From grimoire
Analyzes satellite or aerial imagery for earth science applications including preprocessing, spectral band selection, classification, change detection, and index calculation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/grimoire:apply-remote-sensing-analysisThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Analyze satellite or aerial imagery systematically — preprocessing to remove atmospheric and geometric artifacts, selecting bands appropriate to the application, calculating spectral indices, and validating classifications against ground truth — to extract reliable earth observation data.
Analyze satellite or aerial imagery systematically — preprocessing to remove atmospheric and geometric artifacts, selecting bands appropriate to the application, calculating spectral indices, and validating classifications against ground truth — to extract reliable earth observation data.
Adopted by: NASA, ESA, USGS, FAO, and all major national mapping agencies use standardized remote sensing workflows for operational monitoring programs (Global Forest Watch, Copernicus Land Monitoring, USDA CropScape). IPCC relies on satellite-derived land use change data for national GHG inventories. The Copernicus Emergency Management Service and Global Disaster Alert and Coordination System (GDACS) use remote sensing for rapid disaster response. Impact: Wulder et al. (2022, Remote Sensing of Environment) demonstrated that Landsat time series — analyzed with systematic methods — provides a 50-year global record of land change that no other data source can replicate. Inadequate preprocessing (uncorrected atmospheric haze, misregistered images) corrupts spectral indices by up to 30% and change detection results by orders of magnitude. Systematic analysis converts petabytes of raw satellite data into actionable earth observation products.
Match sensor to application:
| Application | Recommended sensor | Resolution |
|---|---|---|
| Regional land cover mapping | Landsat 8/9 (free), Sentinel-2 (free) | 10-30m |
| Vegetation stress / agriculture | Sentinel-2 (13 bands, 10m visible/NIR) | 10-20m |
| Geological mapping (mineralogy) | ASTER (thermal + SWIR), Sentinel-2 | 15-30m |
| Urban mapping, infrastructure | WorldView-3, Pleiades, Planet | 0.3-3m |
| Vegetation height / structure | LiDAR, GEDI (space-based LiDAR) | variable |
| Sea surface temperature | MODIS, Landsat TIR band | 30-1000m |
Never analyze raw (Level 0/1) imagery — preprocessing is mandatory:
Vegetation:
NDVI = (NIR − Red) / (NIR + Red)
Range: −1 to +1; healthy vegetation: 0.3-0.8; bare soil: 0.1-0.2; water: negative
Water:
NDWI = (Green − NIR) / (Green + NIR) [water bodies]
MNDWI = (Green − SWIR) / (Green + SWIR) [better in urban areas]
Burned area:
NBR = (NIR − SWIR2) / (NIR + SWIR2)
dNBR = pre-fire NBR − post-fire NBR [burn severity]
Geology (iron oxides):
Iron Oxide Ratio = Red / Blue (ASTER or Landsat Band 4/2)
Clay Ratio = SWIR1 / SWIR2 (ASTER Band 5/7)
Built-up area:
NDBI = (SWIR − NIR) / (SWIR + NIR) [built-up vs vegetation]
Two approaches:
Accuracy assessment — mandatory before reporting:
For before/after comparison:
Tools: Google Earth Engine, SNAP (ESA), QGIS Semi-Automatic Classification Plugin (free), ArcGIS Image Analyst.
npx claudepluginhub jeffreytse/grimoire --plugin grimoireProvides geospatial analysis: remote sensing, GIS, spatial ML, satellite imagery processing (Sentinel, Landsat, etc.), vector/raster ops, point clouds, network analysis, cloud-native workflows with 500+ code examples in 8 languages.
Covers geospatial science across remote sensing, GIS, spatial analysis, and ML for Earth observation with code examples in 8 languages. Use for satellite imagery processing, vector/raster operations, spatial statistics, and cloud-native geospatial workflows.
Works with ArcGIS raster and imagery data via ImageryLayer, ImageryTileLayer, pixel filtering, raster functions, multidimensional data, and oriented imagery. For satellite imagery, elevation, and scientific rasters.