You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Note: the conversion of natural lands alerts are already pre-filtered by the SBTN natural lands map
Solution Hints
We can use the current DIST alerts layer we are currently using, applying further filtering based on the datasets below to aggregate by the different drivers.
In the example below, confmask is a confidence filter, selecting for alerts where VEG-DIST-COUNT >= 2 AND VEG-ANOM-MAX > 50
Also, natLands_binary filters for Natural lands using ee.Image('WRI/SBTN/naturalLands/v1/2020').select('natural') (again, similar to the current tooling) and is applied to Wildfires and Conversion to avoid erroneous alerts.
There are some alerts that remain unclassified. The data layers get merged together hierarchically, fire is the highest, then flooding, then crop cycles, then conversion. So, if a pixel is classified as wildfire and conversion, then the final classification is wildfire.
The text was updated successfully, but these errors were encountered:
@yellowcap confirming that we can apply the new Wildfire, Crop cycle, and Flooding layers on top of the current DIST alerts tool. Wildfire, Crop cycle, and Flooding are all simple binary masks that apply to alerts in the range Nov 2023 - Oct 2024 (something to bare in mind for the demo).
Note that the demo below also filters on confidence and natural lands.
Goal
See https://landandcarbon.projects.earthengine.app/view/beta-land-disturbance-alert-classification-system#lon=-51.33205213832403;lat=-29.8785949661847;zoom=16;
Note: the conversion of natural lands alerts are already pre-filtered by the SBTN natural lands map
Solution Hints
We can use the current DIST alerts layer we are currently using, applying further filtering based on the datasets below to aggregate by the different drivers.
In the example below,
confmask
is a confidence filter, selecting for alerts whereVEG-DIST-COUNT >= 2 AND VEG-ANOM-MAX > 50
Also,
natLands_binary
filters for Natural lands usingee.Image('WRI/SBTN/naturalLands/v1/2020').select('natural')
(again, similar to the current tooling) and is applied to Wildfires and Conversion to avoid erroneous alerts.See https://code.earthengine.google.com/06f83955d12f278414f8c53655abdd6d
Wildfire
Crop Cycle
Flooding
Conversion
There are some alerts that remain unclassified. The data layers get merged together hierarchically, fire is the highest, then flooding, then crop cycles, then conversion. So, if a pixel is classified as wildfire and conversion, then the final classification is wildfire.
The text was updated successfully, but these errors were encountered: