geoskill-invasive-species-spread

多时相指数阈值分类检测入侵新增像元，计算面积相对扩散速率，用环境适宜性×距离衰减预测入侵风险。Monitors invasive species spread from multi-temporal classification and risk prediction. 输出新增入侵与风险 GeoTIFF。

入侵物种扩散监测 | Invasive Species Spread Monitoring

Two epochs of remote sensing indices are classified by threshold to obtain the t0 presence zone and the t1 newly invaded zone; spread rate r = (A1−A0)/(A0×Δt); risk prediction = environmental suitability × spread accessibility exp(−d/λ), where d is the Euclidean distance to the nearest known invaded pixel (scipy distance transform) and λ is the dispersal scale (default 5 km). Risk ∈ [0,1].

Use cases: dynamic monitoring of invasive alien species (e.g., smooth cordgrass, Canada goldenrod) and priority ranking for prevention and control.

Dependencies / 依赖

bash

pip install numpy rasterio scipy scikit-learn

Usage / 使用方法

Example 1: synthetic two-epoch scenario (default 5-year interval)

bash

python geoskill-invasive-species-spread.py --bbox 116.0 39.0 117.0 40.0 --synthetic --output-dir ./output

Example 2: real 3-band input (index_t0, index_t1, suitability)

bash

python geoskill-invasive-species-spread.py --input invasive_inputs.tif --output-dir ./real

Example 3: stricter threshold + 3-year interval

bash

python geoskill-invasive-species-spread.py --bbox 116 39 117 40 --synthetic --threshold 0.2 --dt-years 3 --output-dir ./tuned

Example 4: strongly spreading species (large dispersal scale)

bash

python geoskill-invasive-species-spread.py --bbox 121 31 122 32 --synthetic --dispersal-scale 10000 --output-dir ./strong

Example 5: quiet batch run

bash

python geoskill-invasive-species-spread.py --bbox 113 23 114 24 --synthetic --quiet --output-dir ./batch

Output / 输出

File	Format	Description
new_invasion.tif	GeoTIFF (float32)	Newly invaded pixels (0/1), EPSG:4326
invasion_risk.tif	GeoTIFF (float32)	Invasion risk ∈ [0,1]
invasive_params.json	JSON	Area change, spread rate, and risk statistics
output-manifest.json	JSON	Run manifest (input/output/QA/software versions)

Data Source / 数据源 / Source

Local GeoTIFF (3 bands: two-epoch indices + suitability, optional); synthetic mode generates expanding invasion patches and an environmental suitability gradient locally, with no external data source.

Privacy / 隐私声明 / Privacy

• Runs fully offline by default; makes no network requests
• --synthetic mode reads no external data
• All computation is done locally; no user data is uploaded

License / License

MIT

---

---

name: geoskill-invasive-species-spread description: '多时相指数阈值分类检测入侵新增像元，计算面积相对扩散速率，用环境适宜性×距离衰减预测入侵风险。Monitors invasive species spread from multi-temporal classification and risk prediction. 输出新增入侵与风险 GeoTIFF。'

入侵物种扩散监测 | Invasive Species Spread Monitoring

两期遥感指数按阈值分类得到 t0 存在区与 t1 新增入侵区；扩散速率 r = (A1-A0)/(A0×Δt)；风险预测 = 环境适宜性 × 扩散可达性 exp(-d/λ)，其中 d 为到最近已知入侵像元的欧氏距离（scipy 距离变换），λ 为扩散尺度（默认 5 km）。风险 ∈ [0,1]。

适用场景：外来入侵物种（如互花米草、加拿大一枝黄花）动态监测与防控优先级排序。

依赖

bash

pip install numpy rasterio scipy scikit-learn

使用方法

示例 1：合成两期场景（默认 5 年间隔）

bash

python geoskill-invasive-species-spread.py --bbox 116.0 39.0 117.0 40.0 --synthetic --output-dir ./output

示例 2：真实 3 波段输入（index_t0,index_t1,suitability）

bash

python geoskill-invasive-species-spread.py --input invasive_inputs.tif --output-dir ./real

示例 3：更严阈值 + 3 年间隔

bash

python geoskill-invasive-species-spread.py --bbox 116 39 117 40 --synthetic --threshold 0.2 --dt-years 3 --output-dir ./tuned

示例 4：强扩散物种（大扩散尺度）

bash

python geoskill-invasive-species-spread.py --bbox 121 31 122 32 --synthetic --dispersal-scale 10000 --output-dir ./strong

示例 5：静默批量

bash

python geoskill-invasive-species-spread.py --bbox 113 23 114 24 --synthetic --quiet --output-dir ./batch

输出

文件	格式	说明
new_invasion.tif	GeoTIFF (float32)	新增入侵像元（0/1），EPSG:4326
invasion_risk.tif	GeoTIFF (float32)	入侵风险 ∈ [0,1]
invasive_params.json	JSON	面积变化、扩散速率、风险统计
output-manifest.json	JSON	运行清单（输入/输出/QA/软件版本）

数据源 / Source

本地 GeoTIFF（3 波段两期指数+适宜性，可选）；合成模式本地生成扩张型入侵斑块与环境适宜性梯度，无外部数据源。

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