Pakistan 30m land use land cover and carbon storage dataset (1990-2020)

[waleedgeo]

Contact Details

waleedgeo@outlook.com

Dataset description

This dataset, generated from the study titled Urbanization-led land cover change impacts terrestrial carbon storage capacity: A high-resolution remote sensing-based nation-wide assessment in Pakistan (1990–2020), provides high-resolution, national-scale Land Use/Land Cover (LULC) and terrestrial carbon stock maps for Pakistan across four time periods: 1990, 2000, 2010, and 2020. Developed using multi-sensor satellite imagery and advanced classification techniques within Google Earth Engine (GEE), this dataset offers a detailed analysis of land cover transformations driven by urbanization and their effects on carbon storage capacity over 30 years.

The LULC dataset includes nine distinct land cover classes (outlined in the table below). Classification was performed using a hybrid random forest-based machine learning approach, with model training and validation carried out using approximately 40,000 stratified random samples to ensure robust accuracy.

The carbon stock maps were produced using the InVEST model, estimating carbon storage across four major carbon pools: above-ground biomass, below-ground biomass, soil organic carbon, and dead organic matter. The results highlight a substantial reduction in carbon storage capacity due to rapid urban expansion, particularly in major cities such as Karachi and Lahore, where large areas of forest and agricultural land were converted into urban landscapes. The study estimates that Pakistan lost approximately 5% of its carbon storage capacity during this period, while urban areas expanded by over 1040%.

For more details, you can read the full paper here.

Citation

If you find this work useful, please consider citing it as:

Waleed, M., Sajjad, M., & Shazil, M. S. (2024). Urbanization-led land cover change impacts terrestrial carbon storage capacity: A high-resolution remote sensing-based nation-wide assessment in Pakistan (1990–2020). Environmental Impact Assessment Review, 105, 107396. https://doi.org/10.1016/j.eiar.2023.107396

For more details, visit:

• Github Project Repository Link
• Paper DOI Link
• Paper PDF Link:

Note: To bulk download these datasets, please check Zenodo: Link to Zenodo

Contact

For any query, please contact Mirza Waleed at his email or his LinkedIn. For more details, check his personal website.

LULC Classification Table

LULC Class	Class Value	Visual
Forest Cover	1	#54bb19
Agriculture/Cropland	3	#affd08
Rangeland	4	#d1fbb9
Wetlands	5	#652ff3
Barren Lands	6	#fed483
Water Bodies	7	#005ce6
Built-up Areas	8	#e50600
Snow/Ice	9	#fe4fcd

Earth Engine Snippet

// LULC Images
var lulc1990 = ee.Image('projects/pak-var/assets/lulc_pk/img1990');
var lulc2000 = ee.Image('projects/pak-var/assets/lulc_pk/img2000');
var lulc2010 = ee.Image('projects/pak-var/assets/lulc_pk/img2010');
var lulc2020 = ee.Image('projects/pak-var/assets/lulc_pk/img2020');

// Carbon Stock Images
var carbon1990 = ee.Image('projects/pak-var/assets/carbon_pk/img1990').multiply(10);
var carbon2000 = ee.Image('projects/pak-var/assets/carbon_pk/img2000').multiply(10);
var carbon2010 = ee.Image('projects/pak-var/assets/carbon_pk/img2010').multiply(10);
var carbon2020 = ee.Image('projects/pak-var/assets/carbon_pk/img2020').multiply(10);

Sample Code: https://code.earthengine.google.com/0b5a71516494fbac2148398af5db7475

The dataset is also available as interactive Google Earth Engine (GEE) applications.

Link to GEE - LULC App for Pakistan: https://waleedgis.users.earthengine.app/view/pakistan-lulc-1990-2020

Link to GEE - Carbon App for Pakistan: https://waleedgis.users.earthengine.app/view/pakistan-carbon-1990-2020

Enter license information

Creative Commons Attribution 4.0 International

Keywords

land use land cover, pakistan, lulc, carbon, urban, google earth engine, geospatial, remote sensing, gis

Code of Conduct

• [X] I agree to follow this project's Code of Conduct

[asdsaah]

@samapriya Data Download Link: Yes - Uploaded to GEE and working as of 10/23/24. Citation: Yes - reference DOI. License for Use: Yes - data is available under Creative Commons Attribution 4.0 International.

[samapriya]

Sorry this took a bit @waleedgeo This has now been completed and available in the community catalog. Thank you for contributing as always