AKAnna Komarova

Anna Komarova

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Anna Komarova has a total of 51 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Global extent and drivers of tree cover loss quantified with high-resolution satellite data

    by Alexandra Tyukavina, Andrew J. Poulson, Jeffrey Pickering, Bernard Adusei, Matthew C. Hansen, Peter Potapov, Antoine Baggett, Carolina Ortiz Dominguez, Aleksandra Mikus, Andre Oktaviandra, et al.

    Quantifying the drivers of tree cover loss globally provides a synoptic understanding of pressures on the world’s forests. Existing information about tree cover loss drivers relies on maps of coarse spatial and thematic resolution. In this study, we quantified the global extent of tree cover loss in 2018 at the scale of individual disturbances and provided a comprehensive accounting of land use outcomes using a global probability sample of 600 5 × 5–kilometer blocks mapped with high-resolution (3- to 10-meter) satellite data. Out of 277 thousand square kilometers of estimated global tree cover loss, nearly a third (29.0%) was due to long-term conversion of tree cover to other land uses, including conversion of natural tree cover to pasture (15.0%), cropland (6.4%), and nontimber tree plantations (3.8%).


  • Open Access

    An accurate 10 m annual crop map product of maize and soybean across the United States

    by Haijun Li, Xiao-Peng Song, Bernard Adusei, Jeffrey Pickering, Andre Lima, Andrew Poulson, Antoine Baggett, Peter Potapov, Ahmad Khan, Viviana Zalles, et al.

    High-resolution crop maps over large spatial extents are fundamental to many agricultural applications; however, generating high-quality crop maps consistently across space and time remains a challenge. In this study, we improved a workflow for crop mapping and developed an openly available, annual, 10 m spatial resolution maize and soybean map product over the Contiguous United States (CONUS) from 2019 to 2022 (available at https://glad.umd.edu/dataset/mapping-crops-10-m-resolution-united-states, last access: 26 December 2025). We obtained all available Sentinel-2 surface reflectance data between May and October for every year, applied quality assurance, corrected the bidirectional reflectance distribution function (BRDF) effects, and generated 10 d analysis ready data (ARD) composites. We then derived multi-temporal metrics from the 10 d ARD as training features for the national-scale wall-to-wall mapping. We implemented a stratified, two-stage cluster sampling, and then conducted annual field surveys and collected ground data. Utilizing the training data with Sentinel-2 multi-temporal metrics and topographic factors, we trained random forest models generalized for annual maize and soybean classification separately. Validated using field data from the two-stage cluster sample, our annual maps achieved consistent overall accuracies (OA) greater than 95 % with standard errors of less than 1 %. User's accuracies (UAs) and producer's accuracies (PAs) for maize were higher than 91 % and 84 % across the years, and UAs and PAs for soybean were greater than 88 % and 82 %, respectively. To illustrate the substantial improvement of the 10 m map over existing datasets, e.g., the 30 m Cropland Data Layer (CDL), we aggregated the 10 m maps to 30 m spatial resolution and quantified the number of mixed pixels that can be reduced by improving the mapping from 30 to 10 m. The counties with the most maize and soybean production in Iowa, Illinois and Nebraska had the lowest reduction in mixed pixels, ranging from 1 % to 7 %, whereas southern counties had a higher reduction in mixed pixels. Overall, the median percentages of mixed maize and soybean pixels reduction across all counties were 8 % and 9 %, respectively. With more Sentinel-2-like data available from continuous observations and incoming satellite missions, we anticipate that 10 m crop maps will greatly benefit long-term monitoring for agricultural practices from the field to global scales. The dataset is also available at https://doi.org/10.6084/m9.figshare.28934993.v2 (Li et al., 2025).


  • Top Cited Papers
    Open Access

    Rapid monitoring of global land change

    by Amy H. Pickens, Matthew C. Hansen, Zhen Song, Andrew Poulson, Anna Komarova, Antoine Baggett, Theodore Kerr, Aleksandra Mikus, Carolina Ortiz Dominguez, Alexandra Tyukavina, et al.

    Direct human action, principally land use expansion, and natural dynamics, such as fire and drought, drive global land change. Here we present a global land change monitoring system, DIST-ALERT, that rapidly tracks vegetation loss anomalies with 30 m resolution using imagery from Landsat 8/9 and Sentinel-2A/B/C satellites. The alerts capture agricultural expansion, urbanization, logging, mining, fire, drought, landslides, and other dynamics, but without attribution. Identified through a probability sample, 2023 anthropogenic land use conversions totaled 28.6 ± 7.6 Mha (±standard error), half of which replaced long-lived or secondary natural vegetation. Fires resulting in land cover conversion totaled 14.9 ± 4.3 Mha (±standard error). Combined, these dynamics equal 0.3% of the global land surface, equivalent to the area of the state of California. Annual DIST-ALERT summaries of land use expansion and climate-driven land change can serve as a future long-term global environmental data record.


  • Open Access

    An Accurate 10 m Annual Crop Map Product of Maize and Soybean Across the United States

    by Haijun Li, Xiao-Peng Song, Bernard Adusei, Jeffrey Pickering, Andre Lima, Andrew Poulson, Antoine Baggett, Peter Potapov, Ahmad Khan, Viviana Zalles, et al.

    High-resolution crop maps over large spatial extents are fundamental to many agricultural applications; however, generating high-quality crop maps consistently across space and time remains a challenge. In this study, we improved a workflow for crop mapping and developed the first openly available, annual, 10-m spatial resolution maize and soybean maps over the Contiguous United States (CONUS) from 2019 to 2022, available at the Global Land Analysis and Discovery at the University of Maryland (https://glad.umd.edu/projects/mapping-crops-10-m-resolution-united-states). We obtained all available Sentinel-2 surface reflectance data between May and October for every year, applied quality assurance, corrected the bidirectional reflectance distribution function (BRDF) effects, and generated 10-day analysis ready data (ARD) composites. We then derived multi-temporal metrics from the 10-day ARD as training features for the national-scale wall-to-wall mapping. We implemented a stratified, two-stage cluster sampling, and then conducted annual field surveys and collected ground data. Utilizing the training data with Sentinel-2 multi-temporal metrics and topographic factors, we trained random forest models generalized for annual maize and soybean classification separately. Validated using field data from the two-stage cluster sample, our annual maps achieved consistent overall accuracies (OA) greater than 95 % with standard errors of less than 1 %. User’s accuracies (UAs) and producer’s accuracies (PAs) for maize were higher than 91 % and 84 % across the years, and UAs and PAs for soybean were greater than 88 % and 82 %, respectively. To illustrate the substantial improvement of the 10-m map over existing datasets, e.g., the 30-m Cropland Data Layer (CDL), we aggregated the 10-m maps to 30-m spatial resolution and quantified the amount of 30-m mixed pixels that can be reduced at field, regional, and national levels. The counties with the most maize and soybean production in Iowa, Illinois and Nebraska had the lowest reduction in mixed pixels, ranging from 1 % to 10 %, whereas southern counties had a higher reduction in mixed pixels. Overall, the median percentages of mixed maize and soybean pixels reduction across all counties were 14 % and 16 %, respectively. With more Sentinel-2-like data available from continuous observations and incoming satellite missions, we anticipate that 10-m crop maps will greatly benefit long-term monitoring for agricultural practices from the field to global scales. The dataset is also available at https://doi.org/10.6084/m9.figshare.28934993.v1 (Li et al., 2025).


  • Open Access

    Supplementary material to "An Accurate 10 m Annual Crop Map Product of Maize and Soybean Across the United States"

    by Haijun Li, Xiao-Peng Song, Bernard Adusei, Jeffrey Pickering, Andre Lima, Andrew Poulson, Antoine Baggett, Peter Potapov, Ahmad Khan, Viviana Zalles, et al.

  • Open Access

    Spring fires in Russia: results from participatory burned area mapping with Sentinel-2 imagery

    by Igor Glushkov, Ilona Zhuravleva, Jessica L McCarty, Anna Komarova, Alexey Drozdovsky, Marina Drozdovskaya, Vilen Lupachik, Alexey Yaroshenko, Stephen V Stehman, Alexander V Prishchepov

    Human-induced fires play a crucial role in transforming landscapes and contributing to greenhouse gas emissions. Russia is a country where human-induced fires are widespread and form distinctive spring and summer burning cycles. However, spring fires are not well documented and it is unclear which land-cover types are associated with the spread of spring fires. Using Sentinel-2 optical satellite imagery, a wall-to-wall spring burned area data set for 1 January to 15 May 2020 was created for Russia (excluding the Arctic) using a participatory crowdsourcing digitizing approach on an online platform developed specifically for this application. The 2020 spring fire product had a producer accuracy of 85% and user accuracy of 92%. Approximately 13.38 million ha, comprising 1.8% of the study area, were mapped as burned, with the majority of the 2020 spring burned areas in Siberia. Our spring-fire product revealed five times more burned area estimates compared to the burned area estimates from the moderate resolution imaging spectroradiometer (MODIS) MCD64 product. We also found high variability of burned area per active fire pixel across regions of Russia, when compared to MODIS and visible infrared imaging radiometer suite active fire data. Spring fires started to increase from the end of February and reached their maximum by the end of March through the middle of April. Spring fires were associated with arable lands and grasslands as land-cover types, except Siberia, where spring fires were most common in deciduous and needle-leaved forests, followed by arable lands. While spring fires were associated with croplands and grasslands, an estimate for Central Russia showed approximately 75% of spring fires occurred on abandoned agricultural lands. Our study demonstrated the suitability of optical Sentinel-2 imagery for spring fire mapping and the great utility of a participatory mapping approach for fast and accurate mapping as well as engagement of the community.