ABoVE: Gridded 30-meter Aboveground Biomass, Shrub Dominance, North Slope, Alaska, 2007-2016

• Main Authors: Logan Berner, Ken D. Jantz, Scott Goetz
• Format: Dataset
• Language: unknown
• Published: Arctic Data Center 2018
• Subjects: Vegetation Index; Biomass; Vegetation cover; ABoVE; Canada; Brooks Foothills; north slope; Alaska
• Online Access: https://search.dataone.org/view/urn:uuid:60ef7314-8528-4ace-991d-feb1debdfdd4

This dataset includes 30-m gridded estimates of total plant aboveground biomass (AGB), the shrub AGB, and the shrub dominance (shrub/plant AGB) for non-water portions of the Beaufort Coastal Plain and Brooks Foothills ecoregions of the North Slope of Alaska. The estimates were derived by linking biomass harvests from 28 published field site datasets with Normalized Difference Vegetation Index (NDVI) from a regional Landsat mosaic derived from Landsat 5 and 7 satellite imagery. The data cover the period 2007-06-01 to 2016-08-31. The data provided are the best estimates from the described modeling and Monte Carlo approach for each 30-m pixel in the Landsat mosaic at the 50th percentile, and also at the 2.5 and 97.5 percentiles for each data type (plant AGB, shrub AGB, and shrub dominance) which together encompass 95% of predictions. The published field measurements of total plant and shrub AGB used in the modeling were collected between July 1998 and August 2008. The mean and standard error (SE) of plant and shrub AGB were also acquired or computed for the data at each site. The regional Landsat NDVI mosaic was derived from 1,721 summer scenes acquired between 2007 and 2016. Spectral reflectance information was extracted from these 'peak greenness' scenes on a per pixel basis. Empirical AGB-NDVI relationships were developed for the field sites and the relationships were applied to the mosaic. The Monte Carlo uncertainty analysis involved generating 1,000 regional maps of each ecosystem data type, where each map was produced by randomly permuting the underlying field and remote sensing datasets by their uncertainty due to sampling and sensor calibration errors. The data with this dataset are the 50th percentile (best estimates), the 2.5, and 97.5 percentiles of the 1,000 permutations. There are nine data files of mapped AGB and shrub dominance with this dataset in GeoTIFF (.tif) format and one shapefile (.shp) provided in compressed (.zip) format which provides the study locations. The study locations are also provided as a companion file in .kmz format for viewing in Google Earth. A companion file of the published field measurements of total plant and shrub AGB used in the modeling is also available as a .csv file.