GIS for Food Security Analysis Training Course

Participants can create a structured spatial data foundation for food-security analysis.

• Vector, raster and tabular data structures for food-security indicators
• Geodatabases designed around places, households, facilities and seasons
• Coordinate reference systems, datum transformations and positional accuracy
• Metadata records using ISO 19115 core documentation elements
• Unique identifiers connecting survey records with geographic features
• File naming, version control and reproducible project folder structures

Participants can acquire, inspect and document spatial datasets from relevant sources.

• Administrative boundaries sourced from national statistical and mapping agencies
• Rainfall estimates from CHIRPS and satellite precipitation products
• Earth observation imagery from Sentinel-2 and Landsat collections
• Market price, survey and nutrition datasets prepared for joins
• Completeness, consistency and timeliness checks using data-quality rules
• Privacy risks in household locations and sensitive vulnerability attributes

Participants can clean, transform and integrate multi-source datasets for mapping.

• Cleaning field names, coded values and missing observation records
• Geocoding markets, facilities and settlements using validated reference gazetteers
• Spatial joins linking household indicators to administrative polygons
• Raster resampling aligned to target resolution and analysis extent
• Temporal aggregation of monthly observations into agricultural seasons
• Outlier screening for implausible prices, yields and survey coordinates

Participants can produce decision-focused maps that communicate food-security evidence accurately.

• Map purpose statements defining audience, decision and required evidence
• Choropleth classification using quantiles, natural breaks and policy thresholds
• Colour schemes distinguishing severity classes without misleading visual emphasis
• Scale bars, north arrows and coordinate grids used appropriately
• Labelling dense market, settlement and administrative boundary maps
• Map layouts documenting sources, dates, methods and uncertainty notes

Participants can select and prepare satellite imagery for agricultural monitoring questions.

• Electromagnetic spectrum bands relevant to vegetation, water and bare soil
• Sentinel-2 spatial, spectral and temporal resolution trade-offs
• Cloud masking and image compositing for seasonal agricultural monitoring
• Surface reflectance products distinguished from top-of-atmosphere reflectance
• Image mosaics clipped to livelihood zones and assessment areas
• Pixel resolution matched to field size and reporting requirements

Participants can derive and interpret vegetation indicators for crop-condition assessment.

• NDVI calculated from near-infrared and red reflectance bands
• EVI compared with NDVI under dense vegetation conditions
• Seasonal NDVI profiles aligned with local crop phenology
• Vegetation condition index derived from historical NDVI ranges
• Anomaly maps comparing current greenness against baseline periods
• Interpreting low vegetation values alongside rainfall and land-cover context

Participants can classify land cover and evaluate crop-mapping results using reference data.

• Training samples designed across crops, rangeland, water and settlements
• Supervised classification using random forest predictor variables
• Spectral signatures inspected before assigning land-cover classes
• Confusion matrices calculating producer, user and overall accuracy
• Reference data split into training and independent validation samples
• Post-classification filtering removes isolated pixels and classification noise

Participants can map rainfall deficits and relate climate anomalies to seasonal production risks.

• CHIRPS rainfall totals aggregated across planting and growing periods
• Rainfall anomalies calculated against a defined climatological baseline
• Standardized Precipitation Index interpreted across drought timescales
• Dekadal rainfall monitoring aligned with crop-water requirement stages
• Zonal statistics summarising precipitation by livelihood and administrative areas
• Drought maps cross-checked against vegetation and field-report evidence

Participants can analyse biophysical factors affecting agricultural production potential.

• Digital elevation models used to derive slope and aspect
• Slope thresholds screening unsuitable land for mechanised cultivation
• Soil texture, depth and drainage attributes prepared as raster layers
• FAO land evaluation factors matched to specified crop requirements
• Euclidean distance surfaces representing proximity to irrigation infrastructure
• Constraint masks excluding protected areas, water bodies and settlements

Participants can model physical access to markets and essential services.

• Road networks cleaned for connectivity, direction and travel impedance
• Travel speeds assigned by road class, surface and seasonal conditions
• Cost-distance surfaces model walking access beyond mapped roads
• Network service areas identifying populations within market travel thresholds
• Least-cost paths linking remote settlements to distribution locations
• Accessibility results compared with observed prices and dietary diversity

Participants can integrate household survey indicators with geographic analysis units.

• Food Consumption Score and coping indicators joined to locations
• Sampling weights retained when summarising geographically referenced survey results
• Point displacement risks assessed for confidential household survey coordinates
• Aggregation methods selected for villages, wards and livelihood zones
• Rate calculations use population denominators and suppression thresholds
• Survey maps distinguish sampled findings from area-wide estimates

Participants can test and interpret spatial clustering in food-security indicators.

• Spatial autocorrelation assessed using Global Moran's I statistic
• Local Moran's I identifies high-high and low-low clusters
• Getis-Ord Gi* maps statistically significant hot and cold spots
• Spatial weights matrices defined by distance, contiguity or neighbours
• Multiple-testing cautions applied when interpreting local cluster results
• Cluster findings validated against livelihoods, hazards and data coverage

Participants can build transparent composite risk maps from multiple food-security indicators.

• Risk model structure separates hazard, exposure and vulnerability components
• Indicator normalisation converts variables to comparable measurement scales
• Analytic Hierarchy Process derives and checks criterion weights
• Weighted linear combination produces composite food-security risk scores
• Sensitivity analysis tests alternative weights and classification thresholds
• Risk categories documented with assumptions, limitations and intended decisions

Participants can identify and justify priority locations for programme action.

• Targeting criteria combine severity, population, access and operational feasibility
• Population rasters estimate people within high-risk spatial zones
• Overlay analysis identifies communities facing multiple concurrent deprivations
• Exclusion criteria account for security constraints and inaccessible terrain
• Priority rankings compared with existing programme coverage and partner presence
• Targeting maps distinguish needs estimates from beneficiary selection decisions

Participants can analyse temporal signals supporting early-warning monitoring.

• Time-series tables structured by location, indicator and observation date
• Seasonal decomposition separates trend, seasonal pattern and residual variation
• Rolling averages reduce short-term noise in market-price monitoring
• Threshold alerts defined from historical distributions and operational triggers
• Lag analysis compares rainfall, vegetation and staple-price movements
• Early-warning dashboards specify update frequency, owners and escalation actions

Participants can use Google Earth Engine workflows for scalable indicator production.

• Image collections filtered by date, geography and cloud properties
• JavaScript functions calculate NDVI and seasonal composite imagery
• Reducers summarise raster indicators within administrative boundary features
• Feature collections export tabular zonal statistics for reporting
• Batch processing creates repeatable outputs across multiple assessment areas
• Script comments document data sources, parameters and processing decisions

Participants can evaluate analytical uncertainty and communicate limitations responsibly.

• Positional, thematic, temporal and attribute uncertainty distinguished clearly
• Ground-truth observations used to validate remotely sensed interpretations
• Error propagation considered when combining multiple uncertain input layers
• Confidence statements matched to evidence quality and spatial resolution
• Sensitive maps reviewed for protection risks and unintended consequences
• Methods notes specify reproducibility requirements, assumptions and data gaps

Participants can convert spatial analysis into usable food-security reporting products.

• Map series designed for national, subnational and field-level decisions
• Analytical narratives connect geographic patterns with operational implications
• Situation reports present indicators, maps, caveats and recommended actions
• Web map layers organised with clear names and access permissions
• Standard operating procedures define data updates and quality checks
• Decision briefings distinguish observed evidence, assumptions and proposed responses

Participants can build, review and present a complete food-security spatial assessment with maps, methods and recommendations.

• Define an assessment question, geographic scope and decision audience
• Assemble crop, rainfall, market, population and vulnerability datasets
• Build a documented workflow using spatial and remote-sensing methods
• Produce risk, access or crop-condition maps with quality checks
• Review assumptions, uncertainty, targeting implications and ethical considerations
• Present findings, recommendations and reproducible analysis documentation to peers