Python for Hydrological and Geospatial Analysis Training Course

Participants can select, transform and document coordinate reference systems for catchment analysis.

• Coordinate reference systems, datums, ellipsoids and projected coordinate distortions
• EPSG identifiers and WKT definitions inspected with pyproj
• Transform point, vector and raster data between appropriate CRSs
• Assess UTM, national grids and equal-area projections for catchments
• Detect axis-order, unit and datum errors in imported datasets
• Document coordinate assumptions within data dictionaries and processing logs

Participants can process, validate and analyse vector datasets for hydrological applications.

• GeoPandas GeoDataFrames: geometry columns, attributes, indexes and joins
• Read Shapefile, GeoPackage and GeoJSON with Fiona-backed drivers
• Spatial predicates for contains, intersects, within and nearest operations
• Clip administrative boundaries and dissolve subcatchments by reporting units
• Calculate geodesic and projected area, length and centroid measures
• Validate geometries and repair self-intersections using Shapely methods

Participants can prepare aligned raster datasets for terrain, imagery and hydrological calculations.

• Rasterio datasets: bands, transforms, nodata masks and metadata
• Windowed reading and writing for large satellite raster collections
• Resample grids using nearest, bilinear and cubic convolution methods
• Align pixel grids, extents and CRSs before raster calculations
• Derive raster statistics by polygon using raster masking
• Build Cloud-Optimized GeoTIFF outputs with compression and overviews

Participants can derive terrain surfaces and drainage networks from prepared elevation data.

• Acquire and inspect DEM tiles, vertical units and void values
• Fill sinks using priority-flood depression treatment before flow routing
• Calculate slope, aspect and hillshade from terrain derivatives
• D8 flow direction and accumulation grids for drainage extraction
• Threshold contributing area to define preliminary stream networks
• Compare derived channels against mapped hydrography and imagery

Participants can delineate catchments and calculate physical descriptors for hydrological analysis.

• Snap pour points to high-accumulation cells using distance tolerances
• Delineate watersheds and nested subbasins from conditioned elevation grids
• Compute drainage density, relief and hypsometric catchment descriptors
• Extract longest flow paths and travel-distance raster surfaces
• Review boundaries against ridgelines, culverts and engineered drainage
• Export catchment polygons with provenance fields and area checks

Participants can clean, quality-check and structure rainfall and streamflow records.

• Parse station records with timezone, units and sampling-interval controls
• Construct regular time indexes and identify missing observation periods
• Apply range, persistence and rate-of-change quality-control flags
• Aggregate sub-hourly rainfall and discharge to operational reporting periods
• Separate water years, seasons and event windows with pandas
• Create station metadata tables linking gauges, basins and sensors

Participants can process multidimensional climate datasets for basin-scale hydrological use.

• Open NetCDF and GRIB datasets with xarray and CF metadata
• Select dimensions, coordinates and variables from multidimensional climate cubes
• Apply temporal resampling, rolling windows and groupby climatologies
• Convert precipitation fluxes and temperatures using explicit unit checks
• Subset gridded data by basin masks and bounding geometries
• Write analysis-ready NetCDF files with coordinates and encoding

Participants can derive water, vegetation and land-surface indicators from satellite imagery.

• Interpret spectral reflectance, radiometric scaling and quality-assurance bands
• Load Sentinel-2 and Landsat scenes through STAC catalog searches
• Mask clouds, shadows and snow using scene classification flags
• Calculate NDVI, NDWI and MNDWI with band-math expressions
• Create seasonal composites using median reducers and observation counts
• Assess sensor resolution, revisit intervals and mixed-pixel limitations

Participants can create and assess land-cover classifications for catchment analysis.

• Prepare labelled samples from reference polygons and classified imagery
• Random Forest classification using spectral bands and terrain predictors
• Split training and validation samples without spatial data leakage
• Calculate confusion matrices, producer accuracy and user accuracy
• Post-process isolated pixels with majority filters and minimum mapping units
• Summarise land-cover change by subcatchment and reporting period

Participants can estimate basin rainfall from gauges and gridded precipitation products.

• Retrieve gauge, radar and satellite precipitation data for basin analysis
• Apply Thiessen polygons and inverse-distance weighting to gauge networks
• Calculate areal rainfall from gridded products and basin masks
• Compare IMERG estimates against gauges using bias and error metrics
• Bias-correct daily precipitation using ratio and quantile-mapping approaches
• Document missing gauges, interpolation choices and uncertainty bounds

Participants can quantify hydrograph behaviour, flow events and runoff characteristics.

• Plot hydrographs with precipitation, discharge and threshold annotations
• Calculate baseflow using digital filters and recession analysis
• Identify events using inter-event time definitions and flow thresholds
• Derive peak flow, runoff volume, lag time and runoff coefficient
• Construct flow-duration curves and exceedance probability tables
• Estimate annual maxima series for flood-frequency screening analyses

Participants can implement a conceptual rainfall-runoff model for a defined catchment.

• Represent catchment storage using conceptual rainfall-runoff model structures
• Implement a soil-moisture accounting model with Python functions
• Route excess rainfall through linear reservoirs and unit hydrographs
• Specify parameters, initial conditions and forcing-data requirements explicitly
• Run baseline simulations across calibration and validation periods
• Compare model structure assumptions against catchment process evidence

Participants can calibrate hydrological models and report performance with appropriate metrics.

• Calibrate hydrological parameters using grid search and optimisation objectives
• Calculate Nash-Sutcliffe efficiency, KGE, percent bias and RMSE values
• Separate calibration, validation and test periods without temporal leakage
• Inspect residual hydrographs, peak errors and low-flow performance
• Conduct one-at-a-time sensitivity analysis for influential model parameters
• Report parameter bounds, objective functions and reproducible calibration settings

Participants can quantify and communicate uncertainty in hydrological modelling and scenario results.

• Distinguish input, parameter, structural and observational uncertainty sources
• Generate parameter ensembles using Monte Carlo sampling methods
• Propagate precipitation uncertainty through repeated model simulations
• Summarise prediction intervals, exceedance probabilities and ensemble spread
• Perform scenario analysis for land-cover change and climate perturbations
• Communicate uncertainty without overstating forecast precision or model confidence

Participants can prepare transparent flood screening maps and exposure summaries.

• Combine terrain, flow accumulation and imperviousness for screening maps
• Map flood-prone corridors using HAND and relative elevation methods
• Overlay exposed roads, facilities and population datasets with hazard zones
• Rank subcatchments using weighted criteria and transparent scoring rules
• Distinguish screening outputs from hydraulic inundation model results
• Create response-ready maps with scale, north arrow and data dates

Participants can produce basin drought indicators from climate, streamflow and imagery datasets.

• Calculate SPI and SPEI from precipitation and evapotranspiration series
• Derive Standardized Streamflow Index values from discharge records
• Map vegetation stress using NDVI anomalies and seasonal baselines
• Define drought thresholds, durations and severity classes for reporting
• Compare meteorological, agricultural and hydrological drought indicators
• Produce basin drought bulletins with anomalies and confidence notes

Participants can communicate hydrological findings through clear maps, figures and automated reports.

• Design hydrographs, hyetographs and duration curves for technical audiences
• Build static maps with Matplotlib, scale bars and classified legends
• Use contextily basemaps while managing projection and attribution requirements
• Create interactive Folium maps with layers, popups and controls
• Generate tables and figures automatically from parameterised Python scripts
• Apply cartographic hierarchy, colour conventions and accessible labels

Participants can package hydrological analyses for maintenance, review and repeated execution.

• Structure reusable workflows with functions, configuration files and logging
• Manage dependencies using conda environments and pinned package versions
• Add unit tests for transformations, units and watershed outputs
• Profile bottlenecks and apply Dask to larger raster computations
• Record lineage through input inventories, metadata and processing logs
• Package scripts, outputs and README instructions for peer review

Participants can build, review and present a complete Python-based hydrological and geospatial basin analysis.

• Frame a basin management question with decision criteria and scope
• Prepare terrain, hydroclimate, remote-sensing and vector input datasets
• Delineate catchments, derive indicators and document processing assumptions
• Model rainfall-runoff response and evaluate performance against observed flows
• Map flood or drought indicators with stated uncertainty limitations
• Review code, data lineage, maps and findings against acceptance criteria
• Present decision findings, methods and recommended next analytical actions