GIS for Environmental Research Training Course

Participants can select and structure vector, raster and tabular data for environmental studies.

• Vector geometry types matched to environmental features and measurement scales
• Raster grids interpreted through cell size, extent, bands and pixel values
• Attribute schemas designed for field observations, monitoring indicators and provenance
• Feature classes, shapefiles, GeoPackages and geodatabases compared for project use
• Relational joins connecting survey tables, sampling sites and spatial features
• Data dictionaries specifying field definitions, domains, units and accepted values

Participants can assign, transform and verify coordinate reference systems for reliable analysis.

• Geographic and projected coordinate reference systems distinguished by intended measurement
• Datum selection assessed against source data, region and positional accuracy
• EPSG codes used to document coordinate reference system definitions
• On-the-fly projection checked before distance, area and overlay calculations
• Georeferencing scanned maps using ground control points and residual errors
• Transformation methods selected when datasets use different geodetic datums

Participants can source environmental spatial data and record its fitness for purpose.

• Primary, secondary and derived datasets assessed for research suitability
• Satellite, aerial, sensor and field data sources compared by resolution
• Metadata records capture lineage, collection date, scale, licence and limitations
• ISO 19115 metadata elements applied to environmental spatial data records
• Open data licences reviewed before data sharing, publication and reuse
• Field collection protocols linked to identifiers, coordinates and observation timestamps

Participants can prepare environmental datasets using systematic spatial and attribute quality checks.

• Topology rules identify gaps, overlaps, slivers and invalid polygon boundaries
• Attribute validation detects missing values, duplicates, outliers and coding errors
• Spatial joins checked against feature geometry, predicates and expected records
• Raster no-data values distinguished from valid zero measurements
• Field coordinates screened for impossible locations, precision and datum mismatches
• Quality-control logs record corrections, source versions and unresolved issues

Participants can integrate field observations with GIS layers for environmental analysis.

• Sampling points imported from GPS, spreadsheets and mobile data collection systems
• Unique identifiers maintain links between field forms, photographs and spatial features
• Survey effort represented through transects, plots, routes and observation periods
• Field observations aggregated into indicators without obscuring sampling variation
• Temporal fields standardised for seasonal comparisons and repeat monitoring
• Positional accuracy recorded for handheld GPS, mobile devices and digitised features

Participants can use vector analysis methods to quantify spatial relationships and environmental exposure.

• Buffer zones estimate influence areas around rivers, roads and protected sites
• Spatial overlay combines habitats, land tenure and development proposal layers
• Intersect and union operations calculate affected area by environmental category
• Dissolve functions aggregate fragmented features using defined ecological attributes
• Nearest-neighbour analysis measures proximity to infrastructure, hazards and services
• Area-weighted summaries calculate land-cover proportions within reporting zones

Participants can derive terrain and drainage variables from digital elevation models.

• Digital elevation model resolution evaluated against catchment and site-scale questions
• Slope, aspect and curvature derived for erosion and habitat assessments
• Fill-sink processing prepares elevation surfaces for hydrological modelling
• Flow direction and accumulation identify likely drainage pathways
• Watersheds delineated from pour points, streams and elevation data
• Topographic wetness index interpreted for soil moisture and runoff potential

Participants can process raster datasets and calculate environmental indices for spatial interpretation.

• Raster algebra combines environmental variables using explicit mathematical expressions
• Reclassification converts continuous surfaces into policy-relevant suitability classes
• Zonal statistics summarise raster values within management and sampling areas
• NDVI calculated from red and near-infrared reflectance bands
• Cell statistics compare temperature, rainfall and vegetation datasets through time
• Resampling methods selected according to categorical or continuous raster values

Participants can select satellite imagery appropriate for environmental monitoring objectives.

• Spectral signatures distinguish water, vegetation, soil and built surfaces
• Spatial, spectral, temporal and radiometric resolution compared for study requirements
• Landsat and Sentinel imagery selected using revisit period and band availability
• Cloud cover, haze and shadows assessed before image acquisition
• Surface reflectance products distinguished from top-of-atmosphere reflectance products
• Image composites assembled using seasonal windows and valid-pixel criteria

Participants can prepare satellite imagery for classification and environmental change analysis.

• Band stacks assembled from multispectral imagery using consistent spatial alignment
• Cloud and shadow masks remove contaminated pixels before analysis
• Clipping imagery reduces processing extent to the defined study area
• Atmospheric correction rationale assessed when comparing imagery across acquisition dates
• Spectral indices generated using calibrated reflectance bands and documented formulas
• Image mosaics checked for seamlines, date differences and radiometric inconsistency

Participants can create land-cover classifications using suitable training data and class definitions.

• Classification schemes define mutually exclusive classes and minimum mapping units
• Training samples selected across spectral variation and known land-cover conditions
• Supervised classification applies labelled samples to multispectral image bands
• Random forest classification compared with maximum likelihood classification assumptions
• Unsupervised clustering evaluated for exploratory land-cover pattern identification
• Post-classification filtering removes isolated pixels while retaining meaningful features

Participants can assess classification quality and report uncertainty using reference data.

• Reference samples separated from training data for independent accuracy assessment
• Confusion matrices compare classified labels against observed reference classes
• Overall, producer's and user's accuracy calculated from classification results
• Stratified random sampling allocates validation points across mapped classes
• Positional, thematic and temporal uncertainty documented in research reporting
• Validation findings used to revise class definitions and training samples

Participants can identify and quantify environmental change using multi-date spatial datasets.

• Image differencing measures change between comparable raster acquisition dates
• Post-classification comparison identifies transitions among land-cover classes
• Change matrices quantify gains, losses and persistent environmental categories
• Seasonal effects separated from persistent change using matched acquisition periods
• Time-series charts summarise vegetation-index trends at monitoring locations
• Change thresholds justified using sensor noise, validation data and research purpose

Participants can create and assess interpolated environmental surfaces from point measurements.

• Interpolation suitability assessed against sampling density, distribution and environmental processes
• Inverse distance weighting estimates surfaces from nearby measured observations
• Ordinary kriging models spatial autocorrelation through variograms and neighbourhood parameters
• Cross-validation compares predicted values with withheld monitoring observations
• Prediction error maps communicate uncertainty alongside interpolated environmental surfaces
• Interpolation outputs constrained using catchment boundaries, elevation or land-cover masks

Participants can build transparent suitability models for environmental planning decisions.

• Decision criteria derived from policy requirements, ecological evidence and stakeholder priorities
• Constraints distinguish prohibited locations from lower-ranked but permissible locations
• Standardisation converts incompatible environmental variables into comparable suitability scores
• Weighted linear combination combines criterion scores using documented importance weights
• Analytic Hierarchy Process supports pairwise weighting of competing decision criteria
• Sensitivity analysis tests how changing weights alters recommended locations

Participants can test and interpret spatial patterns while avoiding unsupported conclusions.

• Spatial autocorrelation explains why nearby observations may not be independent
• Global Moran's I tests overall clustering of environmental indicator values
• Hot spot analysis identifies statistically concentrated high and low values
• Kernel density estimation maps intensity of incidents, observations and detections
• Modifiable areal unit problem considered when selecting reporting boundaries
• Ecological fallacy avoided when interpreting aggregated environmental spatial data

Participants can produce clear maps and technical outputs for environmental decision-makers.

• Map purpose determines projection, scale, classification and visual hierarchy
• Colour schemes selected for sequential, diverging and categorical environmental variables
• Legends, labels and annotations communicate units, dates and source information
• Map layouts include north arrows, scale bars, credits and coordinate grids
• Uncertainty represented through confidence classes, hatching and explanatory notes
• Technical reports connect methods, findings, limitations and decision implications

Participants can document GIS workflows so environmental analyses can be reviewed and repeated.

• Project folder structures separate raw data, processed data, outputs and documentation
• File naming conventions encode dates, versions, geography and processing status
• Processing models record inputs, parameters, outputs and analytical sequence
• Version logs track dataset updates, methodological decisions and reviewer comments
• Metadata and methods notes support internal review and external audit
• Data-sharing plans address access controls, sensitive locations and retention requirements

Participants can build, review and present a complete environmental GIS analysis addressing a defined research question.

• Research question, study extent and decision context confirmed before analysis
• Source datasets prepared with documented coordinate systems, metadata and quality checks
• Vector, raster or remote-sensing methods selected to answer the question
• Results validated through reference data, peer review and uncertainty assessment
• Final maps communicate findings, limitations, assumptions and recommended actions
• Project presentation defends methodological choices before technical and nontechnical stakeholders