Global Forest Extent in 2000 and 2020 with Annual Fire-Related Forest Loss (2001-2023) at 0.25° Resolution

Supporting datasets of upscaled forest extent and annual forest loss due to fire for Wallace and Ng et al. (2025)John M. Wallace, Chan-Pang Ng, Jinhyuk E. Kim, Aodhan Sweeney, and Siyu Zhao (2025). 21st Century Extratropical Wildfires: A Multi-scale, Global Survey, [Manuscript submitted to PNAS]

Users who use the 0.25 degree datasets provided, be sure to cite the appropriate datasets and papers provided in the explanation of each file below.

Data from the "Global forest loss due to fire" found at https://glad.umd.edu/dataset/Fire_GFL (Tyukavina et al. 2022) is aggregated to the 0.25 degree resolution and provided in "global_forest_loss_fire_total_annual_2001-23_fraction_025deg.nc". The file contains fraction of 0.25 degree gridcell where forest loss due to fire occured for each year throughout the record. The band number goes from 0-23, where each band number is associated with the year of forest loss (e.g. band 1 = 2001,  band 2 = 2002, ... band 23 = 2023). Band 0 is an exception and is the sum of all the bands representing total forest loss due to fire throughout the record.To analyze the loss of forest cover in respect to initial forest cover from 2000, we first define forest cover using the tree cover in 2000 dataset found at https://data.globalforestwatch.org/documents/941f17325a494ed78c4817f9bb20f33a/explore (Hansen et al. 2013). We also aggregated tree cover data into 0.25 deg gridcells in similar fashion provided as "hansen_tree_cover_025deg.nc". This 0.25 deg dataset is not used in the analysis, but we provide it nonetheless.

Using the original 30 m tree cover dataset, we define a pixel to be forest if the 30 m pixel had tree cover greater than or equal to 30%. We then aggregate to the 0.25 deg gridcells which is then defined as F2000 in the manuscript. We note that F2000 is not the same forest cover that defines forests in Tyukavina et al. (2022), since they did not define forest cover with a single tree cover threshhold. However, given we aggregate the data to 0.25 deg, we assume that this accounts for variability in defined forest cover. This file is provided as "hansen_forest_cover_025deg.nc"

To complement the analysis and to understand the change in forest extent while accounting for regrowth of forest after loss, we use forest extent in 2000 and 2020 datasets at https://glad.umd.edu/dataset/GLCLUC2020 (Potapov et al. 2022). Forest extent are binary values in the original product 0=no forest, 1=forest present. We simply aggregate the pixels into 0.25 deg gridcells to get fraction of forest extent. These files are provided in two separate years as "forest_extent_2000_025deg.nc" for the year 2000, and "forest_extent_2020_025deg.nc" for the year 2020.

References

Tyukavina, A., Potapov, P., Hansen, M.C., Pickens, A., Stehman, S., Turubanova, S., Parker, D., Zalles, V., Lima, A., Kommareddy, I., Song, X-P, Wang, L. and Harris, N. (2022) Global trends of forest loss due to fire, 2001-2019. Frontiers in Remote Sensing https://doi.org/10.3389/frsen.2022.825190 

Hansen, M. C., Potapov, P. V., Moore, R., Hancher, M., Turubanova, S. A., Tyukavina, A., Thau, D., Stehman, S. V., Goetz, S. J., Loveland, T. R., Kommareddy, A., Egorov, A., Chini, L., Justice, C. O., & Townshend, J. R. G. (2013). High-Resolution Global Maps of 21st-Century Forest Cover Change. Science, 342(6160), 850–853. https://doi.org/10.1126/science.1244693

Potapov P., Hansen M.C., Pickens A., Hernandez-Serna A., Tyukavina A., Turubanova S., Zalles V., Li X., Khan A., Stolle F., Harris N., Song X.-P., Baggett A., Kommareddy I., Kommareddy A. (2022) The global 2000-2020 land cover and land use change dataset derived from the Landsat archive: first results. Frontiers in Remote Sensing [https://doi.org/10.3389/frsen.2022.856903]