Code for large scale forecasting the potential distribution of Heracleum Sosnowskyi on the territory of Russia under climate change.
Source code for paper Large scale forecasting the potential distribution of Heracleum Sosnowskyi on the territory of Russia under the climate change
We propose a machine learning approach based on the Random Forest model for forecasting the potential distribution of Heracleum Sosnowskyi. This research aims to establish the possible habitat suitability of HS in current and future climate conditions across the territory of European part of Russia.
Map demonstrate points with Heracleum Sosnowskyi that were obtained from open sources
Clone this repository
Install R packages
- biomod
- spThin
- biomod2
- ggplot2
- gridExtra
- raster
- rasterVis
- maptools
The CSV file contains the coordinates of the location of the Heracleum Sosnowskyi and the parameters (soil variables, bioclim data) used to train the Random Forest model
CSV file: Occurrence points
Climatic variables were collected from the Worldclim project
Soil data were downloaded from the SoilGrids database
Source code of paper to conduct Random Forect model training, reproduce results and plots contatins in src.R
file - Code
Trained model stored in models
folder - Model
Code to forecast future dictribution of Heracleum Sosnowskyi under different climate scenarios - Code
ROC-AUC, MDG and MDA plots created with python.
To reproduce plots install python packages
- matplotlib
- numpy
- seaborn
- sklearn
- pandas
- ipython
- jupyter
Open ROC-AUC plots.ipynb
file with Jupyter-notebook
Distributed under the CC0 1.0 license. See LICENSE
for more information.