Google Earth Engine Crop Monitoring and Drought Assessment Training Course

Course overview

Agriculture programmes, commercial farms, insurers and food-security teams need timely evidence of where crops are stressed, how conditions compare with previous seasons, and which locations require field verification or response. Manual image downloads and spreadsheet-based monitoring are too slow for large districts or repeated assessments. This course equips participants to use Google Earth Engine to analyse satellite imagery and rainfall data at scale, producing defensible crop-condition and drought evidence for operational planning, early warning and reporting.

Participants learn the Google Earth Engine Code Editor, JavaScript-based image processing workflows, image collections, geometries, filtering, masking and reduction methods. They calculate and interpret NDVI, EVI and NDWI; create seasonal vegetation composites; analyse rainfall anomalies using CHIRPS data; and build drought indicators that compare current conditions with historical baselines. The course also covers cropland masking, zonal statistics for administrative units or farm blocks, charting time series, and exporting maps and tables for use in reports, dashboards and GIS workflows.

Instruction combines guided demonstrations with daily coding labs using an agriculture and drought-monitoring case area. Participants progressively build a reusable Google Earth Engine crop-monitoring script that imports an area of interest, generates cloud-masked Sentinel-2 vegetation indices, calculates rainfall and vegetation anomalies, summarises results by reporting unit and exports decision-ready outputs. They leave with an annotated script, a drought-assessment map layout, a tabular results export and an implementation plan for applying the workflow to their own geography.

The course is designed for professionals who need to interpret and produce geospatial evidence rather than become remote-sensing researchers. It is particularly suited to teams working in agricultural development, food security, climate adaptation, extension services, agribusiness monitoring and humanitarian early warning.

Course objectives

By the end of this course, participants will be able to:

• Configure a Google Earth Engine Code Editor workspace and organise scripts, assets and areas of interest
• Filter Sentinel-2 image collections by date, location and cloud conditions for crop-monitoring analysis
• Apply cloud and shadow masking to create analysis-ready optical satellite composites
• Calculate NDVI, EVI and NDWI layers and interpret their agricultural significance
• Build dekadal and seasonal vegetation time series for farms, districts or livelihood zones
• Calculate CHIRPS rainfall totals, anomalies and standardised seasonal departures from a baseline
• Produce zonal statistics and ranked drought-stress tables for administrative or operational reporting units
• Export reproducible map layers, charts, CSV tables and an annotated Earth Engine monitoring script

Benefits of attending

Target competencies

Who should attend

• Agricultural Monitoring Officers — who need repeatable satellite-based evidence of crop condition across large areas
• Food Security Analysts — who assess seasonal production risks and support early-warning briefings
• GIS and Remote Sensing Officers — who need to operationalise Google Earth Engine workflows for agriculture programmes
• Climate Adaptation Specialists — who track rainfall deficits and vegetation response for resilience planning
• Agribusiness Field Operations Managers — who require comparable crop-performance indicators across production zones
• Humanitarian Programme Officers — who need spatial evidence to target field verification and drought-response activities

Requirements and prerequisites

Participants should be comfortable using a web browser, managing files and folders, and working with basic tables in Excel or Google Sheets. Familiarity with maps, coordinates, administrative boundaries, crop calendars or agricultural seasons is helpful, but not essential. The course introduces Google Earth Engine JavaScript syntax from the ground up, so prior programming experience is not required; complete beginners should expect to write and adapt short scripts during every day of the course. Participants need a laptop, a reliable internet connection and a Google account that can access Google Earth Engine. Prior experience with GIS software such as QGIS is useful but not assumed.

Training methodology

The course uses short instructor-led concept sessions followed by guided work in the Google Earth Engine Code Editor. Each day, participants adapt working JavaScript examples to an agricultural case area, inspect intermediate outputs and troubleshoot common issues such as cloud contamination, inappropriate date windows and mismatched boundaries. Case discussions connect satellite indicators to crop calendars, rainfall performance and field verification decisions. Small-group reviews test how maps could be interpreted in an early-warning meeting. The final session is an application workshop in which participants refine their own crop-monitoring workflow and implementation plan.

Course outline

Day 1: Earth Engine foundations for agricultural monitoring

• Google Earth Engine data catalogue, Code Editor and asset management
• JavaScript variables, functions and server-side versus client-side operations
• Importing feature collections and defining farm, district and livelihood-zone areas of interest
• Coordinate reference systems, scale and pixel resolution in agricultural analysis
• Image, ImageCollection and FeatureCollection data structures
• Filtering satellite collections by boundary, date and metadata
• Agricultural monitoring questions, crop calendars and indicator selection

Workshop: Participants create an Earth Engine project that imports a reporting boundary, filters Sentinel-2 imagery for a defined growing season and saves a reusable starter script.

Day 2: Satellite imagery and vegetation condition

• Sentinel-2 spectral bands and their relevance to crop observation
• QA60 and S2 Cloud Probability approaches to cloud and shadow masking
• Median, percentile and quality mosaic seasonal composite methods
• NDVI calculation and interpretation for crop canopy development
• EVI calculation for high-biomass crop conditions
• NDWI calculation for vegetation moisture and surface-water context
• False-colour visualisation, legends and map-layer inspection

Workshop: Participants build a cloud-masked Sentinel-2 composite and generate NDVI, EVI and NDWI map layers for a selected agricultural season.

Day 3: Rainfall, anomalies and drought indicators

• CHIRPS rainfall data structure and temporal aggregation
• Dekadal, monthly and seasonal rainfall total calculations
• Long-term baseline construction for rainfall comparison
• Absolute and percentage rainfall anomaly methods
• Vegetation condition index and historical NDVI comparison
• Combining rainfall and vegetation signals for drought interpretation
• Limits of satellite indicators and requirements for field validation

Workshop: Participants calculate seasonal CHIRPS rainfall anomalies and compare them with NDVI departures to identify candidate drought-stress locations.

Day 4: Crop monitoring statistics and operational outputs

• Cropland masks and land-cover considerations for agricultural summaries
• Reducing raster indicators over districts, farms and management blocks
• Mean, median, percentile and area-threshold zonal statistics
• Time-series charting with ui.Chart for rainfall and vegetation trends
• Threshold design for crop stress classification
• Ranking reporting units by vegetation and rainfall conditions
• Exporting GeoTIFF, CSV and chart outputs to Google Drive

Workshop: Participants produce a ranked district table, vegetation time-series chart and exported crop-stress map for an operational reporting cycle.

Day 5: Reproducible drought assessment workflow

• Structuring reusable Earth Engine functions and parameter blocks
• Managing season dates, baseline periods and indicator thresholds
• Quality assurance checks for imagery, masks, boundaries and summary tables
• Interpreting conflicting rainfall and vegetation signals
• Map design for early-warning briefs and management decisions
• Integrating Earth Engine exports with QGIS and spreadsheet reporting
• Workflow documentation, ownership and update schedules

Workshop: Participants complete and present an annotated crop-monitoring and drought-assessment script, map output, results table and 90-day implementation plan.

Tools & standards covered

Google Earth Engine Code Editor, Google Earth Engine JavaScript API, QGIS, Google Drive

A typical training day

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions