Install
openclaw skills install @ruiduobao/geoskill-sar-wind-speedSAR 海面风场反演:基于简化 CMOD5/CMOD7 经验地球物理模型,由后向散射系数 σ⁰ 与风向二分法数值反演海面风速场
openclaw skills install @ruiduobao/geoskill-sar-wind-speedRetrieves sea-surface wind speed at 10 m height from the SAR backscatter coefficient σ⁰, using a simplified CMOD (C-band Model) empirical geophysical model:
σ⁰_dB(U, φ, θ) = [a0 + a1·θ] + [s0 + s1·θ]·U·M(φ)
M(φ) = 1 + m1·cosφ + m2·cos2φ
where U is the wind speed (m/s), θ is the incidence angle, φ is the angle between the wind direction and the radar look direction, and M(φ) is the azimuthal modulation (upwind > crosswind > downwind). Since σ⁰_dB increases monotonically with U, the retrieval solves per-pixel roots using a vectorized bisection method and supports two sets of empirical coefficients: CMOD5 / CMOD7.
pip install 'numpy' 'rasterio'
python geoskill-sar-wind-speed.py --bbox 121.0 30.0 122.0 31.0 --wind-dir 45 --cmod cmod5 --synthetic --output-dir ./out
python geoskill-sar-wind-speed.py --bbox 121.0 30.0 122.0 31.0 --wind-dir 90 --cmod cmod7 --synthetic --output-dir ./out
python geoskill-sar-wind-speed.py --input sigma0_db.tif --wind-dir 225 --incidence-angle 35 --output-dir ./out
python geoskill-sar-wind-speed.py --input sigma0_db.tif --wind-dir 180 --radar-azimuth 90 --output-dir ./out
| File | Format | Description |
|---|---|---|
wind_speed.tif | GeoTIFF | Retrieved wind speed field (m/s) |
retrieval_params.json | JSON | Model coefficients, wind direction, incidence angle, and other parameters |
output-manifest.json | JSON | Run manifest (including RMSE/correlation-coefficient QA for synthetic mode) |
A local σ⁰ (dB) GeoTIFF, or a simulated scene generated by CMOD forward modeling of a --synthetic spatially varying wind field.
--synthetic mode requires no network at all.MIT
从 SAR 后向散射系数 σ⁰ 反演海面 10 m 高度风速。采用简化的 CMOD(C-band Model) 经验模型:
σ⁰_dB(U, φ, θ) = [a0 + a1·θ] + [s0 + s1·θ]·U·M(φ)
M(φ) = 1 + m1·cosφ + m2·cos2φ
其中 U 为风速 (m/s),θ 为入射角,φ 为风向与雷达视线夹角,M(φ) 为方位向调制 (迎风 > 侧风 > 顺风)。σ⁰_dB 对 U 单调递增,反演时用向量化二分法逐像元求根, 支持 CMOD5 / CMOD7 两套经验系数。
pip install 'numpy' 'rasterio'
python geoskill-sar-wind-speed.py --bbox 121.0 30.0 122.0 31.0 --wind-dir 45 --cmod cmod5 --synthetic --output-dir ./out
python geoskill-sar-wind-speed.py --bbox 121.0 30.0 122.0 31.0 --wind-dir 90 --cmod cmod7 --synthetic --output-dir ./out
python geoskill-sar-wind-speed.py --input sigma0_db.tif --wind-dir 225 --incidence-angle 35 --output-dir ./out
python geoskill-sar-wind-speed.py --input sigma0_db.tif --wind-dir 180 --radar-azimuth 90 --output-dir ./out
| 文件 | 格式 | 说明 |
|---|---|---|
wind_speed.tif | GeoTIFF | 反演风速场 (m/s) |
retrieval_params.json | JSON | 模型系数、风向、入射角等参数 |
output-manifest.json | JSON | 运行清单(含合成模式 RMSE/相关系数 QA) |
本地 σ⁰ (dB) GeoTIFF,或 --synthetic 空间变化风场经 CMOD 正演的模拟场景。
--synthetic 模式完全无网络。MIT