GIS and Remote Sensing for Early Warning Training Course

Participants can organise spatial datasets for consistent early warning analysis and map production.

• Vector, raster and tabular data structures for hazard monitoring
• Coordinate reference systems, datum selection and projection-related map distortion
• Geodatabases structured around hazards, assets, populations and administrative boundaries
• Metadata fields recording source, date, resolution, accuracy and restrictions
• Data dictionaries defining attributes, codes, units and update responsibilities
• Version control procedures for operational layers and incident map products

Participants can assess whether geospatial data are fit for specified warning decisions.

• Positional, thematic, temporal and logical consistency quality dimensions
• Completeness checks for settlement, road, facility and population datasets
• Ground-truth observations compared against mapped features and classifications
• Error matrices used to assess remotely sensed classification accuracy
• Uncertainty statements matched to map scale and decision consequence
• Quality assurance logs recording checks, corrections, approvals and publication dates

Participants can select suitable satellite and ancillary data sources for different hazard questions.

• Spatial, spectral, temporal and radiometric resolution trade-offs
• Sentinel-1 radar imagery for cloud-affected flood monitoring
• Sentinel-2 optical imagery for vegetation, burn scars and surface water
• Landsat thermal bands for land-surface temperature assessment
• MODIS products supporting regional fire and drought monitoring
• Gauge, radar, forecast and satellite data combined for validation

Participants can prepare satellite imagery for visual interpretation and quantitative analysis.

• Radiometric calibration converting digital numbers into physical reflectance values
• Atmospheric correction methods for comparable multi-date optical imagery
• Speckle filtering applied to synthetic aperture radar observations
• Cloud, shadow and snow masking before optical image analysis
• Image mosaicking and clipping to defined operational monitoring areas
• False-colour composites designed for water, vegetation and burned land

Participants can derive terrain and drainage indicators for flood and landslide assessment.

• Digital elevation model selection based on terrain and mapping scale
• Slope, aspect and curvature derived for terrain instability screening
• Flow direction and accumulation calculated from conditioned elevation surfaces
• Watershed delineation using pour points and drainage networks
• Topographic Wetness Index applied to saturated-area susceptibility assessment
• Stream buffers and floodplain zones generated for exposure screening

Participants can map flood extent and analyse affected areas using multiple evidence sources.

• Normalized Difference Water Index used for optical water delineation
• SAR backscatter thresholds applied to floodwater classification
• Pre-event and post-event imagery compared through change detection
• Permanent water masks separated from newly inundated land
• Flood extent intersected with buildings, roads and critical facilities
• Gauge thresholds linked with observed inundation and operational impacts

Participants can analyse drought indicators and identify areas requiring early action.

• Normalized Difference Vegetation Index interpreted against seasonal baselines
• Vegetation Condition Index calculated from historical NDVI ranges
• Standardized Precipitation Index used for rainfall deficit classification
• Soil-moisture anomalies combined with rainfall and vegetation indicators
• Crop calendars aligned with monitoring windows and warning thresholds
• Administrative summaries identify districts with persistent drought stress

Participants can develop a landslide susceptibility screening model from spatial conditioning factors.

• Slope gradient, geology and land cover compiled as conditioning factors
• Distance-to-road and distance-to-stream layers calculated for susceptibility
• Rainfall intensity-duration thresholds linked to documented landslide occurrence
• Weighted overlay methods used for transparent susceptibility scoring
• Landslide inventories separated into training and validation datasets
• Susceptibility classes communicated with limitations and monitoring priorities

Participants can identify active fire indicators and assess burned-area impacts.

• VIIRS active-fire detections interpreted with confidence and latency attributes
• Fire Weather Index components linked to ignition and spread conditions
• Normalized Burn Ratio calculated from pre-fire and post-fire imagery
• Differenced Normalized Burn Ratio used for burn-severity classification
• Fuel, slope and settlement proximity combined for priority monitoring
• Burned areas intersected with watersheds, roads and protected areas

Participants can translate storm forecasts into mapped areas of probable impact.

• Track forecasts mapped with forecast cones and timing uncertainty
• Wind swaths generated from intensity, radius and directional assumptions
• Storm-surge zones overlaid with elevation and coastal settlement data
• Rainfall forecasts aggregated by catchment and administrative boundary
• Critical infrastructure exposure assessed within forecast impact zones
• Scenario maps distinguish observed conditions from forecast probabilities

Participants can build exposure and vulnerability layers that support impact estimates.

• Population grids compared with census polygons and settlement footprints
• Critical facilities classified by function, capacity and service dependency
• Building exposure estimated through footprints, occupancy and construction types
• Social vulnerability indicators selected using transparent inclusion criteria
• Network analysis identifies access constraints to hospitals and shelters
• Exposure databases updated after construction, displacement and service changes

Participants can combine hazard, exposure and vulnerability evidence into priority areas for action.

• Risk matrices link likelihood, consequence and decision escalation levels
• Raster overlay combines hazard intensity with exposure and vulnerability
• Multi-criteria decision analysis documents weights, scores and assumptions
• Hotspot analysis identifies statistically significant clusters of potential impact
• Priority rankings distinguish life safety, service disruption and economic loss
• Sensitivity testing shows effects of alternative weights and thresholds

Participants can design trigger logic connecting forecast hazards with anticipated impacts and actions.

• Impact-based forecasting connects hazard probability with exposed vulnerable elements
• Trigger matrices define thresholds, confidence levels, actions and owners
• Forecast-based financing thresholds linked to pre-agreed early actions
• Lead-time analysis identifies feasible preparedness and evacuation measures
• Alert levels aligned with national emergency management arrangements
• Decision logs record evidence, assumptions, authorisation and issued messages

Participants can produce clear map and message products for operational warning communication.

• Cartographic hierarchy directs attention to hazard, impact and required action
• Colour schemes distinguish alert levels while supporting colour-vision accessibility
• Scale bars, north arrows and dates support map interpretation
• Common Alerting Protocol fields structure multi-channel warning messages
• Protective-action instructions matched to locations, populations and lead times
• Uncertainty language states confidence, update timing and information limitations

Participants can specify operational information products for incident monitoring and coordination.

• Dashboard indicators selected from agreed operational decision requirements
• Situation reports distinguish confirmed observations from preliminary assessments
• Web map layers organised by hazard, impact, response and reference
• Update frequencies assigned according to hazard speed and data latency
• Incident geodatabases preserve timestamps, sources and verification status
• Briefing templates summarise affected locations, trends, actions and gaps

Participants can establish responsible data-sharing arrangements for geospatial early warning operations.

• Data-sharing agreements specify access, ownership, retention and permitted uses
• Personal data risks assessed for household, health and displacement records
• Minimum data standards support exchange between agencies and partners
• Open Geospatial Consortium services support interoperable map publication
• Community observations validated without exposing reporters to unnecessary risk
• Access controls protect sensitive infrastructure and vulnerable population locations

Participants can build, review and present a complete GIS and remote-sensing early warning workflow for a defined multi-hazard area.

• Hazard scenario defined with geographic extent, lead time and consequences
• Data inventory documents sources, resolution, update cycles and limitations
• Monitoring indicators selected from satellite, forecast and field observations
• Exposure and vulnerability analysis identifies priority populations and assets
• Trigger matrix links evidence thresholds with authorised early actions
• Warning map and CAP message prepared for specified recipient groups
• Peer review tests assumptions, uncertainty statements and operational feasibility
• Implementation roadmap assigns owners, data updates and review milestones