1. ABOUT THE DATASET -------------------- Title: Forest cover change Tanzania 1987 to 2021 Creator(s): Nike Doggart 1,2,3 Wilson Ancelm Mugasha 4, Aloyce Mpiri 5, Theron Morgan-Brown 2, Susannah M Sallu 3 and Dominick V Spracklen 1 Organisation(s): 1 Institute for Climate and Atmospheric Science, School of Earth and Environment, University of Leeds, Leeds LS2 9JT, United Kingdom 2 Tanzania Forest Conservation Group, PO Box 23410, Dar es Salaam, Tanzania 3 Sustainability Research Institute, School of Earth and Environment, University of Leeds, Leeds LS2 9JT, United Kingdom 4 Department of Forest Resources Assessment and Management, Sokoine University of Agriculture, Morogoro, Tanzania 5 Tanzania Forestry Research Institute, PO Box 1854, Morogoro, Tanzania Rights-holder(s): Copyright 2023 Nike Doggart Publication Year: 2023 Description: This dataset includes data used to determine rates of forest cover change in Tanzania between 1987 and 2021. Also included are data from 180 Field Survey Sample Plots from 10 sampling clusters on village land, in Tanzania. Data was collected in 2021. Cite as: Nike Doggart, Wilson Ancelm Mugasha, Aloyce Mpiri, Theron Morgan-Brown, Susannah M Sallu, Dominick V Spracklen (2023): Forest cover change Tanzania 1987 to 2021. [Dataset]. https://doi.org/10.5518/1295 Related publication: Doggart, N., Mugasha, WA., Mpiri, A., Morgan-Brown, T., Sallu, S.M., Spracklen, D.V., Submitted. 2023. Agricultural fallows are the main driver of natural forest regeneration in Tanzania. Environmental Research Letters. Contact: ndoggart@tfcg.or.tz 2. TERMS OF USE --------------- Copyright 2023. Nike Doggart. Unless otherwise stated, this dataset is licensed under a Creative Commons Attribution 4.0 International Licence: https://creativecommons.org/licenses/by/4.0/.] 3. PROJECT AND FUNDING INFORMATION ---------------------------------- This research was carried out as part of the DECAF project. This work received funding from: - the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 771492) - the Swiss Agency for Development and Cooperation through the Conserving Forests through sustainable forest-based Enterprise Support in Tanzania (CoForEST) - the UK’s Economic and Social Research Council funding to Sallu (ES/R009708/1) - Natural Environment Research Council funding to Spracklen Grant (NE/M003574/1) - Spracklen also acknowledges a Philip Leverhulme Prize. 4. CONTENTS ----------- File 1. GEE_Script_landcover_change_1987to2021 Script author: Theron Morgan-Brown This script is designed for use in Google Earth Engine. It is used to view the randomly selected remote sensing sample points (RSSPs) using the least cloudy images from Landsat 5, 7 and 8, Sentinel 2 and data from PALSAR 1 and 2. Images are loaded with the oldest images at the top. The images were reviewed visually and the RSSP was then classified based on land cover classes described in Supplementary File 1. The data (File 4) collected using this script was then used to calculate rates of deforestation, forest persistence, and regeneration. File 2. GEE_script_regeneration_mapping Script author: Theron Morgan-Brown This script is designed for use in Google Earth Engine. The script uses PALSAR data to identify areas of natural forest regeneration. It was used to help identify the Field Survey Sample Plots. File 3. R_script_biomass_and_sp_richness Script author: Wilson A. Mugasha This R script calculates the biomass and species richness for each of the Field Survey Sample Plots (FSSPs). It is used in conjunction with Files 6 - 9. File 4. Landcover_change_trajectory_results Results of the land cover change trajectory analysis for the randomly selected Remote Sensing Sample Points (RSSPs). This data was used to determine the deforestation and regeneration rates. Number of cases: 500 Description of the variables Column A: Plot. Plot identification Columns B - BS: For each year (as indicated in Row 1), plots are classified by class and sub-class (as indicated in Row 2) using the land cover / land use classes described in the Supplementary File 1. Codes are provided in the 'Description' worksheet of the Excel file. Columns BT - CA: a count of number of years that the plot remained in the respective land cover / land use. Column CC: the earliest year in the analysis Column CD: the number of deforestation events Column CE: the year of the most recent deforestation event for those plots that transitioned from forest in 1987 to non-forest in 2021. Column CF: the year in which regeneration was first detected. Column CG: the number of years that regeneration has continued. Columns CH - DD: Classification of the land cover change trajectory of the plot. Column DE: the status of the plot's land cover / land use in 2021. Column DF - DG: the land cover change trajectory code for the plot. Codes and explained in the description worksheet of the Excel file. Column DH: the location of the plot in longitude / latitude. File 5. all_regen This file presents results for each of the 180 Field Survey Sample Plots based on the vegetation plots and key informant interviews. Number of variables: 18 Number of cases: 180 Fields plot_id: the plot identification code district: the administrative district in which the plot was located vegetation: vegetation type. BA = Acacia bushland, BT = Bushland thicket, CL = Lowland forest, MW = Miombo Woodland. all_agb: above ground tree biomass including stumps and remnant trees Mg / ha norem_agb: above ground tree biomass including stumps but excluding remnant trees Mg / ha norem_nostump_agb: above ground tree biomass excluding stumps and excluding remnant trees Mg / ha spr: species richness time: years since the most recent regeneration period began charcoal: 1 = absent, 2 = present livestock: grazing intensity score. See Table 1 of the main paper for details. fire: burning intensity score. See Table 1 of the main paper for details. Cultivation: 1 = absent, 2 = present tree_cutting: 1 = absent, 2 = present conservation: 1 = absent, 2 = present precipitation: Mean annual precipitation (mm) deforestation_driver: the proximate deforestation driver, prior to regeneration. See Table 1 of the main paper for details. Stems: stem density. Number of stems / hecatre Files 6 - 9 These four files include the results of the field survey sample plot (FSSPs) vegetation plots. 15 m diameter vegetation plots were assessed at each of the 180 FSSPs. They are designed as inputs to the r script, provided in File 3. The R script includes information explaining how they are linked. Data from the structured interviews is available, upon reasonable request, from the lead author. 5. METHODS ---------- Methods are described in the related publication and its supplementary file 1.