Search

Key message Evolving societal demands and accelerated ecological dynamics due to global change are rapidly altering forest ecosystems and their services. This has prompted the need for advancing forest inventorying and monitoring initatives to expand their scope, improve data collection, foster scientific understanding, and better inform policy responses. Here, we discuss the collaborative processes followed to develop an Advanced Inventorying and Monitoring (AIM) system for Swiss forests. Further, we provide the key messages that emerged from this process which can be of interest to those involved in similar processes at the national/international level.

Key message: Evolving societal demands and accelerated ecological dynamics due to global change are rapidly altering forest ecosystems and their services. This has prompted the need for advancing forest inventorying and monitoring initatives to expand their scope, improve data collection, foster scientific understanding, and better inform policy responses. Here, we discuss the collaborative processes followed to develop an Advanced Inventorying and Monitoring (AIM) system for Swiss forests. Further, we provide the key messages that emerged from this process which can be of interest to those involved in similar processes at the national/international level. Forests are under pressure and going through rapid changes. However, current inventorying and monitoring (IM) programs are often either disjointed, too narrow in their scope and/or do not operate at fine enough temporal resolutions, which may hinder scientific understanding, the timely supply of information, fast decision making, and may result in the sub-optimal use of resources. For these reasons, there is an urgent need for Advanced Forest Inventorying and Monitoring (AIM) programs to (i) achieve expanded relevance (by augmenting data/information across ecosystem properties and trophic levels), (ii) have increased temporal resolution (by tailored data collection frequency), and (iii) make use of technological advances (by incorporating novel tools and technologies). The Advanced Inventorying and Monitoring for Swiss Forests (SwissAIM) initiative was launched in 2020 to address these needs. SwissAIM builds upon the foundation offered by the existing programs (e.g., national forest inventory, long-term forest ecosystem research, biodiversity monitoring). It aims to offer a collaborative and adaptive framework to enable integrated data collection, evaluation, interpretation, analysis, and modeling. Ideally, it will result in a more responsive system with respect to current and predicted biotic/abiotic stressors that will challenge Swiss forests. Developing such a system implies identifying the information needs of different stakeholders (e.g., science, policy, practice), related technical requirements, and governance frameworks. Here, we present (i) the main features of the SwissAIM initiative (vision, scientific questions and variables, governance and engagement), (ii) the main outcomes of the participatory design process (measurements, sampling, and plot design), (iii) the potential transferability of AIM initiatives outside Switzerland (timing, relevance, practicability), and (iv) the key messages that emerged (i.e., need for advancement, integration and transdisciplinarity, statistical underpinning). Since similar needs related to forest inventorying and monitoring are emerging throughout Europe and elsewhere, the objective of this opinion paper is to share our experience and promote a dialog with those interested in developing AIM initiatives in other countries and regions. [ABSTRACT FROM AUTHOR]

Accurate forest assessment is essential to detect and tackle deforestation, especially in emerging economies. In Colombia, three different geo-spatial data sources are available for forest monitoring: the European Space Agency (ESA), the Institute for Hydrology, Meteorology and Environmental Studies (IDEAM), and the Global Forest Change Data (GFCD) from the University of Maryland. These information sources have distinct characteristics, purposes, and coverage, and their peculiarities can lead to marked differences in the results when they are used to produce forest cover maps. In this study, we determine the optimal forest threshold for GFCD and assess the accuracy of the three data sources in mapping forests, on the basis of a stratified sample of sites, with Colombian ecoregions used as strata. At each site, the classification into forest or non-forest, according to one of the sources, is compared with reference data collected through Google Earth imagery and landscape photographs. Accuracy measures are produced at both the ecoregion and national level. IDEAM and GFCD prove to be quite accurate in most cases, and each of them turns out to be the best forest map in about half of the ecoregions. GFCD's optimal threshold is found to be equal to 90% in almost all those ecoregions for which it represents the best performing data set. [ABSTRACT FROM AUTHOR]

Abstract The timely and accurate monitoring of forest resources is becoming of increasing importance in light of the multi-functionality of these ecosystems and their increasing vulnerability to climate change. Remote sensing observations of tree cover and systematic ground observations from National Forest Inventories (NFIs) represent the two major sources of information to assess forest area and use. The specificity of two methods is calling for an in-depth analysis of their strengths and weaknesses and for the design of novel methods emerging from the integration of satellite and surface data. On this specific debate, a recent paper by Breidenbach et al. published in this journal suggests that the detection of a recent increase in EU forest harvest rate—as reported in Nature by Ceccherini et al.—is largely due to technical limitations of satellite-based mapping. The article centers on the difficulty of the approaches to estimate wood harvest based on remote sensing. However, it does not discuss issues with the robustness of validation approaches solely based on NFIs. Here we discuss the use of plot data as a validation set for remote sensing products, discussing potentials and limitations of both NFIs and remote sensing, and how they can be used synergistically. Finally, we highlight the need to collect in situ data that is both relevant and compatible with remote sensing products within the European Union.

The spatial and temporal distribution of windthrows in the forest zone of Western Siberia for the period 2001–2020 is considered. Windthrow data are obtained from Landsat and Sentinel-2 satellite images, Global Forest Change project data, and high spatial resolution images from open mapping services. Data from the network of weather stations, images from meteorological satellites, and information on hazardous weather events observed in the region are used to specify dates and time of storm events. The compiled database includes 25 774 elementary damaged areas, which refer to 265 different windthrows. These windthrows were caused by 158 different storm events. The data are available at https://doi.org/10.6084/m9.figshare. 19582786.v1. The total area of windthrows is 508.3 km2 or 0.04% of the total forested area, which is four times less than in the European Russia (ER) for the same period. More than 67% of windthrows included in the database were caused by tornadoes. They also account for 25.4% of the total area, which is much more than in the ER. Squall-induced windthrows account for 30% of cases and 40% of the total area, and windthrows caused by nonconvective events account for 2.5% of cases and 34.5% of the total area. The maximum density of windthrows is found in the Kemerovo oblast, as well as in the west of Tomsk oblast and in the southeast of the Khanty-Mansi autonomous okrug. The largest number of windthrows was recorded in 2007 and 2010. Tornadoes and squalls, which cause windthrows, are most often observed in June, and windthrows caused by nonconvective events were observed in October–November in the Ural and Kuznetsk Alatau mountains. Several large outbreaks of squalls and tornadoes are identified. However, with respect to the area of forest damage, they are 10–15 times less than the largest similar events on ER. The area of windthrows in Western Siberia is 77.5 times less than the area of burns during the same period, with a ratio of 3.5/1 for ER. [ABSTRACT FROM AUTHOR]

Protected areas (PAs) offer safe havens for threatened species, but their effectiveness is jeopardised due to climate change and habitat fragmentation in their surroundings. Species are forced into the unhospitable matrix in search of more favourable areas as climate conditions change, leading to negative effects on biodiversity. For red-listed forest species, the accelerating climate change can strengthen the negative effects of logging around PAs. In this work, we examine climate-change and land-use threats to biologically valuable PAs with red-listed species occurrences in Finland. Where the highest combined threats occur, do they coincide with accumulations of red-listed species, and how much do they vary between different boreal zones? This provides important knowledge for conservation planning in a managed matrix amid climate change. We use three spatial scales (100 m, 1 km, 10 km) to explore where the effects of clear-cutting have been the most intense during this millennium: right outside the PA border, exposing PAs to local-scale edge effects, or at landscape scales disrupting species movements due to degraded connectivity and increased fragmentation. For each PA, we also quantify how the January mean temperature (TJan), growing degree days (GDD5) and water balance (WAB) conditions are predicted to change by the end of the century under the RCP4.5 scenario. Additionally, for these variables, we estimate the velocity of the changes within valuable PAs. The proportion of forests which underwent clear-cutting was the highest in the more productive southern boreal zone and at the landscape scale of 10 km around the PAs. The combined threats of land use and climate change were evident, as there were positive correlations between logging levels around PAs with red-listed species and both the velocity values of GDD5 and TJan in the PAs. TJan is changing particularly fast in the northern boreal zone, where the coldest current climate conditions are projected to disappear by the end of the century. GDD5 in turn is changing fastest in the topographically flat southern boreal zone. GDD5 and TJan had generally a positive relationship with the magnitude of clear-cuts in a 10 km buffer zone. However, the direction of this relationship varied between the boreal zones. In climate-wise conservation planning for red-listed boreal forest species, multiple elements are needed. First and foremost, valuable sites for the red-listed species in the PA network should be identified and the climate change and land use-based threats to them measured. This information provides an elementary basis for additional threat analyses focusing on species habitat requirements and dispersal ecology. Together, this information can be used to assess the potential of different conservation measures to mitigate the combined climate-change and land-use risks, and ultimately to improve the conservation status of red-listed forest species.

The timely and accurate monitoring of forest resources is becoming of increasing importance in light of the multi-functionality of these ecosystems and their increasing vulnerability to climate change. Remote sensing observations of tree cover and systematic ground observations from National Forest Inventories (NFIs) represent the two major sources of information to assess forest area and use. The specificity of two methods is calling for an in-depth analysis of their strengths and weaknesses and for the design of novel methods emerging from the integration of satellite and surface data. On this specific debate, a recent paper by Breidenbach et al. published in this journal suggests that the detection of a recent increase in EU forest harvest rate—as reported in Nature by Ceccherini et al.—is largely due to technical limitations of satellite-based mapping. The article centers on the difficulty of the approaches to estimate wood harvest based on remote sensing. However, it does not discuss issues with the robustness of validation approaches solely based on NFIs. Here we discuss the use of plot data as a validation set for remote sensing products, discussing potentials and limitations of both NFIs and remote sensing, and how they can be used synergistically. Finally, we highlight the need to collect in situ data that is both relevant and compatible with remote sensing products within the European Union. [ABSTRACT FROM AUTHOR]

Is forest harvesting increasing in Europe? There is scientific debate about methodological approach and data regarding clearcut increment in Europe but, besides the discussion, there is a general agreement about the need to collect reliable scientifically robust remote sensing data for sound forest policy-making.

Two recent global maps of tree restoration potential have identified vast regions where tree cover could be increased, ranging from 0.9 to 2.3 billion hectares. Both maps, however, emphasized dryland regions, with arid biomes making up 36%–42% of potential restoration area. Dryland biomes have repeatedly been recognized as inappropriate regions for expanding tree cover due to the risks of biodiversity loss, water overconsumption, and fire, so maps that highlight these regions for restoration must sustain careful scrutiny. Here, I show that both recent attempts to map restoration potential in arid regions have been hindered by underlying errors in the global tree cover maps they used. Systematic underestimates of existing sparse tree cover led directly to large overestimates of the potential for tree recovery in drylands. The Atlas of Forest Landscape Restoration Opportunities (Laestadius et al., Unasylva, 2011, 62, 47) overestimated tree restoration potential across a third of arid biomes by between 7% and 20% (55–166 million hectares [Mha]). Similarly, Bastin, Finegold, Garcia, Mollicone, et al. (Science, 2019, 365, 76) overestimated tree restoration potential across all arid biomes by 33%–45% (316–440 Mha). These inaccuracies limit the utility of this research for policy decisions in drylands and overstate the potential for tree planting to address climate change. Given this long‐standing but underappreciated challenge in mapping global tree cover, I propose various ways forward that keep this lesson in mind. To better monitor and restore tree cover, I call for re‐interpretation and correction of existing global maps, and for a new focus on quantifying sparse tree cover in drylands and other systems. [ABSTRACT FROM AUTHOR]

Is forest harvesting increasing in Europe? There is scientific debate about methodological approach and data regarding clearcut increment in Europe but, besides the discussion, there is a general agreement about the need to collect reliable scientifically robust remote sensing data for sound forest policymaking. [ABSTRACT FROM AUTHOR]

Розглянуто методичні засади використання глобальної карти змін лісового покриву Global Forest Change (GFC) для картографування та визначення площі лісів на відносно невеликих територіях в умовах Українського Полісся. Для цього виконано порівняння точності класифікації та інтервальних оцінок площі лісового покриву, отриманих за результатами дешифрування мультиспектрального супутникового знімка RapidEye та карти GFC. Для створення опорного набору даних застосовано випадкову вибірку обсягом понад 4000 спостережень і непараметричний метод класифікації Random Forest (RF). За результатами статистичного аналізу встановлено, що на точність класифікації найбільший вплив мають три спектральні канали супутникового знімка (NIR, Red, Red-Edge) і чотири допоміжні топографічні змінні (X, Y, висота н.р.м., ухил). Для оцінювання показників точності класифікації використано додатковий набір даних обсягом близько 2000 спостережень. Це дозволило встановити помилку класифікаційної моделі, яка становить 1,6 %. Підтверджено, що мультиспектральні супутникові знімки RapidEye є важливим джерелом геопросторової інформації про лісові екосистеми і дають змогу створювати тематичні карти з високою точністю. Внаслідок дослідження виявлено, що карта GFC забезпечує досить високу точність (близько 75-85 %) дешифрування вкритих лісовою рослинністю ділянок, а її дані узгоджуються з фактичними показниками зімкнутості деревостанів. На основі цього можна стверджувати про можливість і доцільність застосування карти GFC для картографування лісового покриву на території Українського Полісся. Загалом треба визнати, що створені на основі даних GFC лісові карти можуть бути доречними за відсутності точніших картографічних матеріалів.

Numerous studies have shown that intact tropical forests account for half of the total terrestrial sink for anthropogenic carbon dioxide. Here, we analyzed and compared changes in three main tropical forest regions from 2000 to 2014, based on time-series analysis and landscape metrics derived from moderate-resolution imaging spectroradiometer data. We examined spatial-pattern changes in percentage of tree cover and net primary production (NPP) for three tropical forest regions—Amazon basin, Congo basin, and Southeast Asia. The results show that: the Amazon basin region had the largest tropical forest area and total NPP and a better continuity of TC distribution; the Southeast Asia region exhibited a sharp decrease in NPP and had comparatively separate spatial patterns of both TC and NPP; and the Congo basin region exhibited a dramatic increase in NPP and had better aggregation of forest NPP distribution. Results also show that aggregative patterns likely correlate with high NPP values. [ABSTRACT FROM AUTHOR]

Global tree cover products face challenges in accurately predicting tree cover across biophysical gradients, such as precipitation or agricultural cover. To generate a natural forest cover map for Costa Rica, biases in tree cover estimation in the most widely used tree cover product (the Global Forest Change product (GFC) were quantified and corrected, and the impact of map biases on estimates of forest cover and fragmentation was examined. First, a forest reference dataset was developed to examine how the difference between reference and GFC-predicted tree cover estimates varied along gradients of precipitation and elevation, and nonlinear statistical models were fit to predict the bias. Next, an agricultural land cover map was generated by classifying Landsat and ALOS PalSAR imagery (overall accuracy of 97%) to allow removing six common agricultural crops from estimates of tree cover. Finally, the GFC product was corrected through an integrated process using the nonlinear predictions of precipitation and elevation biases and the agricultural crop map as inputs. The accuracy of tree cover prediction increased by ≈29% over the original global forest change product (the R2 rose from 0.416 to 0.538). Using an optimized 89% tree cover threshold to create a forest/nonforest map, we found that fragmentation declined and core forest area and connectivity increased in the corrected forest cover map, especially in dry tropical forests, protected areas, and designated habitat corridors. By contrast, the core forest area decreased locally where agricultural fields were removed from estimates of natural tree cover. This research demonstrates a simple, transferable methodology to correct for observed biases in the Global Forest Change product. The use of uncorrected tree cover products may markedly over- or underestimate forest cover and fragmentation, especially in tropical regions with low precipitation, significant topography, and/or perennial agricultural production.

Global Forest Change datasets have the potential to assist countries with national forest measuring, reporting and verification (MRV) requirements. This paper assesses the accuracy of the Global Forest Change data against nationally derived forest change data by comparing the forest loss estimates from the global data with the equivalent data from Guyana for the period 2001–2017. To perform a meaningful comparison between these two datasets, the initial year 2000 forest state needs first to be matched to the definition of forest land cover appropriate to a local national setting. In Guyana, the default definition of 30% tree cover overestimates forest area is by 483,000 ha (18.15%). However, by using a tree canopy cover (i.e., density of tree canopy coverage metric) threshold of 94%, a close match between the Guyana-MRV non-forest area and the Global Forest Change dataset is achieved with a difference of only 24,210 ha (0.91%) between the two maps. A complimentary analysis using a two-stage stratified random sampling design showed the 94% tree canopy cover threshold gave a close correspondence (R2 = 0.98) with the Guyana-MRV data, while the Global Forest Change default setting of 30% tree canopy cover threshold gave a poorer fit (R2 = 0.91). Having aligned the definitions of forest for the Global Forest Change and the Guyana-MRV products for the year 2000, we show that over the period 2001–2017 the Global Forest Change data yielded a 99.34% overall Correspondence with the reference data and a 94.35% Producer’s Accuracy. The Guyana-MRV data yielded a 99.36% overall Correspondence with the reference data and a 95.94% Producer’s Accuracy. A year-by-year analysis of change from 2001–2017 shows that in some years, the Global Forest Change dataset underestimates change, and in other years, such as 2016 and 2017, change is detected that is not forest loss or gain, hence the apparent overestimation. The conclusion is that, when suitably calibrated for percentage tree cover, the Global Forest Change datasets give a good first approximation of forest loss (and, probably, gains). However, in countries with large areas of forest cover and low levels of deforestation, these data should not be relied upon to provide a precise annual loss/gain or rate of change estimate for audit purposes without using independent high-quality reference data.

Satellite imagery of 25−30 m spatial resolution has been recognized as an effective tool for monitoring the spatial and temporal dynamics of forest cover at different scales. However, the precise mapping of forest cover over fragmented landscapes is complicated and requires special consideration. We have evaluated the performance of four global forest products of 25−30 m spatial resolution within three flatland subregions of Ukraine that have different forest cover patterns. We have explored the relationship between tree cover extracted from the global forest change (GFC) and relative stocking density of forest stands and justified the use of a 40% tree cover threshold for mapping forest in flatland Ukraine. In contrast, the canopy cover threshold for the analogous product Landsat tree cover continuous fields (LTCCF) is found to be 25%. Analysis of the global forest products, including discrete forest masks Global PALSAR-2/PALSAR Forest/Non-Forest Map (JAXA FNF) and GlobeLand30, has revealed a major misclassification of forested areas under severe fragmentation patterns of landscapes. The study also examined the effectiveness of forest mapping over fragmented landscapes using dense time series of Landsat images. We collected 1548 scenes of Landsat 8 Operational Land Imager (OLI) for the period 2014−2016 and composited them into cloudless mosaics for the following four seasons: yearly, summer, autumn, and April−October. The classification of images was performed in Google Earth Engine (GEE) Application Programming Interface (API) using random forest (RF) classifier. As a result, 30 m spatial resolution forest mask for flatland of Ukraine was created. The user’s and producer’s accuracy were estimated to be 0.910 ± 0.015 and 0.880 ± 0.018, respectively. The total forest area for the flatland Ukraine is 9440.5 ± 239.4 thousand hectares, which is 3% higher than official data. In general, we conclude that the Landsat-derived forest mask performs well over fragmented landscapes if forest cover of the territory is higher than 10−15%.

Copyright of Biological Resources & Nature Management is the property of National University of Life & Environmental Sciences of Ukraine and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Global tree cover products face challenges in accurately predicting tree cover across biophysical gradients, such as precipitation or agricultural cover. To generate a natural forest cover map for Costa Rica, biases in tree cover estimation in the most widely used tree cover product (the Global Forest Change product (GFC) were quantified and corrected, and the impact of map biases on estimates of forest cover and fragmentation was examined. First, a forest reference dataset was developed to examine how the difference between reference and GFC-predicted tree cover estimates varied along gradients of precipitation and elevation, and nonlinear statistical models were fit to predict the bias. Next, an agricultural land cover map was generated by classifying Landsat and ALOS PalSAR imagery (overall accuracy of 97%) to allow removing six common agricultural crops from estimates of tree cover. Finally, the GFC product was corrected through an integrated process using the nonlinear predictions of precipitation and elevation biases and the agricultural crop map as inputs. The accuracy of tree cover prediction increased by ≈29% over the original global forest change product (the R2 rose from 0.416 to 0.538). Using an optimized 89% tree cover threshold to create a forest/nonforest map, we found that fragmentation declined and core forest area and connectivity increased in the corrected forest cover map, especially in dry tropical forests, protected areas, and designated habitat corridors. By contrast, the core forest area decreased locally where agricultural fields were removed from estimates of natural tree cover. This research demonstrates a simple, transferable methodology to correct for observed biases in the Global Forest Change product. The use of uncorrected tree cover products may markedly over- or underestimate forest cover and fragmentation, especially in tropical regions with low precipitation, significant topography, and/or perennial agricultural production. [ABSTRACT FROM AUTHOR]

Global Forest Change datasets have the potential to assist countries with national forest measuring, reporting and verification (MRV) requirements. This paper assesses the accuracy of the Global Forest Change data against nationally derived forest change data by comparing the forest loss estimates from the global data with the equivalent data from Guyana for the period 2001–2017. To perform a meaningful comparison between these two datasets, the initial year 2000 forest state needs first to be matched to the definition of forest land cover appropriate to a local national setting. In Guyana, the default definition of 30% tree cover overestimates forest area is by 483,000 ha (18.15%). However, by using a tree canopy cover (i.e., density of tree canopy coverage metric) threshold of 94%, a close match between the Guyana-MRV non-forest area and the Global Forest Change dataset is achieved with a difference of only 24,210 ha (0.91%) between the two maps. A complimentary analysis using a two-stage stratified random sampling design showed the 94% tree canopy cover threshold gave a close correspondence (R2 = 0.98) with the Guyana-MRV data, while the Global Forest Change default setting of 30% tree canopy cover threshold gave a poorer fit (R2 = 0.91). Having aligned the definitions of forest for the Global Forest Change and the Guyana-MRV products for the year 2000, we show that over the period 2001–2017 the Global Forest Change data yielded a 99.34% overall Correspondence with the reference data and a 94.35% Producer's Accuracy. The Guyana-MRV data yielded a 99.36% overall Correspondence with the reference data and a 95.94% Producer's Accuracy. A year-by-year analysis of change from 2001–2017 shows that in some years, the Global Forest Change dataset underestimates change, and in other years, such as 2016 and 2017, change is detected that is not forest loss or gain, hence the apparent overestimation. The conclusion is that, when suitably calibrated for percentage tree cover, the Global Forest Change datasets give a good first approximation of forest loss (and, probably, gains). However, in countries with large areas of forest cover and low levels of deforestation, these data should not be relied upon to provide a precise annual loss/gain or rate of change estimate for audit purposes without using independent high-quality reference data. [ABSTRACT FROM AUTHOR]

Satellite imagery of 25–30 m spatial resolution has been recognized as an effective tool for monitoring the spatial and temporal dynamics of forest cover at different scales. However, the precise mapping of forest cover over fragmented landscapes is complicated and requires special consideration. We have evaluated the performance of four global forest products of 25–30 m spatial resolution within three flatland subregions of Ukraine that have different forest cover patterns. We have explored the relationship between tree cover extracted from the global forest change (GFC) and relative stocking density of forest stands and justified the use of a 40% tree cover threshold for mapping forest in flatland Ukraine. In contrast, the canopy cover threshold for the analogous product Landsat tree cover continuous fields (LTCCF) is found to be 25%. Analysis of the global forest products, including discrete forest masks Global PALSAR-2/PALSAR Forest/Non-Forest Map (JAXA FNF) and GlobeLand30, has revealed a major misclassification of forested areas under severe fragmentation patterns of landscapes. The study also examined the effectiveness of forest mapping over fragmented landscapes using dense time series of Landsat images. We collected 1548 scenes of Landsat 8 Operational Land Imager (OLI) for the period 2014–2016 and composited them into cloudless mosaics for the following four seasons: yearly, summer, autumn, and April–October. The classification of images was performed in Google Earth Engine (GEE) Application Programming Interface (API) using random forest (RF) classifier. As a result, 30 m spatial resolution forest mask for flatland of Ukraine was created. The user's and producer's accuracy were estimated to be 0.910 ± 0.015 and 0.880 ± 0.018, respectively. The total forest area for the flatland Ukraine is 9440.5 ± 239.4 thousand hectares, which is 3% higher than official data. In general, we conclude that the Landsat-derived forest mask performs well over fragmented landscapes if forest cover of the territory is higher than 10–15%. [ABSTRACT FROM AUTHOR]

The global population of wild tigers remains dangerously low at fewer than 3500 individuals. Habitat loss, along with poaching, can undermine the international target recovery of doubling the number of wild tigers by 2022. Using a new satellite-based monitoring system, we analyzed 14 years of forest loss data within the 76 landscapes (ranging from 278 to 269,983 km(2)) that have been prioritized for conservation of wild tigers. Our analysis provides an update of the status of tiger habitat and describes new applications of technology to detect precisely where forest loss is occurring in order to curb future habitat loss. Across the 76 landscapes, forest loss was far less than anticipated (79,597 ± 22,629 km(2), 7.7% of remaining habitat) over the 14-year study period (2001-2014). Habitat loss was unevenly distributed within a subset of 29 landscapes deemed most critical for doubling wild tiger populations: 19 showed little change (1.5%), whereas 10 accounted for more than 98% (57,392 ± 16,316 km(2)) of habitat loss. Habitat loss in source population sites within 76 landscapes ranged from no loss to 435 ± 124 km(2) ([Formula: see text], SD = 89, total = 1676 ± 476 km(2)). Doubling the tiger population by 2022 requires moving beyond tracking annual changes in habitat. We highlight near-real-time forest monitoring technologies that provide alerts of forest loss at relevant spatial and temporal scales to prevent further erosion.