GIS and Remote Sensing for Agriculture Training Course

Participants can design a spatial data structure for an agricultural monitoring project.

• Feature classes for fields, farms, irrigation assets, and sampling points
• Raster layers for imagery, rainfall, elevation, and soil properties
• Attribute tables linking field identifiers, crop records, and survey results
• Unique identifiers for joining production, input, and farmer datasets
• Geodatabases, folder conventions, and controlled file naming
• Data dictionaries specifying fields, units, codes, and allowed values

Participants can prepare GPS-derived agricultural data for accurate spatial analysis.

• WGS 84 coordinates, projected systems, and local mapping requirements
• UTM zones and distortion implications for agricultural area calculations
• GPS waypoints, tracks, polygons, and positional accuracy records
• Field boundary capture using vertices, offsets, and satellite basemaps
• Mobile data forms for crop observations and geotagged photographs
• Quality checks for duplicate points, missing attributes, and implausible locations

Participants can source, document and assess fitness of agricultural spatial datasets.

• Open satellite, climate, elevation, soil, and administrative boundary sources
• Sentinel-2 acquisition dates, cloud cover, bands, and processing levels
• Field survey records, extension registers, and farmer-reported information
• Metadata elements covering source, date, projection, resolution, and limitations
• Data licensing, consent, privacy, and farmer location confidentiality
• Fitness-for-purpose assessment using currency, coverage, accuracy, and scale

Participants can use QGIS to organise, query and display agricultural spatial data.

• QGIS project setup, layer properties, symbology, and map units
• Layer filtering for crop types, seasons, irrigation schemes, and districts
• Attribute queries using logical operators, null values, and coded fields
• Spatial selection by location, intersection, containment, and proximity
• Joins and relates between field polygons and seasonal production tables
• Map layouts with scale bars, legends, north arrows, and source notes

Participants can calculate and analyse field-level agricultural indicators in GIS.

• Digitising field boundaries from imagery using consistent editing rules
• Polygon topology checks for gaps, overlaps, slivers, and duplicates
• Field area calculation using suitable projected coordinate reference systems
• Buffer analysis around irrigation canals, roads, markets, and water points
• Overlay operations combining parcels, soil units, slopes, and protected areas
• Zonal statistics linking raster conditions to individual field polygons

Participants can interpret how crop and soil conditions appear in multispectral imagery.

• Electromagnetic spectrum regions used for vegetation, moisture, and soil assessment
• Spectral signatures of healthy crops, stressed vegetation, bare soil, and water
• Red, near-infrared, and shortwave infrared band responses
• Atmospheric effects, haze, aerosols, clouds, and cloud shadows
• Phenology, planting dates, canopy development, and harvest timing
• Sensor selection based on field size, revisit interval, and budget

Participants can prepare satellite imagery for reliable agricultural interpretation.

• Surface reflectance products and atmospheric correction rationale
• Cloud masks, shadow masks, and scene-quality assessment procedures
• Clipping imagery to project boundaries and agricultural areas
• Band stacking, mosaicking, resampling, and pixel alignment checks
• Temporal composites using median, maximum, and minimum pixel values
• Image metadata review for acquisition conditions and processing history

Participants can calculate and interpret spectral indices for agricultural monitoring.

• NDVI calculation from near-infrared and red reflectance bands
• EVI for dense vegetation and reduced atmospheric sensitivity
• NDWI and moisture indices for irrigation and water-stress assessment
• Bare Soil Index interpretation for exposed soil conditions
• Index thresholds calibrated with crop stage and field observations
• Time-series profiles distinguishing seasonal crops, perennial crops, and fallow land

Participants can assess crop condition trends against seasonal baselines.

• Dekadal and monthly vegetation-index time series for crop monitoring
• Seasonal baselines built from comparable historical observation periods
• Anomaly calculation using departure from median and standardized values
• Rainfall estimates compared with vegetation response and crop calendars
• Early warning indicators for delayed planting, dry spells, and poor biomass
• Interpretation limits caused by clouds, mixed pixels, and management differences

Participants can create a supervised classification for agricultural land-cover categories.

• Classification scheme design for crops, fallow, rangeland, water, and settlements
• Training samples selected from field records, imagery, and reference maps
• Supervised classification using Random Forest and spectral predictor bands
• Object-based and pixel-based classification method selection
• Post-classification editing using expert rules and ancillary datasets
• Class aggregation aligned with reporting requirements and decision needs

Participants can validate agricultural classifications and report uncertainty clearly.

• Reference sample design using random, stratified, and systematic approaches
• Confusion matrices comparing classified labels with reference observations
• Producer's accuracy, user's accuracy, overall accuracy, and kappa interpretation
• Area-adjusted estimates for land-cover class reporting
• Ground-truth protocols for crop type, growth stage, and management observations
• Validation limitations arising from timing mismatch and sample bias

Participants can create a transparent land-suitability model for a specified crop.

• Digital elevation models for slope, aspect, flow direction, and drainage
• Soil texture, pH, organic carbon, and depth data interpretation
• FAO land suitability classes and factor-rating approaches
• Reclassification of slope, soil, rainfall, and access criteria
• Weighted overlay using documented criteria weights and sensitivity checks
• Suitability constraints from erosion risk, flooding, and protected land

Participants can analyse spatial water conditions relevant to agricultural production.

• Irrigation scheme mapping using canals, pumps, fields, and command areas
• Surface water detection from multispectral imagery and water indices
• Watershed delineation using flow accumulation and stream networks
• Drought indicators combining rainfall anomalies, vegetation anomalies, and field reports
• Water-point accessibility analysed through distance, terrain, and road constraints
• Irrigation performance indicators mapped by season and field block

Participants can combine mapped crop area with production evidence for estimation.

• Crop area estimation from classified maps and field boundary datasets
• Yield data cleaning for units, outliers, missing values, and season labels
• Production estimates calculated from crop area and yield assumptions
• Sampling frames for crop-cutting surveys and area measurement
• Regression concepts linking vegetation indices, rainfall, and observed yield
• Uncertainty statements for area, yield, production, and extrapolation limits

Participants can assess rangeland condition and livestock-service accessibility using spatial data.

• Rangeland cover classes distinguishing pasture, shrubs, bare ground, and invasive species
• Vegetation productivity trends for seasonal grazing-condition assessment
• Grazing pressure proxies using settlements, water points, and livestock routes
• Travel-time analysis for veterinary services, markets, and water access
• Pastoral mobility corridors mapped from local knowledge and terrain
• Conflict-sensitive mapping of shared resources and seasonal access arrangements

Participants can apply basic spatial statistics and sampling logic to agricultural datasets.

• Spatial autocorrelation and its consequences for agricultural survey interpretation
• Point pattern mapping for pest outbreaks, animal disease, and input distribution
• Kernel density estimation for clustered agricultural service demand
• Stratified sampling using agroecological zones, crop types, and accessibility
• Interpolation concepts for rainfall, soil measurements, and field observations
• Outlier investigation using maps, attribute distributions, and local knowledge

Participants can produce maps and interpretation notes for agricultural decision makers.

• Map purpose defined for operational, strategic, reporting, or early-warning decisions
• Visual hierarchy through colour, classification, labels, and annotation
• Colour schemes suitable for crop classes, anomalies, and accessibility surfaces
• Classification methods including natural breaks, quantiles, and fixed thresholds
• Interpretation notes separating observed evidence, assumptions, and recommendations
• Technical map review covering projection, sources, dates, scale, and limitations

Participants can establish a repeatable and accountable agricultural geospatial workflow.

• Standard operating procedures for seasonal imagery acquisition and processing
• Version control principles for datasets, scripts, maps, and analysis outputs
• Quality assurance checkpoints before publication, reporting, and operational decisions
• Metadata registers for datasets, processing steps, and responsible officers
• Data-sharing protocols for farmer data, partner data, and public products
• Work plans linking crop calendars, satellite schedules, validation, and reporting

Participants can build, review and present a complete GIS and remote sensing analysis for an agricultural resource management decision.

• Problem statement defining agricultural decision, users, geography, and seasonal timeframe
• Data inventory documenting imagery, field observations, boundaries, and supporting layers
• Analysis workflow combining preprocessing, indices, classification, and spatial overlay
• Validation plan using reference samples, field evidence, and accuracy metrics
• Decision map portfolio presenting findings, limitations, and recommended actions
• Peer review against technical quality, reproducibility, and decision usefulness