Documents
Detect forest disturbance from multi-temporal NDVI. Use when the user wants to analyze changes, detect hazards, or generate assessment reports.
Detect forest disturbance from multi-temporal NDVI. Use when the user wants to analyze changes, detect hazards, or generate assessment reports.
Detect forest disturbance from multi-temporal NDVI.
python scripts/forest_disturbance_alert.py --baseline baseline.tif --current current.tif
python scripts/forest_disturbance_alert.py --baseline baseline.tif --current current.tif --threshold 0.25
python scripts/forest_disturbance_alert.py --baseline baseline.tif --current current.tif --output-dir my_output
| Argument | Required | Default | Description |
|---|---|---|---|
--baseline | Yes | — | Path to baseline (undisturbed) NDVI raster |
--current | Yes | — | Path to current NDVI raster (same grid as --baseline) |
--threshold | No | 0.2 | NDVI drop threshold (current − baseline < −threshold is flagged) |
--output-dir, -o | No | disturbance-output | Directory to write outputs |
| File | Description |
|---|---|
disturbance-report.json | Machine-readable disturbance stats (affected area, severity classes) |
report.html | Human-readable HTML report |
output-manifest.json | Run metadata + result summary |
| Code | Meaning |
|---|---|
| 0 | Success |
| 2 | Argument error |
| 7 | Processing failure |
本 skill 可自动从 Microsoft Planetary Computer 下载数据 (无需 API key):
python forest_disturbance_alert.py --bbox 116,39,117,40 --date-range 2024-06-01,2024-06-30 --output-dir
--bbox W,S,E,N: WGS-84 边界框 (西, 南, 东, 北)--date-range START,END: 日期范围 (YYYY-MM-DD,YYYY-MM-DD)--aoi-file : 替代 --bbox 的 GeoJSON 多边形--cache-dir : 缓存目录 (默认 ~/.geoskill_cache)当用户只给 --bbox + --date-range (没有 --baseline) 时，skill 自动下载数据。
当用户给 --baseline 时，走原文件路径 (向后兼容)。
Monitor forest canopy vitality decline, drought stress, pest damage, or wind throw from multi-temporal spectral indices. Distinguishes short-term fluctuations from persistent decline using historical baselines, persistence state machines, and climate attribution. Use when assessing forest health, detecting anomalies, or planning field sampling.
Compute forest fire burn severity from pre/post-fire NIR and SWIR imagery using differenced Normalized Burn Ratio (dNBR). Classifies severity into unburned, low, moderate, and high categories. Use when the user wants to assess burn severity, map fire damage, or generate burn severity reports.
Monitor surface disturbance at mining sites using multi-temporal optical/SAR/DEM data. Detect bare land, pits, dumps, roads, and vegetation removal. Track disturbance objects across years, identify boundary violations, and generate area statistics.
Assess crop health from NDVI time series. Use when the user wants to analyze changes, detect hazards, or generate assessment reports.
CHM tree height and crown width allometric volume, NDVI/NDRE health grading and SAR biomass fusion for precision forestry
Estimate forest carbon stock from remote sensing data using BEF, allometric equations, or IPCC Tier 1/2 methods. Includes Monte Carlo uncertainty analysis. Supports raster (GeoTIFF) and tabular (CSV) inputs.