Skip to content

Running Analysis

Once your inputs are configured and validated, you're ready to run the analysis.

Pre-Run Checklist

Before clicking Run, verify:

  • Rasters added with correct years
  • Output directory specified
  • Validation passed (no FAIL items)
  • Desired outputs selected in Options tab
  • Legend configured (optional but recommended)

Starting the Analysis

  1. Click the green Run Analysis button
  2. The button will be disabled during processing
  3. Monitor progress in the progress bar and log panel
+------------------------------------------+
|  [Validate Inputs]    [Run Analysis]     |
+------------------------------------------+
|  ████████████░░░░░░░░░░░░░  45%          |
+------------------------------------------+
|  Processing interval 2010-2015...        |
|  Computing transition matrix...          |
|  Writing net_gross_change_2010_2015.csv  |
+------------------------------------------+

Progress Monitoring

Progress Bar

The progress bar shows overall completion based on:

  • Number of rasters processed
  • Number of analysis blocks completed
  • Total blocks = (raster width / 256) × (raster height / 256) × number of rasters

Log Panel

The log panel displays real-time messages:

Message Type Example
Start Starting analysis with 3 rasters
Processing Processing interval 2010-2015
Output Writing area_by_class.csv
Raster added Added change_frequency.tif to project
Completion Analysis complete

Processing Steps

The analysis follows this sequence:

graph TD
    A[Start Analysis] --> B[Compute max class ID]
    B --> C[For each raster: count area by class]
    C --> D[For each interval: compute transitions]
    D --> E[Compute first-to-last transitions]
    E --> F[Generate change frequency raster]
    F --> G[Generate hotspot rasters]
    G --> H[Generate charts]
    H --> I[Analysis Complete]

Detailed Processing Order

  1. Initialize

    • Load all rasters
    • Apply AOI mask if specified
    • Set up NoData handling
  2. Area by Class

    • For each raster, count pixels per class
    • Apply area factor for unit conversion
    • Write area_by_class.csv
  3. Interval Metrics (for each consecutive year pair)

    • Count changed vs. unchanged pixels
    • Compute gain/loss per class
    • Build transition matrix
    • Rank top transitions
    • Calculate interval intensity
    • Write interval CSV files
  4. First-to-Last Comparison

    • Compare first and last year only
    • Generate comprehensive transition summary
    • Write transition_matrix_first_last_*.csv
  5. Change Frequency Raster

    • Count transitions per pixel across all intervals
    • Write change_frequency.tif
    • Add to QGIS project
  6. Hotspot Rasters (for each interval)

    • Extract changed pixel locations
    • Apply kernel density estimation
    • Write change_hotspot_*.tif
    • Add to QGIS project
  7. Charts (if enabled)

    • Generate HTML files with Plotly.js
    • Write to charts/ subdirectory

During Processing

Do Not Interrupt

Avoid closing QGIS or the plugin panel during analysis. If interrupted:

  • Partial outputs may be incomplete
  • Some files may not be written
  • Rasters may not be added to the project

Memory Considerations

The plugin uses block-based processing (256×256 pixel blocks) to handle large rasters efficiently:

  • Memory usage stays bounded regardless of raster size
  • Processing time scales with total pixel count
  • Very large rasters may take longer but won't exhaust memory

After Completion

When analysis finishes:

  1. Log shows "Analysis complete"
  2. Progress bar reaches 100%
  3. Run Analysis button re-enables

Checking Outputs

In QGIS:

  • Change frequency raster appears in Layers panel
  • Hotspot rasters appear in Layers panel

In Output Directory:

output_directory/
├── area_by_class.csv
├── net_gross_change_2010_2015.csv
├── transition_matrix_2010_2015.csv
├── top_transitions_2010_2015.csv
├── change_frequency.tif
├── change_hotspot_2010_2015.tif
├── change_intensity.csv
└── charts/
    ├── area_by_class.html
    ├── net_gross_change.html
    └── ...

Re-Running Analysis

To run analysis again with different settings:

  1. Modify inputs, legend, or options as needed
  2. Re-run validation if inputs changed
  3. Click Run Analysis

Files Overwritten

Running analysis again will overwrite existing files in the output directory. Use a new output directory to preserve previous results.

Troubleshooting

Analysis Won't Start

Possible causes:

  • Validation has FAIL items → Fix and re-validate
  • No output directory specified → Set directory in Inputs tab
  • No rasters added → Add at least one raster

Analysis Stops Unexpectedly

Check:

  • QGIS message bar for errors
  • Log panel for error messages
  • Disk space in output directory

Missing Outputs

If some outputs are missing:

  • Verify they were enabled in Options tab
  • Check log for any errors during that output
  • For charts: verify plotly.min.js exists in vendor/js/

Slow Processing

For large datasets:

  • Disable hotspots (most intensive)
  • Disable charts
  • Use an AOI to limit the analysis area
  • Consider downsampling rasters for initial exploration

Output Details

For detailed information about each output: