exacerbated pressures on populations dependent on forest resources for their subsistence. This study employs a comprehensive methodological approach, integrating advanced remote sensing techniques, including image classification, mapping, and detailed landscape analysis, to quantify alterations in forest cover within the LCPB during the pandemic period. Our findings reveal a consistent trend of declining forested area, characterized by processes of attrition and dissection observed throughout various study phases, spanning from May 2019 to November 2023. This reduction in forest cover, notably more pronounced in the vicinity of Lubumbashi city and the northern zone of the LCPB, proved to be less pronounced between November 2019 and September 2020, underscoring the influence of COVID-19 pandemic-induced confinement measures on forest management practices in the region. However, subsequent to this period of restriction, deforestation activity intensified, leading to significant landscape transformations within the LCPB, primarily attributable to expanded human activities, consequently resulting in a notable decrease in the proportion of land occupied by these natural ecosystems. Consequently, the size of the largest forest patch declined substantially, decreasing from 14.62% to 8.20% between May 2019 and November 2023, thereby fostering a heightened density of forest edges over time. Our findings provide a significant contribution to understanding the complex interactions between the COVID-19 pandemic and deforestation phenomena, emphasizing the urgent need to adopt adaptive management strategies and appropriate conservation measures in response to current economic challenges.

Correction: Miletić et al. Quantifying Forest Cover Loss as a Response to Drought and Dieback of Norway Spruce and Evaluating Sensitivity of Various Vegetation Indices Using Remote Sensing. Forests 2024, 15, 662

In the original publication [...]

Quantifying Forest Cover Loss as a Response to Drought and Dieback of Norway Spruce and Evaluating Sensitivity of Various Vegetation Indices Using Remote Sensing

The Norway spruce is one of the most important tree species in Europe. This tree species has been put under considerable pressure due to the ongoing impacts of climate change. Meanwhile, frequent droughts and pest outbreaks are reported as the main reason for its dieback, resulting in severe forest cover loss. Such was the case with Norway spruce forests within the Kopaonik National Park (NP) in Serbia. This study aims to quantify, spatially and temporally, forest cover loss and to evaluate the sensitivity of various vegetation indices (VIs) in detecting drought-induced response and predicting the dieback of Norway spruce due to long-lasting drought effects in the Kopaonik NP. For this purpose, we downloaded and processed a large number of Landsat 7 (ETM+), Landsat 8 (OLI), and Sentinel 2 (MSI) satellite imagery acquired from 2009 to 2022. Our results revealed that forest cover loss was mainly driven by severe drought in 2011 and 2012, which was later significantly influenced by bark beetle outbreaks. Furthermore, various VIs proved to be very useful in monitoring and predicting forest health status. In summary, the drought-induced response detected using various VIs provides valuable insights into the dynamics of forest cover change, with implications for monitoring and conservation efforts of Norway spruce forests in the Kopaonik NP.

Losing lemurs: Declining populations and land cover changes over space and time.

Forest loss and degradation due to land cover changes imperil biodiversity worldwide. Subtropical and tropical ecosystems experience high deforestation rates, negatively affecting species like primates. Madagascar's endemic lemurs face exceptionally high risks of population declines and extirpation. We examined how short-term land cover changes within a fragmented landscape in southeastern Madagascar impacted the density of lemur species. Using line transects, we assessed density changes in nine lemur species across five forest fragments. Diurnal surveys were conducted monthly from 2015 to 2019 on 35 transects (total effort = 1268 km). Additionally, 21 transects were surveyed nocturnally in 2015 and 2016 (total effort = 107.5 km). To quantify forest cover changes, we generated land use/land cover (LULC) maps from Sentinel-2 imagery using supervised classification for each year. For the LULC maps, we overlayed species-specific buffers around all transects and calculated the proportion of land cover classes within them. We observed declines in the annual densities of four diurnal and cathemeral lemur species between 2015 and 2019, with species-specific declines of up to 80% (Varecia variegata). While the density of two nocturnal species decreased, one increased fivefold (Cheirogaleus major) between 2015 and 2016. By 2019, Grassland was the dominant land type (50%), while Paddy Fields had the smallest coverage (1.03%). Mature Agricultural Land increased the most (63.37%), while New Agricultural Land decreased the most (-66.36%). Unexpectedly, we did not find evidence that higher forest cover supported a higher lemur population density within sampled areas, but we found support for the negative impact of degraded land cover types on three lemur species. Our study underscores the urgent need to address land-use changes and their repercussions for primate populations in tropical ecosystems. The diverse responses of lemur species to modified habitats highlight the complexity of these impacts and emphasize the importance of targeted conservation efforts.

Patterns of forest cover loss in the terrestrial Key Biodiversity Areas in the Philippines: critical habitat conservation priorities

The Philippines, home to over 20,000 endemic species of plants and animals, is facing a biodiversity crisis due to the constant decrease of forest cover. The Key Biodiversity Area (KBA) approach was developed to conserve species threatened with extinction using a site-based conservation strategy to select globally important sites using threshold-based criteria for species irreplaceability and vulnerability. This study investigates the applicability of remotely sensed data through geospatial analysis to quantify forest cover loss of the 101 terrestrial KBAs in the country between 2001 and 2019. Results showed that the study sites had 4.5 million hectares (ha) of forest in the year 2000. However, these sites have lost about 270,000 ha of forest in nearly two decades, marking a steady decline with an annual deforestation rate of 14,213 ha per year in these terrestrial KBAs. The majority of the study sites (58) had a high percentage of forest loss (>3.13%), and these should be prioritized for conservation. By the year 2030, it is forecast that a total of 331 thousand ha of forest will be lost unless there is a transformational change in the country’s approach to dealing with deforestation. The results of this study provide relevant data and information in forest habitat in near real-time monitoring to assess the impact and effectiveness of forest governance and approaches within these critical habitats.

Proximity and abundance of mother trees affects recruitment patterns in a long‐term tropical forest restoration study

Remnant trees and forest fragments in agricultural landscapes can be important sources of propagules to facilitate forest recovery. However, many studies simply quantify forest cover in the surrounding landscape as a percentage, with little attention given to species composition, and subsequently fail to detect an effect on recruitment patterns. We assessed the relative importance of the spatial distribution and life‐history traits of 77 tree species on recruitment patterns at a landscape scale in a well‐replicated long‐term restoration study in southern Costa Rica. We censused and mapped potential mother trees in a 100‐m buffer surrounding eight replicate restoration plots and quantified respective tree recruits within each plot. We assessed how mother tree abundance, species life‐history characteristics (seed size, dispersal mode), tree size (DBH, height) and distance to restoration plot affected recruitment at coarse (plot: 50 × 50 m) and fine (quadrat: 3 × 3 m) spatial scales. The presence of a mother tree within 100 m of a restoration plot resulted in a 10‐fold increase in potential mean recruitment. Mother tree abundance was also an important driver of recruit density, and particularly so for large‐seeded (≥ 5 mm) zoochorous species with a fivefold increase in recruit density across the observed mother tree abundance range. An interaction between mother tree abundance and proximity demonstrated that the effect of mother tree abundance on recruit density was important but waned with increasing distance from restoration plots. At the fine spatial scale, proximity was uniformly important; height and DBH of the closest potential mother tree also affected recruit abundance but responses differed by seed size. Results highlight the importance of remnant vegetation composition to the recovery of adjacent degraded habitats, underscoring the outsized role nearby remnant forest and isolated trees can play for the persistence of localized biodiversity.

Avaliação da estrutura da paisagem visando à conservação da biodiversidade em paisagem urbanizada

Considerando que a expansão urbana e o aumento das atividades antrópicas geram perda e fragmentação dos hábitats naturais, ameaçando a biodiversidade, a conservação dos remanescentes florestais urbanos é de extrema importância para a persistência de espécies de fauna e flora nativas. Sendo assim, o objetivo deste estudo foi avaliar a estrutura da paisagem de área urbanizada no bioma Mata Atlântica, utilizando técnicas de geoprocessamento e fundamentos de Ecologia de Paisagem, visando a fornecer subsídios à conservação florestal na paisagem urbana. Um mapa de uso e cobertura foi utilizado para quantificar a cobertura florestal e para avaliar a estrutura da paisagem. As métricas de paisagem quantificaram área (AREA), forma (SHAPE) e proximidade (NEAR) dos remanescentes florestais, fornecendo indicadores da estrutura da paisagem. A cobertura florestal representa 22,9% da área de estudo, sendo apenas 3,6% dos fragmentos maiores que 50 ha. A grande maioria dos fragmentos (83%) tem área menor que 10 ha. Apenas três remanescentes são maiores que 500 ha, sendo que o maior deles localiza-se na Floresta Nacional de Ipanema. A maioria dos fragmentos florestais (80%) localiza-se em propriedades particulares. A paisagem foi diagnosticada como altamente antropizada e fragmentada, porém com remanescentes próximos entre si (NEAR 50 m), potencializando a movimentação das espécies na matriz antrópica. Cenários de exclusão dos pequenos fragmentos demonstraram a fragilidade da matriz sem o apoio deles, pois os pequenos remanescentes contribuem para a conexão entre os maiores. O campo antrópico de vegetação pioneira é a segunda classe de uso de maior cobertura na área de estudo (20,4%), atrás somente da área urbanizada (25%). Ações para a manutenção da biodiversidade em paisagens antropizadas, como as da área de estudo, devem priorizar projetos de restauração de campos antrópicos, incentivos à conservação de remanescentes em propriedades particulares e a manutenção de pequenos fragmentos na paisagem, a fim de promover a conexão entre as grandes áreas florestais. Os planejamentos urbano e ambiental das cidades devem ser alinhados e baseados em estudos científicos sobre a espacialização da vegetação no ambiente urbano.

Object-Based Classification of Forest Disturbance Types in the Conterminous United States

Forest ecosystems provide critical ecosystem goods and services, and any disturbance-induced changes can have cascading impacts on natural processes and human socioeconomic systems. Forest disturbance frequency, intensity, and spatial and temporal scale can be altered by changes in climate and human activity, but without baseline forest disturbance data, it is impossible to quantify the magnitude and extent of these changes. Methodologies for quantifying forest cover change have been developed at the regional-to-global scale via several approaches that utilize data from high (e.g., IKONOS, Quickbird), moderate (e.g., Landsat) and coarse (e.g., Moderate Resolution Imaging Spectroradiometer (MODIS)) spatial resolution satellite imagery. While detection and quantification of forest cover change is an important first step, attribution of disturbance type is critical missing information for establishing baseline data and effective land management policy. The objective here was to prototype and test a semi-automated methodology for characterizing high-magnitude (>50% forest cover loss) forest disturbance agents (stress, fire, stem removal) across the conterminous United States (CONUS) from 2003–2011 using the existing University of Maryland Landsat-based Global Forest Change Product and Web-Enabled Landsat Data (WELD). The Forest Cover Change maps were segmented into objects based on temporal and spatial adjacency, and object-level spectral metrics were calculated based on WELD reflectance time series. A training set of objects with known disturbance type was developed via high-resolution imagery and expert interpretation, ingested into a Random Forest classifier, which was then used to attribute disturbance type to all 15,179,430 forest loss objects across CONUS. Accuracy assessments of the resulting classification was conducted with an independent dataset consisting of 4156 forest loss objects. Overall accuracy was 88.1%, with the highest omission and commission errors observed for fire (32.8%) and stress (31.9%) disturbances, respectively. Of the total 172,686 km2 of forest loss, 83.75% was attributed to stem removal, 10.92% to fire and 5.33% to stress. The semi-automated approach described in this paper provides a promising framework for the systematic characterization and monitoring of forest disturbance regimes.

Evaluación ambiental integrada de áreas inscritas en el programa federal de Pago por Servicios Ambientales Hidrológicos. Caso de estudio: Ajusco, México

Since the specific problem of deforestation on an international level, they have created compensation mechanisms for ecosystem services, which have recently been increasing interest in the academic and government sectors, however, a lack of information on their potential effects is detected, especially from the aspects of integrative approach and spatial. Therefore, this work presents the case study of the community of San Miguel and Santo Tomas Ajusco, located on the suburbs of México City, because the community has participated in the federal program payments for hydrological environmental services (phes) since 2004. Based on the construction of longitudinal profiles (which are incorporated thematic geographic information layers) we analyzed if there is correspondence between the potential of ecosystem services and enrolled in the program areas. In addition, the effectiveness of the program regarding deforestation detected in the area was observed. For the profiling, it is noticed the exercise of combining the different layers selected for analysis, forming a two-dimensional model that allows us to appreciate the altitudinal gradient of the area with its biophysical components (vegetation and land use, soil science, geology and landforms highlights) with information environmental Services (ES). This facilitates the analysis of all these variables and has the advantage that it can be adapted to the number of variables we have, since they must only be incorporated to the profile (with pretreatment data).The profiles helped to identify whether registered in the phes program areas correspond to the most important areas for the provision of environmental services; however, a complementary work was developed consisting of a map which would bring together the most important areas for the contribution of es; This was generated by applying the statistical model of Jenks’ Natural Breaks “(which allows to discriminate classes by the similarity of values that compose them), in the layers of ES that it had (potential infiltration, carbon stock in forests, surface runoff and provision of habitat), establishing quantitative 5 for each category. The criterion model allow standardize classes; therefore, quantitative categories were matched to (very low, low, medium, high and very high) qualitative categories. Then the “high” and “very high” categories were selected and merged into a map with software tools Arc. map 10.1. denominating as “high priority areas for ecosystem services”.ES mapping was deployed with phes polygons and correspondence between sites on phes and areas with greatest potential for ES (except for an area of 236.7 hectares) was found. However, in many cases, the highest values of es not correspond to water resources, as they are more associated with vegetation and biodiversity (carbon stock in forestsand provision of habitat), which helps to conclude that the phes program should consider other variables and not be restricted to the es associated with hydrology. Two studies on quantifying forest cover were reviewed in the area (one covers from 1986 to 2010, with a projection to 2030 and the other from 2006 to 2014). The information regarding sites on phes Program and the result shows that deforestation has remained revised (although the rate decreased in the period 2006-2014) and deforested areas within enrolled in the program areas phes are greater than those who have not been. Although factors that could contribute to the process are identified, it is concluded that a program like phes implemented in an isolated manner, has no bearing on the problem of deforestation. Finally, the importance of obtaining the map “high priority areas for ecosystem services”, the most environmentally important areas are identified and this helps manage phes expanding areas and get other resources or protection status in areas with natural ecosystems.

Linking Forest Cover to Water Quality: A Multivariate Analysis of Large Monitoring Datasets

Forested catchments are generally assumed to provide higher quality water. However, this hypothesis must be validated in various contexts as interactions between multiple land use and land cover (LULC) types, ecological variables and water quality variables render this relationship highly complex. This paper applies a straightforward multivariate approach on a typical large monitoring dataset of a highly managed and densely populated area (Wallonia, Belgium; 10-year dataset), quantifying forest cover effects on nine physico-chemical water quality variables. Results show that forest cover explains about one third of the variability of water quality and is positively correlated with higher quality water. When controlling for spatial autocorrelation, forest cover still explains 9% of water quality. Unlike needle-leaved forest cover, broad-leaved forest cover presents an independent effect from ecological variables and explains independently 4.8% of water quality variability while it shares 5.8% with cropland cover. This study demonstrates clear independent effects of forest cover on water quality, and presents a method to tease out independent LULC effects from typical large multivariate monitoring datasets. Further research on explanatory variables, spatial distribution effects and water quality datasets could lead to effective strategies to mitigate pollution and reach legal targets.

PRIORITY AREAS FOR FOREST CONSERVATION IN AN URBAN LANDSCAPE AT THE TRANSITION BETWEEN ATLANTIC FOREST AND CERRADO

ABSTRACT Urbanization and agriculture growth are some of the major causes of natural ecosystems depletion and biodiversity loss. Conservation efforts can be developed through the prioritization of areas for forest conservation in order to minimize this process. Here, we establish conservation strategies based on a spatial analysis of forest fragments in an urban landscape at the transition between two important Brazilian biodiversity hotspots: Atlantic Forest and Cerrado. A high-resolution mapping of forest patches was used to quantify forest cover and to provide spatial analysis. We developed a Forest Conservation Priority Index (FCPI) derived from landscape metrics to select priority areas for forest conservation efforts. We used area, shape and proximity metrics as landscape structure indicators. Landscape metrics were classified and we attributed a rank and weights for them to calculate the FCPI. Forest covers 17% of the study area. 60% of the forest patches comprise less than one hectare, 95% less than 10 ha and only 1% more than 50 ha. The largest fragments (> 100ha) are all long and narrow. Also, fewer than half the patches are connected to others. Regions classified as high and very high priority for forest conservation are localized at greater slopes, along rivers and on private lands. Our index allowed the prioritization of forest fragments in an urban landscape, directing efforts of conservation. Creating protected areas and restoration plans are necessary for the better situation of the natural ecosystem. Land-use planning must resolve human demands and conservation of this important ecosystem.

Forest cover changes in Bhutan: Revisiting the forest transition

Reforestation is observed in several tropical countries, mostly in marginal lands including mountain environments. Bhutan, a landlocked and mountainous country characterized by a long tradition of environmental preservation, is considered as one of the few tropical countries which has experienced a recent forest transition, i.e., a shift from net loss to net increase of forest cover. However, doubts persist regarding the contemporary evolution of forest cover in Bhutan as two recent studies reported opposite trends. The objective of this study was to assess whether a forest transition did indeed recently take place in Bhutan. We used topographically corrected Landsat images from 1990, 2001/2002, and 2011 to detect and quantify forest cover change trajectories over Bhutan. Results of the land cover classifications confirmed that forest cover in Bhutan remained very stable, with a forested area covering more than 60% of the country. For both time periods, gross and net forest cover changes were remarkably small. A trend reversal occurred as a net forest loss of 91 km2 during the 1990s was followed by a net forest gain of 52 km2 during the following decade. This small transition resulted from both an increase in rates of forest gain and a decrease in rates of forest loss, mostly affecting the broadleaf forest ecosystem of southern low-elevation districts. Northern districts covered by coniferous forests did not contribute to the forest transition. A spatially explicit comparison of our forest cover change maps with two previous studies highlighted the importance of a robust validation of change detection results and of a topographic correction of remote sensing images in rugged environments, as performed in this study. These recent forest cover changes can be explained by socio-economic and political changes associated with the opening of the country to trade, which induced a gradual transition from subsistence shifting cultivation to market-oriented agriculture. This study confirmed that the high forest cover of Bhutan was maintained since 1990, with a slight increase during the 2000s.

Using the Landsat data archive to assess long-term regional forest dynamics assessment in Eastern Europe, 1985-2012

Abstract. Dramatic political and economic changes in Eastern European countries following the dissolution of the “Eastern Bloc” and the collapse of the Soviet Union greatly affected land-cover and land-use trends. In particular, changes in forest cover dynamics may be attributed to the collapse of the planned economy, agricultural land abandonment, economy liberalization, and market conditions. However, changes in forest cover are hard to quantify given inconsistent forest statistics collected by different countries over the last 30 years. The objective of our research was to consistently quantify forest cover change across Eastern Europe from 1985 until 2012 using the complete Landsat data archive. We developed an algorithm for processing imagery from different Landsat platforms and sensors (TM and ETM+), aggregating these images into a common set of multi-temporal metrics, and mapping annual gross forest cover loss and decadal gross forest cover gain. Our results show that forest cover area increased from 1985 to 2012 by 4.7% across the region. Average annual gross forest cover loss was 0.41% of total forest cover area, with a statistically significant increase from 1985 to 2012. Most forest disturbance recovered fast, with only 12% of the areas of forest loss prior to 1995 not being recovered by 2012. Timber harvesting was the main cause of forest loss. Logging area declined after the collapse of socialism in the late 1980s, increased in the early 2000s, and decreased in most countries after 2007 due to the global economic crisis. By 2012, Central and Baltic Eastern European countries showed higher logging rates compared to their Western neighbours. Comparing our results with official forest cover and change estimates showed agreement in total forest area for year 2010, but with substantial disagreement between Landsat-based and official net forest cover area change. Landsat-based logging areas exhibit strong relationship with reported roundwood production at national scale. Our results allow national and sub-national level analysis of forest cover extent, change, and logging intensity and are available on-line as a baseline for further analyses of forest dynamics and its drivers.

Eastern Europe's forest cover dynamics from 1985 to 2012 quantified from the full Landsat archive

In the former “Eastern Bloc” countries, there have been dramatic changes in forest disturbance and forest recovery rates since the collapse of the Soviet Union, due to the transition to open-market economies, and the recent economic crisis. Unfortunately though, Eastern European countries collected their forest statistics inconsistently, and their boundaries have changed, making it difficult to analyze forest dynamics over time. Our goal here was to consistently quantify forest cover change across Eastern Europe since the 1980s based on the Landsat image archive. We developed an algorithm to simultaneously process data from different Landsat platforms and sensors (TM and ETM+) to map annual forest cover loss and decadal forest cover gain. We processed 59,539 Landsat images for 527 footprints across Eastern Europe and European Russia. Our results were highly accurate, with gross forest loss producer's and user's accuracy of >88% and >89%, respectively, and gross forest gain producer's and user's accuracy of >75% and >91%, based on a sample of probability-based validation points. We found substantial changes in the forest cover of Eastern Europe. Net forest cover increased from 1985 to 2012 by 4.7% across the region, but decreased in Estonia and Latvia. Average annual gross forest cover loss was 0.41% of total forest cover area, with a statistically significant increase from 1985 to 2012. Timber harvesting was the main cause of forest loss, accompanied by some insect defoliation and forest conversion, while only 7.4% of the total forest cover loss was due to large-scale wildfires and windstorms. Overall, the countries of Eastern Europe experienced constant levels or declines in forest loss after the collapse of socialism in the late 1980s, but a pronounced increase in loss in the early 2000s. By the late 2000s, however, the global economic crisis coincided with reduced timber harvesting in most countries, except Poland, Czech Republic, Slovakia, and the Baltic states. Most forest disturbance did not result in a permanent forest loss during our study period. Indeed, forest generally recovered fast and only 12% of the areas of forest loss prior to 1995 had not yet recovered by 2012. Our results allow national and sub-national level analysis and are available on-line (http://glad.geog.umd.edu/europe/) to serve as a baseline for further analyses of forest dynamics and its drivers.

Forest cover change over four decades in the Blue Nile Basin, Ethiopia: comparison of three watersheds

The objective of this study was to quantify forest cover changes in three watersheds (Gilgel Abbay (1,646 km2), Birr (980 km2), and Upper-Didesa (1,980 km2) of the Blue Nile Basin between 1957 and 2001. Four land cover maps were produced for each watershed for 1957/1958, 1975, 1986, and 2000/2001. Nine different types of land cover were identified, five of which were forest cover classes. Between 1957 and 2001, the total forest cover increased in Gilgel Abbay (from 10 to 22 % cover) and decreased in Birr (from 29 to 22 % cover) as well as in Upper-Didesa (from 89 to 45 % cover). The increase in Gilgel Abbay was primarily due to the expansion of eucalyptus plantations. Natural forest cover decreased in all three watersheds. Wooded grassland decreased by two-thirds, dry/moist mixed forests decreased by half, and riverine forests had disappeared by 1975 in Gilgel Abbay and Birr. Major deforestation had already taken place in the northern watersheds, Gilgel Abbay and Birr, before the 1960s and 1970s, while in the southern watershed, Upper-Didesa, much of the deforestation occurred after 1975. The southern watershed still remained by far the most forested watershed in 2001 despite the strong ongoing deforestation. The changes in forest cover could affect natural resource management, greenhouse gas emissions, water resources, and agricultural production including coffee production. The patterns of change are different in the three watersheds. We therefore recommend further studies of the local conditions and drivers of change as the basis for designing effective policy to halt further loss of natural forest, which offers a wealth of ecosystem services.

A Sample-Based Forest Monitoring Strategy Using Landsat, AVHRR and MODIS Data to Estimate Gross Forest Cover Loss in Malaysia between 1990 and 2005

Insular Southeast Asia is a hotspot of humid tropical forest cover loss. A sample-based monitoring approach quantifying forest cover loss from Landsat imagery was implemented to estimate gross forest cover loss for two eras, 1990–2000 and 2000–2005. For each time interval, a probability sample of 18.5 km × 18.5 km blocks was selected, and pairs of Landsat images acquired per sample block were interpreted to quantify forest cover area and gross forest cover loss. Stratified random sampling was implemented for 2000–2005 with MODIS-derived forest cover loss used to define the strata. A probability proportional to x (πpx) design was implemented for 1990–2000 with AVHRR-derived forest cover loss used as the x variable to increase the likelihood of including forest loss area in the sample. The estimated annual gross forest cover loss for Malaysia was 0.43 Mha/yr (SE = 0.04) during 1990–2000 and 0.64 Mha/yr (SE = 0.055) during 2000–2005. Our use of the πpx sampling design represents a first practical trial of this design for sampling satellite imagery. Although the design performed adequately in this study, a thorough comparative investigation of the πpx design relative to other sampling strategies is needed before general design recommendations can be put forth.

Time-series analysis of multi-resolution optical imagery for quantifying forest cover loss in Sumatra and Kalimantan, Indonesia

Monitoring loss of humid tropical forests via remotely sensed imagery is critical for a number of environmental monitoring objectives, including carbon accounting, biodiversity, and climate modeling science applications. Landsat imagery, provided free of charge by the U.S. Geological Survey Center for Earth Resources Observation and Science (USGS/EROS), enables consistent and timely forest cover loss updates from regional to biome scales. The Indonesian islands of Sumatra and Kalimantan are a center of significant forest cover change within the humid tropics with implications for carbon dynamics, biodiversity maintenance and local livelihoods. Sumatra and Kalimantan feature poor observational coverage compared to other centers of humid tropical forest change, such as Mato Grosso, Brazil, due to the lack of ongoing acquisitions from nearby ground stations and the persistence of cloud cover obscuring the land surface. At the same time, forest change in Indonesia is transient and does not always result in deforestation, as cleared forests are rapidly replaced by timber plantations and oil palm estates. Epochal composites, where single best observations are selected over a given time interval and used to quantify change, are one option for monitoring forest change in cloudy regions. However, the frequency of forest cover change in Indonesia confounds the ability of image composite pairs to quantify all change. Transient change occurring between composite periods is often missed and the length of time required for creating a cloud-free composite often obscures change occurring within the composite period itself. In this paper, we analyzed all Landsat 7 imagery with <50% cloud cover and data and products from the Moderate Resolution Imaging Spectroradiometer (MODIS) to quantify forest cover loss for Sumatra and Kalimantan from 2000 to 2005. We demonstrated that time-series approaches examining all good land observations are more accurate in mapping forest cover change in Indonesia than change maps based on image composites. Unlike other time-series analyses employing observations with a consistent periodicity, our study area was characterized by highly unequal observation counts and frequencies due to persistent cloud cover, scan line corrector off (SLC-off) gaps, and the absence of a complete archive. Our method accounts for this variation by generating a generic variable space. We evaluated our results against an independent probability sample-based estimate of gross forest cover loss and expert mapped gross forest cover loss at 64 sample sites. The mapped gross forest cover loss for Sumatra and Kalimantan was 2.86% of the land area, or 2.86Mha from 2000 to 2005, with the highest concentration having occurred in Riau and Kalimantan Tengah provinces.

Reply to Reams et al.: Quantifying forest cover change at local and global scales

Estimates derived from ground data-rich programs such as the US Forest Service Forest Information and Analysis (FIA) program contribute valuable information to quantifying forest dynamics. The results of our study, Hansen et al. (1), comparing gross forest cover loss (GFCL) globally, are not intended to replace estimates such as the southeast US net-change estimate reported by Reams et al. (2). The result that we reported for the United States (table 3 in ref. 1) is a national estimate of GFCL, and Reams et al. (2) do not provide estimates for other US regions or a national estimate of net or gross change that can be compared with our result. Estimates of forest change vary depending on the definition of forest (including forest cover vs. forest use), the time period assessed, the primary data, the change dynamic quantified, and the region evaluated. For example, whereas Reams et al. (2) report no net change in forest area for the southeast United States for 1997–2007, Drummond and Loveland (3) report 2.1 million hectares of net forest loss (5.7% of forest cover) between 1973 and 2000 for the six southeastern US ecoregions that they evaluated and net forest loss of 3.7 million hectares (4.1% of forest cover) for eastern US forests.

Mapping tropical forest cover and deforestation using synthetic aperture radar (SAR) images

The changes in forest cover in many parts of the world lead to increase the accumulation of atmospheric carbon and thus accelerate the process of global warming. Optical remote sensing has been used to map and quantify deforestation but its application is limited because of the presence of cloud coverage on the images. Recent availability of several space-borne synthetic aperture radar (SAR) missions has widened the scope of utilizing radar images for monitoring of forest cover change. The objective of this investigation is to examine the capability of SAR data to assess and map deforestation. The study area is located at the tropical forest region of southeastern Bangladesh. Shuttle Imaging Radar-C (SIR-C) data of 1994 and Advanced Land Observation Satellite (ALOS) PALSAR data of 2007 were used in this study. ALOS PALSAR data were orthorectified with Shuttle Radar Topographic Mission digital elevation model data. Image to image geometric registration was done between the two SAR scenes. Study area was clipped and separated as subsets. SIR-C data (L- and C-bands) was in dual polarization (HH and HV) and PALSAR (L-band) was in quad-polarization (HH, HV, VH, and VV). Five different categories of land covers (forest, upland soil/shrubs, lowland soil, settlements, and water/wetlands) were recognized on both SAR scenes. An additional class representing forest re-growth could be identified only on SIR-C image. Both the images were classified using maximum likelihood algorithms. The classification accuracy was computed from the randomly selected independent validation pixels. The accuracy for forest is more than 83% except users accuracy computed for PALSAR image. Forest was reduced from 18,000 to 13,800 ha in the region during the study period. The results of this study will be useful for understanding the applicability of SAR to map and quantify forest cover changes in the tropics.

An assessment of Hawaiian dry forest condition with fine resolution remote sensing

Remote sensing offers the potential to spatially map forest cover quickly and reliably for inventory purposes. We developed a new image analysis approach using an integrated methodology of “object-based” image classification techniques and field-based measurements to quantify forest cover in a degraded dry forest ecosystem on the leeward side of the Island of Hawaii. This new approach explicitly recognized the transitional areas between tree crowns and tree shades (tree shadows) as a unique class and fully utilized them for the quantification of canopy cover. Object-oriented classification of Ikonos-2 satellite images allowed delineation of tree shades and crowns and the transitional areas between them from objects with similar reflectance and size that were surrounding the trees. These included patches of fountain ( Pennisetum setaceum) and kikuyu ( Pennisetum clandestinum) grass, lava outcrops and lava–grass mixtures. Crown-shade transitions were clearly differentiated in spite of their wide range of spectral values and reflectance similarities with areas of lava–grass mixture. Segments representing tree shades and dark lava outcrops were also classified into their respective classes even if they were contiguous. The image estimates of canopy cover using the tree shade plus transition classes were linearly related with field estimates of canopy cover ( R 2 = 0.86 and slope = 0.976). Based on this relationship, dry forest cover throughout the 2627-ha area was estimated at 7.7 ± 1.9%. An immediate application of this new approach is to select and delineate areas with higher canopy cover in order to concentrate ecological restoration and conservation efforts.