用 dNBR（NIR/SWIR 差分归一化烧伤比）判定五级烧伤严重度，结合火后多期 NDVI 恢复曲线估算恢复轨迹、恢复斜率与恢复年限，输出严重度 GeoTIFF、恢复轨迹 JSON 与恢复年限栅格。Post-fire recovery from dNBR severity and NDVI time series.
1 次安装
编程
geoskill-fire-weather-index
试用基于加拿大 FWI 系统，由气象要素（温度/湿度/风速/降水）逐日递推计算 FFMC、DMC、DC、ISI、BUI、FWI 六分量，输出火险等级栅格与时序
它能做什么
基于加拿大 FWI 系统，由气象要素（温度/湿度/风速/降水）逐日递推计算 FFMC、DMC、DC、ISI、BUI、FWI 六分量，输出火险等级栅格与时序
技能文档
火险天气指数 | Fire Weather Index
This skill implements the Canadian Forest Fire Weather Index System (Van Wagner 1987). From daily noon meteorological observations (air temperature °C, relative humidity %, wind speed km/h, 24h precipitation mm), six components are recursively updated day by day: FFMC (Fine Fuel Moisture Code), DMC (Duff Moisture Code), DC (Drought Code), ISI (Initial Spread Index), BUI (Buildup Index), and FWI (Fire Weather Index). The recurrence formulas are consistent with the open-source cffdrs library.
FFMC is highly sensitive to precipitation (a single heavy rainfall event can bring it down quickly); DMC/DC are cumulative drought indicators that grow day by day with potential evapotranspiration; ISI/BUI/FWI jointly reflect spread potential and energy release. The final outputs are the final-day six-component rasters, a fire danger class raster (Low/Moderate/High/Very High/Extreme), and a daily spatial-mean time series JSON.
Application scenarios: forest and grassland fire danger rating forecasting, drought–fire coupling analysis, and post-hoc review of the meteorological context of historical fires.
Dependencies / 依赖
pip install 'numpy' 'rasterio' 'scipy'
Usage / 使用方法
Basic usage
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 [other options]
Example 1 (synthetic data, offline)
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 --synthetic --output-dir ./out
Example 2 (bbox only; auto-synthesized, offline)
python geoskill-fire-weather-index.py --bbox 121.0 31.0 122.0 32.0 --n-dates 45 --output-dir ./out
Example 3 (specify the number of time steps and silence)
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 --synthetic --n-dates 60 --output-dir ./out --quiet
Example 4 (real meteorological raster input; 4 bands = temperature / humidity / wind speed / precipitation)
python geoskill-fire-weather-index.py --input meteo_day.tif --output-dir ./out
Example 5 (tiny area)
python geoskill-fire-weather-index.py --bbox 116.39 39.90 116.40 39.91 --output-dir ./out --quiet
Output / 输出
| File | Format | Description |
|---|---|---|
fwi_components.tif | GeoTIFF | Final-day six-component stack (6 bands: FFMC/DMC/DC/ISI/BUI/FWI) |
fwi_danger_class.tif | GeoTIFF | Final-day fire danger class (1=Low … 5=Extreme) |
fwi_timeseries.json | JSON | Daily six-component spatial-mean time series |
output-manifest.json | JSON | Run manifest |
Data Source / 数据源 / Source
- Real mode: local multi-band meteorological GeoTIFF (4 bands = temperature / humidity / wind speed / precipitation).
- Synthetic mode (
--syntheticor bbox-only): locally generates a physically consistent meteorological time series containing "dry, hot, windy → heavy rainfall" events; no network required.
Privacy / 隐私声明 / Privacy
- Runs offline by default;
--syntheticmode requires no network at all. - All processing is performed locally; user data is never uploaded.
License / License
MIT
name: geoskill-fire-weather-index description: '基于加拿大 FWI 系统，由气象要素（温度/湿度/风速/降水）逐日递推计算 FFMC、DMC、DC、ISI、BUI、FWI 六分量，输出火险等级栅格与时序'
火险天气指数 | Fire Weather Index
本 skill 实现加拿大林火天气指数系统（Canadian Forest Fire Weather Index System，Van Wagner 1987），由每日正午气象观测（气温 ℃、相对湿度 %、风速 km/h、24h 降水 mm）逐日递推六个分量：FFMC（细可燃物湿度码）、DMC（腐殖质 湿度码）、DC（干旱码）、ISI（初始蔓延指数）、BUI（累积指数）与 FWI（火险 天气指数）。递推公式与 cffdrs 开源库一致。
FFMC 对降水高度敏感（一次强降水即可快速回落）；DMC/DC 为累积型干旱指标， 随潜在蒸散逐日增长；ISI/BUI/FWI 综合反映蔓延潜力与能量释放。最终输出末日 六分量栅格、火险等级栅格（Low/Moderate/High/Very High/Extreme）与逐日空间 均值时序 JSON。
应用场景：森林草原火险等级预报、干旱-火险耦合分析、历史火灾气象背景复盘。
依赖
pip install 'numpy' 'rasterio' 'scipy'
使用方法
基本用法
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 [其他参数]
示例 1（合成数据，离线）
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 --synthetic --output-dir ./out
示例 2（仅给 bbox，自动合成，离线）
python geoskill-fire-weather-index.py --bbox 121.0 31.0 122.0 32.0 --n-dates 45 --output-dir ./out
示例 3（指定时间步长与静默）
python geoskill-fire-weather-index.py --bbox 116.0 39.0 117.0 40.0 --synthetic --n-dates 60 --output-dir ./out --quiet
示例 4（真实气象栅格输入，4 波段=温度/湿度/风速/降水）
python geoskill-fire-weather-index.py --input meteo_day.tif --output-dir ./out
示例 5（极小区域）
python geoskill-fire-weather-index.py --bbox 116.39 39.90 116.40 39.91 --output-dir ./out --quiet
输出
| 文件 | 格式 | 说明 |
|---|---|---|
fwi_components.tif | GeoTIFF | 末日六分量栈（6 波段：FFMC/DMC/DC/ISI/BUI/FWI） |
fwi_danger_class.tif | GeoTIFF | 末日火险等级（1=Low … 5=Extreme） |
fwi_timeseries.json | JSON | 逐日六分量空间均值时序 |
output-manifest.json | JSON | 运行清单 |
数据源 / Source
- 真实模式：本地多波段气象 GeoTIFF（4 波段 = 温度/湿度/风速/降水）。
- 合成模式（
--synthetic或仅--bbox）：本地生成含"干热大风→强降水"事件的物理一致气象时序，无需网络。
隐私声明 / Privacy
- 默认离线运行，
--synthetic模式完全无网络。 - 所有处理在本地完成，不上传用户数据。
License
MIT
相关技能
融合 SPI（标准化降水指数，Gamma 分布拟合+正态反演）与 VHI（植被健康指数）的综合干旱分级。Combined drought grading fusing SPI (Gamma-fit standardized precipitation index) and VHI (vegetation health index). 输出干旱等级/SPI 栅格 + 面积统计 JSON。
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.
1 次安装
基于百分位阈值检测热浪、寒潮与暴雨等极端天气事件，统计强度、持续时间与空间范围，输出事件清单 JSON 与空间分布栅格。Percentile-threshold detection of heatwaves, cold spells, and heavy rainfall, reporting intensity, duration, and spatial extent with an event-list JSON and spatial raster.
热浪影响评估：逐像元 P90 分位数阈值 + 连续≥3天连通域热浪检测，Stull/简化湿球温度健康风险分级，人口暴露与脆弱性制图。Heatwave impact assessment: per-pixel P90 threshold with connected-run detection, wet-bulb temperature health risk, population exposure and vulnerability mapping. 输出热浪日数/暴露人口/脆弱性/湿球风险 GeoTIFF + 事件清单 JSON。
多压力因子归一化加权叠加得综合影响指数，独立概率模型估算累积效应，按阈值划分5 级影响等级。Assesses environmental impact grades by multi-factor overlay and cumulative effects. 输出影响指数与等级 GeoTIFF。