Skip to content

Commit 5ae897f

Browse files
ghidalgo3ghidalgo3
authored
Venezuela Damage 2026 Collection (#374)
* Venezuela Damage 2026 Collection * Venezuela Predicted Damage Dataset * move collection
1 parent 7cef77d commit 5ae897f

4 files changed

Lines changed: 175 additions & 0 deletions

File tree

‎datasets/predicted-damage-venezuela-2026/README.md‎

Lines changed: 17 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,17 @@
1+
# Venezuela Damage 2026
2+
3+
## First-time publishing
4+
5+
First you need to validate the STAC collection with `pctasks dataset validate-collection [path-to-template.json]`, fix any validation errors.
6+
7+
Then submit the collection ingestion with `pctasks dataset ingest-collection -d datasets/predicted-damage-venezuela-2026/dataset.yaml -s -a registry pccomponents`
8+
9+
Get the workflow ID and then watch it with: `pctasks runs status $WORKFLOW_ID --watch`.
10+
It must succeed.
11+
12+
Verify that it was successful with `curl "https://planetarycomputer.microsoft.com/api/stac/v1/collections/predicted-damage-venezuela-2026"`
13+
14+
## Updating
15+
16+
Simply add `-u` to the command.
17+
`pctasks dataset ingest-collection -d datasets/conus404/dataset.yaml -u -s -a registry pccomponents`

‎datasets/predicted-damage-venezuela-2026/collection/description.md‎

Lines changed: 11 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,11 @@
1+
Following the June 2026 earthquakes that struck Catia La Mar, Venezuela, the [Microsoft AI for Good Lab](AI For Good Lab - Microsoft Research) ran their building damage assessment AI models to map and classify the buildings affected by the earthquakes. The models were applied to satellite imagery captured on 25 June 2026, producing building footprints with predicted damage across the affected area.
2+
3+
This collection mirrors the [Venezuela Earthquakes: Building Damage Assessment in Catia La Mar](https://data.humdata.org/dataset/venezuela-earthquakes-catia-la-mar) dataset published on the Humanitarian Data Exchange (HDX) by the Microsoft AI for Good Lab and made available under the [Creative Commons Attribution 4.0 International (CC-BY-4.0)](https://creativecommons.org/licenses/by/4.0/) license.
4+
5+
## Methodology
6+
7+
The Microsoft AI for Good Lab ran their damage assessment AI models on satellite images taken 25 June 2026 and have mapped out the affected buildings.
8+
9+
## Caveats
10+
11+
Please note that the source imagery has orthorectification quality issues, which may cause building footprints to be misaligned with existing basemaps.

‎datasets/predicted-damage-venezuela-2026/collection/template.json‎

Lines changed: 134 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,134 @@
1+
{
2+
"type": "Collection",
3+
"id": "predicted-damage-venezuela-2026",
4+
"stac_version": "1.0.0",
5+
"description": "{{ collection.description }}",
6+
"links": [
7+
{
8+
"rel": "cite-as",
9+
"href": "https://www.microsoft.com/en-us/research/group/ai-for-good-research-lab/",
10+
"type": "text/html",
11+
"title": "AI For Good Lab Harnessing AI to help solve some of the world’s greatest challenges"
12+
},
13+
{
14+
"rel": "license",
15+
"href": "https://creativecommons.org/licenses/by/4.0/",
16+
"type": "text/html",
17+
"title": "Creative Commons Attribution 4.0 International"
18+
},
19+
{
20+
"rel": "describedby",
21+
"href": "https://planetarycomputer.microsoft.com/dataset/predicted-damage-venezuela-2026",
22+
"type": "text/html",
23+
"title": "Human readable dataset overview and reference"
24+
}
25+
],
26+
"stac_extensions": [
27+
"https://stac-extensions.github.io/table/v1.2.0/schema.json",
28+
"https://stac-extensions.github.io/storage/v2.0.0/schema.json"
29+
],
30+
"item_assets": {
31+
"data": {
32+
"type": "application/geopackage+sqlite3",
33+
"roles": [
34+
"data"
35+
],
36+
"title": "Predicted building damage footprints",
37+
"description": "GeoPackage of building footprints with predicted earthquake damage for Catia La Mar, Venezuela.",
38+
"table:storage_options": {
39+
"account_name": "ai4edataeuwest"
40+
}
41+
}
42+
},
43+
"msft:region": "westeurope",
44+
"table:columns": [
45+
{
46+
"name": "geometry",
47+
"type": "byte_array",
48+
"description": "Building footprint polygons"
49+
}
50+
],
51+
"msft:container": "ai4good",
52+
"msft:storage_account": "ai4edataeuwest",
53+
"msft:short_description": "AI-predicted building damage footprints for Catia La Mar, Venezuela, following the June 2026 earthquakes.",
54+
"title": "Predicted Building Damage: Venezuela Earthquakes (Catia La Mar) 2026",
55+
"extent": {
56+
"spatial": {
57+
"bbox": [
58+
[
59+
-73.354,
60+
0.647,
61+
-59.7589,
62+
12.2019
63+
]
64+
]
65+
},
66+
"temporal": {
67+
"interval": [
68+
[
69+
"2026-06-25T00:00:00Z",
70+
"2026-06-25T23:59:59Z"
71+
]
72+
]
73+
}
74+
},
75+
"license": "CC-BY-4.0",
76+
"keywords": [
77+
"Venezuela",
78+
"Catia La Mar",
79+
"Earthquake",
80+
"Building damage",
81+
"Damage assessment",
82+
"Buildings",
83+
"Microsoft AI for Good Lab"
84+
],
85+
"providers": [
86+
{
87+
"name": "Microsoft AI for Good Lab",
88+
"roles": [
89+
"producer",
90+
"licensor"
91+
],
92+
"url": "https://www.microsoft.com/en-us/research/group/ai-for-good-research-lab/"
93+
},
94+
{
95+
"name": "Microsoft",
96+
"roles": [
97+
"host",
98+
"processor"
99+
],
100+
"url": "https://planetarycomputer.microsoft.com"
101+
}
102+
],
103+
"assets": {
104+
"thumbnail": {
105+
"href": "https://ai4edatasetspublicassets.blob.core.windows.net/assets/pc_thumbnails/venezuela2026.png",
106+
"type": "image/png",
107+
"title": "Predicted Building Damage Thumbnail",
108+
"roles": [
109+
"thumbnail"
110+
]
111+
},
112+
"data": {
113+
"href": "https://ai4edataeuwest.blob.core.windows.net/ai4good/venezuela2026/predicted_damage_catia_la_mar_footprints.gpkg",
114+
"type": "application/geopackage+sqlite3",
115+
"title": "Predicted building damage footprints",
116+
"description": "GeoPackage of building footprints with predicted earthquake damage for Catia La Mar, Venezuela.",
117+
"table:storage_options": {
118+
"account_name": "ai4edataeuwest"
119+
},
120+
"roles": [
121+
"data"
122+
]
123+
}
124+
},
125+
"storage:schemes": {
126+
"azure": {
127+
"type": "ms-azure",
128+
"platform": "https://{account}.blob.core.windows.net",
129+
"account": "ai4edataeuwest",
130+
"container": "ai4good",
131+
"region": "westeurope"
132+
}
133+
}
134+
}

‎datasets/predicted-damage-venezuela-2026/dataset.yaml‎

Lines changed: 13 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,13 @@
1+
id: predicted-damage-venezuela-2026
2+
image: ${{ args.registry }}/pctasks-task-base:latest
3+
4+
args:
5+
- registry
6+
7+
collections:
8+
- id: predicted-damage-venezuela-2026
9+
template: ${{ local.path(./collection) }}
10+
class: pctasks.dataset.collection:PremadeItemCollection
11+
asset_storage: []
12+
chunk_storage:
13+
uri: "blob://cpdataeuwest/cpdata-etl-data/chunks"

0 commit comments

Comments
 (0)