[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M739VB7NR55EHVCGF1AY9YVP":3,"skill-same:01M739VB7NR55EHVCGF1AY9YVP":111},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":27,"commitSha":28,"skillDirectory":29,"forksCount":30,"repositoryHtmlUrl":31,"repositoryHomepageUrl":6,"ownerHtmlUrl":32,"ownerBlogUrl":33,"ownerDescription":6,"updatedAt":34,"install":35,"resources":48},"01M739VB7NR55EHVCGF1AY9YVP","geomaster",null,"Comprehensive geospatial science skill covering remote sensing, GIS, spatial analysis, machine learning for earth observation, and 30+ scientific domains. Supports satellite imagery processing (Sentinel, Landsat, MODIS, SAR, hyperspectral), vector and raster data operations, spatial statistics, point cloud processing, network analysis, cloud-native workflows (STAC, COG, Planetary Computer), and 8 programming languages (Python, R, Julia, JavaScript, C++, Java, Go, Rust) with 500+ code examples. Use for remote sensing workflows, GIS analysis, spatial ML, Earth observation data processing, terrain analysis, hydrological modeling, marine spatial analysis, atmospheric science, and any geospatial computation task.","qhjqhj00\u002Fresearch-skills-pool","qhjqhj00","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F29684092?v=4","01M72AE0ZGXVM5C57N512DGQG8","Tommy Chien","01M72AE107ESJMHT7QC5SXX144",3,"MIT","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\r\nname: geomaster\r\ndescription: Comprehensive geospatial science skill covering remote sensing, GIS, spatial analysis, machine learning for earth observation, and 30+ scientific domains. Supports satellite imagery processing (Sentinel, Landsat, MODIS, SAR, hyperspectral), vector and raster data operations, spatial statistics, point cloud processing, network analysis, cloud-native workflows (STAC, COG, Planetary Computer), and 8 programming languages (Python, R, Julia, JavaScript, C++, Java, Go, Rust) with 500+ code examples. Use for remote sensing workflows, GIS analysis, spatial ML, Earth observation data processing, terrain analysis, hydrological modeling, marine spatial analysis, atmospheric science, and any geospatial computation task.\r\nlicense: MIT License\r\nmetadata:\r\n    skill-author: K-Dense Inc.\r\n---\r\n\r\n# GeoMaster\r\n\r\nComprehensive geospatial science skill covering GIS, remote sensing, spatial analysis, and ML for Earth observation across 70+ topics with 500+ code examples in 8 programming languages.\r\n\r\n## Installation\r\n\r\n```bash\r\n# Core Python stack (conda recommended)\r\nconda install -c conda-forge gdal rasterio fiona shapely pyproj geopandas\r\n\r\n# Remote sensing & ML\r\nuv pip install rsgislib torchgeo earthengine-api\r\nuv pip install scikit-learn xgboost torch-geometric\r\n\r\n# Network & visualization\r\nuv pip install osmnx networkx folium keplergl\r\nuv pip install cartopy contextily mapclassify\r\n\r\n# Big data & cloud\r\nuv pip install xarray rioxarray dask-geopandas\r\nuv pip install pystac-client planetary-computer\r\n\r\n# Point clouds\r\nuv pip install laspy pylas open3d pdal\r\n\r\n# Databases\r\nconda install -c conda-forge postgis spatialite\r\n```\r\n\r\n## Quick Start\r\n\r\n### NDVI from Sentinel-2\r\n\r\n```python\r\nimport rasterio\r\nimport numpy as np\r\n\r\nwith rasterio.open('sentinel2.tif') as src:\r\n    red = src.read(4).astype(float)   # B04\r\n    nir = src.read(8).astype(float)   # B08\r\n    ndvi = (nir - red) \u002F (nir + red + 1e-8)\r\n    ndvi = np.nan_to_num(ndvi, nan=0)\r\n\r\n    profile = src.profile\r\n    profile.update(count=1, dtype=rasterio.float32)\r\n\r\n    with rasterio.open('ndvi.tif', 'w', **profile) as dst:\r\n        dst.write(ndvi.astype(rasterio.float32), 1)\r\n```\r\n\r\n### Spatial Analysis with GeoPandas\r\n\r\n```python\r\nimport geopandas as gpd\r\n\r\n# Load and ensure same CRS\r\nzones = gpd.read_file('zones.geojson')\r\npoints = gpd.read_file('points.geojson')\r\n\r\nif zones.crs != points.crs:\r\n    points = points.to_crs(zones.crs)\r\n\r\n# Spatial join and statistics\r\njoined = gpd.sjoin(points, zones, how='inner', predicate='within')\r\nstats = joined.groupby('zone_id').agg({\r\n    'value': ['count', 'mean', 'std', 'min', 'max']\r\n}).round(2)\r\n```\r\n\r\n### Google Earth Engine Time Series\r\n\r\n```python\r\nimport ee\r\nimport pandas as pd\r\n\r\nee.Initialize(project='your-project')\r\nroi = ee.Geometry.Point([-122.4, 37.7]).buffer(10000)\r\n\r\ns2 = (ee.ImageCollection('COPERNICUS\u002FS2_SR_HARMONIZED')\r\n      .filterBounds(roi)\r\n      .filterDate('2020-01-01', '2023-12-31')\r\n      .filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20)))\r\n\r\ndef add_ndvi(img):\r\n    return img.addBands(img.normalizedDifference(['B8', 'B4']).rename('NDVI'))\r\n\r\ns2_ndvi = s2.map(add_ndvi)\r\n\r\ndef extract_series(image):\r\n    stats = image.reduceRegion(ee.Reducer.mean(), roi.centroid(), scale=10, maxPixels=1e9)\r\n    return ee.Feature(None, {'date': image.date().format('YYYY-MM-dd'), 'ndvi': stats.get('NDVI')})\r\n\r\nseries = s2_ndvi.map(extract_series).getInfo()\r\ndf = pd.DataFrame([f['properties'] for f in series['features']])\r\ndf['date'] = pd.to_datetime(df['date'])\r\n```\r\n\r\n## Core Concepts\r\n\r\n### Data Types\r\n\r\n| Type | Examples | Libraries |\r\n|------|----------|-----------|\r\n| Vector | Shapefile, GeoJSON, GeoPackage | GeoPandas, Fiona, GDAL |\r\n| Raster | GeoTIFF, NetCDF, COG | Rasterio, Xarray, GDAL |\r\n| Point Cloud | LAS, LAZ | Laspy, PDAL, Open3D |\r\n\r\n### Coordinate Systems\r\n\r\n- **EPSG:4326** (WGS 84) - Geographic, lat\u002Flon, use for storage\r\n- **EPSG:3857** (Web Mercator) - Web maps only (don't use for area\u002Fdistance!)\r\n- **EPSG:326xx\u002F327xx** (UTM) - Metric calculations, \u003C1% distortion per zone\r\n- Use `gdf.estimate_utm_crs()` for automatic UTM detection\r\n\r\n```python\r\n# Always check CRS before operations\r\nassert gdf1.crs == gdf2.crs, \"CRS mismatch!\"\r\n\r\n# For area\u002Fdistance calculations, use projected CRS\r\ngdf_metric = gdf.to_crs(gdf.estimate_utm_crs())\r\narea_sqm = gdf_metric.geometry.area\r\n```\r\n\r\n### OGC Standards\r\n\r\n- **WMS**: Web Map Service - raster maps\r\n- **WFS**: Web Feature Service - vector data\r\n- **WCS**: Web Coverage Service - raster coverage\r\n- **STAC**: Spatiotemporal Asset Catalog - modern metadata\r\n\r\n## Common Operations\r\n\r\n### Spectral Indices\r\n\r\n```python\r\ndef calculate_indices(image_path):\r\n    \"\"\"NDVI, EVI, SAVI, NDWI from Sentinel-2.\"\"\"\r\n    with rasterio.open(image_path) as src:\r\n        B02, B03, B04, B08, B11 = [src.read(i).astype(float) for i in [1,2,3,4,5]]\r\n\r\n    ndvi = (B08 - B04) \u002F (B08 + B04 + 1e-8)\r\n    evi = 2.5 * (B08 - B04) \u002F (B08 + 6*B04 - 7.5*B02 + 1)\r\n    savi = ((B08 - B04) \u002F (B08 + B04 + 0.5)) * 1.5\r\n    ndwi = (B03 - B08) \u002F (B03 + B08 + 1e-8)\r\n\r\n    return {'NDVI': ndvi, 'EVI': evi, 'SAVI': savi, 'NDWI': ndwi}\r\n```\r\n\r\n### Vector Operations\r\n\r\n```python\r\n# Buffer (use projected CRS!)\r\ngdf_proj = gdf.to_crs(gdf.estimate_utm_crs())\r\ngdf['buffer_1km'] = gdf_proj.geometry.buffer(1000)\r\n\r\n# Spatial relationships\r\nintersects = gdf[gdf.geometry.intersects(other_geometry)]\r\ncontains = gdf[gdf.geometry.contains(point_geometry)]\r\n\r\n# Geometric operations\r\ngdf['centroid'] = gdf.geometry.centroid\r\ngdf['simplified'] = gdf.geometry.simplify(tolerance=0.001)\r\n\r\n# Overlay operations\r\nintersection = gpd.overlay(gdf1, gdf2, how='intersection')\r\nunion = gpd.overlay(gdf1, gdf2, how='union')\r\n```\r\n\r\n### Terrain Analysis\r\n\r\n```python\r\ndef terrain_metrics(dem_path):\r\n    \"\"\"Calculate slope, aspect, hillshade from DEM.\"\"\"\r\n    with rasterio.open(dem_path) as src:\r\n        dem = src.read(1)\r\n\r\n    dy, dx = np.gradient(dem)\r\n    slope = np.arctan(np.sqrt(dx**2 + dy**2)) * 180 \u002F np.pi\r\n    aspect = (90 - np.arctan2(-dy, dx) * 180 \u002F np.pi) % 360\r\n\r\n    # Hillshade\r\n    az_rad, alt_rad = np.radians(315), np.radians(45)\r\n    hillshade = (np.sin(alt_rad) * np.sin(np.radians(slope)) +\r\n                 np.cos(alt_rad) * np.cos(np.radians(slope)) *\r\n                 np.cos(np.radians(aspect) - az_rad))\r\n\r\n    return slope, aspect, hillshade\r\n```\r\n\r\n### Network Analysis\r\n\r\n```python\r\nimport osmnx as ox\r\nimport networkx as nx\r\n\r\n# Download and analyze street network\r\nG = ox.graph_from_place('San Francisco, CA', network_type='drive')\r\nG = ox.add_edge_speeds(G).add_edge_travel_times(G)\r\n\r\n# Shortest path\r\norig = ox.distance.nearest_nodes(G, -122.4, 37.7)\r\ndest = ox.distance.nearest_nodes(G, -122.3, 37.8)\r\nroute = nx.shortest_path(G, orig, dest, weight='travel_time')\r\n```\r\n\r\n## Image Classification\r\n\r\n```python\r\nfrom sklearn.ensemble import RandomForestClassifier\r\nimport rasterio\r\nfrom rasterio.features import rasterize\r\n\r\ndef classify_imagery(raster_path, training_gdf, output_path):\r\n    \"\"\"Train RF and classify imagery.\"\"\"\r\n    with rasterio.open(raster_path) as src:\r\n        image = src.read()\r\n        profile = src.profile\r\n        transform = src.transform\r\n\r\n    # Extract training data\r\n    X_train, y_train = [], []\r\n    for _, row in training_gdf.iterrows():\r\n        mask = rasterize([(row.geometry, 1)],\r\n                        out_shape=(profile['height'], profile['width']),\r\n                        transform=transform, fill=0, dtype=np.uint8)\r\n        pixels = image[:, mask > 0].T\r\n        X_train.extend(pixels)\r\n        y_train.extend([row['class_id']] * len(pixels))\r\n\r\n    # Train and predict\r\n    rf = RandomForestClassifier(n_estimators=100, max_depth=20, n_jobs=-1)\r\n    rf.fit(X_train, y_train)\r\n\r\n    prediction = rf.predict(image.reshape(image.shape[0], -1).T)\r\n    prediction = prediction.reshape(profile['height'], profile['width'])\r\n\r\n    profile.update(dtype=rasterio.uint8, count=1)\r\n    with rasterio.open(output_path, 'w', **profile) as dst:\r\n        dst.write(prediction.astype(rasterio.uint8), 1)\r\n\r\n    return rf\r\n```\r\n\r\n## Modern Cloud-Native Workflows\r\n\r\n### STAC + Planetary Computer\r\n\r\n```python\r\nimport pystac_client\r\nimport planetary_computer\r\nimport odc.stac\r\n\r\n# Search Sentinel-2 via STAC\r\ncatalog = pystac_client.Client.open(\r\n    \"https:\u002F\u002Fplanetarycomputer.microsoft.com\u002Fapi\u002Fstac\u002Fv1\",\r\n    modifier=planetary_computer.sign_inplace,\r\n)\r\n\r\nsearch = catalog.search(\r\n    collections=[\"sentinel-2-l2a\"],\r\n    bbox=[-122.5, 37.7, -122.3, 37.9],\r\n    datetime=\"2023-01-01\u002F2023-12-31\",\r\n    query={\"eo:cloud_cover\": {\"lt\": 20}},\r\n)\r\n\r\n# Load as xarray (cloud-native!)\r\ndata = odc.stac.load(\r\n    list(search.get_items())[:5],\r\n    bands=[\"B02\", \"B03\", \"B04\", \"B08\"],\r\n    crs=\"EPSG:32610\",\r\n    resolution=10,\r\n)\r\n\r\n# Calculate NDVI on xarray\r\nndvi = (data.B08 - data.B04) \u002F (data.B08 + data.B04)\r\n```\r\n\r\n### Cloud-Optimized GeoTIFF (COG)\r\n\r\n```python\r\nimport rasterio\r\nfrom rasterio.session import AWSSession\r\n\r\n# Read COG directly from cloud (partial reads)\r\nsession = AWSSession(aws_access_key_id=..., aws_secret_access_key=...)\r\nwith rasterio.open('s3:\u002F\u002Fbucket\u002Fpath.tif', session=session) as src:\r\n    # Read only window of interest\r\n    window = ((1000, 2000), (1000, 2000))\r\n    subset = src.read(1, window=window)\r\n\r\n# Write COG\r\nwith rasterio.open('output.tif', 'w', **profile,\r\n                   tiled=True, blockxsize=256, blockysize=256,\r\n                   compress='DEFLATE', predictor=2) as dst:\r\n    dst.write(data)\r\n\r\n# Validate COG\r\nfrom rio_cogeo.cogeo import cog_validate\r\ncog_validate('output.tif')\r\n```\r\n\r\n## Performance Tips\r\n\r\n```python\r\n# 1. Spatial indexing (10-100x faster queries)\r\ngdf.sindex  # Auto-created by GeoPandas\r\n\r\n# 2. Chunk large rasters\r\nwith rasterio.open('large.tif') as src:\r\n    for i, window in src.block_windows(1):\r\n        block = src.read(1, window=window)\r\n\r\n# 3. Dask for big data\r\nimport dask.array as da\r\ndask_array = da.from_rasterio('large.tif', chunks=(1, 1024, 1024))\r\n\r\n# 4. Use Arrow for I\u002FO\r\ngdf.to_file('output.gpkg', use_arrow=True)\r\n\r\n# 5. GDAL caching\r\nfrom osgeo import gdal\r\ngdal.SetCacheMax(2**30)  # 1GB cache\r\n\r\n# 6. Parallel processing\r\nrf = RandomForestClassifier(n_jobs=-1)  # All cores\r\n```\r\n\r\n## Best Practices\r\n\r\n1. **Always check CRS** before spatial operations\r\n2. **Use projected CRS** for area\u002Fdistance calculations\r\n3. **Validate geometries**: `gdf = gdf[gdf.is_valid]`\r\n4. **Handle missing data**: `gdf['geometry'] = gdf['geometry'].fillna(None)`\r\n5. **Use efficient formats**: GeoPackage > Shapefile, Parquet for large data\r\n6. **Apply cloud masking** to optical imagery\r\n7. **Preserve lineage** for reproducible research\r\n8. **Use appropriate resolution** for your analysis scale\r\n\r\n## Detailed Documentation\r\n\r\n- **[Coordinate Systems](references\u002Fcoordinate-systems.md)** - CRS fundamentals, UTM, transformations\r\n- **[Core Libraries](references\u002Fcore-libraries.md)** - GDAL, Rasterio, GeoPandas, Shapely\r\n- **[Remote Sensing](references\u002Fremote-sensing.md)** - Satellite missions, spectral indices, SAR\r\n- **[Machine Learning](references\u002Fmachine-learning.md)** - Deep learning, CNNs, GNNs for RS\r\n- **[GIS Software](references\u002Fgis-software.md)** - QGIS, ArcGIS, GRASS integration\r\n- **[Scientific Domains](references\u002Fscientific-domains.md)** - Marine, hydrology, agriculture, forestry\r\n- **[Advanced GIS](references\u002Fadvanced-gis.md)** - 3D GIS, spatiotemporal, topology\r\n- **[Big Data](references\u002Fbig-data.md)** - Distributed processing, GPU acceleration\r\n- **[Industry Applications](references\u002Findustry-applications.md)** - Urban planning, disaster management\r\n- **[Programming Languages](references\u002Fprogramming-languages.md)** - Python, R, Julia, JS, C++, Java, Go, Rust\r\n- **[Data Sources](references\u002Fdata-sources.md)** - Satellite catalogs, APIs\r\n- **[Troubleshooting](references\u002Ftroubleshooting.md)** - Common issues, debugging, error reference\r\n- **[Code Examples](references\u002Fcode-examples.md)** - 500+ examples\r\n\r\n---\r\n\r\n**GeoMaster covers everything from basic GIS operations to advanced remote sensing and machine learning.**\r\n","# GeoMaster\r\n\r\nComprehensive geospatial science skill covering GIS, remote sensing, spatial analysis, and ML for Earth observation across 70+ topics with 500+ code examples in 8 programming languages.\r\n\r\n## Installation\r\n\r\n```bash\r\n# Core Python stack (conda recommended)\r\nconda install -c conda-forge gdal rasterio fiona shapely pyproj geopandas\r\n\r\n# Remote sensing & ML\r\nuv pip install rsgislib torchgeo earthengine-api\r\nuv pip install scikit-learn xgboost torch-geometric\r\n\r\n# Network & visualization\r\nuv pip install osmnx networkx folium keplergl\r\nuv pip install cartopy contextily mapclassify\r\n\r\n# Big data & cloud\r\nuv pip install xarray rioxarray dask-geopandas\r\nuv pip install pystac-client planetary-computer\r\n\r\n# Point clouds\r\nuv pip install laspy pylas open3d pdal\r\n\r\n# Databases\r\nconda install -c conda-forge postgis spatialite\r\n```\r\n\r\n## Quick Start\r\n\r\n### NDVI from Sentinel-2\r\n\r\n```python\r\nimport rasterio\r\nimport numpy as np\r\n\r\nwith rasterio.open('sentinel2.tif') as src:\r\n    red = src.read(4).astype(float)   # B04\r\n    nir = src.read(8).astype(float)   # B08\r\n    ndvi = (nir - red) \u002F (nir + red + 1e-8)\r\n    ndvi = np.nan_to_num(ndvi, nan=0)\r\n\r\n    profile = src.profile\r\n    profile.update(count=1, dtype=rasterio.float32)\r\n\r\n    with rasterio.open('ndvi.tif', 'w', **profile) as dst:\r\n        dst.write(ndvi.astype(rasterio.float32), 1)\r\n```\r\n\r\n### Spatial Analysis with GeoPandas\r\n\r\n```python\r\nimport geopandas as gpd\r\n\r\n# Load and ensure same CRS\r\nzones = gpd.read_file('zones.geojson')\r\npoints = gpd.read_file('points.geojson')\r\n\r\nif zones.crs != points.crs:\r\n    points = points.to_crs(zones.crs)\r\n\r\n# Spatial join and statistics\r\njoined = gpd.sjoin(points, zones, how='inner', predicate='within')\r\nstats = joined.groupby('zone_id').agg({\r\n    'value': ['count', 'mean', 'std', 'min', 'max']\r\n}).round(2)\r\n```\r\n\r\n### Google Earth Engine Time Series\r\n\r\n```python\r\nimport ee\r\nimport pandas as pd\r\n\r\nee.Initialize(project='your-project')\r\nroi = ee.Geometry.Point([-122.4, 37.7]).buffer(10000)\r\n\r\ns2 = (ee.ImageCollection('COPERNICUS\u002FS2_SR_HARMONIZED')\r\n      .filterBounds(roi)\r\n      .filterDate('2020-01-01', '2023-12-31')\r\n      .filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20)))\r\n\r\ndef add_ndvi(img):\r\n    return img.addBands(img.normalizedDifference(['B8', 'B4']).rename('NDVI'))\r\n\r\ns2_ndvi = s2.map(add_ndvi)\r\n\r\ndef extract_series(image):\r\n    stats = image.reduceRegion(ee.Reducer.mean(), roi.centroid(), scale=10, maxPixels=1e9)\r\n    return ee.Feature(None, {'date': image.date().format('YYYY-MM-dd'), 'ndvi': stats.get('NDVI')})\r\n\r\nseries = s2_ndvi.map(extract_series).getInfo()\r\ndf = pd.DataFrame([f['properties'] for f in series['features']])\r\ndf['date'] = pd.to_datetime(df['date'])\r\n```\r\n\r\n## Core Concepts\r\n\r\n### Data Types\r\n\r\n| Type | Examples | Libraries |\r\n|------|----------|-----------|\r\n| Vector | Shapefile, GeoJSON, GeoPackage | GeoPandas, Fiona, GDAL |\r\n| Raster | GeoTIFF, NetCDF, COG | Rasterio, Xarray, GDAL |\r\n| Point Cloud | LAS, LAZ | Laspy, PDAL, Open3D |\r\n\r\n### Coordinate Systems\r\n\r\n- **EPSG:4326** (WGS 84) - Geographic, lat\u002Flon, use for storage\r\n- **EPSG:3857** (Web Mercator) - Web maps only (don't use for area\u002Fdistance!)\r\n- **EPSG:326xx\u002F327xx** (UTM) - Metric calculations, \u003C1% distortion per zone\r\n- Use `gdf.estimate_utm_crs()` for automatic UTM detection\r\n\r\n```python\r\n# Always check CRS before operations\r\nassert gdf1.crs == gdf2.crs, \"CRS mismatch!\"\r\n\r\n# For area\u002Fdistance calculations, use projected CRS\r\ngdf_metric = gdf.to_crs(gdf.estimate_utm_crs())\r\narea_sqm = gdf_metric.geometry.area\r\n```\r\n\r\n### OGC Standards\r\n\r\n- **WMS**: Web Map Service - raster maps\r\n- **WFS**: Web Feature Service - vector data\r\n- **WCS**: Web Coverage Service - raster coverage\r\n- **STAC**: Spatiotemporal Asset Catalog - modern metadata\r\n\r\n## Common Operations\r\n\r\n### Spectral Indices\r\n\r\n```python\r\ndef calculate_indices(image_path):\r\n    \"\"\"NDVI, EVI, SAVI, NDWI from Sentinel-2.\"\"\"\r\n    with rasterio.open(image_path) as src:\r\n        B02, B03, B04, B08, B11 = [src.read(i).astype(float) for i in [1,2,3,4,5]]\r\n\r\n    ndvi = (B08 - B04) \u002F (B08 + B04 + 1e-8)\r\n    evi = 2.5 * (B08 - B04) \u002F (B08 + 6*B04 - 7.5*B02 + 1)\r\n    savi = ((B08 - B04) \u002F (B08 + B04 + 0.5)) * 1.5\r\n    ndwi = (B03 - B08) \u002F (B03 + B08 + 1e-8)\r\n\r\n    return {'NDVI': ndvi, 'EVI': evi, 'SAVI': savi, 'NDWI': ndwi}\r\n```\r\n\r\n### Vector Operations\r\n\r\n```python\r\n# Buffer (use projected CRS!)\r\ngdf_proj = gdf.to_crs(gdf.estimate_utm_crs())\r\ngdf['buffer_1km'] = gdf_proj.geometry.buffer(1000)\r\n\r\n# Spatial relationships\r\nintersects = gdf[gdf.geometry.intersects(other_geometry)]\r\ncontains = gdf[gdf.geometry.contains(point_geometry)]\r\n\r\n# Geometric operations\r\ngdf['centroid'] = gdf.geometry.centroid\r\ngdf['simplified'] = gdf.geometry.simplify(tolerance=0.001)\r\n\r\n# Overlay operations\r\nintersection = gpd.overlay(gdf1, gdf2, how='intersection')\r\nunion = gpd.overlay(gdf1, gdf2, how='union')\r\n```\r\n\r\n### Terrain Analysis\r\n\r\n```python\r\ndef terrain_metrics(dem_path):\r\n    \"\"\"Calculate slope, aspect, hillshade from DEM.\"\"\"\r\n    with rasterio.open(dem_path) as src:\r\n        dem = src.read(1)\r\n\r\n    dy, dx = np.gradient(dem)\r\n    slope = np.arctan(np.sqrt(dx**2 + dy**2)) * 180 \u002F np.pi\r\n    aspect = (90 - np.arctan2(-dy, dx) * 180 \u002F np.pi) % 360\r\n\r\n    # Hillshade\r\n    az_rad, alt_rad = np.radians(315), np.radians(45)\r\n    hillshade = (np.sin(alt_rad) * np.sin(np.radians(slope)) +\r\n                 np.cos(alt_rad) * np.cos(np.radians(slope)) *\r\n                 np.cos(np.radians(aspect) - az_rad))\r\n\r\n    return slope, aspect, hillshade\r\n```\r\n\r\n### Network Analysis\r\n\r\n```python\r\nimport osmnx as ox\r\nimport networkx as nx\r\n\r\n# Download and analyze street network\r\nG = ox.graph_from_place('San Francisco, CA', network_type='drive')\r\nG = ox.add_edge_speeds(G).add_edge_travel_times(G)\r\n\r\n# Shortest path\r\norig = ox.distance.nearest_nodes(G, -122.4, 37.7)\r\ndest = ox.distance.nearest_nodes(G, -122.3, 37.8)\r\nroute = nx.shortest_path(G, orig, dest, weight='travel_time')\r\n```\r\n\r\n## Image Classification\r\n\r\n```python\r\nfrom sklearn.ensemble import RandomForestClassifier\r\nimport rasterio\r\nfrom rasterio.features import rasterize\r\n\r\ndef classify_imagery(raster_path, training_gdf, output_path):\r\n    \"\"\"Train RF and classify imagery.\"\"\"\r\n    with rasterio.open(raster_path) as src:\r\n        image = src.read()\r\n        profile = src.profile\r\n        transform = src.transform\r\n\r\n    # Extract training data\r\n    X_train, y_train = [], []\r\n    for _, row in training_gdf.iterrows():\r\n        mask = rasterize([(row.geometry, 1)],\r\n                        out_shape=(profile['height'], profile['width']),\r\n                        transform=transform, fill=0, dtype=np.uint8)\r\n        pixels = image[:, mask > 0].T\r\n        X_train.extend(pixels)\r\n        y_train.extend([row['class_id']] * len(pixels))\r\n\r\n    # Train and predict\r\n    rf = RandomForestClassifier(n_estimators=100, max_depth=20, n_jobs=-1)\r\n    rf.fit(X_train, y_train)\r\n\r\n    prediction = rf.predict(image.reshape(image.shape[0], -1).T)\r\n    prediction = prediction.reshape(profile['height'], profile['width'])\r\n\r\n    profile.update(dtype=rasterio.uint8, count=1)\r\n    with rasterio.open(output_path, 'w', **profile) as dst:\r\n        dst.write(prediction.astype(rasterio.uint8), 1)\r\n\r\n    return rf\r\n```\r\n\r\n## Modern Cloud-Native Workflows\r\n\r\n### STAC + Planetary Computer\r\n\r\n```python\r\nimport pystac_client\r\nimport planetary_computer\r\nimport odc.stac\r\n\r\n# Search Sentinel-2 via STAC\r\ncatalog = pystac_client.Client.open(\r\n    \"https:\u002F\u002Fplanetarycomputer.microsoft.com\u002Fapi\u002Fstac\u002Fv1\",\r\n    modifier=planetary_computer.sign_inplace,\r\n)\r\n\r\nsearch = catalog.search(\r\n    collections=[\"sentinel-2-l2a\"],\r\n    bbox=[-122.5, 37.7, -122.3, 37.9],\r\n    datetime=\"2023-01-01\u002F2023-12-31\",\r\n    query={\"eo:cloud_cover\": {\"lt\": 20}},\r\n)\r\n\r\n# Load as xarray (cloud-native!)\r\ndata = odc.stac.load(\r\n    list(search.get_items())[:5],\r\n    bands=[\"B02\", \"B03\", \"B04\", \"B08\"],\r\n    crs=\"EPSG:32610\",\r\n    resolution=10,\r\n)\r\n\r\n# Calculate NDVI on xarray\r\nndvi = (data.B08 - data.B04) \u002F (data.B08 + data.B04)\r\n```\r\n\r\n### Cloud-Optimized GeoTIFF (COG)\r\n\r\n```python\r\nimport rasterio\r\nfrom rasterio.session import AWSSession\r\n\r\n# Read COG directly from cloud (partial reads)\r\nsession = AWSSession(aws_access_key_id=..., aws_secret_access_key=...)\r\nwith rasterio.open('s3:\u002F\u002Fbucket\u002Fpath.tif', session=session) as src:\r\n    # Read only window of interest\r\n    window = ((1000, 2000), (1000, 2000))\r\n    subset = src.read(1, window=window)\r\n\r\n# Write COG\r\nwith rasterio.open('output.tif', 'w', **profile,\r\n                   tiled=True, blockxsize=256, blockysize=256,\r\n                   compress='DEFLATE', predictor=2) as dst:\r\n    dst.write(data)\r\n\r\n# Validate COG\r\nfrom rio_cogeo.cogeo import cog_validate\r\ncog_validate('output.tif')\r\n```\r\n\r\n## Performance Tips\r\n\r\n```python\r\n# 1. Spatial indexing (10-100x faster queries)\r\ngdf.sindex  # Auto-created by GeoPandas\r\n\r\n# 2. Chunk large rasters\r\nwith rasterio.open('large.tif') as src:\r\n    for i, window in src.block_windows(1):\r\n        block = src.read(1, window=window)\r\n\r\n# 3. Dask for big data\r\nimport dask.array as da\r\ndask_array = da.from_rasterio('large.tif', chunks=(1, 1024, 1024))\r\n\r\n# 4. Use Arrow for I\u002FO\r\ngdf.to_file('output.gpkg', use_arrow=True)\r\n\r\n# 5. GDAL caching\r\nfrom osgeo import gdal\r\ngdal.SetCacheMax(2**30)  # 1GB cache\r\n\r\n# 6. Parallel processing\r\nrf = RandomForestClassifier(n_jobs=-1)  # All cores\r\n```\r\n\r\n## Best Practices\r\n\r\n1. **Always check CRS** before spatial operations\r\n2. **Use projected CRS** for area\u002Fdistance calculations\r\n3. **Validate geometries**: `gdf = gdf[gdf.is_valid]`\r\n4. **Handle missing data**: `gdf['geometry'] = gdf['geometry'].fillna(None)`\r\n5. **Use efficient formats**: GeoPackage > Shapefile, Parquet for large data\r\n6. **Apply cloud masking** to optical imagery\r\n7. **Preserve lineage** for reproducible research\r\n8. **Use appropriate resolution** for your analysis scale\r\n\r\n## Detailed Documentation\r\n\r\n- **[Coordinate Systems](references\u002Fcoordinate-systems.md)** - CRS fundamentals, UTM, transformations\r\n- **[Core Libraries](references\u002Fcore-libraries.md)** - GDAL, Rasterio, GeoPandas, Shapely\r\n- **[Remote Sensing](references\u002Fremote-sensing.md)** - Satellite missions, spectral indices, SAR\r\n- **[Machine Learning](references\u002Fmachine-learning.md)** - Deep learning, CNNs, GNNs for RS\r\n- **[GIS Software](references\u002Fgis-software.md)** - QGIS, ArcGIS, GRASS integration\r\n- **[Scientific Domains](references\u002Fscientific-domains.md)** - Marine, hydrology, agriculture, forestry\r\n- **[Advanced GIS](references\u002Fadvanced-gis.md)** - 3D GIS, spatiotemporal, topology\r\n- **[Big Data](references\u002Fbig-data.md)** - Distributed processing, GPU acceleration\r\n- **[Industry Applications](references\u002Findustry-applications.md)** - Urban planning, disaster management\r\n- **[Programming Languages](references\u002Fprogramming-languages.md)** - Python, R, Julia, JS, C++, Java, Go, Rust\r\n- **[Data Sources](references\u002Fdata-sources.md)** - Satellite catalogs, APIs\r\n- **[Troubleshooting](references\u002Ftroubleshooting.md)** - Common issues, debugging, error reference\r\n- **[Code Examples](references\u002Fcode-examples.md)** - 500+ examples\r\n\r\n---\r\n\r\n**GeoMaster covers everything from basic GIS operations to advanced remote sensing and machine learning.**\r\n",{"name":5,"license":24,"metadata":25,"description":7},"MIT License",{"skill-author":26},"K-Dense Inc.","master","c247302c836de6438d746b45670990eb71691166","skills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster",1,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool","https:\u002F\u002Fgithub.com\u002Fqhjqhj00","chien.io","2026-05-11 15:07:59",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add qhjqhj00\u002Fresearch-skills-pool --skill geomaster",{"tool":41,"command":42},"bunx","bunx skills add qhjqhj00\u002Fresearch-skills-pool --skill geomaster",{"tool":44,"command":45},"pnpm","pnpm dlx skills add qhjqhj00\u002Fresearch-skills-pool --skill geomaster","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002FSKILL.md",true,[49,55,59,63,67,71,75,79,83,87,91,95,99,103,107],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"README.md","REFERENCE","参考文档",3323,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002FREADME.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"references\u002Fadvanced-gis.md",11121,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fadvanced-gis.md",{"path":60,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":61,"sourceHtmlUrl":62},"references\u002Fbig-data.md",8818,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fbig-data.md",{"path":64,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Fcode-examples.md",13023,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fcode-examples.md",{"path":68,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":69,"sourceHtmlUrl":70},"references\u002Fcoordinate-systems.md",8893,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fcoordinate-systems.md",{"path":72,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":73,"sourceHtmlUrl":74},"references\u002Fcore-libraries.md",6866,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fcore-libraries.md",{"path":76,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":77,"sourceHtmlUrl":78},"references\u002Fdata-sources.md",9331,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fdata-sources.md",{"path":80,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":81,"sourceHtmlUrl":82},"references\u002Fgis-software.md",9284,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fgis-software.md",{"path":84,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Findustry-applications.md",13491,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Findustry-applications.md",{"path":88,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Fmachine-learning.md",13245,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fmachine-learning.md",{"path":92,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":93,"sourceHtmlUrl":94},"references\u002Fprogramming-languages.md",11417,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fprogramming-languages.md",{"path":96,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":97,"sourceHtmlUrl":98},"references\u002Fremote-sensing.md",10474,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fremote-sensing.md",{"path":100,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Fscientific-domains.md",11027,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fscientific-domains.md",{"path":104,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Fspecialized-topics.md",11114,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Fspecialized-topics.md",{"path":108,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Ftroubleshooting.md",11219,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002FK-Dense-AI--scientific-agent-skills--scientific-skills__geomaster\u002Freferences\u002Ftroubleshooting.md",[112,119,126,133,140,147],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":19},"01M73JJB6EQPFPPVQK2Y9N67Q0","transformer-lens-interpretability","Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F130-transformer-lens\u002FSKILL.md",[],[],{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":19},"01M73JHTANRRWEE9HP98TPEHAY","wiki-vitepress","Packages generated wiki Markdown into a VitePress static site with dark theme, dark-mode Mermaid diagrams with click-to-zoom, and production build output. Use when the user wants to create a browsable website from generated wiki pages.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F126-wiki-vitepress\u002FSKILL.md",[],[],{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":19},"01M73JHJBMVFW46F1Y8K3SBF43","phoenix-observability","Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications with detailed traces, running evaluations on datasets, or monitoring production AI systems with real-time insights.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F124-phoenix\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":19},"01M73JHFPA3A2806PMP42ZN86Q","frontend-ui-dark-ts","Build dark-themed React applications using Tailwind CSS with custom theming, glassmorphism effects, and Framer Motion animations. Use when creating dashboards, admin panels, or data-rich interfaces with a refined dark aesthetic.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F123-frontend-ui-dark-ts\u002FSKILL.md",[],[],{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":19},"01M73JFVBVK57NRJS49TT2VDCT","tensorboard","Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F113-tensorboard\u002FSKILL.md",[],[],{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":151,"licenseFileUrl":6,"sourceHtmlUrl":152,"categories":153,"keywords":154,"publishedAt":19},"01M73JFF6PK6ZCN41C89J7HS2B","markdown-mermaid-writing","Comprehensive markdown and Mermaid diagram writing skill. Use when creating any scientific document, report, analysis, or visualization. Establishes text-based diagrams as the default documentation standard with full style guides (markdown + mermaid), 24 diagram type references, and 9 document templates.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F110-markdown-mermaid-writing\u002FSKILL.md",[],[]]