ESA SNAP SAR Image Processing and Terrain Correction Training Course

Course overview

Synthetic Aperture Radar (SAR) data can reveal flood extent, ground deformation, crop condition and infrastructure change when cloud cover or darkness makes optical imagery unreliable. However, raw Sentinel-1 scenes are not immediately decision-ready: they contain radar geometry, speckle, acquisition-mode differences, terrain effects and radiometric distortions that can lead to false interpretation. This course equips GIS and remote sensing professionals to build defensible SAR processing workflows in ESA SNAP, producing spatially accurate, analysis-ready outputs for operational mapping and monitoring.

Participants work through the complete ESA SNAP workflow for Sentinel-1 GRD and SLC products. They learn to inspect SAR metadata; apply precise orbit files, thermal-noise removal and border-noise correction; calibrate imagery to sigma nought and gamma nought; manage speckle; perform multilooking, subsetting and terrain correction; select DEMs, projections and resampling methods; and export outputs for GIS analysis. The programme also introduces image co-registration, coherence, change detection, radiometric terrain flattening and quality checks needed to distinguish meaningful surface change from processing artefacts.

Instruction combines guided demonstrations with repeated hands-on processing in ESA SNAP using realistic land, flood and infrastructure-monitoring cases. Each participant builds a documented Sentinel-1 processing chain, creates terrain-corrected backscatter products, and completes a capstone workflow for a selected monitoring scenario. They leave with exported GeoTIFF outputs, SNAP Graph Processing Tool (GPT) workflow files, processing notes, quality-control criteria and an implementation plan that can be adapted to their organisation's imagery, DEMs and reporting requirements.

Course objectives

By the end of this course, participants will be able to:

• Inspect Sentinel-1 GRD and SLC metadata to select suitable acquisitions, polarisations and processing paths
• Apply precise orbit files, thermal-noise removal and border-noise correction in ESA SNAP
• Calibrate SAR imagery to beta nought, sigma nought and gamma nought backscatter products
• Configure speckle filters, multilooking and spatial subsetting for a defined mapping objective
• Perform Range-Doppler terrain correction using an appropriate DEM, map projection and resampling method
• Create co-registered image pairs and coherence layers for surface-change and deformation screening
• Build reusable SNAP Graph Processing Tool workflows for batch Sentinel-1 preprocessing
• Produce a documented terrain-corrected GeoTIFF package with quality checks and processing provenance

Benefits of attending

Target competencies

Who should attend

• GIS Analysts — who need to integrate cloud-independent Sentinel-1 products into mapping and spatial decision workflows
• Remote Sensing Analysts — who process radar imagery and need reliable calibration, filtering and terrain-correction methods
• Earth Observation Scientists — who require reproducible SAR workflows for land, water, agriculture or infrastructure studies
• Environmental Monitoring Officers — who need repeatable methods for flood, land-cover and environmental change assessment
• Disaster Risk and Emergency Mapping Specialists — who must generate usable radar-derived layers when optical imagery is unavailable
• Geospatial Data Engineers — who support automated satellite-data pipelines and need to operationalise SNAP GPT workflows

Requirements and prerequisites

Participants should be comfortable working with spatial datasets and should understand basic GIS concepts including raster cells, coordinate reference systems, map projections and GeoTIFF files. Familiarity with any desktop GIS, such as QGIS or ArcGIS, is useful because participants will inspect and export spatial layers. No prior SAR processing experience is required, and no programming, advanced mathematics or interferometry background is assumed. Complete beginners in remote sensing can attend, but should expect to spend additional time learning SAR terminology, Sentinel-1 product types and the interpretation of radar backscatter.

Training methodology

The course alternates short instructor-led explanations with guided processing in ESA SNAP. Participants inspect real Sentinel-1 scenes, follow parameter choices through SNAP's processing operators, and compare outputs produced with different filters, DEMs and terrain-correction settings. Case exercises focus on flood mapping, land-surface change and terrain-affected backscatter. Small-group reviews use visual checks and metadata to diagnose artefacts. On the final day, each participant completes and documents a processing workflow, receives structured feedback, and prepares a practical plan for applying it to an organisational use case.

Course outline

Day 1: SAR foundations and ESA SNAP workspace

• Active microwave sensing principles and radar backscatter
• Sentinel-1 mission, acquisition modes and polarisation options
• GRD, SLC and OCN product distinctions
• SAR geometry: range, azimuth, incidence angle and layover
• ESA SNAP interface, product explorer and raster viewer
• Band metadata, tie-point grids and image-information panels
• Opening, organising and inspecting Sentinel-1 SAFE products

Workshop: Participants inspect two Sentinel-1 scenes in ESA SNAP and produce an acquisition-selection note identifying product type, polarisation, incidence-angle range and intended use.

Day 2: Sentinel-1 data preparation

• Copernicus Data Space Ecosystem search criteria
• Acquisition-date, orbit-direction and relative-orbit selection
• Downloading and validating Sentinel-1 SAFE archives
• Precise orbit file application in SNAP
• Thermal-noise removal for Sentinel-1 GRD products
• Border-noise correction and edge-pixel inspection
• Spatial subsetting and band selection for efficient processing

Workshop: Participants prepare a study-area subset from a Sentinel-1 GRD scene and save a parameter record for orbit, noise-removal and band-selection steps.

Day 3: Radiometric calibration and backscatter interpretation

• Digital numbers and calibrated radar backscatter
• Beta nought, sigma nought and gamma nought definitions
• Calibration operator settings in ESA SNAP
• Linear power values and decibel conversion
• VV, VH, HH and HV polarisation interpretation
• Incidence-angle effects on backscatter comparison
• Histogram, profile and pixel-value inspection

Workshop: Participants calibrate a dual-polarised scene to sigma nought and gamma nought, then create a comparison chart explaining the selected output for a mapping task.

Day 4: Speckle reduction and spatial detail management

• Speckle formation and its effect on SAR interpretation
• Single-look complex data and multilooking concepts
• Boxcar, Lee, Refined Lee and Gamma Map filters
• Filter window-size selection and edge preservation
• Equivalent number of looks as a quality indicator
• Trade-offs between noise reduction and spatial resolution
• Visual and statistical assessment of filtered outputs

Workshop: Participants test three speckle filters on the same calibrated image and produce a recommendation matrix based on noise reduction, boundary retention and target visibility.

Day 5: Terrain correction for map-accurate SAR products

• Radar terrain distortions: foreshortening, layover and shadow
• Range-Doppler Terrain Correction operator
• DEM selection: SRTM, Copernicus DEM and local elevation models
• DEM coverage, voids and vertical-reference considerations
• Coordinate reference system and output-pixel spacing choices
• Nearest-neighbour, bilinear and bicubic resampling methods
• Geolocation checks against vector and optical reference data

Workshop: Participants terrain-correct a calibrated Sentinel-1 subset with two DEM options and produce a map-overlay comparison with documented positional findings.

Day 6: Radiometric terrain effects and analysis-ready backscatter

• Topographic modulation of SAR backscatter
• Radiometric Terrain Flattening operator in ESA SNAP
• Gamma nought terrain-flattened output selection
• Local incidence angle and projected local incidence angle
• Masking radar shadow and layover areas
• No-data handling and output extent control
• GeoTIFF export settings and layer naming conventions

Workshop: Participants create a terrain-flattened, terrain-corrected backscatter product and deliver a GIS-ready GeoTIFF with shadow and layover masks.

Day 7: Co-registration, coherence and interferometric preparation

• Temporal baselines, perpendicular baselines and coherence
• Selecting compatible Sentinel-1 SLC acquisition pairs
• Back-geocoding workflow for SLC co-registration
• DEM-assisted co-registration settings
• Interferogram formation and phase representation
• Coherence estimation and window-size choices
• Recognising decorrelation and registration artefacts

Workshop: Participants co-register a Sentinel-1 SLC pair, generate a coherence layer and write a short interpretation of areas with high and low coherence.

Day 8: Change detection and thematic SAR applications

• Multi-temporal backscatter comparison methods
• Image differencing and log-ratio change indicators
• Threshold selection using reference samples
• Water and flood mapping from low-backscatter responses
• Vegetation and agricultural monitoring with VV and VH signals
• Built-environment change interpretation limitations
• Combining SAR outputs with vector, DEM and optical data in QGIS

Workshop: Participants create a preliminary flood-change layer from two terrain-corrected scenes and validate its threshold against reference imagery and terrain context.

Day 9: Reproducible SNAP workflows and quality assurance

• SNAP Graph Builder for chained processing steps
• Graph Processing Tool command-line execution
• Parameterising input files, output folders and processing variables
• Batch processing design for multiple Sentinel-1 scenes
• Metadata retention and processing-provenance records
• Quality-control checklist for calibrated and terrain-corrected outputs
• GDAL inspection and GeoTIFF validation utilities

Workshop: Participants build and run a reusable SNAP GPT graph that preprocesses a Sentinel-1 GRD scene and generates a quality-control log.

Day 10: Operational SAR workflow capstone

• Defining a SAR monitoring question and acceptance criteria
• Choosing a processing chain for the operational objective
• Selecting acquisitions, DEMs and output specifications
• Executing calibration, filtering and terrain-correction stages
• Reviewing artefacts, uncertainty and fitness for purpose
• Preparing map outputs and technical processing notes
• Implementation planning for organisational deployment

Workshop: Participants complete a capstone Sentinel-1 processing workflow and present a terrain-corrected output package, GPT graph, quality-check record and workplace application plan.

Tools & standards covered

ESA SNAP, Sentinel-1 Toolbox (S1TBX), QGIS, GDAL

A typical training day

08:30 – 10:30	First session
10:30 – 10:45	Refreshment break
10:45 – 12:30	Second session
12:30 – 13:30	Lunch and networking
13:30 – 15:00	Third session
15:00 – 15:15	Refreshment break
15:15 – 16:30	Workshop and daily review

Live online deliveries follow the same structure in the East Africa Time zone, with shorter screen blocks and longer breaks.

What the fee includes

How you can take this course

Certification

Participants who complete the full five days receive the Skillset Development Certificate of Completion, stating the course title, course code, dates and delivery format — suitable for professional-development records and employer reimbursement.

Frequently asked questions