BigQuery to Earth Engine automation with Cloud Function

1. Sign-up Earth Engine

Sign up with the project owner/admin email account

• Earth Engine Sign-up

2. Provision artifacts for demo

Go to cloud shell, run the following commands

cd $HOME
git clone https://github.com/GoogleCloudPlatform/earthengine-bigquery-automation-demo.git
cd $HOME/earthengine-bigquery-automation-demo && chmod +x start.sh
./start.sh

If something goes wrong, run ./start/sh again.

Set permissions

• Sign up with the service account with Earth Engine
• Earth Engine Service account Sign-up
• Go to Service Accounts under IAM in your cloud project
• Identify the service account named 'App Engine default service account' and copy the email address
• In Earth Engine registration sign-up screen, enter the service account email in box: Register a new service account for XXXX, click "Register service account" button, the service account will show up in the Current service accounts list

Validation steps:

• Go to BigQuery console, check the dataset and table created.
• Go to Cloud Scheduler screen, run the job task manually
• Go to Cloud storage, and check the PROJECT_ID-bq_export_bucket, verify one csv file created with current date as file name( Month-DD-YYYY)

3. Open Earth Engine Code Editor and connect to GCP project

• Web Code Editor When prompted user, use the project owner who has signed-up the earth engine. Click the user/profile Icon on top right screen, Click Choose Cloud Project, choose "select existing project", choose the cloud project from dropdown list Validation steps:
• Code Editor console, click the Asset tab, verify there is asset listed under PROJECT_ID with name of MONTH-DD-YYYY

4. Earth Engine analysis with in Code Editor

run base script in Code Editor

Create a script

• Click Script tab in Code Editor console. Create a new Script, give name as app.js (you may need to create a folder first when prompt to do so)
• Copy the contents of cloud shell $HOME/earthengine-bigquery-automation-demo/ee_app.js to app.js in Code Editor screen
• Change line 10 with proper values for PROJECT_ID

• Click Save and Run button in Code Editor console
• Verify no error, and you can see the polygon areas created.

run advanced analytics script in Code Editor

Create a script

• Click Script tab in Code Editor console. Create a new Script, give name as adavanceAnalytics.js
• Copy the contents of cloud shell $HOME/earthengine-bigquery-automation-demoo/ee_advanceAnlytics.js to advanceAnalytics.js in Code Editor screen
• Change line 9 and line 13 with proper values for PROJECT_ID

• Click Save and Run button in Code Editor console
• Verify no error, and you can see the polygon areas created.
• Click Tasks tab( with yellow color in top right panel), click Run button.
• After task run completed, go to Cloud Storage bucket PROJECT_ID-ee_export_bucket and verify file created

5. Create Earth Engine apps

Update link for app.js

• From advanceAnalytics.js in Code Editor, click Get Link dropdown, click Copy script path,
• Switch to app.js in Code Editor, in line 14, replace the ADVANCE_URL with script path just copied. Save app.js

Create apps for app.js

• Click apps from app.js Code Editor
• Click New App
• Provide App Name, Choose Existing Project
• In source code step, choose Repository script Path.
• In next screen, type in app.js.
• Click Next, and then Publish
• In next screen, there will be URL link below the app-name. Validate the URL works after a minute

Set permissions

• Go to Service Accounts under IAM in your cloud project
• Identify the service account named 'Service Account for Earth Engine App' and copy the email address
• On the IAM page, click Add
• Set New Principal: copied email address and Role: Earth Engine Resource Viewer

Validations:

• Open the app link
• Click "Advanced Analytics" button in right panel
• A separate window will open
• Click the Run button,
• Check on/off different Tree Cover layers

Congratulations! Demo completed successfully