I want to make a susceptibility map of an area by using ArcGIS Pro and R.

I am studying forest fires, so considering the independent variables are precipitation, air temperature, surface temperature, altitude, slope, etc., the dependent variables are fire occurrence, and fire does not occur. I have obtained the fire position information, and randomly sample a certain number of non-fire positions and information. Meanwhile, through these points, information about fire-related factors was obtained. The analysis of these data was performed in R using logistic regression and random forest methods, and I have obtained the importance of each fire-related factor.

Currently, in R I can predict the probabilities of fires and non-fires, which I calculate from the model and the properties of the known points, but I want to extend this prediction to the entire study area. I have tried converting the raster layer of the study area into a point layer in ArcGIS and extracting the information of various fire-related factors through these points, and then importing these points into R for prediction. However, due to a large amount of data, some data are missing. I am currently planning to superimpose the face information of each relevant factor in R for prediction. I don't know if it is feasible.

  • 1
    Welcome to GIS SE. As a new user, please take the Tour, which emphasizes the importance of asking One question per Question. "Map the whole research" is not a focused Question -- You don't even indicate which software you will use or what the map should contain. Please Edit the Question to answer the questions "What have you tried?" and " Where are you stuck?" This will help increase focus.
    – Vince
    Aug 3 at 13:17
  • Have you researched how to construct forest fire susceptibility maps?
    – Spacedman
    Aug 3 at 14:36
  • I have not done relevant research before, but I have read some relevant research papers. In terms of susceptible map, there is more research on landslides, and there is relatively little content on the production of forest fire susceptible maps. But the two types of graphs are very similar, the difference lies in what factors are considered.
    – bangmin li
    Aug 3 at 16:07


Browse other questions tagged or ask your own question.