Myanmar Earthquake: Naypyidaw Building Damage Assessment

This dataset is part of Myanmar Earthquake
  • 70+ Downloads
  • This dataset updates: Never
Interactive Data
Additional information
Time Period of the Dataset [?]
March 31, 2025-March 31, 2025 ... More
Modified [?]
1 April 2025
Dataset Added on HDX [?]
1 April 2025 Less
Expected Update Frequency
Never
Location
Source
Microsoft AI for Good Lab
Methodology

Microsoft AI4G Lab ran their damage assessment AI models on images provided by Planet and have mapped out the affected buildings.

Caveats / Comments

While the data provides a valuable first look, it should serve as a preliminary guide and will require on-the-ground verification for a complete understanding.

File Format
Visibility
Public
Export metadata for this dataset: JSON | CSV
[{"date": "2024-11-04", "value": 0}, {"date": "2024-11-11", "value": 0}, {"date": "2024-11-18", "value": 0}, {"date": "2024-11-25", "value": 0}, {"date": "2024-12-02", "value": 0}, {"date": "2024-12-09", "value": 0}, {"date": "2024-12-16", "value": 0}, {"date": "2024-12-23", "value": 0}, {"date": "2024-12-30", "value": 0}, {"date": "2025-01-06", "value": 0}, {"date": "2025-01-13", "value": 0}, {"date": "2025-01-20", "value": 0}, {"date": "2025-01-27", "value": 0}, {"date": "2025-02-03", "value": 0}, {"date": "2025-02-10", "value": 0}, {"date": "2025-02-17", "value": 0}, {"date": "2025-02-24", "value": 0}, {"date": "2025-03-03", "value": 0}, {"date": "2025-03-10", "value": 0}, {"date": "2025-03-17", "value": 0}, {"date": "2025-03-24", "value": 0}, {"value": 102, "date": "2025-03-31"}, {"value": 34, "date": "2025-04-07"}, {"value": 9, "date": "2025-04-14"}, {"date": "2025-04-21", "value": 0}]

Downloads

Related Showcases

There are no showcases for this dataset.


Data and Resources [1]
  • naypyidaw_myanmar_results_03313025_skysat.gpkgGeopackage (4.6M)
    Modified: 1 April 2025

    The images cover Naypyidaw, Myanmar.

    Analysis was done on Planet imagery captured 03/31/2025.

    -19,837 buildings with damage fraction between 0% and 20% -19 buildings with damage fraction between 40% and 60% -14 buildings with damage fraction between 60% and 80% -24 buildings with damage fraction between 80% and 100% -70 buildings with damage fraction between 20% and 40%

    While the data provides a valuable first look, it should serve as a preliminary guide and will require on-the-ground verification for a complete understanding.

    The result file contains the following fields for each building footprint: -damage_pct_0m​ – the fraction of the building footprint's area that is classified as damaged by our model -damaged​ – 1 if damage_pct_0m > 0 else 0 -unknown_pct – fraction of the pixels within the building footprint that we think are obstructed (clouds, smoke, haze, too dark to evaluate)