Ruth DeFries

Land use has generally been considered a local environmental issue, but it is becoming a force of global importance. Worldwide changes to forests, farmlands, waterways, and air are being driven by the need to provide food, fiber, water, and shelter to more than six billion people. Global croplands, pastures, plantations, and urban areas have expanded in recent decades, accompanied by large increases in energy, water, and fertilizer consumption, along with considerable losses of biodiversity. Such changes in land use have enabled humans to appropriate an increasing share of the planet's resources, but they also potentially undermine the capacity of ecosystems to sustain food production, maintain freshwater and forest resources, regulate climate and air quality, and ameliorate infectious diseases. We face the challenge of managing trade-offs between immediate human needs and maintaining the capacity of the biosphere to provide goods and services in the long term.

Burn, Baby, Burn Wildfires can have dramatic and devastating effects on landscapes and human structures and are important agents in environmental transformation. Their impacts on nonanthropocentric aspects of the environment, such as ecosystems, biodiversity, carbon reserves, and climate, are often overlooked. Bowman et al. (p. 481 ) review what is known and what is needed to develop a holistic understanding of the role of fire in the Earth system, particularly in view of the pervasive impact of fires and the likelihood that they will become increasingly difficult to control as climate changes.

New burned area datasets and top-down constraints from atmospheric concentration measurements of pyrogenic gases have decreased the large uncertainty in fire emissions estimates. However, significant gaps remain in our understanding of the contribution of deforestation, savanna, forest, agricultural waste, and peat fires to total global fire emissions. Here we used a revised version of the Carnegie-Ames-Stanford-Approach (CASA) biogeochemical model and improved satellite-derived estimates of area burned, fire activity, and plant productivity to calculate fire emissions for the 1997–2009 period on a 0.5° spatial resolution with a monthly time step. For November 2000 onwards, estimates were based on burned area, active fire detections, and plant productivity from the MODerate resolution Imaging Spectroradiometer (MODIS) sensor. For the partitioning we focused on the MODIS era. We used maps of burned area derived from the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) and Along-Track Scanning Radiometer (ATSR) active fire data prior to MODIS (1997–2000) and estimates of plant productivity derived from Advanced Very High Resolution Radiometer (AVHRR) observations during the same period. Average global fire carbon emissions according to this version 3 of the Global Fire Emissions Database (GFED3) were 2.0 Pg C year−1 with significant interannual variability during 1997–2001 (2.8 Pg C year−1 in 1998 and 1.6 Pg C year−1 in 2001). Globally, emissions during 2002–2007 were relatively constant (around 2.1 Pg C year−1) before declining in 2008 (1.7 Pg C year−1) and 2009 (1.5 Pg C year−1) partly due to lower deforestation fire emissions in South America and tropical Asia. On a regional basis, emissions were highly variable during 2002–2007 (e.g., boreal Asia, South America, and Indonesia), but these regional differences canceled out at a global level. During the MODIS era (2001–2009), most carbon emissions were from fires in grasslands and savannas (44%) with smaller contributions from tropical deforestation and degradation fires (20%), woodland fires (mostly confined to the tropics, 16%), forest fires (mostly in the extratropics, 15%), agricultural waste burning (3%), and tropical peat fires (3%). The contribution from agricultural waste fires was likely a lower bound because our approach for measuring burned area could not detect all of these relatively small fires. Total carbon emissions were on average 13% lower than in our previous (GFED2) work. For reduced trace gases such as CO and CH4, deforestation, degradation, and peat fires were more important contributors because of higher emissions of reduced trace gases per unit carbon combusted compared to savanna fires. Carbon emissions from tropical deforestation, degradation, and peatland fires were on average 0.5 Pg C year−1. The carbon emissions from these fires may not be balanced by regrowth following fire. Our results provide the first global assessment of the contribution of different sources to total global fire emissions for the past decade, and supply the community with an improved 13-year fire emissions time series.

This paper on reports the production of a 1 km spatial resolution land cover classification using data for 1992-1993 from the Advanced Very High Resolution Radiometer (AVHRR). This map will be included as an at-launch product of the Moderate Resolution Imaging Spectroradiometer (MODIS) to serve as an input for several algorithms requiring knowledge of land cover type. The methodology was derived from a similar effort to create a product at 8 km spatial resolution, where high resolution data sets were interpreted in order to derive a coarse-resolution training data set. A set of 37 294 x 1 km pixels was used within a hierarchical tree structure to classify the AVHRR data into 12 classes. The approach taken involved a hierarchy of pair-wise class trees where a logic based on vegetation form was applied until all classes were depicted. Multitemporal AVHRR metrics were used to predict class memberships. Minimum annual red reflectance, peak annual Normalized Difference Vegetation Index (NDVI), and minimum channel three brightness temperature were among the most used metrics. Depictions of forests and woodlands, and areas of mechanized agriculture are in general agreement with other sources of information, while classes such as low biomass agriculture and high-latitude broadleaf forest are not. Comparisons of the final product with regional digital land cover maps derived from high-resolution remotely sensed data reveal general agreement, except for apparently poor depictions of temperate pastures within areas of agriculture. Distinguishing between forest and non-forest was achieved with agreements ranging from 81 to 92% for these regional subsets. The agreements for all classes varied from an average of 65% when viewing all pixels to an average of 82% when viewing only those 1 km pixels consisting of greater than 90% one class within the high-resolution data sets.

The Millennium Ecosystem Assessment (MA) introduced a new framework for analyzing social–ecological systems that has had wide influence in the policy and scientific communities. Studies after the MA are taking up new challenges in the basic science needed to assess, project, and manage flows of ecosystem services and effects on human well-being. Yet, our ability to draw general conclusions remains limited by focus on discipline-bound sectors of the full social–ecological system. At the same time, some polices and practices intended to improve ecosystem services and human well-being are based on untested assumptions and sparse information. The people who are affected and those who provide resources are increasingly asking for evidence that interventions improve ecosystem services and human well-being. New research is needed that considers the full ensemble of processes and feedbacks, for a range of biophysical and social systems, to better understand and manage the dynamics of the relationship between humans and the ecosystems on which they rely. Such research will expand the capacity to address fundamental questions about complex social–ecological systems while evaluating assumptions of policies and practices intended to advance human well-being through improved ecosystem services.

The first Moderate Resolution Imaging Spectroradiometer (MODIS) instrument is planned for launch by NASA in 1998. This instrument will provide a new and improved capability for terrestrial satellite remote sensing aimed at meeting the needs of global change research. The MODIS standard products will provide new and improved tools for moderate resolution land surface monitoring. These higher order data products have been designed to remove the burden of certain common types of data processing from the user community and meet the more general needs of global-to-regional monitoring, modeling, and assessment. The near-daily coverage of moderate resolution data from MODIS, coupled with the planned increase in high-resolution sampling from Landsat 7, will provide a powerful combination of observations. The full potential of MODIS will be realized once a stable and well-calibrated time-series of multispectral data has been established. In this paper the proposed MODIS standard products for land applications are described along with the current plans for data quality assessment and product validation.

Abstract Aim To analyse the conservation status of tropical dry forests at the global scale, by combining a newly developed global distribution map with spatial data describing different threats, and to identify the relative exposure of different forest areas to such threats. Location Global assessment. Methods We present a new global distribution map of tropical dry forest derived from the recently developed MODIS Vegetation Continuous Fields (VCF) product, which depicts percentage tree cover at a resolution of 500 m, combined with previously defined maps of biomes. This distribution map was overlaid with spatial data to estimate the exposure of tropical dry forests to a number of different threats: climate change, habitat fragmentation, fire, human population density and conversion to cropland. The extent of tropical dry forest currently protected was estimated by overlaying the forest map with a global data set of the distribution of protected areas. Results It is estimated that 1,048,700 km 2 of tropical dry forest remains, distributed throughout the three tropical regions. More than half of the forest area (54.2%) is located within South America, the remaining area being almost equally divided between North and Central America, Africa and Eurasia, with a relatively small proportion (3.8%) occurring within Australasia and Southeast Asia. Overall, c. 97% of the remaining area of tropical dry forest is at risk from one or more of the threats considered, with highest percentages recorded for Eurasia. The relative exposure to different threats differed between regions: while climate change is relatively significant in the Americas, habitat fragmentation and fire affect a higher proportion of African forests, whereas agricultural conversion and human population density are most influential in Eurasia. Evidence suggests that c. 300,000 km 2 of tropical dry forest now coincide with some form of protected area, with 71.8% of this total being located within South America. Main conclusions Virtually all of the tropical dry forests that remain are currently exposed to a variety of different threats, largely resulting from human activity. Taking their high biodiversity value into consideration, this indicates that tropical dry forests should be accorded high conservation priority. The results presented here could be used to identify which forest areas should be accorded highest priority for conservation action. In particular, the expansion of the global protected area network, particularly in Mesoamerica, should be given urgent consideration.

Abstract Phenological differences among vegetation types, reflected in temporal variations in the Normalized Difference Vegetation Index (NDVI) derived from satellite data, have been used to classify land cover at continental scales. Extending this technique to global scales raises several issues: identifying land cover types that are spectrally distinct and applicable at the global scale; accounting for phasing of seasons in different parts of the world; validating results in the absence of reliable information on global land cover; and acquiring high quality global data sets of satellite sensor data for input to land cover classifications. For this study, a coarse spatial resolution (one by one degree) data set of monthly NDVI values for 1987 was used to explore these methodological issues. A result of a supervised, maximum likelihood classification of eleven cover types is presented to illustrate the feasibility of using satellite sensor data to increase the accuracy of global land cover information, although the result has not been validated systematically. Satellite sensor data at finer spatial resolutions that include other bands in addition to NDVI, as well as methodologies to better identify and describe gradients between cover types, could increase the accuracy of results of global land cover data sets derived from satellite sensor data.

The first results of the Moderate Resolution Imaging Spectroradiometer (MODIS) vegetation continuous field algorithm's global percent tree cover are presented. Percent tree cover per 500-m MODIS pixel is estimated using a supervised regression tree algorithm. Data derived from the MODIS visible bands contribute the most to discriminating tree cover. The results show that MODIS data yield greater spatial detail in the characterization of tree cover compared to past efforts using AVHRR data. This finer-scale depiction should allow for using successive tree cover maps in change detection studies at the global scale. Initial validation efforts show a reasonable relationship between the MODIS-estimated tree cover and tree cover from validation sites.

Humans and their ancestors are unique in being a fire-making species, but 'natural' (i.e. independent of humans) fires have an ancient, geological history on Earth. Natural fires have influenced biological evolution and global biogeochemical cycles, making fire integral to the functioning of some biomes. Globally, debate rages about the impact on ecosystems of prehistoric human-set fires, with views ranging from catastrophic to negligible. Understanding of the diversity of human fire regimes on Earth in the past, present and future remains rudimentary. It remains uncertain how humans have caused a departure from 'natural' background levels that vary with climate change. Available evidence shows that modern humans can increase or decrease background levels of natural fire activity by clearing forests, promoting grazing, dispersing plants, altering ignition patterns and actively suppressing fires, thereby causing substantial ecosystem changes and loss of biodiversity. Some of these contemporary fire regimes cause substantial economic disruptions owing to the destruction of infrastructure, degradation of ecosystem services, loss of life, and smoke-related health effects. These episodic disasters help frame negative public attitudes towards landscape fires, despite the need for burning to sustain some ecosystems. Greenhouse gas-induced warming and changes in the hydrological cycle may increase the occurrence of large, severe fires, with potentially significant feedbacks to the Earth system. Improved understanding of human fire regimes demands: (1) better data on past and current human influences on fire regimes to enable global comparative analyses, (2) a greater understanding of different cultural traditions of landscape burning and their positive and negative social, economic and ecological effects, and (3) more realistic representations of anthropogenic fire in global vegetation and climate change models. We provide an historical framework to promote understanding of the development and diversification of fire regimes, covering the pre-human period, human domestication of fire, and the subsequent transition from subsistence agriculture to industrial economies. All of these phases still occur on Earth, providing opportunities for comparative research.

Abstract. The net flux of carbon from land use and land-cover change (LULCC) accounted for 12.5% of anthropogenic carbon emissions from 1990 to 2010. This net flux is the most uncertain term in the global carbon budget, not only because of uncertainties in rates of deforestation and forestation, but also because of uncertainties in the carbon density of the lands actually undergoing change. Furthermore, there are differences in approaches used to determine the flux that introduce variability into estimates in ways that are difficult to evaluate, and not all analyses consider the same types of management activities. Thirteen recent estimates of net carbon emissions from LULCC are summarized here. In addition to deforestation, all analyses considered changes in the area of agricultural lands (croplands and pastures). Some considered, also, forest management (wood harvest, shifting cultivation). None included emissions from the degradation of tropical peatlands. Means and standard deviations across the thirteen model estimates of annual emissions for the 1980s and 1990s, respectively, are 1.14 ± 0.23 and 1.12 ± 0.25 Pg C yr−1 (1 Pg = 1015 g carbon). Four studies also considered the period 2000–2009, and the mean and standard deviations across these four for the three decades are 1.14 ± 0.39, 1.17 ± 0.32, and 1.10 ± 0.11 Pg C yr−1. For the period 1990–2009 the mean global emissions from LULCC are 1.14 ± 0.18 Pg C yr−1. The standard deviations across model means shown here are smaller than previous estimates of uncertainty as they do not account for the errors that result from data uncertainty and from an incomplete understanding of all the processes affecting the net flux of carbon from LULCC. Although these errors have not been systematically evaluated, based on partial analyses available in the literature and expert opinion, they are estimated to be on the order of ± 0.5 Pg C yr−1.

Intensive mechanized agriculture in the Brazilian Amazon grew by >3.6 million hectares (ha) during 2001–2004. Whether this cropland expansion resulted from intensified use of land previously cleared for cattle ranching or new deforestation has not been quantified and has major implications for future deforestation dynamics, carbon fluxes, forest fragmentation, and other ecosystem services. We combine deforestation maps, field surveys, and satellite-based information on vegetation phenology to characterize the fate of large (>25-ha) clearings as cropland, cattle pasture, or regrowing forest in the years after initial clearing in Mato Grosso, the Brazilian state with the highest deforestation rate and soybean production since 2001. Statewide, direct conversion of forest to cropland totaled >540,000 ha during 2001–2004, peaking at 23% of 2003 annual deforestation. Cropland deforestation averaged twice the size of clearings for pasture (mean sizes, 333 and 143 ha, respectively), and conversion occurred rapidly; >90% of clearings for cropland were planted in the first year after deforestation. Area deforested for cropland and mean annual soybean price in the year of forest clearing were directly correlated ( R 2 = 0.72), suggesting that deforestation rates could return to higher levels seen in 2003–2004 with a rebound of crop prices in international markets. Pasture remains the dominant land use after forest clearing in Mato Grosso, but the growing importance of larger and faster conversion of forest to cropland defines a new paradigm of forest loss in Amazonia and refutes the claim that agricultural intensification does not lead to new deforestation.

Conversion of land to grow crops, raise animals, obtain timber, and build cities is one of the foundations of human civilization. While land use provides these essential ecosystem goods, it alters a range of other ecosystem functions, such as the provisioning of freshwater, regulation of climate and biogeochemical cycles, and maintenance of soil fertility. It also alters habitat for biological diversity. Balancing the inherent trade-offs between satisfying immediate human needs and maintaining other ecosystem functions requires quantitative knowledge about ecosystem responses to land use. These responses vary according to the type of land-use change and the ecological setting, and have local, short-term as well as global, long-term effects. Land-use decisions ultimately weigh the need to satisfy human demands and the unintended ecosystem responses based on societal values, but ecological knowledge can provide a basis for assessing the trade-offs.

Interactions between distant places are increasingly widespread and influential, often leading to unexpected outcomes with profound implications for sustainability.Numerous sustainability studies have been conducted within a particular place with little attention to the impacts of distant interactions on sustainability in multiple places.Although distant forces have been studied, they are usually treated as exogenous variables and feedbacks have rarely been considered.To understand and integrate various distant interactions better, we propose an integrated framework based on telecoupling, an umbrella concept that refers to socioeconomic and environmental interactions over distances.The concept of telecoupling is a logical extension of research on coupled human and natural systems, in which interactions occur within particular geographic locations.The telecoupling framework contains five major interrelated components, i.e., coupled human and natural systems, flows, agents, causes, and effects.We illustrate the framework using two examples of distant interactions associated with trade of agricultural commodities and invasive species, highlight the implications of the framework, and discuss research needs and approaches to move research on telecouplings forward.The framework can help to analyze system components and their interrelationships, identify research gaps, detect hidden costs and untapped benefits, provide a useful means to incorporate feedbacks as well as trade-offs and synergies across multiple systems (sending, receiving, and spillover systems), and improve the understanding of distant interactions and the effectiveness of policies for socioeconomic and environmental sustainability from local to global levels.

The global distribution of C 3 and C 4 plants is required for accurately simulating exchanges of CO 2 , water, and energy between the land surface and atmosphere. It is also important to know the C 3 /C 4 distribution for simulations of the carbon isotope composition of atmospheric CO 2 owing to the distinct fractionations displayed by each photosynthetic type. Large areas of the land surface are spatial and temporal mosaics of both photosynthetic types. We developed an approach for capturing this heterogeneity by combining remote sensing products, physiological modeling, a spatial distribution of global crop fractions, and national harvest area data for major crop types. Our C 3 /C 4 distribution predicts the global coverage of C 4 vegetation to be 18.8 million km 2 , while C 3 vegetation covers 87.4 million km 2 . We incorporated our distribution into the SiB2 model and simulated carbon fluxes for each photosynthetic type. The gross primary production (GPP) of C 4 plants is 35.3 Pg C yr −1 , or ∼23% of total GPP, while that of C 3 plants is 114.7 Pg C yr −1 . The assimilation‐weighted terrestrial discrimination against 13 CO 2 is −16.5‰. If the terrestrial component of the carbon sink is proportional to GPP, this implies a net uptake of 2.4 Pg C yr −1 on land and 1.4 Pg C yr −1 in the ocean using a 13 C budgeting approach and average carbon cycle parameter values for the 1990s. We also simulated the biomass of each photosynthetic type using the CASA model. The simulated biomass values of C 3 and C 4 vegetation are 389.3 and 18.6 Pg C, respectively.

Fire is an essential Earth system process that alters ecosystem and atmospheric composition. Here we assessed long-term fire trends using multiple satellite data sets. We found that global burned area declined by 24.3 ± 8.8% over the past 18 years. The estimated decrease in burned area remained robust after adjusting for precipitation variability and was largest in savannas. Agricultural expansion and intensification were primary drivers of declining fire activity. Fewer and smaller fires reduced aerosol concentrations, modified vegetation structure, and increased the magnitude of the terrestrial carbon sink. Fire models were unable to reproduce the pattern and magnitude of observed declines, suggesting that they may overestimate fire emissions in future projections. Using economic and demographic variables, we developed a conceptual model for predicting fire in human-dominated landscapes.

Protected areas are one of the cornerstones for conserving the world's remaining biodiversity, most of which occurs in tropical forests. We use multiple sources of satellite data to estimate the extent of forest habitat and loss over the last 20 years within and surrounding 198 of the most highly protected areas (IUCN status 1 and 2) located throughout the world's tropical forests. In the early 1980s, surrounding habitat in the 50-km unprotected or less highly protected “buffers” enhanced the protected areas' effective size and their capacity to conserve richness of forest-obligate species above the hypothetical case of complete isolation. However, in nearly 70% of the surrounding buffers, the area of forest habitat declined during the last 20 years, while 25% experienced declines within their administrative boundaries. The loss of habitat occurred in all tropical regions, but protected areas in South and Southeast Asia were most severely affected because of relatively low surrounding forest habitat in the early 1980s and high subsequent loss, particularly in dry tropical forests. The future ability of protected areas to maintain current species richness depends on integrating reserve management within the land use dynamics of their larger regional settings.

Carbon fluxes from tropical deforestation and regrowth are highly uncertain components of the contemporary carbon budget, due in part to the lack of spatially explicit and consistent information on changes in forest area. We estimate fluxes for the 1980s and 1990s using subpixel estimates of percent tree cover derived from coarse (National Oceanic and Atmospheric Administration's Advanced Very High Resolution Radiometer) satellite data in combination with a terrestrial carbon model. The satellite-derived estimates of change in forest area are lower than national reports and remote-sensing surveys from the United Nations Food and Agriculture Organization Forest Resource Assessment (FRA) in all tropical regions, especially for the 1980s. However, our results indicate that the net rate of tropical forest clearing increased approximately 10% from the 1980s to 1990s, most notably in southeast Asia, in contrast to an 11% reduction reported by the FRA. We estimate net mean annual carbon fluxes from tropical deforestation and regrowth to average 0.6 (0.3-0.8) and 0.9 (0.5-1.4) petagrams (Pg).yr(-1) for the 1980s and 1990s, respectively. Compared with previous estimates of 1.9 (0.6-2.5) Pg.yr(-1) based on FRA national statistics of changes in forest area, this alternative estimate suggests less "missing" carbon from the global carbon budget but increasing emissions from tropical land-use change.

Forest cover is an important input variable for assessing changes to carbon stocks, climate and hydrological systems, biodiversity richness, and other sustainability science disciplines. Despite incremental improvements in our ability to quantify rates of forest clearing, there is still no definitive understanding on global trends. Without timely and accurate forest monitoring methods, policy responses will be uninformed concerning the most basic facts of forest cover change. Results of a feasible and cost-effective monitoring strategy are presented that enable timely, precise, and internally consistent estimates of forest clearing within the humid tropics. A probability-based sampling approach that synergistically employs low and high spatial resolution satellite datasets was used to quantify humid tropical forest clearing from 2000 to 2005. Forest clearing is estimated to be 1.39% (SE 0.084%) of the total biome area. This translates to an estimated forest area cleared of 27.2 million hectares (SE 2.28 million hectares), and represents a 2.36% reduction in area of humid tropical forest. Fifty-five percent of total biome clearing occurs within only 6% of the biome area, emphasizing the presence of forest clearing "hotspots." Forest loss in Brazil accounts for 47.8% of total biome clearing, nearly four times that of the next highest country, Indonesia, which accounts for 12.8%. Over three-fifths of clearing occurs in Latin America and over one-third in Asia. Africa contributes 5.4% to the estimated loss of humid tropical forest cover, reflecting the absence of current agro-industrial scale clearing in humid tropical Africa.

Background: Forest, grass, and peat fires release approximately 2 petagrams of carbon into the atmosphere each year, influencing weather, climate, and air quality.Objective: We estimated the annual global mortality attributable to landscape fire smoke (LFS).Methods: Daily and annual exposure to particulate matter ≤ 2.5 μm in aerodynamic diameter (PM2.5) from fire emissions was estimated globally for 1997 through 2006 by combining outputs from a chemical transport model with satellite-based observations of aerosol optical depth. In World Health Organization (WHO) subregions classified as sporadically affected, the daily burden of mortality was estimated using previously published concentration–response coefficients for the association between short-term elevations in PM2.5 from LFS (contrasted with 0 μg/m3 from LFS) and all-cause mortality. In subregions classified as chronically affected, the annual burden of mortality was estimated using the American Cancer Society study coefficient for the association between long-term PM2.5 exposure and all-cause mortality. The annual average PM2.5 estimates were contrasted with theoretical minimum (counterfactual) concentrations in each chronically affected subregion. Sensitivity of mortality estimates to different exposure assessments, counterfactual estimates, and concentration–response functions was evaluated. Strong La Niña and El Niño years were compared to assess the influence of interannual climatic variability.Results: Our principal estimate for the average mortality attributable to LFS exposure was 339,000 deaths annually. In sensitivity analyses the interquartile range of all tested estimates was 260,000–600,000. The regions most affected were sub-Saharan Africa (157,000) and Southeast Asia (110,000). Estimated annual mortality during La Niña was 262,000, compared with 532,000 during El Niño.Conclusions: Fire emissions are an important contributor to global mortality. Adverse health outcomes associated with LFS could be substantially reduced by curtailing burning of tropical rainforests, which rarely burn naturally. The large estimated influence of El Niño suggests a relationship between climate and the burden of mortality attributable to LFS.

Abstract. Long term, high quality estimates of burned area are needed for improving both prognostic and diagnostic fire emissions models and for assessing feedbacks between fire and the climate system. We developed global, monthly burned area estimates aggregated to 0.5° spatial resolution for the time period July 1996 through mid-2009 using four satellite data sets. From 2001–2009, our primary data source was 500-m burned area maps produced using Moderate Resolution Imaging Spectroradiometer (MODIS) surface reflectance imagery; more than 90% of the global area burned during this time period was mapped in this fashion. During times when the 500-m MODIS data were not available, we used a combination of local regression and regional regression trees developed over periods when burned area and Terra MODIS active fire data were available to indirectly estimate burned area. Cross-calibration with fire observations from the Tropical Rainfall Measuring Mission (TRMM) Visible and Infrared Scanner (VIRS) and the Along-Track Scanning Radiometer (ATSR) allowed the data set to be extended prior to the MODIS era. With our data set we estimated that the global annual area burned for the years 1997–2008 varied between 330 and 431 Mha, with the maximum occurring in 1998. We compared our data set to the recent GFED2, L3JRC, GLOBCARBON, and MODIS MCD45A1 global burned area products and found substantial differences in many regions. Lastly, we assessed the interannual variability and long-term trends in global burned area over the past 13 years. This burned area time series serves as the basis for the third version of the Global Fire Emissions Database (GFED3) estimates of trace gas and aerosol emissions.

Land use is expanding and intensifying in the unprotected lands surrounding many of the world's protected areas. The influence of this land use change on ecological processes is poorly understood. The goal of this paper is to draw on ecological theory to provide a synthetic framework for understanding how land use change around protected areas may alter ecological processes and biodiversity within protected areas and to provide a basis for identifying scientifically based management alternatives. We first present a conceptual model of protected areas embedded within larger ecosystems that often include surrounding human land use. Drawing on case studies in this Invited Feature, we then explore a comprehensive set of ecological mechanisms by which land use on surrounding lands may influence ecological processes and biodiversity within reserves. These mechanisms involve changes in ecosystem size, with implications for minimum dynamic area, species–area effect, and trophic structure; altered flows of materials and disturbances into and out of reserves; effects on crucial habitats for seasonal and migration movements and population source/sink dynamics; and exposure to humans through hunting, poaching, exotics species, and disease. These ecological mechanisms provide a basis for assessing the vulnerability of protected areas to land use. They also suggest criteria for designing regional management to sustain protected areas in the context of surrounding human land use. These design criteria include maximizing the area of functional habitats, identifying and maintaining ecological process zones, maintaining key migration and source habitats, and managing human proximity and edge effects.

From 2006 to 2010, deforestation in the Amazon frontier state of Mato Grosso decreased to 30% of its historical average (1996-2005) whereas agricultural production reached an all-time high. This study combines satellite data with government deforestation and production statistics to assess land-use transitions and potential market and policy drivers associated with these trends. In the forested region of the state, increased soy production from 2001 to 2005 was entirely due to cropland expansion into previously cleared pasture areas (74%) or forests (26%). From 2006 to 2010, 78% of production increases were due to expansion (22% to yield increases), with 91% on previously cleared land. Cropland expansion fell from 10 to 2% of deforestation between the two periods, with pasture expansion accounting for most remaining deforestation. Declining deforestation coincided with a collapse of commodity markets and implementation of policy measures to reduce deforestation. Soybean profitability has since increased to pre-2006 levels whereas deforestation continued to decline, suggesting that antideforestation measures may have influenced the agricultural sector. We found little evidence of direct leakage of soy expansion into cerrado in Mato Grosso during the late 2000s, although indirect land-use changes and leakage to more distant regions are possible. This study provides evidence that reduced deforestation and increased agricultural production can occur simultaneously in tropical forest frontiers, provided that land is available and policies promote the efficient use of already-cleared lands (intensification) while restricting deforestation. It remains uncertain whether government- and industry-led policies can contain deforestation if future market conditions favor another boom in agricultural expansion.

The research community needs to develop analytical tools for projecting future trends and evaluating the success of interventions as well as indicators to monitor biological, physical, and social changes.

The Amazon Basin is one of the world's most important bioregions, harboring a rich array of plant and animal species and offering a wealth of goods and services to society. For years, ecological science has shown how large-scale forest clearings cause declines in biodiversity and the availability of forest products. Yet some important changes in the rainforests, and in the ecosystem services they provide, have been underappreciated until recently. Emerging research indicates that land use in the Amazon goes far beyond clearing large areas of forest; selective logging and other canopy damage is much more pervasive than once believed. Deforestation causes collateral damage to the surrounding forests – through enhanced drying of the forest floor, increased frequency of fires, and lowered productivity. The loss of healthy forests can degrade key ecosystem services, such as carbon storage in biomass and soils, the regulation of water balance and river flow, the modulation of regional climate patterns, and the amelioration of infectious diseases. We review these newly revealed changes in the Amazon rainforests and the ecosystem services that they provide.

Over the past 50 years, human agents of deforestation have changed in ways that have potentially important implications for conservation efforts. We characterized these changes through a meta-analysis of case studies of land-cover change in the tropics. From the 1960s to the 1980s, small-scale farmers, with state assistance, deforested large areas of tropical forest in Southeast Asia and Latin America. As globalization and urbanization increased during the 1980s, the agents of deforestation changed in two important parts of the tropical biome, the lowland rainforests in Brazil and Indonesia. Well-capitalized ranchers, farmers, and loggers producing for consumers in distant markets became more prominent in these places and this globalization weakened the historically strong relationship between local population growth and forest cover. At the same time, forests have begun to regrow in some tropical uplands. These changing circumstances, we believe, suggest two new and differing strategies for biodiversity conservation in the tropics, one focused on conserving uplands and the other on promoting environmental stewardship in lowlands and other areas conducive to industrial agriculture.

Does the intensification of agriculture reduce cultivated areas and, in so doing, spare some lands by concentrating production on other lands? Such sparing is important for many reasons, among them the enhanced abilities of released lands to sequester carbon and provide other environmental services. Difficulties measuring the extent of spared land make it impossible to investigate fully the hypothesized causal chain from agricultural intensification to declines in cultivated areas and then to increases in spared land. We analyze the historical circumstances in which rising yields have been accompanied by declines in cultivated areas, thereby leading to land-sparing. We use national-level United Nations Food and Agricultural Organization data on trends in cropland from 1970–2005, with particular emphasis on the 1990–2005 period, for 10 major crop types. Cropland has increased more slowly than population during this period, but paired increases in yields and declines in cropland occurred infrequently, both globally and nationally. Agricultural intensification was not generally accompanied by decline or stasis in cropland area at a national scale during this time period, except in countries with grain imports and conservation set-aside programs. Future projections of cropland abandonment and ensuing environmental services cannot be assumed without explicit policy intervention.

The continuous fields Moderate Resolution Imaging Spectroradiometer (MODIS) land cover products are 500-m sub-pixel representations of basic vegetation characteristics including tree, herbaceous and bare ground cover. Our previous approach to deriving continuous fields used a linear mixture model based on spectral endmembers of forest, grassland and bare ground training. We present here a new approach for estimating percent tree cover employing continuous training data over the whole range of tree cover. The continuous training data set is derived by aggregating high-resolution tree cover to coarse scales and is used with multi-temporal metrics based on a full year of coarse resolution satellite data. A regression tree algorithm is used to predict the dependent variable of tree cover based on signatures from the multi-temporal metrics. The automated algorithm was tested globally using Advanced Very High Resolution Radiometer (AVHRR) data, as a full year of MODIS data has not yet been collected. A root mean square error (rmse) of 9.06% tree cover was found from the global training data set. Preliminary MODIS products are also presented, including a 250-m map of the lower 48 United States and 500-m maps of tree cover and leaf type for North America. Results show that the new approach used with MODIS data offers an improved characterization of land cover.

Hydrological ProcessesVolume 18, Issue 11 p. 2183-2186 Invited Commentary Land-use change and hydrologic processes: a major focus for the future R. DeFries, Corresponding Author R. DeFries [email protected] Department of Geography and Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD, USADepartment of Geography, 2181 LeFrak Hall, University of Maryland, College Park, MD 20742, USA.===Search for more papers by this authorK. N. Eshleman, K. N. Eshleman University of Maryland Center for Environmental Science, Appalachian Laboratory, Frostburg, MD, USASearch for more papers by this author R. DeFries, Corresponding Author R. DeFries [email protected] Department of Geography and Earth System Science Interdisciplinary Center, University of Maryland, College Park, MD, USADepartment of Geography, 2181 LeFrak Hall, University of Maryland, College Park, MD 20742, USA.===Search for more papers by this authorK. N. Eshleman, K. N. Eshleman University of Maryland Center for Environmental Science, Appalachian Laboratory, Frostburg, MD, USASearch for more papers by this author First published: 16 July 2004 https://doi.org/10.1002/hyp.5584Citations: 358AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Citing Literature Volume18, Issue1115 August 2004Pages 2183-2186 RelatedInformation

Summary Accurate assessment of the spatial extent of forest cover is a crucial requirement for quantifying the sources and sinks of carbon from the terrestrial biosphere. In the more immediate context of the United Nations Framework Convention on Climate Change, implementation of the Kyoto Protocol calls for estimates of carbon stocks for a baseline year as well as for subsequent years. Data sources from country level statistics and other ground‐based information are based on varying definitions of ‘forest’ and are consequently problematic for obtaining spatially and temporally consistent carbon stock estimates. By combining two datasets previously derived from the Advanced Very High Resolution Radiometer (AVHRR) at 1 km spatial resolution, we have generated a prototype global map depicting percentage tree cover and associated proportions of trees with different leaf longevity (evergreen and deciduous) and leaf type (broadleaf and needleleaf). The product is intended for use in terrestrial carbon cycle models, in conjunction with other spatial datasets such as climate and soil type, to obtain more consistent and reliable estimates of carbon stocks. The percentage tree cover dataset is available through the Global Land Cover Facility at the University of Maryland at http://glcf.umiacs.umd.edu .

C hanges in land cover and in the way people use the land have become recognized over the last 15 years as important global environmental changes in their own right (Turner 2002).They are also intertwined in many ways with other environmental issues, such as climate change and carbon cycle, loss of biodiversity, sustainability of agriculture, and provision of safe drinking water.The international scientific community has created new interdisciplinary research programs to understand the multiple causes and consequences of land-cover and land-use change (Lambin et al. 2003).There has been a concomitant rapid expansion in the availability of data and information.However, there has not yet been a systematic examination, using global and regional observations, of the status and trends in terrestrial and coastal land-cover or related important ecosystem processes.The information needs for such a synthesis are diverse.Remote sensing has an important contribution to make in documenting the actual change in land cover on regional and global spatial scales from the mid-1970s (Achard et al. 2002, DeFries et al. 2002, Lambin et al. 2003).It also has a role to play in evaluating indices of change in ecological processes, such as net primary production and rainfall use efficiency (Prince et al. 1998).Remote sensing information is found in a widely scattered literature, some of it refereed, some in the gray literature, and some unpublished as yet.There is also an obvious need for good inventory data and statistics about land cover and land-cover change at subnational, national, and in-ternational scales, augmented by a need for subnational and national indicators of condition, status, and trends of the global environment.Finally, there is a need to determine the interrelationships of remotely sensed and statistical inventory data, to integrate heterogenous data sources.The tremendous investment in scientific analysis of remote sensing data over the last decade, and the profusion of studies based on other data sources, provides a basis for a synthesis.Although information is not complete globally, several products are now available that depict the land cover of Earth globally in the 1990s and in 2000-2001.The same is true for snapshots of many important regions with substantial land-

Protected areas throughout the world are key for conserving biodiversity, and land use is key for providing food, fiber, and other ecosystem services essential for human sustenance. As land use change isolates protected areas from their surrounding landscapes, the challenge is to identify management opportunities that maintain ecological function while minimizing restrictions on human land use. Building on the case studies in this Invited Feature and on ecological principles, we identify opportunities for regional land management that maintain both ecological function in protected areas and human land use options, including preserving crucial habitats and migration corridors, and reducing dependence of local human populations on protected area resources. Identification of appropriate and effective management opportunities depends on clear definitions of: (1) the biodiversity attributes of concern; (2) landscape connections to delineate particular locations with strong ecological interactions between the protected area and its surrounding landscape; and (3) socioeconomic dynamics that determine current and future use of land resources in and around the protected area.

Abstract Classification trees are a powerful alternative to more traditional approaches of land cover classification. Trees provide a hierarchical and nonlinear classification method and are suited to handling non-parametric training data as well as categorical or missing data. By revealing the predictive hierarchical structure of the independent variables, the tree allows for great flexibility in data analysis and interpretation. In this Letter, we compare a tree' s performance to that of a maximum likelihood classifier using a 1° by 1° global data sel. The tree's accuracy in classifying a validation dala set is comparable to that when using maximum likelihood (82 per cent). The tree also may be used to reduce the dimensionality of data sets and to find those metrics that are most useful for discriminating among cover types.

In recent decades the rate and geographic extent of land-use and land-cover change has increased throughout the world's humid tropical forests. The pan-tropical geography of forest change is a challenge to assess, and improved estimates of the human footprint in the tropics are critical to understanding potential changes in biodiversity. We combined recently published and new satellite observations, along with images from Google Earth and a literature review, to estimate the contemporary global extent of deforestation, selective logging, and secondary regrowth in humid tropical forests. Roughly 1.4% of the biome was deforested between 2000 and 2005. As of 2005, about half of the humid tropical forest biome contained 50% or less tree cover. Although not directly comparable to deforestation, geographic estimates of selective logging indicate that at least 20% of the humid tropical forest biome was undergoing some level of timber harvesting between 2000 and 2005. Forest recovery estimates are even less certain, but a compilation of available reports suggests that at least 1.2% of the humid tropical forest biome was in some stage of long-term secondary regrowth in 2000. Nearly 70% of the regrowth reports indicate forest regeneration in hilly, upland, and mountainous environments considered marginal for large-scale agriculture and ranching. Our estimates of the human footprint are conservative because they do not resolve very small-scale deforestation, low-intensity logging, and unreported secondary regrowth, nor do they incorporate other impacts on tropical forest ecosystems, such as fire and hunting. Our results highlight the enormous geographic extent of forest change throughout the humid tropics and the considerable limitations of the science and technology available for such a synthesis.

We use data from two satellites and a terrestrial carbon model to quantify the impact of urbanization on the carbon cycle and food production in the US as a result of reduced net primary productivity (NPP). Our results show that urbanization is taking place on the most fertile lands and hence has a disproportionately large overall negative impact on NPP. Urban land transformation in the US has reduced the amount of carbon fixed through photosynthesis by 0.04 pg per year or 1.6% of the pre-urban input. The reduction is enough to offset the 1.8% gain made by the conversion of land to agricultural use, even though urbanization covers an area less than 3% of the land surface in the US and agricultural lands approach 29% of the total land area. At local and regional scales, urbanization increases NPP in resource-limited regions and through localized warming "urban heat" contributes to the extension of the growing season in cold regions. In terms of biologically available energy, the loss of NPP due to urbanization of agricultural lands alone is equivalent to the caloric requirement of 16.5 million people, or about 6% of the US population.

Drainage of peatlands and deforestation have led to large-scale fires in equatorial Asia, affecting regional air quality and global concentrations of greenhouse gases. Here we used several sources of satellite data with biogeochemical and atmospheric modeling to better understand and constrain fire emissions from Indonesia, Malaysia, and Papua New Guinea during 2000–2006. We found that average fire emissions from this region [128 ± 51 (1σ) Tg carbon (C) year −1 , T = 10 12 ] were comparable to fossil fuel emissions. In Borneo, carbon emissions from fires were highly variable, fluxes during the moderate 2006 El Niño more than 30 times greater than those during the 2000 La Niña (and with a 2000–2006 mean of 74 ± 33 Tg C yr −1 ). Higher rates of forest loss and larger areas of peatland becoming vulnerable to fire in drought years caused a strong nonlinear relation between drought and fire emissions in southern Borneo. Fire emissions from Sumatra showed a positive linear trend, increasing at a rate of 8 Tg C year −2 (approximately doubling during 2000–2006). These results highlight the importance of including deforestation in future climate agreements. They also imply that land manager responses to expected shifts in tropical precipitation may critically determine the strength of climate–carbon cycle feedbacks during the 21st century.

Abstract An accurate estimate of carbon fluxes associated with tropical deforestation from the last two decades is needed to balance the global carbon budget. Several studies have already estimated carbon emissions from tropical deforestation, but the estimates vary greatly and are difficult to compare due to differences in data sources, assumptions, and methodologies. In this paper, we review the different estimates and datasets, and the various challenges associated with comparing them and with accurately estimating carbon emissions from deforestation. We performed a simulation study over legal Amazonia to illustrate some of these major issues. Our analysis demonstrates the importance of considering land‐cover dynamics following deforestation, including the fluxes from reclearing of secondary vegetation, the decay of product and slash pools, and the fluxes from regrowing forest. It also suggests that accurate carbon‐flux estimates will need to consider historical land‐cover changes for at least the previous 20 years. However, this result is highly sensitive to estimates of the partitioning of cleared carbon into instantaneous burning vs. long‐timescale slash pools. We also show that carbon flux estimates based on ‘committed flux’ calculations, as used by a few studies, are not comparable with the ‘annual balance’ calculation method used by other studies.

Ecosystems are self-regulating systems that provide societies with food, water, timber, and other resources. As demands for resources increase, management decisions are replacing self-regulating properties. Counter to previous technical approaches that applied simple formulas to estimate sustainable yields of single species, current research recognizes the inherent complexity of ecosystems and the inability to foresee all consequences of interventions across different spatial, temporal, and administrative scales. Ecosystem management is thus more realistically seen as a “wicked problem” that has no clear-cut solution. Approaches for addressing such problems include multisector decision-making, institutions that enable management to span across administrative boundaries, adaptive management, markets that incorporate natural capital, and collaborative processes to engage diverse stakeholders and address inequalities. Ecosystem management must avoid two traps: falsely assuming a tame solution and inaction from overwhelming complexity. An incremental approach can help to avoid these traps.

This study uses a global terrestrial carbon cycle model (the Carnegie‐Ames‐Stanford Approach (CASA) model), a satellite‐derived map of existing vegetation, and global maps of natural vegetation to estimate the effects of human‐induced land cover change on carbon emissions to the atmosphere and net primary production. We derived two maps approximating global land cover that would exist for current climate in the absence of human disturbance of the landscape, using a procedure that minimizes disagreements between maps of existing and natural vegetation that represent artifacts in the data. Similarly, we simulated monthly fields of the Normalized Difference Vegetation Index, required as input to CASA, for the undisturbed land cover case. Model results estimate total carbon losses from human‐induced land cover changes of 182 and 199 Pg for the two simulations, compared with an estimate of 124 Pg for total flux between 1850 and 1990 [ Houghton , 1999], suggesting that land cover change prior to 1850 accounted for approximately one‐third of total carbon emissions from land use change. Estimates of global carbon loss from the two independent methods, the modeling approach used in this paper and the accounting approach of Houghton [1999], are comparable taking into account carbon losses from agricultural expansion prior to 1850 estimated at 48–57 Pg. However, estimates of regional carbon losses vary considerably, notably in temperate midlatitudes where our estimates indicate higher cumulative carbon loss. Overall, land cover changes reduced global annual net primary productivity (NPP) by approximately 5%, with large regional variations. High‐input agriculture in North America and Europe display higher annual NPP than the natural vegetation that would exist in the absence of cropland. However, NPP has been depleted in localized areas in South Asia and Africa by up to 90%. These results provide initial crude estimates, limited by the spatial resolution of the data sets used as input to the model and by the lack of information about transient changes in land cover. The results suggest that a modeling approach can be used to estimate spatially‐explicit effects of land cover change on biosphere‐atmosphere interactions.

In response to the United Nations Framework Convention on Climate Change (UNFCCC) process investigating the technical issues surrounding the ability to reduce greenhouse gas (GHG) emissions from deforestation in developing countries, this paper reviews technical capabilities for monitoring deforestation and estimating emissions. Implementation of policies to reduce emissions from deforestation require effective deforestation monitoring systems that are reproducible, provide consistent results, meet standards for mapping accuracy, and can be implemented at the national level. Remotely sensed data supported by ground observations are key to effective monitoring. Capacity in developing countries for deforestation monitoring is well-advanced in a few countries and is a feasible goal in most others. Data sources exist to determine base periods in the 1990s as historical reference points. Forest degradation (e.g. from high impact logging and fragmentation) also contribute to greenhouse gas emissions but it is more technically challenging to measure than deforestation. Data on carbon stocks, which are needed to estimate emissions, cannot currently be observed directly over large areas with remote sensing. Guidelines for carbon accounting from deforestation exist and are available in approved Intergovernmental Panel on Climate Change (IPCC) reports and can be applied at national scales in the absence of forest inventory or other data. Key constraints for implementing programs to monitor greenhouse gas emissions from deforestation are international commitment of resources to increase capacity, coordination of observations to ensure pan-tropical coverage, access to free or low-cost data, and standard and consensual protocols for data interpretation and analysis.

Seasonal changes in the greenness of vegetation, described in remotely sensed data as changes in the normalized difference vegetation index (NDVI) throughout the year, have been the basis for discriminating between cover types in previous attempts to derive land cover from AVHRR data at global and continental scales. Several researchers have suggested and applied the use of metrics, such as maximum NDVI or length of growing season derived from a temporal profile of 10-day or monthly NDVI values, as an alternative to classifying cover types from the monthly NDVI values directly. This study examines the use of metrics derived from the NDVI temporal profile, as well as metrics derived from observations in red, infrared, and thermal bands, to improve discrimination between 12 cover types on a global scale. According to separability measures calculated from Bhattacharya distances, average separabilities improved by using 12 of the 16 metrics tested (1.97) compared to separabilities using 12 monthly NDVI values alone (1.88). Separabilities improved from poor to good in 20 out of 25 pairs of cover types with poor separability. Percentage of pixels correctly classified in a maximum likelihood classifications also improved by using the metrics from 76% to 86%. Overall, the most robust metrics for discriminating between cover types were: mean NDVI, maximum NDVI, NDVI amplitude, AVHRR Band 2 (near-infrared reflectance) and Band 1 (red reflectance) corresponding to the time of maximum NDVI, and maximum land surface temperature. Deciduous and evergreen vegetation can be distinguished by mean NDVI, maximum NDVI, NDVI amplitude, and maximum land surface temperature. Needleleaf and broadleaf vegetation can be distinguished by either mean NDVI and NDVI amplitude or maximum NDVI and NDVI amplitude.

Abstract Systematic, operational, long‐term observations of the terrestrial carbon cycle (including its interactions with water, energy and nutrient cycles and ecosystem dynamics) are important for the prediction and management of climate, water resources, food resources, biodiversity and desertification. To contribute to these goals, a terrestrial carbon observing system requires the synthesis of several kinds of observation into terrestrial biosphere models encompassing the coupled cycles of carbon, water, energy and nutrients. Relevant observations include atmospheric composition (concentrations of CO 2 and other gases); remote sensing; flux and process measurements from intensive study sites; in situ vegetation and soil monitoring; weather, climate and hydrological data; and contemporary and historical data on land use, land use change and disturbance (grazing, harvest, clearing, fire). A review of model–data synthesis tools for terrestrial carbon observation identifies ‘nonsequential’ and ‘sequential’ approaches as major categories, differing according to whether data are treated all at once or sequentially. The structure underlying both approaches is reviewed, highlighting several basic commonalities in formalism and data requirements. An essential commonality is that for all model–data synthesis problems, both nonsequential and sequential, data uncertainties are as important as data values themselves and have a comparable role in determining the outcome. Given the importance of data uncertainties, there is an urgent need for soundly based uncertainty characterizations for the main kinds of data used in terrestrial carbon observation. The first requirement is a specification of the main properties of the error covariance matrix. As a step towards this goal, semi‐quantitative estimates are made of the main properties of the error covariance matrix for four kinds of data essential for terrestrial carbon observation: remote sensing of land surface properties, atmospheric composition measurements, direct flux measurements, and measurements of carbon stores.

Trans-boundary haze events in Southeast Asia are associated with large forest and peatland fires in Indonesia. These episodes of extreme air pollution usually occur during drought years induced by climate anomalies from the Pacific (El Niño Southern Oscillation) and Indian Oceans (Indian Ocean Dipole). However, in June 2013--a non-drought year--Singapore's 24-hr Pollutants Standards Index reached an all-time record 246 (rated "very unhealthy"). Here, we show using remote sensing, rainfall records and other data, that the Indonesian fires behind the 2013 haze followed a two-month dry spell in a wetter-than-average year. These fires were short-lived (one week) and limited to a localized area in Central Sumatra (1.6% of Indonesia): burning an estimated 163,336 ha, including 137,044 ha (84%) on peat. Most burning was confined to deforested lands (82%; 133,216 ha). The greenhouse gas (GHG) emissions during this brief, localized event were considerable: 172 ± 59 Tg CO2-eq (or 31 ± 12 Tg C), representing 5-10% of Indonesia's mean annual GHG emissions for 2000-2005. Our observations show that extreme air pollution episodes in Southeast Asia are no longer restricted to drought years. We expect major haze events to be increasingly frequent because of ongoing deforestation of Indonesian peatlands.

In September–October 2015, El Niño and positive Indian Ocean Dipole conditions set the stage for massive fires in Sumatra and Kalimantan (Indonesian Borneo), leading to persistently hazardous levels of smoke pollution across much of Equatorial Asia. Here we quantify the emission sources and health impacts of this haze episode and compare the sources and impacts to an event of similar magnitude occurring under similar meteorological conditions in September–October 2006. Using the adjoint of the GEOS-Chem chemical transport model, we first calculate the influence of potential fire emissions across the domain on smoke concentrations in three receptor areas downwind—Indonesia, Malaysia, and Singapore—during the 2006 event. This step maps the sensitivity of each receptor to fire emissions in each grid cell upwind. We then combine these sensitivities with 2006 and 2015 fire emission inventories from the Global Fire Assimilation System (GFAS) to estimate the resulting population-weighted smoke exposure. This method, which assumes similar smoke transport pathways in 2006 and 2015, allows near real-time assessment of smoke pollution exposure, and therefore the consequent morbidity and premature mortality, due to severe haze. Our approach also provides rapid assessment of the relative contribution of fire emissions generated in a specific province to smoke-related health impacts in the receptor areas. We estimate that haze in 2015 resulted in 100 300 excess deaths across Indonesia, Malaysia and Singapore, more than double those of the 2006 event, with much of the increase due to fires in Indonesia's South Sumatra Province. The model framework we introduce in this study can rapidly identify those areas where land use management to reduce and/or avoid fires would yield the greatest benefit to human health, both nationally and regionally.

Increasing food production and mitigating climate change are two primary but seemingly contradictory objectives for tropical landscapes. This special feature examines synergies and trade-offs among these objectives. Four themes emerge from the papers: the important roles of both forest and agriculture sectors for climate mitigation in tropical countries; the minor contribution from deforestation-related agricultural expansion to overall food production at global and continental scales; the opportunities for synergies between improved food production and reductions in greenhouse gas emissions through diversion of agricultural expansion to already-cleared lands, improved soil, crop, and livestock management, and agroforestry; and the need for targeted policy and management interventions to make these synergistic opportunities a reality. We conclude that agricultural intensification is a key factor to meet dual objectives of food production and climate mitigation, but there is no single panacea for balancing these objectives in all tropical landscapes. Place-specific strategies for sustainable land use emerge from assessments of current land use, demographics, and other biophysical and socioeconomic characteristics, using a whole-landscape, multisector perspective.

Emissions from landscape fires affect both climate and air quality. This study uses satellite-derived fire estimates and atmospheric modelling to quantify the effects on health from fire emissions in southeast Asia from 1997 to 2006. Strong El Nino years are found to increase the incidence of fires, in addition to those caused by anthropogenic land use change, leading to an additional 200 days per year when the WHO atmospheric particle target is exceeded and increase adult mortality by 2%. Reducing regional deforestation and degradation, and thereby forest fires caused by land use change would therefore improve public health. Emissions from landscape fires affect both climate and air quality1. Here, we combine satellite-derived fire estimates and atmospheric modelling to quantify health effects from fire emissions in southeast Asia from 1997 to 2006. This region has large interannual variability in fire activity owing to coupling between El Niño-induced droughts and anthropogenic land-use change2,3. We show that during strong El Niño years, fires contribute up to 200 μg m−3 and 50 ppb in annual average fine particulate matter (PM2.5) and ozone surface concentrations near fire sources, respectively. This corresponds to a fire contribution of 200 additional days per year that exceed the World Health Organization 50 μg m−3 24-hr PM2.5 interim target4 and an estimated 10,800 (6,800–14,300)-person (∼ 2%) annual increase in regional adult cardiovascular mortality. Our results indicate that reducing regional deforestation and degradation fires would improve public health along with widely established benefits from reducing carbon emissions, preserving biodiversity and maintaining ecosystem services.

A long history of household-level research has provided important local-level insights into climate adaptation strategies in the agricultural sector. It remains unclear to what extent these strategies are generalizable or vary across regions. In this study we ask about three potential key factors influencing farming households’ ability to adapt: access to weather information, household and agricultural production-related assets, and participation in local social institutions. We use a 12-country data set from sub-Saharan Africa and South Asia to explore the links between these three potential drivers of agricultural change and the likelihood that farmers made farm-associated changes, such as adopting improved crop varieties, increasing fertilizer use, investing in improved land management practices, and changing the timing of agricultural activities. We find evidence that access to weather information, assets, and participation in social institutions are associated with households that have reported making farming changes in recent years, although these results vary across countries and types of practices. Understanding these drivers and outcomes of farm-associated changes across different socio-economic and environmental conditions is critical for ongoing dialogues for climate-resilient strategies and policies for increasing the adaptive capacity of smallholders under climate change.

Protected areas (PAs) often depend on landscapes surrounding them to maintain flows of organisms, water, nutrients, and energy. Park managers have little authority over the surrounding landscape although land use change and infrastructure development can have major impacts on the integrity of a PA. The need for scientifically-based regional-scale land use planning around protected areas is acute in human-dominated landscapes to balance conservation goals with livelihood needs for fuelwood, fodder, and other ecosystem services. As a first step, we propose the designation of a “zone of interaction” (ZOI) around PAs that encompasses hydrologic, ecological, and socioeconomic interactions between a PA and the surrounding landscape. We illustrate the concept by delineating the ZOI in three Indian PAs – Kanha, Ranthambore, and Nagarahole – using remote sensing, population census, and field data. The ZOI in Ranthambore is three times the size of the park and is largely defined by the socioeconomic interactions with surrounding villages. Ranthambore is located in headwaters and wildlife corridors are largely severed. In Nagarahole, the ZOI is more than seven times larger than the park and includes upstream watershed and elephant corridors. Kanha’s ZOI is approximately four times larger than the park and is mostly defined by contiguous surrounding forest. The three examples highlight the differing extents of ZOIs when applying equivalent criteria, even though all are located in densely-populated landscapes. Quantitative understanding of which activities (e.g. collection of forest products, grazing, road construction, tourism development) and which locations within the ZOI are most crucial to conservation goals will enable improved land use planning around PAs in human-dominated landscapes.

Attribution of the causes of atmospheric trace gas and aerosol variability often requires the use of high resolution time series of anthropogenic and natural emissions inventories.Here we developed an approach for representing synoptic-and diurnal-scale temporal variability in fire emissions for the Global Fire Emissions Database version 3 (GFED3).We disaggregated monthly GFED3 emissions during 2003-2009 to a daily time step using Moderate Resolution Imaging Spectroradiometer (MODIS)-derived measurements of active fires from Terra and Aqua satellites.In parallel, mean diurnal cycles were constructed from Geostationary Operational Environmental Satellite (GOES) Wildfire Automated Biomass Burning Algorithm (WF_ABBA) active fire observations.Daily variability in fires varied considerably across different biomes, with short but intense periods of daily emissions in boreal ecosystems and lower intensity (but more continuous) periods of burning in savannas.These patterns were consistent with earlier field and modeling work characterizing fire behavior dynamics in different ecosystems.On diurnal timescales, our analysis of the GOES WF_ABBA active fires indicated that fires in savannas, grasslands, and croplands occurred earlier in the day as compared to fires in nearby forests.Comparison with Total Carbon Column Observing Network (TCCON) and Measurements of Pollution in the Troposphere (MOPITT) column CO observations provided evidence that including daily variability in emissions moderately improved atmospheric model simulations, particularly during the fire season and near regions with high levels of biomass burning.The high temporal resolution estimates of fire emissions developed here may ultimately reduce uncertainties related to fire contributions to atmospheric trace gases and aerosols.Important future directions include reconciling top-down and bottom up estimates of fire radiative power and integrating burned area and active fire time series from multiple satellite sensors to improve daily emissions estimates.

Abstract. A globally integrated carbon observation and analysis system is needed to improve the fundamental understanding of the global carbon cycle, to improve our ability to project future changes, and to verify the effectiveness of policies aiming to reduce greenhouse gas emissions and increase carbon sequestration. Building an integrated carbon observation system requires transformational advances from the existing sparse, exploratory framework towards a dense, robust, and sustained system in all components: anthropogenic emissions, the atmosphere, the ocean, and the terrestrial biosphere. The paper is addressed to scientists, policymakers, and funding agencies who need to have a global picture of the current state of the (diverse) carbon observations. We identify the current state of carbon observations, and the needs and notional requirements for a global integrated carbon observation system that can be built in the next decade. A key conclusion is the substantial expansion of the ground-based observation networks required to reach the high spatial resolution for CO2 and CH4 fluxes, and for carbon stocks for addressing policy-relevant objectives, and attributing flux changes to underlying processes in each region. In order to establish flux and stock diagnostics over areas such as the southern oceans, tropical forests, and the Arctic, in situ observations will have to be complemented with remote-sensing measurements. Remote sensing offers the advantage of dense spatial coverage and frequent revisit. A key challenge is to bring remote-sensing measurements to a level of long-term consistency and accuracy so that they can be efficiently combined in models to reduce uncertainties, in synergy with ground-based data. Bringing tight observational constraints on fossil fuel and land use change emissions will be the biggest challenge for deployment of a policy-relevant integrated carbon observation system. This will require in situ and remotely sensed data at much higher resolution and density than currently achieved for natural fluxes, although over a small land area (cities, industrial sites, power plants), as well as the inclusion of fossil fuel CO2 proxy measurements such as radiocarbon in CO2 and carbon-fuel combustion tracers. Additionally, a policy-relevant carbon monitoring system should also provide mechanisms for reconciling regional top-down (atmosphere-based) and bottom-up (surface-based) flux estimates across the range of spatial and temporal scales relevant to mitigation policies. In addition, uncertainties for each observation data-stream should be assessed. The success of the system will rely on long-term commitments to monitoring, on improved international collaboration to fill gaps in the current observations, on sustained efforts to improve access to the different data streams and make databases interoperable, and on the calibration of each component of the system to agreed-upon international scales.

Commodity crop expansion, for both global and domestic urban markets, follows multiple land change pathways entailing direct and indirect deforestation, and results in various social and environmental impacts. Here we compare six published case studies of rapid commodity crop expansion within forested tropical regions. Across cases, between 1.7% and 89.5% of new commodity cropland was sourced from forestlands. Four main factors controlled pathways of commodity crop expansion: (i) the availability of suitable forestland, which is determined by forest area, agroecological or accessibility constraints, and land use policies, (ii) economic and technical characteristics of agricultural systems, (iii) differences in constraints and strategies between small-scale and large-scale actors, and (iv) variable costs and benefits of forest clearing. When remaining forests were unsuitable for agriculture and/or policies restricted forest encroachment, a larger share of commodity crop expansion occurred by conversion of existing agricultural lands, and land use displacement was smaller. Expansion strategies of large-scale actors emerge from context-specific balances between the search for suitable lands; transaction costs or conflicts associated with expanding into forests or other state-owned lands versus smallholder lands; net benefits of forest clearing; and greater access to infrastructure in already-cleared lands. We propose five hypotheses to be tested in further studies: (i) land availability mediates expansion pathways and the likelihood that land use is displaced to distant, rather than to local places; (ii) use of already-cleared lands is favored when commodity crops require access to infrastructure; (iii) in proportion to total agricultural expansion, large-scale actors generate more clearing of mature forests than smallholders; (iv) property rights and land tenure security influence the actors participating in commodity crop expansion, the form of land use displacement, and livelihood outcomes; (v) intensive commodity crops may fail to spare land when inducing displacement. We conclude that understanding pathways of commodity crop expansion is essential to improve land use governance.

Over the last several decades, voluntary certification programs have become a key approach to promote sustainable supply chains for agricultural commodities. These programs provide premiums and other benefits to producers for adhering to environmental and labor practices established by the certifying entities. Following the principles of Cochrane Reviews used in health sciences, we assess evidence to evaluate whether voluntary certification of tropical agricultural commodities (bananas, cocoa, coffee, oil palm, and tea) has achieved environmental benefits and improved economic and social outcomes for small-scale producers at the level of the farm household. We reviewed over 2600 papers in the peer-review literature and identified 24 cases of unique combinations of study area, certification program, and commodity in 16 papers that rigorously analyzed differences between treatment (certified households) and control groups (uncertified households) for a wide range of response variables. Based on analysis of 347 response variables reported in these papers, we conclude that certification is associated on average with positive outcomes for 34% of response variables, no significant difference for 58% of variables, and negative outcomes for 8% of variables. No significant differences were observed for different categories of responses (environmental, economic and social) or for different commodities (banana, coffee and tea), except negative outcomes were significantly less for environmental than other outcome categories (p = 0.01). Most cases (20 out of 24) investigated coffee certification and response variables were inconsistent across cases, indicating the paucity of studies to conduct a conclusive meta-analysis. The somewhat positive results indicate that voluntary certification programs can sometimes play a role in meeting sustainable development goals and do not support the view that such programs are merely greenwashing. However, results also indicate that certification is not a panacea to improve social outcomes or overall incomes of smallholder farmers. Rigorous analysis, standardized criteria, and independent evaluation are needed to assess effectiveness of certification programs in the future.

Land use is central to addressing sustainability issues, including biodiversity conservation, climate change, food security, poverty alleviation, and sustainable energy. In this paper, we synthesize knowledge accumulated in land system science, the integrated study of terrestrial social-ecological systems, into 10 hard truths that have strong, general, empirical support. These facts help to explain the challenges of achieving sustainability in land use and thus also point toward solutions. The 10 facts are as follows: 1) Meanings and values of land are socially constructed and contested; 2) land systems exhibit complex behaviors with abrupt, hard-to-predict changes; 3) irreversible changes and path dependence are common features of land systems; 4) some land uses have a small footprint but very large impacts; 5) drivers and impacts of land-use change are globally interconnected and spill over to distant locations; 6) humanity lives on a used planet where all land provides benefits to societies; 7) land-use change usually entails trade-offs between different benefits-"win-wins" are thus rare; 8) land tenure and land-use claims are often unclear, overlapping, and contested; 9) the benefits and burdens from land are unequally distributed; and 10) land users have multiple, sometimes conflicting, ideas of what social and environmental justice entails. The facts have implications for governance, but do not provide fixed answers. Instead they constitute a set of core principles which can guide scientists, policy makers, and practitioners toward meeting sustainability challenges in land use.

Nutrient content must be better integrated into planning

Air pollution in many of India's cities exceeds national and international standards, and effective pollution control strategies require knowledge of the sources that contribute to air pollution and their spatiotemporal variability. In this study, we examine the influence of a single pollution source, outdoor biomass burning, on particulate matter (PM) concentrations, surface visibility, and aerosol optical depth (AOD) from 2007 to 2013 in three of the most populous Indian cities. We define the upwind regions, or “airsheds,” for the cities by using atmospheric back trajectories from the HYSPLIT model. Using satellite fire radiative power (FRP) observations as a measure of fire activity, we target pre-monsoon and post-monsoon fires upwind of the Delhi National Capital Region and pre-monsoon fires surrounding Bengaluru and Pune. We find varying contributions of outdoor fires to different air quality metrics. For the post-monsoon burning season, we find that a subset of local meteorological variables (air temperature, humidity, sea level pressure, wind speed and direction) and FRP as the only pollution source explained 39% of variance in Delhi station PM10 anomalies, 77% in visibility, and 30% in satellite AOD; additionally, per unit increase in FRP within the daily airshed (1000 MW), PM10 increases by 16.34 μg m−3, visibility decreases by 0.155 km, and satellite AOD increases by 0.07. In contrast, for the pre-monsoon burning season, we find less significant contributions from FRP to air quality in all three cities. Further, we attribute 99% of FRP from post-monsoon outdoor fires within Delhi's average airshed to agricultural burning. Our work suggests that although outdoor fires are not the dominant air pollution source in India throughout the year, post-monsoon fires contribute substantially to regional air pollution and high levels of population exposure around Delhi. During 3-day blocks of extreme PM2.5 in the 2013 post-monsoon burning season, which coincided with statistically significant high fire activity, concentrations in Delhi averaged 304 μg m−3, or more than 1000% above the 24-h PM2.5 guideline (25 μg m−3) of the World Health Organization. These results suggest that providing viable alternatives to agricultural residue burning could help improve post-monsoon air quality for a growing population of 63 million (39% in urban areas) within Delhi's airshed.

Since at least the 1980s, many farmers in northwest India have switched to mechanized combine harvesting to boost efficiency. This harvesting technique leaves abundant crop residue on the fields, which farmers typically burn to prepare their fields for subsequent planting. A key question is to what extent the large quantity of smoke emitted by these fires contributes to the already severe pollution in Delhi and across other parts of the heavily populated Indo-Gangetic Plain located downwind of the fires. Using a combination of observed and modeled variables, including surface measurements of PM2.5, we quantify the magnitude of the influence of agricultural fire emissions on surface air pollution in Delhi. With surface measurements, we first derive the signal of regional PM2.5 enhancements (i.e. the pollution above an anthropogenic baseline) during each post-monsoon burning season for 2012–2016. We next use the Stochastic Time-Inverted Lagrangian Transport model (STILT) to simulate surface PM2.5 using five fire emission inventories. We reproduce up to 25% of the weekly variability in total observed PM2.5 using STILT. Depending on year and emission inventory, our method attributes 7.0%–78% of the maximum observed PM2.5 enhancements in Delhi to fires. The large range in these attribution estimates points to the uncertainties in fire emission parameterizations, especially in regions where thick smoke may interfere with hotspots of fire radiative power. Although our model can generally reproduce the largest PM2.5 enhancements in Delhi air quality for 1–3 consecutive days each fire season, it fails to capture many smaller daily enhancements, which we attribute to the challenge of detecting small fires in the satellite retrieval. By quantifying the influence of upwind agricultural fire emissions on Delhi air pollution, our work underscores the potential health benefits of changes in farming practices to reduce fires.

Models of atmospheric composition rely on fire emissions inventories to reconstruct and project impacts of biomass burning on air quality, public health, climate, ecosystem dynamics, and land-atmosphere exchanges. Many such global inventories use satellite measurements of active fires and/or burned area from the Moderate Resolution Imaging Spectroradiometer (MODIS). However, differences across inventories in the interpretation of satellite imagery, the emissions factors assumed for different components of smoke, and the adjustments made for small and obscured fires can result in large regional differences in fire emissions estimates across inventories. Using Google Earth Engine, we leverage 15 years (2003–2017) of MODIS observations and 6 years (2012–2017) of observations from the higher spatial resolution Visible Imaging Infrared Radiometer Suite (VIIRS) sensor to develop metrics to quantify five major sources of spatial bias or uncertainty in the inventories: (1) primary reliance on active fires versus burned area, (2) cloud/haze burden on the ability of satellites to “see” fires, (3) fragmentation of burned area, (4) roughness in topography, and (5) small fires, which are challenging to detect. Based on all these uncertainties, we devise comprehensive “relative fire confidence scores,” mapped globally at 0.25° × 0.25° spatial resolution over 2003–2017. We then focus on fire activity in Indonesia as a case study to analyze how the choice of a fire emissions inventory affects model estimates of smoke-induced health impacts across Equatorial Asia. We use the adjoint of the GEOS-Chem chemical transport model and apply emissions of particulate organic carbon and black carbon (OC + BC smoke) from five global inventories: Global Fire Emissions Database (GFEDv4s), Fire Inventory from NCAR (FINNv1.5), Global Fire Assimilation System (GFASv1.2), Quick Fire Emissions Dataset (QFEDv2.5r1), and Fire Energetics and Emissions Research (FEERv1.0-G1.2). We find that modeled monthly smoke PM2.5 in Singapore from 2003 to 2016 correlates with observed smoke PM2.5, with r ranging from 0.64–0.84 depending on the inventory. However, during the burning season (July to October) of high fire intensity years (e.g., 2006 and 2015), the magnitude of mean Jul-Oct modeled smoke PM2.5 can differ across inventories by >20 μg m−3 (>500%). Using the relative fire confidence metrics, we deduce that uncertainties in this region arise primarily from the small, fragmented fire landscape and very poor satellite observing conditions due to clouds and thick haze at this time of year. Indeed, we find that modeled smoke PM2.5 using GFASv1.2, which adjusts for fires obscured by clouds and thick haze and accounts for peatland emissions, is most consistent with observations in Singapore, as well as in Malaysia and Indonesia. Finally, we develop an online app called FIRECAM for end-users of global fire emissions inventories. The app diagnoses differences in emissions among the five inventories and gauges the relative uncertainty associated with satellite-observed fires on a regional basis.

Groundwater depletion will reduce cropping intensity in India, and canal irrigation cannot fully substitute for this loss.

Global, monthly, 1° by 1° biophysical land surface datasets for 1982–90 were derived from data collected by the Advanced Very High Resolution Radiometer (AVHRR) on board the NOAA-7, -9, and -11 satellites. The AVHRR data are adjusted for sensor degradation, volcanic aerosol effects, cloud contamination, short-term atmospheric effects (e.g., water vapor and aerosol effects ⩽2 months), solar zenith angle variations, and missing data. Interannual variation in the data is more realistic as a result. The following biophysical parameters are estimated: fraction of photosynthetically active radiation absorbed by vegetation, vegetation cover fraction, leaf area index, and fraction of green leaves. Biophysical retrieval algorithms are tested and updated with data from intensive remote sensing experiments. The multiyear vegetation datasets are consistent spatially and temporally and are useful for studying spatial, seasonal, and interannual variability in the biosphere related to the hydrological cycle, the energy balance, and biogeochemical cycles. The biophysical data are distributed via the Internet by the Goddard Distributed Active Archive Center as a precursor to the International Satellite Land Surface Climatology Project (ISLSCP) Initiative II. Release of more extensive, higher-resolution datasets (0.25° by 0.25°) over longer time periods (1982–97/98) is planned for ISLSCP Initiative II.

Fractional vegetation cover (συ) is needed in the modeling of the land–atmosphere exchanges of momentum, energy, water, and trace gases. From global 1-km, 10-day composite Advanced Very High Resolution Radiometer normalized difference vegetation index (NDVI) data from April 1992 to March 1993, global 1-km συ is derived based on the annual maximum NDVI value for each pixel in comparison with the NDVI value that corresponds to 100% vegetation cover for each International Geosphere–Biosphere Program land cover type. This dataset is pixel dependent but season independent, with the seasonal variation of vegetation greenness in a pixel accounted for by the leaf area index. The authors’ algorithm is found to be insensitive to the use of a specific land cover classification. In comparison with an independent dataset derived by DeFries et al. by using a more sophisticated statistical approach, the current dataset has a similar spatial distribution but systematically smaller συ (particularly over shrublands and barren land cover). It also gives συ values that overall are consistent with those derived from higher-resolution aircraft and satellite data over Arizona and field-survey data over Germany.

The Vegetative Cover Conversion (VCC) product is designed to serve as a global alarm for land cover change caused by anthropogenic activities and extreme natural events. MODIS 250 m surface reflectance data availability was limited both spatially and temporally in the first year after launch due to processing system constraints. To address this situation, the VCC algorithms were applied to available MODIS 250 m Level 1B radiance data to test the VCC change detection algorithms presented in this paper. Five data sets of MODIS Level 1B 250 m data were collected for the year 2000, representing: (1) Idaho–Montana wildfires; (2) the Cerro Grande prescribed fire in New Mexico; (3) flood in Cambodia; (4) Thailand–Laos flood retreat; and (5) deforestation in southern Brazil. Decision trees are developed for each of the VCC change detection methods for each of these six cases. These decision trees are to be used for updating the look-up tables required by the VCC production code. For these change detection cases, the VCC change detection methods worked reasonably well. In the Idaho–Montana wildfire case, a fire perimeter polygon data set compiled by the USDA Forest Service was used to validate the output of the VCC change detection methods. Although the VCC output identified only 32% of the burned pixels within the ground observed Idaho–Montana fire perimeter polygons, the detection accuracy of the VCC output did reach 99% when the VCC product is considered as an alarm system identifying the occurrence of the change in an area. For other cases, the detection accuracy in per-pixel terms of the VCC output ranges from 55% to 90% against reference change bitmaps that were created by image interpretation. Look-up tables created with AVHRR and Landsat Thematic Mapper data require modifications for the MODIS data due to differences in radiometric response between MODIS and the heritage instruments. The applications presented in this paper also evaluate the relative performance of each of the five change detection methods used as VCC algorithms. Conclusions reached in this paper will be used for future refinement of the VCC product.

Nelson, G. C., E. Bennett, A. A. Berhe, K. Cassman, R. DeFries, T. Dietz, A. Dobermann, A. Dobson, A. Janetos, M. Levy, D. Marco, N. Nakicenovic, B. O’Neill, R. Norgaard, G. Petschel-Held, D. Ojima, P. Pingali, R. Watson, and M. Zurek. 2006. Anthropogenic drivers of ecosystem change: an overview. Ecology and Society 11(2): 29.

As an alternative to the traditional approach of using predefined classification schemes with discrete numbers of cover types to describe the geographic distribution of vegetation over the Earth's land surface, we apply a linear mixture model to derive global continuous fields of percentage woody vegetation, herbaceous vegetation and bare ground from 8 km Advanced Very High Resolution Radiometer (AVHRR) Pathfinder Land data. Linear discriminants for input into the mixture model are derived from 30 metrics representing the annual phenological cycle, using training data derived from a global network of scenes acquired by Landsat. We test the stability and robustness of the method by assessing the consistency of results derived independently for each year in the 1982 to 1994 AVHRR data set. For those forested locations where land cover variability would not be expected, the percentage woody estimates displayed standard deviations over the 12 years of less than 10%. Problems with the method occur in high latitudes where snow cover in some years and not others produces inconsistencies in the continuous fields. Overall, the results suggest that the method produces fairly consistent results despite apparent problems with artifacts in the multi-year AVHRR data set due to calibration problems, aerosols and other atmospheric effects, bidirectional effects, changes in equatorial crossing time, and other factors. Comparison of continuous fields with other land cover data sets derived from remote sensing suggests 69% to 84% agreement in the per cent woody field, with the highest agreement when per cent woody is averaged over the 12 years. In comparison with regional data sets for the US and Bolivia, the method overestimates per cent woody vegetation for grassland and sparsely wooded locations. We conclude that the method, with possible refinements and more sophisticated methods to include multiple endmembers, improved estimates of endmember values and nonlinear responses of vegetation to proportional cover, can potentially be used to indicate changes in land cover characteristics over time using multi-year data sets as inputs when perfect calibration and consistency between years cannot be assumed.

Operational monitoring of land cover from satellite data will require automated procedures for analyzing large volumes of data. We propose multiple criteria for assessing algorithms for this task. In addition to standard classification accuracy measures, we propose criteria to account for computational resources required by the algorithms, stability of the algorithms, and robustness to noise in the training data. We also propose that classification accuracy take account, through estimation of misclassification costs, of unequal consequences to the user depending on which cover types are confused. In this article, we apply these criteria to three variants of decision tree classifiers, a standard decision tree implemented in C5.0 and two techniques recently proposed in the machine learning literature known as “bagging” and “boosting.” Each of these algorithms are applied to two data sets, a global land cover classification from 8 km AVHRR data and a Landsat Thematic Mapper scene in Peru. Results indicate comparable accuracy of the three variants of the decision tree algorithms on the two data sets, with boosting providing marginally higher accuracies. The bagging and boosting algorithms, however, are both substantially more stable and more robust to noise in the training data compared with the standard C5.0 decision tree. The bagging algorithm is most costly in terms of computational resources while the standard decision tree is least costly. The results illustrate that the choice of the most suitable algorithm requires consideration of a suite of criteria in addition to the traditional accuracy measures and that there are likely to be trade-offs between algorithm performance and required computational resources.

The geographic distribution of vegetation over the Earth's land surface is traditionally described using classification schemes with discrete numbers of vegetation types. When such land cover data sets are used as boundary conditions in Earth system models, abrupt boundaries and unrealistic homogeneity are introduced into parameter estimates. This paper proposes an alternative approach to describe global land cover with continuous fields of vegetation characteristics. A linear mixture model is applied to 1‐km advanced very high resolution radiometer data to estimate proportional cover for three important vegetation characteristics: life form (percent woody vegetation, percent herbaceous vegetation, and percent bare ground), leaf type (percent needleleaf and percent broadleaf), and leaf duration (percent evergreen and percent deciduous). Linear discriminants for input into the mixture model are derived from 30 metrics representing the annual phenological cycle. Through comparison with training data derived from a global network of Landsat multispectral scanner scenes, we conclude that the linear assumption implicit in the linear mixture model is not severely violated. The linear relationships between percent cover as determined from the training data and the linear discriminants are used to estimate end‐member values, and the mixture model is applied to derive the seven layers of global continuous fields. The availability of Moderate Resolution Imaging Spectroradiometer data in the future holds promise for refining the simple technique used in this paper to derive improved global continuous fields.

Global land surface characteristics are important boundary conditions for global models that describe exchanges of water, energy, and carbon dioxide between the atmosphere and biosphere. Existing data sets of global land cover are based on classification schemes that characterize each grid cell as a discrete vegetation type. Consequently, parameter fields derived from these data sets are dependent on the particular scheme and the number of vegetation types it includes. The functional controls on exchanges of water, energy, and carbon dioxide between the atmosphere and biosphere are now well enough understood that it is increasingly feasible to model these exchanges using a small number of vegetation characteristics that either are related to or closely related to the functional controls. Ideally, these characteristics would be mapped as continuous distributions to capture mixtures and gradients in vegetation within the cell size of the model. While such an approach makes it more difficult to build models from detailed observations at a small number of sites, it increases the potential for capturing functionally important variation within, as well as between, vegetation types. Globally, the vegetation characteristics that appear to be most important in controlling fluxes of water, energy, and carbon dioxide include (1) growth form (tree, shrub, herb), (2) seasonality of woody vegetation (deciduous, evergreen), (3) leaf type (broadleaf, coniferous), (4) photosynthetic pathway of nonwoody vegetation (C 3 , C 4 ), (5) longevity (annual, perennial), and (6) type and intensity of disturbance (e.g., cultivation, fire history). Many of these characteristics can be obtained through remote sensing, though some require ground‐based information. The minimum number and the identity of the required land surface characteristics almost certainly vary with the intended objective, but the philosophy of driving models with continuous distributions of a small number of land surface characteristics is likely to be applicable to a broad range of problems.

Fires in South America cause forest degradation and contribute to carbon emissions associated with land use change. We investigated the relationship between year-to-year changes in fire activity in South America and sea surface temperatures. We found that the Oceanic Niño Index was correlated with interannual fire activity in the eastern Amazon, whereas the Atlantic Multidecadal Oscillation index was more closely linked with fires in the southern and southwestern Amazon. Combining these two climate indices, we developed an empirical model to forecast regional fire season severity with lead times of 3 to 5 months. Our approach may contribute to the development of an early warning system for anticipating the vulnerability of Amazon forests to fires, thus enabling more effective management with benefits for climate and air quality.

A simulation experiment was carried out to analyse the effects of the modulation transfer function on our ability to estimate the proportions of land cover within a pixel by linear mixture modelling. In the simulated landscape the proportion of each land cover type in every pixel was known exactly. The standard error of the estimate (SEE) between percentages derived from mixture modelling and the actual land cover percentages was 11%. Substantial improvements in estimating the percentages can be obtained simply by deriving estimates for pixels of twice the original dimensions, the SEE dropping to 4.16%, though this is with the obvious consequence of a final product with a coarser spatial resolution. Alternatively by deconvolving the input bands using a linear approximation of the point spread function the SEE can be reduced by almost as much, namely to 5.11%. If we combine the two approaches, by first doconvolving the bands, estimating the percentages and then aggregating resultant pixels to twice their original linear dimensions, the SEE drops to 2.24%.

Abstract Fire‐driven deforestation is the major source of carbon emissions from Amazonia. Recent expansion of mechanized agriculture in forested regions of Amazonia has increased the average size of deforested areas, but related changes in fire dynamics remain poorly characterized. We estimated the contribution of fires from the deforestation process to total fire activity based on the local frequency of active fire detections from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensors. High‐confidence fire detections at the same ground location on 2 or more days per year are most common in areas of active deforestation, where trunks, branches, and stumps can be piled and burned many times before woody fuels are depleted. Across Amazonia, high‐frequency fires typical of deforestation accounted for more than 40% of the MODIS fire detections during 2003–2007. Active deforestation frontiers in Bolivia and the Brazilian states of Mato Grosso, Pará, and Rondônia contributed 84% of these high‐frequency fires during this period. Among deforested areas, the frequency and timing of fire activity vary according to postclearing land use. Fire usage for expansion of mechanized crop production in Mato Grosso is more intense and more evenly distributed throughout the dry season than forest clearing for cattle ranching (4.6 vs. 1.7 fire days per deforested area, respectively), even for clearings >200 ha in size. Fires for deforestation may continue for several years, increasing the combustion completeness of cropland deforestation to nearly 100% and pasture deforestation to 50–90% over 1–3‐year timescales typical of forest conversion. Our results demonstrate that there is no uniform relation between satellite‐based fire detections and carbon emissions. Improved understanding of deforestation carbon losses in Amazonia will require models that capture interannual variation in the deforested area that contributes to fire activity and variable combustion completeness of individual clearings as a function of fire frequency or other evidence of postclearing land use.

Abstract Human modification of the landscape potentially affects exchanges of energy and water between the terrestrial biosphere and the atmosphere. This study develops a possible scenario for land cover in the year 2050 based on results from the IMAGE 2 (Integrated Model to Assess the Greenhouse Effect) model, which projects land‐cover changes in response to demographic and economic activity. We use the land‐cover scenario as a surface boundary condition in a biophysically‐based land‐surface model coupled to a general circulation model for a 15‐years simulation with prescribed sea surface temperature and compare with a control run using current land cover. To assess the sensitivity of climate to anthropogenic land‐cover change relative to the sensitivity to decadal‐scale interannual variations in vegetation density, we also carry out two additional simulations using observed normalized difference vegetation index (NDVI) from relatively low (1982–83) and high (1989–90) years to describe the seasonal phenology of the vegetation. In the past several centuries, large‐scale land‐cover change occurred primarily in temperate latitudes through conversion of forests and grassland to highly productive cropland and pasture. Several studies in the literature indicate that past changes in surface climate resulting from this conversion had a cooling effect owing to changes in vegetation morphology (increased albedo). In contrast, this study indicates that future land‐cover change, likely to occur predominantly in the tropics and subtropics, has a warming effect governed by physiological rather than morphological mechanisms. The physiological mechanism is to reduce carbon assimilation and consequently latent relative to sensible heat flux resulting in surface temperature increases up to 2 °C and drier hydrologic conditions in locations where land cover was altered in the experiment. In addition, in contrast to an observed decrease in diurnal temperature range (DTR) over land expected with greenhouse warming, results here suggest that future land‐cover conversion in tropics could increase the DTR resulting from decreased evaporative cooling during the daytime. For grid cells with altered land cover, the sensitivity of surface temperature to future anthropogenic land‐cover change is generally within the range induced by decadal‐scale interannual variability in vegetation density in temperate latitudes but up to 1.5 °C warmer in the tropics.

Comparisons of MODIS inputs appropriate to mapping land cover at different scales are made using global training data and a SAFARI 2000 validation database from western Zambia. Multiple single‐date images, 40‐day composites and multitemporal annual metrics from the MODIS sensor are tested in mapping percent tree cover. While the metrics outperform the composites at the global scale and are comparable to composites at the continental scale, composites outperform the metrics at the local/regional scale of the Zambia test area. Multiple single‐date MODIS imagery are best at mapping the test area, and this points to their utility in mapping at the local/regional scale. However, the overall difference between inputs is less than 1% in terms of standard error. This implies that the metrics and composites, with appropriate training data, can come close to replicating the spatial detail present in single‐date images. Comparing the Zambia test area data with a subset of the global MODIS percent tree cover map in a validation exercise shows that the general tree cover distribution is well represented. However, the global signatures underestimate Kalahari woodlands on sands and do poorly in overestimating tree cover in some dambos and pans. Further work will aim at refining the global signal using multiple validation sites.

The global change research community needs to renew its social contract with society by moving beyond a focus on biophysical limits and toward solution-oriented research to provide realistic, context-specific pathways to a sustainable future. A focus on planetary opportunities is based on the premise that societies adapt to change and have historically implemented solutions—for example, to protect watersheds, improve food security, and reduce harmful atmospheric emissions. Daunting social and biophysical challenges for achieving a sustainable future demand that the global change research community work to provide underpinnings for workable solutions at multiple scales of governance. Global change research must reorient itself from a focus on biophysically oriented, global-scale analysis of humanity's negative impact on the Earth system to consider the needs of decisionmakers from household to global scales.

Measured and modeled point spread functions (PSF) of sensor systems indicate that a significant portion of the recorded signal of each pixel of a satellite image originates from outside the area represented by that pixel. This hinders the ability to derive surface information from satellite images on a per-pixel basis. In this study, the impact of the PSF of the Moderate Resolution Imaging Spectroradiometer (MODIS) 250 m bands was assessed using four images representing different landscapes. Experimental results showed that though differences between pixels derived with and without PSF effects were small on the average, the PSF generally brightened dark objects and darkened bright objects. This impact of the PSF lowered the performance of a support vector machine (SVM) classifier by 5.4% in overall accuracy and increased the overall root mean square error (RMSE) by 2.4% in estimating subpixel percent land cover. An inversion method based on the known PSF model reduced the signals originating from surrounding areas by as much as 53%. This method differs from traditional PSF inversion deconvolution methods in that the PSF was adjusted with lower weighting factors for signals originating from neighboring pixels than those specified by the PSF model. By using this deconvolution method, the lost classification accuracy due to residual impact of PSF effects was reduced to only 1.66% in overall accuracy. The increase in the RMSE of estimated subpixel land cover proportions due to the residual impact of PSF effects was reduced to 0.64%. Spatial aggregation also effectively reduced the errors in estimated land cover proportion images. About 50% of the estimation errors were removed after applying the deconvolution method and aggregating derived proportion images to twice their dimensional pixel size.

Recent drought events underscore the vulnerability of Amazon forests to understorey fires. The long-term impact of fires on biodiversity and forest carbon stocks depends on the frequency of fire damages and deforestation rates of burned forests. Here, we characterized the spatial and temporal dynamics of understorey fires (1999-2010) and deforestation (2001-2010) in southern Amazonia using new satellite-based estimates of annual fire activity (greater than 50 ha) and deforestation (greater than 10 ha). Understorey forest fires burned more than 85 500 km(2) between 1999 and 2010 (2.8% of all forests). Forests that burned more than once accounted for 16 per cent of all understorey fires. Repeated fire activity was concentrated in Mato Grosso and eastern Pará, whereas single fires were widespread across the arc of deforestation. Routine fire activity in Mato Grosso coincided with annual periods of low night-time relative humidity, suggesting a strong climate control on both single and repeated fires. Understorey fires occurred in regions with active deforestation, yet the interannual variability of fire and deforestation were uncorrelated, and only 2.6 per cent of forests that burned between 1999 and 2008 were deforested for agricultural use by 2010. Evidence from the past decade suggests that future projections of frontier landscapes in Amazonia should separately consider economic drivers to project future deforestation and climate to project fire risk.

We are deeply disturbed by the recent escalation of political assaults on scientists in general and on climate scientists in particular. All citizens should understand some basic scientific facts. There is always some uncertainty associated with scientific conclusions; science never absolutely proves anything. When someone says that society should wait until scientists are absolutely certain before taking any action, it is the same as saying society should never take action. For a problem as potentially catastrophic as climate change, taking no action poses a dangerous risk for our planet.

Oil palm expansion is a major threat to forest conservation in the tropics. Oil palm can also be a sustainable economic alternative if incentives for expansion outside forests are set in place. Consistent methods to monitor the time and location of oil palm expansion and the area converted from different land covers are essential for the success of such incentives. We developed methods to detect and quantify annual land cover changes associated with oil palm expansion in the Peruvian Amazon between 2001 and 2010 at two spatial scales and for two production modes. At the coarse scale, comprising the whole Peruvian Amazon, we used MODIS data to detect forest conversion to large-scale, industrial oil palm plantations based on metrics characterizing temporal changes in vegetation greenness associated with the conversion. At the fine scale, we used data from the satellite sensors Landsat TM/ETM + and ALOS-PALSAR to map and quantify the area from different land covers converted into large and small-scale oil palm plantations annually, in a focus area near the city of Pucallpa. Estimates were obtained from the elaboration and further combination of maps representing oil palm plantations by ages in 2010 and non-oil palm land covers in each year between 2001 and 2010. Validation data were obtained in the field and from geospatial information from previous studies. At the coarse scale, MODIS detected deforestation in 73% of training events larger than 50 ha. Detected events added up to 95% of the training areas. Total area converted to oil palm annually was quantified visually by using data from Landsat TM/ETM + with 96.3% accuracy. At the fine scale, the combination of data from Landsat TM/ETM + and ALOS-PALSAR identified oil palm expansion in areas larger than 5 ha with 94% accuracy and the year of expansion with an uncertainty of ± 1.3 years. This work underscores the need for data from multiple satellite sensors for a comprehensive monitoring of oil palm expansion, considering needs for information not only on the area expanded but also the time of conversion and land cover transitions associated with large- and small-scale plantations.

Fires associated with agricultural and plantation development in Indonesia impact ecosystem services and release emissions into the atmosphere that degrade regional air quality and contribute to greenhouse gas concentrations. In this study, we estimate the relative contributions of the oil palm, timber (for wood pulp and paper), and logging industries in Sumatra and Kalimantan to land cover change, fire activity, and regional population exposure to smoke concentrations. Concessions for these three industries cover 21% and 49% of the land area in Sumatra and Kalimantan respectively, with the highest overall area in lowlands on mineral soils instead of more carbon-rich peatlands. In 2012, most remaining forest area was located in logging concessions for both islands, and for all combined concessions, there was higher remaining lowland and peatland forest area in Kalimantan (45% and 46%, respectively) versus Sumatra (20% and 27%, respectively). Emissions from all combined concessions comprised 41% of total fire emissions (within and outside of concession boundaries) in Sumatra and 27% in Kalimantan for the 2006 burning season, which had high fire activity relative to decadal emissions. Most fire emissions were observed in concessions located on peatlands and non-forested lowlands, the latter of which could include concessions that are currently under production, cleared in preparation for production, or abandoned lands. For the 2006 burning season, timber concessions from Sumatra (47% of area and 88% of emissions) and oil palm concessions from Kalimantan (33% of area and 67% of emissions) contributed the most to concession-related fire emissions from each island. Although fire emissions from concessions were higher in Kalimantan, emissions from Sumatra contributed 63% of concession-related smoke concentrations for the population-weighted region because fire sources were located closer to population centers. In order to protect regional public health, our results highlight the importance of limiting the use of fire by the timber and oil palm industries, particularly on concessions that contain peatlands and non-forest, by such methods as improving monitoring systems, local-level management, and enforcement of existing fire bans.

Mitigating crop and livestock loss to wildlife and improving compensation distribution are important for conservation efforts in landscapes where people and wildlife co-occur outside protected areas. The lack of rigorously collected spatial data poses a challenge to management efforts to minimize loss and mitigate conflicts. We surveyed 735 households from 347 villages in a 5154 km(2) area surrounding Kanha Tiger Reserve in India. We modeled self-reported household crop and livestock loss as a function of agricultural, demographic and environmental factors, and mitigation measures. We also modeled self-reported compensation received by households as a function of demographic factors, conflict type, reporting to authorities, and wildlife species involved. Seventy-three percent of households reported crop loss and 33% livestock loss in the previous year, but less than 8% reported human injury or death. Crop loss was associated with greater number of cropping months per year and proximity to the park. Livestock loss was associated with grazing animals inside the park and proximity to the park. Among mitigation measures only use of protective physical structures were associated with reduced livestock loss. Compensation distribution was more likely for tiger related incidents, and households reporting loss and located in the buffer. Average estimated probability of crop loss was 0.93 and livestock loss was 0.60 for surveyed households. Estimated crop and livestock loss and compensation distribution were higher for households located inside the buffer. Our approach modeled conflict data to aid managers in identifying potential conflict hotspots, influential factors, and spatially maps risk probability of crop and livestock loss. This approach could help focus allocation of conservation efforts and funds directed at conflict prevention and mitigation where high densities of people and wildlife co-occur.

Urbanization is currently a major force in tropical land use transitions as economic activities aggregate in urban centers, particularly in Asia. This paper examines relationships among urbanization, household energy source, and forest cover at the state level in India using available census, survey, and remote sensing analysis from the 1990s and 2000s. Central questions include (1) how rapidly are urban and rural households switching from traditional to modern fuel sources; and (2) what are the consequences of changing household energy sources for fuelwood demand and forest cover. Country-wide, 30 and 78% of urban and rural households respectively used fuelwood for cooking in 1993. In urban households, the percentage decreased to 22% by 2005 with a shift towards liquefied petroleum gas (LPG). The shift occurred across almost all income classes. In rural areas, the use of LPG increased fourfold but 75% of households still rely on fuelwood. Despite the decline in percentage households using traditional fuels, fuelwood demand continued to increase from 1993 to 2005 at a national scale due to an increasing total number of households. However, 25% of states and union territories experienced declines in rural fuelwood demand and over 70% declines in urban fuelwood demand. Forest cover has remained steady or increased slightly over the time period, reaffirming the conclusion that fuelwood demand may lead to local degradation but not large-scale deforestation. At the state level, increases in percent forest cover between 2000 and 2004 are positively associated with percent of total households that are urban (corresponding to fewer percentage households using wood) but not related to changes in fuelwood demand. Plantations are a primary cause of increases in forest area, where benefits to ecosystem services such as biodiversity and hydrologic function are controversial. Results suggest that households will continue to climb the energy ladder with future urbanization, resulting in substantial development benefits and reduced exposure to indoor air pollution. Implications of reduced fuelwood demand for forest cover are less certain but the limited data suggest that urbanization will promote a transition to increasing forest cover in the Indian context.

This paper reviews the technical capabilities for monitoring deforestation from a pan-tropical perspective in response to the United Nations Framework Convention on Climate Change (UNFCCC) process, which is studying the technical issues surrounding the ability to reduce greenhouse gas emissions from deforestation in developing countries. The successful implementation of such policies requires effective forest monitoring systems that are reproducible, provide consistent results, meet standards for mapping accuracy, and can be implemented from national to pan-tropical levels. Remotely sensed data, supported by ground observations, are crucial to such efforts. Recent developments in global to regional monitoring of forests can contribute to reducing the uncertainties in estimates of emissions from deforestation. Monitoring systems at national levels in developing countries can also benefit from pan-tropical and regional observations, mainly by identifying hot spots of change and prioritizing areas for monitoring at finer spatial scales. A pan-tropical perspective is also required to ensure consistency between different national monitoring systems.

Abstract High-yield agriculture potentially reduces pressure on forests by requiring less land to increase production. Using satellite and field data, we assessed the area deforested by industrial-scale high-yield oil palm expansion in the Peruvian Amazon from 2000 to 2010, finding that 72% of new plantations expanded into forested areas. In a focus area in the Ucayali region, we assessed deforestation for high- and smallholder low-yield oil palm plantations. Low-yield plantations accounted for most expansion overall (80%), but only 30% of their expansion involved forest conversion, contrasting with 75% for high-yield expansion. High-yield expansion minimized the total area required to achieve production but counter-intuitively at higher expense to forests than low-yield plantations. The results show that high-yield agriculture is an important but insufficient strategy to reduce pressure on forests. We suggest that high-yield agriculture can be effective in sparing forests only if coupled with incentives for agricultural expansion into already cleared lands.

Weather variability poses numerous risks to agricultural communities, yet farmers may be able to reduce some of these risks by adapting their cropping practices to better suit changes in weather. However, not all farmers respond to weather variability in the same way. To better identify the causes and consequences of this heterogeneous decision-making, we develop a framework that identifies (1) which socio-economic and biophysical factors are associated with heterogeneous cropping decisions in response to weather variability and (2) which cropping strategies are the most adaptive, considering economic outcomes (e.g., yields and profits). This framework aims to understand how, why, and how effectively farmers adapt to current weather variability; these findings, in turn, may contribute to a more mechanistic and predictive understanding of individual-level adaptation to future climate variability and change. To illustrate this framework, we assessed how 779 farmers responded to delayed monsoon onset in fifteen villages in Gujarat, India during the 2011 growing season, when the monsoon onset was delayed by three weeks. We found that farmers adopted a variety of strategies to cope with delayed monsoon onset, including increasing irrigation use, switching to more drought-tolerant crops, and/or delaying sowing. We found that farmers' access to and choice of strategies varied with their assets, irrigation access, perceptions of weather, and risk aversion. Richer farmers with more irrigation access used high levels of irrigation, and this strategy was associated with the highest yields in our survey sample. Poorer farmers with less secure access to irrigation were more likely to push back planting dates or switch crop type, and economic data suggest that these strategies were beneficial for those who did not have secure access to irrigation. Interestingly, after controlling for assets and irrigation access, we found that cognitive factors, such as beliefs that the monsoon onset date had changed over the last 20 years or risk aversion, were associated with increased adaptation. Our framework illustrates the importance of considering the complexity and heterogeneity of individual decision-making when conducting climate impact assessments or when developing policies to enhance the adaptive capacity of local communities to future climate variability and change.

Improvements in agricultural production have drastically increased grain yields in the past half-century. Despite this growth in productivity and calories available per capita, malnutrition – both undernutrition and, increasingly, overweight – remains pervasive. Though nutrition is critical to human health, it has yet to be systematically integrated into assessments of agricultural and food systems. Using three complementary diversity metrics, we find strong associations between nutritional diversity of national food supplies and key human health outcomes, while controlling for socio-economic factors. For low-income countries the diversity of agricultural goods produced by a country is a strong predictor for food supply diversity; for middle- and high-income countries national income and trade are better predictors. Our results highlight the importance of diversity in national food systems for human health. We provide metrics for agricultural and food security policies to consider nutritional diversity.

The food security of smallholder farmers is vulnerable to climate change and climate variability. Cropping intensity, the number of crops planted annually, can be used as a measure of food security for smallholder farmers given that it can greatly affect net production. Current techniques for quantifying cropping intensity may not accurately map smallholder farms where the size of one field is typically smaller than the spatial resolution of readily available satellite data. We evaluated four methods that use multi-scalar datasets and are commonly used in the literature to assess cropping intensity of smallholder farms: 1) the Landsat threshold method, which identifies if a Landsat pixel is cropped or uncropped during each growing season, 2) the MODIS peak method, which determines if there is a phenological peak in the MODIS Enhanced Vegetation Index time series during each growing season, 3) the MODIS temporal mixture analysis, which quantifies the sub-pixel heterogeneity of cropping intensity using phenological MODIS data, and 4) the MODIS hierarchical training method, which quantifies the sub-pixel heterogeneity of cropping intensity using hierarchical training techniques. Each method was assessed using four criteria: 1) data availability, 2) accuracy across different spatial scales (at aggregate scales 250 × 250 m, 1 × 1 km, 5 × 5 km, and 10 × 10 km), 3) ease of implementation, and 4) ability to use the method over large spatial and temporal scales. We applied our methods to two regions in India (Gujarat and southeastern Madhya Pradesh) that represented diversity in crop type, soils, climatology, irrigation access, cropping intensity, and field size. We found that the Landsat threshold method is the most accurate (R2 ≥ 0.71 and RMSE ≤ 0.14), particularly at smaller scales of analysis. Yet given the limited availability of Landsat data, we find that the MODIS hierarchical training method meets multiple criteria for mapping cropping intensity over large spatial and temporal scales. Furthermore, the adjusted R2 between predicted and validation data generally increased and the RMSE decreased with spatial aggregation ≥ 5 × 5 km (R2 up to 0.97 and RMSE as low as 0.00). Our model accuracy varied based on the region and season of analysis and was lowest during the summer season in Gujarat when there was high sub-pixel heterogeneity due to sparsely cropped agricultural land-cover. While our results specifically apply to our study regions in India, they most likely also apply to smallholder agriculture in other locations across the globe where the same types of satellite data are readily available.

Background: Landscape fires in South America have considerable impacts on ecosystems, air quality and the climate system. We examined long-term trends and interannual variability of forest, savanna and agricultural fires for the continent during 2001–2012 using multiple satellite-derived fire products. Results: The annual number of active fires in tropical forests increased significantly during 2001–2005. Several satellite-derived metrics, including fire persistence, indicated that this trend was mostly driven by deforestation. Fires between 2005 and 2012 had a small decreasing trend and large year-to-year changes that were associated with climate extremes. Fires in savannas and evergreen forests increased in parallel during drought events in 2005, 2007 and 2010, suggesting similar regional climate controls on fire behavior. Deforestation fire intensity (the number of fires per unit of deforested area) increased significantly within the Brazilian Amazon in areas with small-scale deforestation. Conclusion: Fires associated with forest degradation are becoming an increasingly important component of the fire regime and associated carbon emissions.

Decarbonizing the economy must remain a critical priority

Large-scale cattle and crop production are the primary drivers of deforestation in the Amazon today. Such land-use changes can degrade stream ecosystems by reducing connectivity, changing light and nutrient inputs, and altering the quantity and quality of streamwater. This study integrates field data from 12 catchments with satellite-derived information for the 176,000 km(2) upper Xingu watershed (Mato Grosso, Brazil). We quantify recent land-use transitions and evaluate the influence of land management on streamwater temperature, an important determinant of habitat quality in small streams. By 2010, over 40 per cent of catchments outside protected areas were dominated (greater than 60% of area) by agriculture, with an estimated 10,000 impoundments in the upper Xingu. Streams in pasture and soya bean watersheds were significantly warmer than those in forested watersheds, with average daily maxima over 4°C higher in pasture and 3°C higher in soya bean. The upstream density of impoundments and riparian forest cover accounted for 43 per cent of the variation in temperature. Scaling up, our model suggests that management practices associated with recent agricultural expansion may have already increased headwater stream temperatures across the Xingu. Although increased temperatures could negatively impact stream biota, conserving or restoring riparian buffers could reduce predicted warming by as much as fivefold.

Understory fires in Amazon forests alter forest structure, species composition, and the likelihood of future disturbance. The annual extent of fire-damaged forest in Amazonia remains uncertain due to difficulties in separating burning from other types of forest damage in satellite data. We developed a new approach, the Burn Damage and Recovery (BDR) algorithm, to identify fire-related canopy damages using spatial and spectral information from multi-year time series of satellite data. The BDR approach identifies understory fires in intact and logged Amazon forests based on the reduction and recovery of live canopy cover in the years following fire damages and the size and shape of individual understory burn scars. The BDR algorithm was applied to time series of Landsat (1997–2004) and MODIS (2000–2005) data covering one Landsat scene (path/row 226/068) in southern Amazonia and the results were compared to field observations, image-derived burn scars, and independent data on selective logging and deforestation. Landsat resolution was essential for detection of burn scars < 50 ha, yet these small burns contributed only 12% of all burned forest detected during 1997–2002. MODIS data were suitable for mapping medium (50–500 ha) and large (> 500 ha) burn scars that accounted for the majority of all fire-damaged forests in this study. Therefore, moderate resolution satellite data may be suitable to provide estimates of the extent of fire-damaged Amazon forest at a regional scale. In the study region, Landsat-based understory fire damages in 1999 (1508 km2) were an order of magnitude higher than during the 1997–1998 El Niño event (124 km2 and 39 km2, respectively), suggesting a different link between climate and understory fires than previously reported for other Amazon regions. The results in this study illustrate the potential to address critical questions concerning climate and fire risk in Amazon forests by applying the BDR algorithm over larger areas and longer image time series.

Fire disturbance in many tropical forests, including peat swamps, has become more frequent and extensive in recent decades. These fires compromise a variety of ecosystem services, among which mitigating global climate change through carbon storage is particularly important for peat swamps. Indonesia holds the largest amount of tropical peat carbon globally, and mean annual CO2 emissions from decomposition of deforested and drained peatlands and associated fires in Southeast Asia have been estimated at ∼2000 Mt y-1. A key component to understanding and therefore managing fire in the region is identifying the land use/land cover classes associated with fire ignitions. We assess the oft-asserted claim that escaped fires from oil palm concessions and smallholder farms near settlements are the primary sources of fire in a peat-swamp forest area in Central Kalimantan, Indonesia, equivalent to around a third of Kalimantan's total peat area. We use the MODIS Active Fire product from 2000 to 2010 to evaluate the fire origin and spread on the land use/land cover classes of legal, industrial oil palm concessions (the only type of legal concession in the study area), non-forest, and forest, as well as in relation to settlement proximity. We find that most fires (68–71%) originate in non-forest, compared to oil palm concessions (17%–19%), and relatively few (6–9%) are within 5 km of settlements. Moreover, most fires started within oil palm concessions and in close proximity to settlements stay within those boundaries (90% and 88%, respectively), and fires that do escape constitute only a small proportion of all fires on the landscape (2% and 1%, respectively). Similarly, a small proportion of fire detections in forest originate from oil palm concessions (2%) and within close proximity to settlements (2%). However, fire ignition density in oil palm (0.055 ignitions km−2) is comparable to that in non-forest (0.060 km-2 ignitions km-2), which is approximately ten times that in forest (0.006 ignitions km−2). Ignition density within 5 km of settlements is the highest at 0.125 ignitions km−2. Furthermore, increased anthropogenic activity in close proximity to oil palm concessions and settlements produces a detectable pattern of fire activity. The number of ignitions decreases exponentially with distance from concessions; the number of ignitions initially increases with distance from settlements, and, around from 7.2 km, then decreases with distance from settlements. These results refute the claim that most fires originate in oil palm concessions, and that fires escaping from oil palm concessions and settlements constitute a major proportion of fires in this study region. However, there is a potential for these land use types to contribute substantially to the fire landscape if their area expands. Effective fire management in this area should therefore target not just oil palm concessions, but also non-forested, degraded areas where ignitions and fires escaping into forest are most likely to occur.

The Program on Ecosystem Change and Society (PECS), a new initiative within the ICSU global change programs, aims to integrate research on the stewardship of social–ecological systems, the services they generate, and the relationships among natural capital, human wellbeing, livelihoods, inequality and poverty. The vision of PECS is a world where human actions have transformed to achieve sustainable stewardship of social–ecological systems. The goal of PECS is to generate the scientific and policy-relevant knowledge of social–ecological dynamics needed to enable such a shift, including mitigation of poverty. PECS is a coordinating body for diverse independently funded research projects, not a funder of research. PECS research employs a range of transdisciplinary approaches and methods, with comparative, place-based research that is international in scope at the core.

Humanity faces the grand challenge of feeding a growing, more affluent population in the coming decades while reducing the environmental burden of agriculture. Approaches that integrate food security and environmental goals offer promise for achieving a more sustainable global food system, yet little work has been done to link potential solutions with agricultural policies. Taking the case of cereal production in India, we use a process-based crop water model and government data on food production and nutrient content to assess the implications of various crop-shifting scenarios on consumptive water demand and nutrient production. We find that historical growth in wheat production during the rabi (non-monsoon) season has been the main driver of the country's increased consumptive irrigation water demand and that rice is the least water-efficient cereal for the production of key nutrients, especially for iron, zinc, and fiber. By replacing rice areas in each district with the alternative cereal (maize, finger millet, pearl millet, or sorghum) with the lowest irrigation (blue) water footprint (WFP), we show that it is possible to reduce irrigation water demand by 33% and improve the production of protein (+1%), iron (+27%), and zinc (+13%) with only a modest reduction in calories. Replacing rice areas with the lowest total (rainfall + irrigation) WFP alternative cereal or the cereal with the highest nutritional yield (metric tons of protein per hectare or kilograms of iron per hectare) yielded similar benefits. By adopting a similar multidimensional framework, India and other nations can identify food security solutions that can achieve multiple sustainability goals simultaneously.