Arctic vegetation cover fractions derived from Landsat time series (1984-2020) for the greater Mackenzie Delta Region (Western Canadian Arctic)

Data to the publication by Nill et al. (2022) "Arctic shrub expansion revealed by Landsat-derived multitemporal vegetation cover fractions in the Western Canadian Arctic" The dataset features Landsat-derived fractional cover estimates of Arctic plant functional types (shrub, evergreen tree...

Full description

Bibliographic Details
Main Authors: Nill, Leon, Grünberg, Inge, Ullmann, Tobias, Gessner, Matthias, Boike, Julia, Hostert, Patrick
Format: Dataset
Language:English
Published: 2022
Subjects:
land cover
fractional cover
arctic
tundra
shrub
greening
Mackenzie Delta
Landsat
Remote Sensing
Map
Arctic
Canada
North Pole
Tundra
Online Access:https://zenodo.org/record/7055100
https://doi.org/10.5281/zenodo.7055100
Description
Summary:Data to the publication by Nill et al. (2022) "Arctic shrub expansion revealed by Landsat-derived multitemporal vegetation cover fractions in the Western Canadian Arctic" The dataset features Landsat-derived fractional cover estimates of Arctic plant functional types (shrub, evergreen trees, herbaceous, lichen) and other land cover (barren, water) in the greater Mackenzie Delta Region, Canada. We utilized regression-based unmixing based on synthetic training data in order to build multitemporal Kernel Ridge Regression (KRR) models for estimating fractional cover and validated our predictions based on independent very-high-resolution imagery (please be referred to publication for details). Dataset information The fraction cover predictions ("krr-avg") are provided separately for each epoch (1984-1990, 1991-1996, ., 2015-2020) and class/cover type. The decadal change images ("dec-cng") between 1984 and 2020 are provided separately for each class/cover type. The naming convention of the files is as follows: XXXX-XXXX_YYY-YYY_int16-10e3_class-Z-Z XXXX-XXXX = epoch, e.g. 2015-2020 YYY-YYY = dataset ("krr-avg" = fraction cover, "dec-cng" = decadal fraction cover change) Z-Z = class ID and associated class name (sh = shrub, cf = coniferous, hb = herbaceous, lc = lichen, wt = water, br = barren) The fraction cover values are % scaled by 10,000. For instance, a value of 1234 refers to 12.34%. Further image metadata: Datatype: Signed 16-bit integer (Int16) Data format: GeoTiff (.tif) No data value: -9999 Projection: EPSG:3573 with custom central meridian; WKT string: 'PROJCS["WGS 84 / North Pole LAEA Canada",GEOGCS["WGS 84",DATUM["WGS_1984",SPHEROID["WGS ...

Similar Items