GDAEvapotranspiration_EarthCode

EO data processor packaged as a Docker image, built and published through the Insula processor pipeline.

The processor estimates Daily Evapotranspiration from Sentinel-2 and Sentinel-3 SLSTR LST morning acquisition(s) based on the Two-Source Energy Balance Algorithm (TSEB) and machine learning sharpened Land Surface Temperature, utilizing Sen-ET project as baseline.

Structure

• code/ - your processor. Must contain a Dockerfile that builds the image.
• gdaevapotranspiration-earthcode.cwl - the OGC Application Package describing the process (inputs, outputs, command). The dockerPull value __IMAGE__ is filled in automatically at build time; do not edit it.

Base image constraint (important)

The Dockerfile FROM must be a public image (Docker Hub, quay.io, ghcr.io, ...). The build pipeline is public and cannot receive private-registry credentials, so a private base image fails at build time with a 401.

Security scan gate

The pipeline scans the built image (Grype + Trivy) and blocks publishing on HIGH/CRITICAL vulnerabilities, including unfixed ones - most come from the base image. Prefer a slim, freshly patched base (-slim, alpine, distroless), rebuild on its newest patch tag, and keep build tools out of the final stage (multi-stage build). If a genuinely unfixable base CVE still blocks you, contact a pipeline maintainer.

Build and deploy

This repository has no CI of its own. Build and deploy it with the insula-processors-builder CLI (the repo must be public):

insula-processors-builder create --repo-url https://github.com/<you>/gdaevapotranspiration-earthcode

The pipeline clones this repo, builds code/Dockerfile, scans and publishes the image, injects the published image reference into the CWL, and deploys the process to Insula. Iterate by pushing changes and running the command again.