Landsat Geospatial Dataset

In this file, the necessary technical details of the dataset is explained.

Location of the Landsat Dataset Files

The LandSat geospatial dataset files are located under the following directory accessible from multiple clusters:

# DRAC Graham HPC location
/project/rpp-kshook/Model_Output/Landsat/ #rpp-kshook allocation

# UCalgary ARC HPC location
/work/comphyd_lab/data/geospatial-data/Landsat #comphyd_lab allocation

# Perdue University Anvil HPC location
/anvil/datasets/geospatial/NALCMS # Perdue University's Anvil cluster

And the structure of the files is as following:

/path/to/dataset/dir/
├── land_change_2010v2_2015v2_30m_tif.zip
├── land_cover_2005v3_250m_tif.zip
├── land_cover_2010v2_250m_tif.zip
├── land_cover_2010v2_30m_tif.zip
├── land_cover_2015v2_30m_tif.zip
├── land_cover_2020_30m_tif.zip
└── dl-landcover.sh

Spatial and Temporal Extents

The spatial extent of this dataset (so far only NALCMS that is a land cover dataset) covers longitudes from approximately -180 to -50 degress and latitudes from approximately +14 to +84 degress. This dataset is available for 2005 and 2020. Furthermore, one static dataset variable is also included demonstrating the gains and losses of various land covers between 2010 and 2015. Please see the list of variables for more information.

Note

The 250m resolution datasets for 2005 and 2010 are not Landsat products, but MODIS. However, as a temporary measure, these two datasets are included in the landsat workflow in gistool.

Dataset Variables

This variables of this dataset are detailed in the table below:

#

Variable Name (used in gistool)

Description

Comments

1

land-cover

Land cover classes for 2005, 2010, 2015 and 2020

2005 dataset info, 2010-30m dataset info, 2010-250m dataset info, 2015 dataset info, 2020 dataset info

2

land-cover-change

Land cover change (between 2010 and 2015)

2010-2015 difference info

Other relevant information

Land Cover Types

Below the land cover types for each values of the .tif files is detailed based on [1].

#

Class Value (used in gistool)

Land Cover Description

RGB Value

1

1

Temperate or sub-polar needleleaf forest

RGB (0, 61, 0)

2

2

Sub-polar taiga needleleaf forest

RGB (148, 156, 112)

3

3

Tropical or sub-tropical broadleaf evergreen forest

RGB (0, 99, 0)

4

4

Tropical or sub-tropical broadleaf deciduous forest

RGB (30, 171, 5)

5

5

Temperate or sub-polar broadleaf deciduous forest

RGB (20, 140, 61)

6

6

Mixed forest

RGB (92, 117, 43)

7

7

Tropical or sub-tropical shrubland

RGB (179, 158, 43)

8

8

Temperate or sub-polar shrubland

RGB (179, 138, 51)

9

9

Tropical or sub-tropical grassland

RGB (232, 220, 94)

10

10

Temperate or sub-polar grassland

RGB (225, 207, 138)

11

11

Sub-polar or polar shrubland-lichen-moss

RGB (156, 117, 84)

12

12

Sub-polar or polar grassland-lichen-moss

RGB (186, 212, 143)

13

13

Sub-polar or polar barren-lichen-moss

RGB (64, 138, 112)

14

14

Wetland

RGB (107, 163, 138)

15

15

Cropland

RGB (230, 174, 102)

16

16

Barren lands

RGB (168, 171, 174)

17

17

Urban

RGB (220, 33, 38)

18

18

Water

RGB (76, 112, 163)

19

19

Snow and ice

RGB (255, 250, 255)

Also, the details of the above table has been included in the following files: landsat_classes.csv.

Land Cover Changes between 2010 and 2015

The following file contains the following GeoTIFF data:

/path/to/dataset/dir/land_change_2010v2_2015v2_30m_tif.zip
├── NA_NALCMS_2010v2_2015v2_30m_barren_land_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_cropland_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_forest_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_grassland_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_shrubland_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_snow_ice_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_urban_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_water_loss_and_gain.tif
├── NA_NALCMS_2010v2_2015v2_30m_wetland_loss_and_gain.tif
└── NA_NALCMS_land_change_2010v2_2015v2_30m.tif

Based on the metadata also available in the mentioned .zip file, the GeoTIFF data containing the *_land_loss_and_gain.tif phrase in their file name contain “[g]eneral land cover gains an losses […] at the North American scale […] for the following land cover class groups:”

1. Forest
2. Shrubland
3. Grassland
4. Wetland
5. Cropland
6. Barren Land
7. Urban and Built-up
8. Water
9. Snow and Ice

And, the pixel values of the mentioned .tif files are as following:

#

Pixel Value

Value Description

1

1

Gain

2

2

Loss

Furthermore, the GeoTIFF data named NA_NALCMS_land_change_2010v2_2015v2_30m.tif also contains the landcover change data. Each pixel value of the GeoTIFF file contains information regarding transformations between different landcover classes of the dataset at the North American scale. Each pixel values of the file contains three or four digits, with “[t]he first one or two digits in each pixel value show the land cover class in 2010, while the third and four digits show the land cover class in 2015.” An example of digit values are given in the following:

105   = Class 1 to 5    "Temperate or sub-polar needleleaf forest" to "Temperate or sub-polar broadleaf deciduous forest"
206   = Class 2 to 6    "Sub-polar taiga needleleaf forest" to "Mixed forest"
814   = Class 8 to 14   "Temperate or sub-polar shrubland" to "Wetland"
915   = Class 9 to 15   "Tropical or sub-tropical grassland" to "Cropland"
1018  = Class 10 to 18  "Temperate or sub-polar grassland" to "Water"
1502  = Class 15 to 02  "Cropland" to "Sub-polar taiga needleleaf forest"
1716  = Class 17 to 16  "Urban and built-up" to "Barren land"
1913  = Class 19 to 01  "Snow and ice" to "Sub-polar or polar barren-lichen-moss"