This product contains LCMAP CONUS Reference plot location data in a .shp format as well as annual land cover, land use, and change process variables for each reference data plot in a separate .csv table. The same information available in the.csv file is also provided in a .xlsx format. The LCMAP Reference Data Product was utilized for evaluation and validation of the Land Change Monitoring, Assessment, and Projection (LCMAP) land cover and land cover change products. This LCMAP Reference Data Product includes the collection of an independent dataset of 25,000 randomly-distributed and 2,000 stratified random 30-meter by 30-meter samples across the conterminous United States (CONUS). The 2,000 samples were selected using a stratified random sampling process with 4 strata based off the LCMAP Collection 1.0 Science Products. The LCMAP Reference Data Products collected variables related to primary and secondary land use, primary and secondary land cover(s), change processes, and other ancillary variables annually across CONUS from 1984–2021.

This product contains plot location data in a .shp format as well as annual land cover, land use, and change process variables for each reference data plot in a separate .csv table. The same information available in the.csv file is also provided in a .xlsx format. The LCMAP Reference Data Product was utilized for evaluation and validation of the Land Change Monitoring, Assessment, and Projection (LCMAP) land cover and land cover change products. The LCMAP Reference Data Product includes the collection of an independent dataset of 25,000 randomly-distributed 30-meter by 30-meter plots across the conterminous United States (CONUS). This dataset was collected via manual image interpretation to aid in validation of the land cover and land cover change products as well as area estimates. The LCMAP Reference Data Product collected variables related to primary and secondary land use, primary and secondary land cover(s), change processes, and other ancillary variables annually across CONUS from 1984-2018. First posted - May 1, 2020 (available from author) Revised - September 21, 2021 (version 1.1) Revised - November 17, 2021 (version 1.2)

This product contains plot location data for LCMAP Hawaii Reference Data in a .shp format as well as annual land cover, land use, and change process variables for each reference data plot in a separate .csv table. The same information available in the.csv file is also provided in a .xlsx format. The LCMAP Hawaii Reference Data Product was utilized for evaluation and validation of the Land Change Monitoring, Assessment, and Projection (LCMAP) land cover and land cover change products. The LCMAP Hawaii Reference Data Product includes the collection of an independent dataset of 600 30-meter by 30-meter plots across the island chain of Hawaii. The LCMAP Hawaii Reference Data Products collected variables related to primary and secondary land use, primary and secondary land cover(s), change processes, and other ancillary variables annually across Hawaii from 2000-2019. The sites in this dataset were collected via manual image interpretation. These samples were selected using a strat ...

This preliminary product contains annual land cover, land use, and change process variables for reference data plots in a .csv table. The same information available in the.csv file is also provided in a .xlsx format. The LCMAP Preliminary Duplicate Attributes Collection for Reference Data includes the attributes collected from 30-meter by 30-meter plots across the conterminous United States (CONUS) that were randomly selected during data collection to have 2 independent interpretations. These duplicate interpretations were utilized in the quality assurance and quality control processes for the creation of the final, authoritative LCMAP Reference Data Product. These attributes were collected via manual image interpretation. The interpretations collected variables related to primary and secondary land use, primary and secondary land cover(s), change processes, and other ancillary variables annually across CONUS from 1984-2016. The duplicate interpretations were utilized to identify classes and situations that could require additional interpreter training, further quality review, and/or correct identified errors in the preparation of the final dataset. The data in this dataset is considered interim/provisional in nature. The LCMAP Preliminary Duplicate Attributes Collection for Reference Data provides the interpretation attributes for a subset of the 25,000 plots that were utilized for evaluation and validation of the Land Change Monitoring, Assessment, and Projection (LCMAP) land cover and land cover change products.

A validation assessment of Land Cover Monitoring, Assessment, and Projection Collection 1.2 annual land cover products (1985–2018) for the Conterminous United States was conducted with an independently collected reference dataset. Reference data land cover attributes were assigned by trained interpreters for each year of the time series (1984–2018) to a reference sample of 26,971 Landsat resolution (30m x 30m) pixels. The LCMAP and reference dataset labels for each pixel location are displayed here for each year, 1985–2018.

A validation assessment of Land Cover Monitoring, Assessment, and Projection (LCMAP) Collection 1 annual land cover products (1985–2017) for the Conterminous United States was conducted with an independently collected reference data set. Reference data land cover attributes were assigned by trained interpreters for each year of the time series (1984–2018) to a reference sample of 24,971 Landsat resolution (30m x 30m) pixels. These pixels were randomly selected from a sample frame of all pixels in the Landsat Analysis Ready Data (ARD) grid system which fell within the map area (Dwyer et al., 2018). Interpretation used the TimeSync reference data collection tool which visualizes Landsat images and Landsat data values for all usable images in the time series (1984–2018) (Cohen et al., 2010). Interpreters also referred to air photos and high resolution images available in Google Earth as well as several ancillary data layers. The interpreted land cover attributes were crosswalked to ...

This is a collection of data tables supporting the LCMAP CONUS Geographic Assessment v1.0. The data used to generate these tables come from the USGS LCMAP reference dataset and the map products released by LCMAP. Tables include annual land cover class composition and annual rate of land cover change metrics developed with a post-stratified estimator. Other tables including annual gross change of specific types of land covers, cumulative metrics of overall geographic footprint of change, frequency of overall geographic footprint of change, overall area estimates of specific class changes, and all unique changes in land cover classes. All tables cover the time period 1985-2016. All values in these tables are presented in percent and/or square kilometers of CONUS and include standard errors (SE). Tables 1a and 1b are annual land-cover class area estimates. Tables 1a and 1b are annual land-cover class area estimates. Table 1a presents the estimated area of each LCMAP land cover class as a percent of CONUS and Table 1b is the same in square kilometers, both with SEs. The area estimates include the following eight LCMAP land covers classes: Developed, Cropland, Grass/Shrub, Tree Cover, Water, Wetland, Snow/Ice, and Barren (Brown et al., 2020; Zhu et al., 2014).Table 2 is the estimated overall net change in each of the eight LCMAP land-cover classes: Developed, Cropland, Grass/Shrub, Tree Cover, Water, Wetland, Ice/Snow, and Barren (Brown et al., 2020; Zhu et al., 2014). It includes percent and square kilometers of CONUS in 1985 and 2016. The net area that changed between 1985 and 2016 is presented in percent and km2 of CONUS with the standard error of this change. Table 3 represents area estimates for the area that changed a specific number of times during 1985–2016. These areas are presented as both the fraction of LCMAP’s CONUS extent and as the equivalent in square kilometers, with standard errors. The overall footprint of change that occurred between 1985 and 2016 is presented as the percent and km2 of CONUS. Table 4 is the estimated percent and areal extent (km2) of CONUS that experienced either a change, or none, for each year 1986–2016, with associated standard errors. Tables 5a–h present the areal extent estimates (in km2) for specific land-cover class changes with standard error for every year, 1986–2016. The specific land-cover classes include Developed, Cropland, Grass/Shrub, Tree Cover, Water, Wetland, Ice/Snow, and Barren (Brown et al., 2020; Zhu et al., 2014). Table 6a presents area estimates (km2) of CONUS for all land-cover class changes with standard error for every year, 1986–2016. Table 6b presents area estimates (km2) of CONUS that did not change for all eight LCMAP land-cover classes with standard error for every year, 1986–2016. The eight LCMAP land-cover classes include Developed, Cropland, Grass/Shrub, Tree Cover, Water, Wetland, Ice/Snow, and Barren (Brown et al., 2020; Zhu et al., 2014). Table 7 presents the estimated area (km2) of CONUS for four groupings of land-cover class changes with standard error for every year, 1986–2016. The eight LCMAP land-cover classes and class codes include Developed (1), Cropland (2), Grass/Shrub (3), Tree Cover (4), Water (5), Wetland (6), Ice/Snow (7), and Barren (8) (Brown et al., 2020; Zhu et al. 2014). The four groups include “Natural Resource Cycles”, “Increases in Developed and Built-up Land”, “Surface Water Expansion/Contraction”, and “Other”. Table 8 presents the cumulative (1985-2016) area of all specific land cover class changes, 56 in total.

A validation assessment of Land Cover Monitoring, Assessment, and Projection Collection 1.0 annual land cover products (2000–2019) for Hawaii was conducted with an independently collected reference dataset. Reference data land cover attributes were assigned by trained interpreters for each year of the time series (2000–2019) to a reference sample of 600 Landsat resolution (30m x 30m) pixels. The LCMAP and reference dataset labels for each pixel location are displayed here for each year, 2000–2019.

INTRODUCTION

This document outlines the creation of a global inventory of reference samples and Earth Observation (EO) / gridded datasets for the Global Pasture Watch (GPW) initiative. This inventory supports the training and validation of machine-learning models for GPW grassland mapping. This documentation outlines methodology, data sources, workflow, and results.

Keywords: Grassland, Land Use, Land Cover, Gridded Datasets, Harmonization

OBJECTIVES

• Create a global inventory of existing reference samples for land use and land cover (LULC);

• Compile global EO / gridded datasets that capture LULC classes and harmonize them to match the GPW classes;

• Develop automated scripts for data harmonization and integration.

DATA COLLECTION

Datasets incorporated:

Total: 102,355,642 samples

WORKFLOW

Harmonization Process

We harmonized global reference samples and EO/gridded datasets to align with GPW classes, optimizing their integration into the GPW machine-learning workflow.

We considered reference samples derived by visual interpretation with spatial support of at least 30 m (Landsat and Sentinel), that could represent LULC classes for a point or region.

Each dataset was processed using automated Python scripts to download vector files and convert the original LULC classes into the following GPW classes:

0. Other land cover

1. Natural and Semi-natural grassland

2. Cultivated grassland

3. Crops and other related agricultural practices

We empirically assigned a weight to each sample based on the original dataset's class description, reflecting the level of mixture within the class. The weights range from 1 (Low) to 3 (High), with higher weights indicating greater mixture. Samples with low mixture levels are more accurate and effective for differentiating typologies and for validation purposes.

The harmonized dataset includes these columns:

ACKNOWLEDGMENTS

The development of this global inventory of reference samples and EO/gridded datasets relied on valuable contributions from various sources. We would like to express our sincere gratitude to the creators and maintainers of all datasets used in this project.

REFERENCES

• Brown, C.F., Brumby, S.P., Guzder-Williams, B. et al. Dynamic World, Near real-time global 10 m land use land cover mapping. Sci Data 9, 251 (2022). https://doi.org/10.1038/s41597-022-01307-4Van Tricht, K. et al. Worldcereal: a dynamic open-source system for global-scale, seasonal, and reproducible crop and irrigation mapping. Earth Syst. Sci. Data 15, 5491–5515, 10.5194/essd-15-5491-2023 (2023)

• Buchhorn, M.; Smets, B.; Bertels, L.; De Roo, B.; Lesiv, M.; Tsendbazar, N.E., Linlin, L., Tarko, A. (2020): Copernicus Global Land Service: Land Cover 100m: Version 3 Globe 2015-2019: Product User Manual; Zenodo, Geneve, Switzerland, September 2020; doi: 10.5281/zenodo.3938963

• d’Andrimont, R. et al. Harmonised lucas in-situ land cover and use database for field surveys from 2006 to 2018 in the european union. Sci. data 7, 352, 10.1038/s41597-019-0340-y (2020)

• Fritz, S. et al. Geo-Wiki: An online platform for improving global land cover, Environmental Modelling & Software, 31, https://doi.org/10.1016/j.envsoft.2011.11.015 (2012)

• Fritz, S., See, L., Perger, C. et al. A global dataset of crowdsourced land cover and land use reference data. Sci Data 4, 170075 https://doi.org/10.1038/sdata.2017.75 (2017)

• Schneider, M., Schelte, T., Schmitz, F. & Körner, M. Eurocrops: The largest harmonized open crop dataset across the european union. Sci. Data 10, 612, 10.1038/s41597-023-02517-0 (2023)

• Souza, C. M. et al. Reconstructing Three Decades of Land Use and Land Cover Changes in Brazilian Biomes with Landsat Archive and Earth Engine. Remote. Sens. 12, 2735, 10.3390/rs12172735 (2020)

• Stanimirova, R. et al. A global land cover training dataset from 1984 to 2020. Sci. Data 10, 879 (2023)

• Stehman, S. V., Pengra, B. W., Horton, J. A. & Wellington, D. F. Validation of the us geological survey’s land change monitoring, assessment and projection (lcmap) collection 1.0 annual land cover products 1985–2017. Remot Sensing environment 265, 112646, 10.1016/j.rse.2021.112646 (2021).

• Tsendbazar, N. et al. Product validation report (d12-pvr) v 1.1 (2021).

• Tseng, G., Zvonkov, I., Nakalembe, C. L., & Kerner, H. (2021). CropHarvest: A global dataset for crop-type classification. In Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track.