geoskill-climate-zone-classification

基于月均温与月降水栅格的气候区划，支持柯本-盖格（Köppen-Geiger）规则树与简化 Strahler 发生分类，输出气候类别码 GeoTIFF + 面积统计 JSON，可选双期变化检测。Climate zone classification from monthly temperature/precipitation rasters (Köppen-Geiger rule tree & simplified Strahler), with area statistics and optional change detection.

气候区划 | Climate Zone Classification

Performs climate zone classification from monthly mean temperature and monthly precipitation rasters, for climate mapping, ecological zoning, agro-climatic resource assessment, and climate change monitoring. Two schemes are built in:

• Köppen-Geiger: the classic rule tree (Peel et al. 2007 / Kottek et al. 2006). Determines the five main groups A (tropical) / B (arid) / C (temperate) / D (continental) / E (polar) from the coldest/warmest month mean temperature, annual mean temperature, annual precipitation, and the seasonal distribution of precipitation, then subdivides into second and third letters, yielding 30 classes in total (Af tropical rainforest, Am tropical monsoon, Aw/As tropical savanna, BWh/BWk desert, BSh/BSk steppe, Cfa/Cfb/Cfc, Csa/Csb/Csc, Dfa…Dfd, ET tundra, EF ice cap).
• Strahler (simplified genetic classification): aggregates temperature zones and moisture conditions into 10 genetic climate classes (tropical humid / tropical arid / humid subtropical / Mediterranean / marine west coast / humid continental / dry continental / subarctic / tundra / polar ice cap).

Supports a --synthetic mode that generates a simulated climate field zoned along latitude (with physically consistent seasonal temperature cycles and precipitation distributions) for offline validation of the full workflow and class recovery. --input2 enables two-epoch climate zone change detection.

Dependencies / 依赖

bash

pip install 'numpy' 'rasterio' 'scipy'

Usage / 使用方法

Basic usage (bbox only, auto-synthetic, Köppen)

bash

python geoskill-climate-zone-classification.py --bbox 116.0 39.0 117.0 40.0 --output-dir ./out

Example 1: synthetic data Strahler classification

bash

python geoskill-climate-zone-classification.py --bbox 116.0 39.0 117.0 40.0 --synthetic --classification strahler --output-dir ./strahler

Example 2: real 24-band climate rasters (12 temperature + 12 precipitation)

bash

python geoskill-climate-zone-classification.py --input climate.tif --classification koppen --output-dir ./real

Example 3: two-epoch change detection

bash

python geoskill-climate-zone-classification.py --input climate_1990.tif --input2 climate_2020.tif --classification koppen --output-dir ./change

Example 4: synthetic data for a different region

bash

python geoskill-climate-zone-classification.py --bbox 121.0 31.0 122.0 32.0 --synthetic --output-dir ./shanghai

Example 5: Köppen vs Strahler comparison

bash

python geoskill-climate-zone-classification.py --bbox 116 39 117 40 --synthetic --classification koppen --output-dir ./cmp_koppen --quiet
python geoskill-climate-zone-classification.py --bbox 116 39 117 40 --synthetic --classification strahler --output-dir ./cmp_strahler --quiet

Output / 输出

File	Format	Description
climate_zones.tif	GeoTIFF (float32)	Climate class code raster (code table in area_statistics.json), EPSG:4326
area_statistics.json	JSON	Code table + per-class pixel count/fraction/estimated area + optional change detection
output-manifest.json	JSON	Run manifest (inputs/outputs/QA/software versions)

Data Source / 数据源 / Source

• Input: local 24-band GeoTIFF (bands 1-12 = mean monthly temperature for Jan-Dec in °C, bands 13-24 = monthly precipitation for Jan-Dec in mm)
• Synthetic mode: monthly climate field zoned along latitude generated locally; no external data source
• Classification rules: Peel, M. C., Finlayson, B. L., McMahon, T. A. (2007), HESS

Privacy / 隐私声明 / Privacy

• Runs fully offline by default; no network requests are made
• --synthetic mode reads no external data
• All computation is performed locally; user data is never uploaded

License / License

MIT

---

---

name: geoskill-climate-zone-classification description: '基于月均温与月降水栅格的气候区划，支持柯本-盖格（Köppen-Geiger）规则树与简化 Strahler 发生分类，输出气候类别码 GeoTIFF + 面积统计 JSON，可选双期变化检测。Climate zone classification from monthly temperature/precipitation rasters (Köppen-Geiger rule tree & simplified Strahler), with area statistics and optional change detection.'

气候区划 | Climate Zone Classification

对逐月均温与逐月降水栅格执行气候区划分类，用于气候制图、生态分区、农业 气候资源评价与气候变化监测。内置两套方案：

• Köppen-Geiger（柯本-盖格）：经典规则树（Peel et al. 2007 / Kottek et al. 2006）。 由最冷/最热月均温、年均温、年降水及降水季节分配判定 A（热带）/B（干旱）/ C（温带）/D（大陆性）/E（极地）五大主组，并细分第二、第三字母，共 30 个类别 （Af 热带雨林、Am 热带季风、Aw/As 热带稀树草原、BWh/BWk 沙漠、BSh/BSk 草原、 Cfa/Cfb/Cfc、Csa/Csb/Csc、Dfa…Dfd、ET 苔原、EF 冰原）。
• Strahler（简化发生分类）：以气温带与水分条件归并为 10 个发生气候类 （热带多雨/热带干旱/湿润亚热带/地中海/海洋性西海岸/湿润大陆性/干燥大陆性/ 副极地/苔原/极地冰原）。

支持 --synthetic 模式生成沿纬度分带的模拟气候场（含物理一致的季节温降与 降水分配），可离线验证全流程与类别恢复。提供 --input2 做两期气候区变化检测。

依赖

bash

pip install 'numpy' 'rasterio' 'scipy'

使用方法

基本用法（仅给 bbox，自动合成，Köppen）

bash

python geoskill-climate-zone-classification.py --bbox 116.0 39.0 117.0 40.0 --output-dir ./out

示例 1：合成数据 Strahler 分类

bash

python geoskill-climate-zone-classification.py --bbox 116.0 39.0 117.0 40.0 --synthetic --classification strahler --output-dir ./strahler

示例 2：真实 24 波段气候栅格（12 温度 + 12 降水）

bash

python geoskill-climate-zone-classification.py --input climate.tif --classification koppen --output-dir ./real

示例 3：双期变化检测

bash

python geoskill-climate-zone-classification.py --input climate_1990.tif --input2 climate_2020.tif --classification koppen --output-dir ./change

示例 4：不同区域合成

bash

python geoskill-climate-zone-classification.py --bbox 121.0 31.0 122.0 32.0 --synthetic --output-dir ./shanghai

示例 5：Köppen vs Strahler 对比

bash

python geoskill-climate-zone-classification.py --bbox 116 39 117 40 --synthetic --classification koppen --output-dir ./cmp_koppen --quiet
python geoskill-climate-zone-classification.py --bbox 116 39 117 40 --synthetic --classification strahler --output-dir ./cmp_strahler --quiet

输出

文件	格式	说明
climate_zones.tif	GeoTIFF (float32)	气候类别码栅格（码表见 area_statistics.json），EPSG:4326
area_statistics.json	JSON	码表 + 各类别像元数/占比/估算面积 + 可选变化检测
output-manifest.json	JSON	运行清单（输入/输出/QA/软件版本）

数据源 / Source

• 输入：本地 24 波段 GeoTIFF（band 1-12 = 1-12 月均温 °C，band 13-24 = 1-12 月降水 mm）
• 合成模式：本地生成沿纬度分带的月值气候场，无外部数据源
• 分类规则：Peel, M. C., Finlayson, B. L., McMahon, T. A. (2007), HESS

隐私声明 / Privacy

• 默认完全离线运行，不发起任何网络请求
• --synthetic 模式不读取任何外部数据
• 所有计算在本地完成，不上传用户数据

License

MIT