Python for GIS and Spatial Analysis Training Course

Participants can write and debug Python code used in routine GIS processing tasks.

• Variables, strings, numbers and Boolean values used in GIS scripts
• Lists and dictionaries storing layer names, fields and processing settings
• Conditional statements controlling data checks and processing branches
• For loops batch-processing files, records and geographic features
• Functions packaging repeatable GIS tasks with arguments and return values
• Exceptions and logging used to diagnose failed spatial processing steps

Participants can select suitable spatial data models and formats for an analysis requirement.

• Vector geometry types: points, lines, polygons and multipart features
• Raster grids representing imagery, elevation, temperature and classified surfaces
• GeoJSON, GeoPackage, Shapefile and GeoTIFF format selection criteria
• Attribute tables linked to features through identifiers and field definitions
• Metadata records documenting source, date, resolution, projection and limitations
• Data licensing checks for open geographic datasets and imagery sources

Participants can diagnose CRS problems and transform spatial data for valid analysis.

• Geographic and projected coordinate reference systems distinguished by purpose
• EPSG codes used to identify coordinate reference system definitions
• Datum, ellipsoid and coordinate units affecting positional accuracy
• Projection selection based on area, distance, direction and location
• GeoPandas transformations performed with set_crs and to_crs methods
• Spatial layers checked for mismatched CRS before overlay operations

Participants can load, inspect, filter and export vector datasets using GeoPandas.

• GeoPandas GeoDataFrames created from GeoPackage, GeoJSON and shapefile sources
• Geometry columns inspected for type, extent, missing values and validity
• Attribute fields filtered using Boolean masks and pandas query expressions
• Columns renamed, recoded and calculated using pandas vectorised operations
• Feature subsets selected by administrative area, category and date
• Processed layers exported to GeoPackage with defined layer names

Participants can identify and correct common spatial data-quality defects before analysis.

• Invalid polygons identified using geometry validity checks and error reports
• Duplicate features detected through identifiers, geometry comparison and attribute rules
• Missing coordinates, null geometries and empty features handled explicitly
• Topology errors assessed through gaps, overlaps, slivers and self-intersections
• Geometry repair methods compared before replacing source data
• Quality-control logs recording checks, exceptions, corrections and residual risks

Participants can connect business data to geographic locations and assess match quality.

• Tabular CSV data read with explicit field types and encodings
• Attribute joins performed using standardised keys and duplicate checks
• Addresses cleaned through parsing, normalisation and postal-code validation
• Geocoding results assessed using match scores and positional review
• Unmatched records classified for correction, exclusion or manual investigation
• Geocoded point layers linked to operational, customer or programme records

Participants can combine geographic layers using suitable spatial predicates and overlay operations.

• Spatial predicates compared: intersects, within, contains, touches and crosses
• Point-in-polygon joins assigning locations to administrative or service areas
• Buffer zones generated around roads, assets, facilities and hazard sites
• Intersection overlays calculating affected land, population or asset counts
• Difference and union operations identifying changes between geographic boundaries
• Area calculations completed after selecting an appropriate projected CRS

Participants can measure proximity and accessibility using Euclidean and network-based methods.

• Euclidean distance calculated for preliminary proximity screening and prioritisation
• Nearest-feature analysis matching demand points to candidate service locations
• Buffers dissolved to measure overlapping coverage and service gaps
• Travel-time catchments distinguished from straight-line distance measures
• Accessibility indicators calculated for populations within defined service thresholds
• Results interpreted against data quality, transport assumptions and equity concerns

Participants can model routes and service access using network analysis concepts.

• Network nodes, edges, impedance and turn restrictions defined clearly
• Shortest-path calculations applied to inspection routes and emergency access
• Origin-destination matrices measuring travel cost between demand and facilities
• Service-area polygons generated from travel-time or distance thresholds
• Network datasets checked for disconnected segments and missing attributes
• Routing assumptions documented for operational planning and scenario comparison

Participants can inspect, transform and subset raster data using GDAL-based workflows.

• Raster bands, pixels, resolution, extent and nodata values inspected
• GDAL tools used for format conversion, reprojection and clipping
• Raster windows read to process large imagery files efficiently
• Resampling methods compared: nearest neighbour, bilinear and cubic convolution
• Digital elevation models prepared for terrain and hydrological analysis
• Raster metadata checked for acquisition date, sensor and processing level

Participants can derive environmental and terrain indicators from raster datasets.

• Map algebra combines raster layers using arithmetic and conditional expressions
• Slope, aspect and hillshade derived from digital elevation models
• Reclassification groups continuous values into defined analytical classes
• Suitability models combine weighted criteria using documented scoring rules
• Nodata cells managed to prevent distorted summary statistics
• Zonal statistics summarise raster values within administrative or asset polygons

Participants can prepare satellite imagery and derive spectral indicators for monitoring tasks.

• Multispectral bands interpreted across visible, near-infrared and shortwave wavelengths
• Sentinel-2 imagery selected by date, cloud cover and spatial resolution
• Cloud masks applied before calculating land and water indicators
• NDVI calculated to measure vegetation greenness across observation dates
• NDWI and NDBI compared for water and built-up feature detection
• Index outputs validated against reference imagery and known ground conditions

Participants can produce and evaluate basic land-cover classifications and change-detection outputs.

• Training samples prepared from representative land-cover reference areas
• Supervised classification uses labelled pixels and selected spectral predictors
• Random Forest classification parameters assessed for reproducible image outputs
• Confusion matrices report producer accuracy, user accuracy and overall accuracy
• Post-classification comparison identifies transitions between mapped land-cover classes
• Change maps screened for seasonal variation, clouds and registration errors

Participants can apply spatial statistical measures and interpret results with appropriate caution.

• Spatial autocorrelation explains dependence among nearby observations and residuals
• Global Moran's I tests overall clustering of mapped attribute values
• Local Indicators of Spatial Association identify local clusters and outliers
• Getis-Ord Gi* identifies statistically significant high-value and low-value hotspots
• Spatial weights matrices defined using contiguity, distance bands or neighbours
• Multiple testing and modifiable areal unit risks considered during interpretation

Participants can create clear maps that communicate spatial findings without misleading readers.

• Choropleth classification methods compared: quantile, equal interval and natural breaks
• Colour schemes selected for sequential, diverging and categorical data
• Map layouts include title, legend, scale, north arrow and sources
• Proportional symbols communicate counts while normalised rates support comparison
• Context layers chosen to support interpretation without obscuring key patterns
• Matplotlib and GeoPandas used for scripted static map production

Participants can publish exploratory map outputs suitable for review and internal communication.

• Folium maps created with basemaps, markers, polygons and layer controls
• Popup content presents selected attributes, indicators and source references
• Web map performance improved through simplified geometries and filtered attributes
• Interactive layers organised by theme, time period and decision question
• Export formats selected for web viewing, reporting and GIS handover
• Accessibility considerations address colour contrast, labels and alternative descriptions

Participants can automate a repeatable GIS process and document it for team use.

• Parameterized scripts separate business settings from processing logic
• Command-line arguments support repeatable runs with changing input files
• Unit tests check expected geometry counts, fields and output values
• Logging records processing date, inputs, warnings, errors and completion status
• Configuration files store paths, CRS choices, thresholds and classification rules
• README documentation explains installation, inputs, workflow steps and limitations

Participants can scope a spatial analysis project with defensible methods and acceptance criteria.

• Problem statements define decision users, geographic scope and measurable questions
• Data inventories record sources, owners, update frequency and quality limitations
• Analysis plans specify methods, assumptions, outputs and validation criteria
• Risk registers address privacy, bias, positional error and disclosure controls
• Peer code review checks readability, duplication, assumptions and reproducibility
• Technical reports connect spatial findings to recommendations and uncertainty

Participants can build, review and present a complete Python spatial analysis project addressing a defined decision question.

• Capstone brief defines users, geographic question, datasets and decision criteria
• Data pipeline imports, validates and documents vector, raster and tabular inputs
• Spatial analysis combines selected overlay, proximity, raster or statistical methods
• Quality checks test CRS, geometry validity, attributes and analytical assumptions
• Maps and summary tables communicate findings, uncertainty and recommended actions
• Code review confirms reproducibility, documentation quality and maintainable structure