Historical time-series reconstruction benchmark dataset of Landsat bi-monthly aggregates from GLAD ARD-2 at 30-m resolution with stratified sampling based on ESA CCI

Description

Historical time-series reconstruction benchmark dataset presented here is designed for evaluating and comparing the performance of time series reconstruction methods in the context of land cover change detection. The dataset is based on the European Space Agency Climate Change Initiative (ESA CCI) land cover dataset, which has been aggregated into 18 classes to facilitate analysis. The dataset includes information on land cover dynamics from 2000 to 2020, focusing on identifying and characterizing changes in land cover over time.

Data Collection and Processing:

The dataset is derived from the ESA CCI land cover dataset, which provides information on land cover classes at a global scale. The original dataset, containing 37 land cover classes, was aggregated into 18 classes based on similarity. Pixels with stable land cover over the study period and pixels with one or multiple land cover changes were identified and grouped into strata for sampling purposes.

Sampling points were selected using a stratified sampling design, ensuring representation across different land cover classes and change scenarios. Approximately 2600 points were selected from each stratum, resulting in a total of 51,978 sampling points. The selected points were uniformly distributed along the strata, with spatial variations accounted for.

Bimonthly time series data were extracted for each sampling point from 1997 to 2022, capturing temporal dynamics in land cover. Artificial gaps were introduced into the time series data to simulate real-world data loss, allowing for the evaluation of time series reconstruction methods under varying gap densities.

The time series values were extracted from Landsat GLAD imagery using the specified spectral bands, including blue, green, red, NIR, SWIR1, SWIR2, and thermal bands. Additionally, a clear quality band was also extracted.

Data Details

Time Period: 1997-01-01 to 2022-12-31

Type of Data: R data frame / Geopackage points.

Collection/Derivation: Derived from Landsat ARD v2, processed with Scikit-map.

Coordinate Reference System: EPSG:4326

Bounding Box: All the globe

File Format: RDS

Reclassified Classes of ESA CCI Land Cover Dataset

Aggregated Class Code

Aggregated Class Label

Original ESA CCI Classes

10

Cropland rainfed

10, 11, 12

30

Mosaic cropland | natural vegetation

30, 40

50

Tree cover broadleaved evergreen

50

60

Tree cover broadleaved deciduous

60, 61, 62

70

Tree cover needleleaved evergreen

70, 71, 72

80

Tree cover needleleaved deciduous

80, 81, 82

90

Tree cover mixed leaf type

90

100

Mosaic tree and shrub | herbaceous cover

100, 110

120

Shrubland

120, 121, 122

150

Sparse vegetation

150, 151, 152, 153

160

Tree cover flooded

160, 170

180

Shrub or herbaceous cover flooded

180

200

Bare areas

200, 201, 202

In the table, each row represents a reclassified land cover class, identified by a unique code. The 'Original ESA CCI Classes' column lists the specific land cover classes from the European Space Agency Climate Change Initiative dataset that are grouped together to form each broader category. Note that land cover classes not listed in this table were retained in their original value and were not reclassified.

File Format

The dataset comprises observations spanning from January 1997 to November 2022, capturing data for 51,978 samples.

blue.rds: Time series data for the blue spectral band.

green.rds: Time series data for the green spectral band.

red.rds: Time series data for the red spectral band.

nir.rds: Time series data for the near-infrared (NIR) spectral band.

swir1.rds: Time series data for the shortwave infrared 1 (SWIR1) spectral band.

swir2.rds: Time series data for the shortwave infrared 2 (SWIR2) spectral band.

thermal.rds: Time series data for the thermal infrared band.

clear.rds: Time series data for the clear quality band, used for masking out cloudy observations.

How open the files in R:

blue <- readRDS("blue.rds")

To open the files in Python, you need to the pyreadr library:

import pyreadrblue = pyreadr.read_r('blue.rds')