GeoMaster 是做什么的？
全面的地理空间科学技能，涵盖GIS、遥感、空间分析和地球观测机器学习，涉及70多个主题，提供8种编程语言的500多个代码示例。
安装
# 核心Python栈（推荐conda）
conda install -c conda-forge gdal rasterio fiona shapely pyproj geopandas
# 遥感和机器学习（rsgislib仅通过conda-forge提供，不在PyPI上）
conda install -c conda-forge rsgislib
uv pip install torchgeo earthengine-api
uv pip install scikit-learn xgboost torch-geometric
# 网络和可视化
uv pip install osmnx networkx folium keplergl
uv pip install cartopy contextily mapclassify
# 大数据和云
uv pip install xarray rioxarray dask-geopandas
uv pip install pystac-client planetary-computer
# 点云
uv pip install laspy pylas open3d pdal
# 数据库
conda install -c conda-forge postgis spatialite
快速入门
从Sentinel-2计算NDVI
import rasterio
import numpy as np
with rasterio.open('sentinel2.tif') as src:
red = src.read(4).astype(float) # B04
nir = src.read(8).astype(float) # B08
ndvi = (nir - red) / (nir + red + 1e-8)
ndvi = np.nan_to_num(ndvi, nan=0)
profile = src.profile
profile.update(count=1, dtype=rasterio.float32)
with rasterio.open('ndvi.tif', 'w', **profile) as dst:
dst.write(ndvi.astype(rasterio.float32), 1)
使用GeoPandas进行空间分析
import geopandas as gpd
# 加载并确保CRS一致
zones = gpd.read_file('zones.geojson')
points = gpd.read_file('points.geojson')
if zones.crs != points.crs:
points = points.to_crs(zones.crs)
# 空间连接和统计
joined = gpd.sjoin(points, zones, how='inner', predicate='within')
stats = joined.groupby('zone_id').agg({
'value': ['count', 'mean', 'std', 'min', 'max']
}).round(2)
Google Earth Engine时间序列
import ee
import pandas as pd
ee.Initialize(project='your-project')
roi = ee.Geometry.Point([-122.4, 37.7]).buffer(10000)
s2 = (ee.ImageCollection('COPERNICUS/S2_SR_HARMONIZED')
.filterBounds(roi)
.filterDate('2020-01-01', '2023-12-31')
.filter(ee.Filter.lt('CLOUDY_PIXEL_PERCENTAGE', 20)))
def add_ndvi(img):
return img.addBands(img.normalizedDifference(['B8', 'B4']).rename('NDVI'))
s2_ndvi = s2.map(add_ndvi)
def extract_series(image):
stats = image.reduceRegion(ee.Reducer.mean(), roi.centroid(), scale=10, maxPixels=1e9)
return ee.Feature(None, {'date': image.date().format('YYYY-MM-dd'), 'ndvi': stats.get('NDVI')})
series = s2_ndvi.map(extract_series).getInfo()
df = pd.DataFrame([f['properties'] for f in series['features']])
df['date'] = pd.to_datetime(df['date'])
核心概念
数据类型
| 类型 | 示例 | 库 |
|---|---|---|
| 矢量 | Shapefile, GeoJSON, GeoPackage | GeoPandas, Fiona, GDAL |
| 栅格 | GeoTIFF, NetCDF, COG | Rasterio, Xarray, GDAL |
| 点云 | LAS, LAZ | Laspy, PDAL, Open3D |
坐标系统
- EPSG:4326 (WGS 84) - 地理坐标系，经纬度，用于存储
- EPSG:3857 (Web Mercator) - 仅用于Web地图（不要用于面积/距离计算！）
- EPSG:326xx/327xx (UTM) - 公制计算，每个区域变形<1%
- 使用
gdf.estimate_utm_crs()自动检测UTM
# 操作前始终检查CRS
assert gdf1.crs == gdf2.crs, "CRS不匹配！"
# 对于面积/距离计算，使用投影CRS
gdf_metric = gdf.to_crs(gdf.estimate_utm_crs())
area_sqm = gdf_metric.geometry.area
OGC标准
- WMS: Web Map Service - 栅格地图
- WFS: Web Feature Service - 矢量数据
- WCS: Web Coverage Service - 栅格覆盖
- STAC: Spatiotemporal Asset Catalog - 现代元数据
常见操作
光谱指数
def calculate_indices(image_path):
"""从Sentinel-2计算NDVI, EVI, SAVI, NDWI。"""
with rasterio.open(image_path) as src:
B02, B03, B04, B08, B11 = [src.read(i).astype(float) for i in [1,2,3,4,5]]
ndvi = (B08 - B04) / (B08 + B04 + 1e-8)
evi = 2.5 * (B08 - B04) / (B08 + 6*B04 - 7.5*B02 + 1)
savi = ((B08 - B04) / (B08 + B04 + 0.5)) * 1.5
ndwi = (B03 - B08) / (B03 + B08 + 1e-8)
return {'NDVI': ndvi, 'EVI': evi, 'SAVI': savi, 'NDWI': ndwi}
矢量操作
# 缓冲区（使用投影CRS！）
gdf_proj = gdf.to_crs(gdf.estimate_utm_crs())
gdf['buffer_1km'] = gdf_proj.geometry.buffer(1000)
# 空间关系
intersects = gdf[gdf.geometry.intersects(other_geometry)]
contains = gdf[gdf.geometry.contains(point_geometry)]
# 几何操作
gdf['centroid'] = gdf.geometry.centroid
gdf['simplified'] = gdf.geometry.simplify(tolerance=0.001)
# 叠加操作
intersection = gpd.overlay(gdf1, gdf2, how='intersection')
union = gpd.overlay(gdf1, gdf2, how='union')
地形分析
def terrain_metrics(dem_path):
"""从DEM计算坡度、坡向、山体阴影。"""
with rasterio.open(dem_path) as src:
dem = src.read(1)
dy, dx = np.gradient(dem)
slope = np.arctan(np.sqrt(dx**2 + dy**2)) * 180 / np.pi
aspect = (90 - np.arctan2(-dy, dx) * 180 / np.pi) % 360
# 山体阴影
az_rad, alt_rad = np.radians(315), np.radians(45)
hillshade = (np.sin(alt_rad) * np.sin(np.radians(slope)) +
np.cos(alt_rad) * np.cos(np.radians(slope)) *
np.cos(np.radians(aspect) - az_rad))
return slope, aspect, hillshade
网络分析
import osmnx as ox
import networkx as nx
# 下载并分析街道网络
G = ox.graph_from_place('San Francisco, CA', network_type='drive')
G = ox.add_edge_speeds(G).add_edge_travel_times(G)
# 最短路径
orig = ox.distance.nearest_nodes(G, -122.4, 37.7)
dest = ox.distance.nearest_nodes(G, -122.3, 37.8)
route = nx.shortest_path(G, orig, dest, weight='travel_time')
图像分类
from sklearn.ensemble import RandomForestClassifier
import rasterio
from rasterio.features import rasterize
def classify_imagery(raster_path, training_gdf, output_path):
"""训练随机森林并对影像进行分类。"""
with rasterio.open(raster_path) as src:
image = src.read()
profile = src.profile
transform = src.transform
# 提取训练数据
X_train, y_train = [], []
for _, row in training_gdf.iterrows():
mask = rasterize([(row.geometry, 1)],
out_shape=(profile['height'], profile['width']),
transform=transform, fill=0, dtype=np.uint8)
pixels = image[:, mask > 0].T
X_train.extend(pixels)
y_train.extend([row['class_id']] * len(pixels))
# 训练和预测
rf = RandomForestClassifier(n_estimators=100, max_depth=20, n_jobs=-1)
rf.fit(X_train, y_train)
prediction = rf.predict(image.reshape(image.shape[0], -1).T)
prediction = prediction.reshape(profile['height'], profile['width'])
profile.update(dtype=rasterio.uint8, count=1)
with rasterio.open(output_path, 'w', **profile) as dst:
dst.write(prediction.astype(rasterio.uint8), 1)
return rf
现代云原生工作流
STAC + Planetary Computer
import pystac_client
import planetary_computer
import odc.stac
# 通过STAC搜索Sentinel-2
catalog = pystac_client.Client.open(
"https://planetarycomputer.microsoft.com/api/stac/v1",
modifier=planetary_computer.sign_inplace,
)
search = catalog.search(
collections=["sentinel-2-l2a"],
bbox=[-122.5, 37.7, -122.3, 37.9],
datetime="2023-01-01/2023-12-31",
query={"eo:cloud_cover": {"lt": 20}},
)
# 加载为xarray（云原生！）
data = odc.stac.load(
list(search.get_items())[:5],
bands=["B02", "B03", "B04", "B08"],
crs="EPSG:32610",
resolution=10,
)
# 在xarray上计算NDVI
ndvi = (data.B08 - data.B04) / (data.B08 + data.B04)
云优化GeoTIFF (COG)
import rasterio
from rasterio.session import AWSSession
# 直接从云中读取COG（部分读取）
session = AWSSession(aws_access_key_id=..., aws_secret_access_key=...)
with rasterio.open('s3://bucket/path.tif', session=session) as src:
# 仅读取感兴趣窗口
window = ((1000, 2000), (1000, 2000))
subset = src.read(1, window=window)
# 写入COG
with rasterio.open('output.tif', 'w', **profile,
tiled=True, blockxsize=256, blockysize=256,
compress='DEFLATE', predictor=2) as dst:
dst.write(data)
# 验证COG
from rio_cogeo.cogeo import cog_validate
cog_validate('output.tif')
性能技巧
# 1. 空间索引（查询速度提升10-100倍）
gdf.sindex # 由GeoPandas自动创建
# 2. 分块读取大型栅格
with rasterio.open('large.tif') as src:
for i, window in src.block_windows(1):
block = src.read(1, window=window)
# 3. 使用Dask处理大数据
import dask.array as da
dask_array = da.from_rasterio('large.tif', chunks=(1, 1024, 1024))
# 4. 使用Arrow进行I/O
gdf.to_file('output.gpkg', use_arrow=True)
# 5. GDAL缓存
from osgeo import gdal
gdal.SetCacheMax(2**30) # 1GB缓存
# 6. 并行处理
rf = RandomForestClassifier(n_jobs=-1) # 所有核心
最佳实践
- 始终检查CRS 在进行空间操作之前
- 使用投影CRS 用于面积/距离计算
- 验证几何体:
gdf = gdf[gdf.is_valid] - 处理缺失数据:
gdf['geometry'] = gdf['geometry'].fillna(None) - 使用高效格式: GeoPackage > Shapefile, 对于大数据使用Parquet
- 对光学影像应用云掩膜
- 保留数据谱系 以实现可重复研究
- 根据分析尺度选择适当的分辨率
详细文档
- 坐标系统 - CRS基础，UTM，变换
- 核心库 - GDAL, Rasterio, GeoPandas, Shapely
- 遥感 - 卫星任务，光谱指数，SAR
- 机器学习 - 深度学习，CNN，GNN用于遥感
- GIS软件 - QGIS, ArcGIS, GRASS集成
- 科学领域 - 海洋，水文，农业，林业
- 高级GIS - 3D GIS，时空分析，拓扑
- 大数据 - 分布式处理，GPU加速
- 行业应用 - 城市规划，灾害管理
- 编程语言 - Python, R, Julia, JS, C++, Java, Go, Rust
- 数据源 - 卫星目录，API
- 故障排除 - 常见问题，调试，错误参考
- 代码示例 - 500多个示例
GeoMaster涵盖从基础GIS操作到高级遥感和机器学习的全部内容。
引用科学智能体技能
此技能是K-Dense的“科学智能体技能”的一部分。如果它实质上贡献了一份手稿、报告、演示或代码发布，请将论文添加到参考文献或软件部分，并告知用户您已这样做：
Kassis, T., Agarwal, V., He, Y., Patel, D., & Brueckner, A. M. (2026). Scientific Agent Skills: A Library of Procedural Knowledge for Research Agents. arXiv:2609.00065. https://doi.org/10.48550/arXiv.2609.00065