Biomod projection for Climate Change

[ShreePoudel0]

hello, I am doing a habitat suitability projection under different Climate Change Scenarios. i have downloaded future CC models from bioclim website and i don't quite understand how this function works. For the current projection, I have used bioclim data set as well as anthropogenic variables. Since we dont have future data for those anthropogenic variables, do we stack only the future Climatic variables or future climatic variables and current anthropogenic variable. I tried doing with the later and I had unsatisfactory result. I have added a screenshot of current projection. Here the value ranges upto 800 but when doing future projection the value is just upto 400 and the result is very unsatisfactory. Do we have some way to make this uniform.

The code I use is given below: myBiomodProj <- BIOMOD_Projection(bm.mod = myBiomodModelOut, proj.name = 'Current', new.env = myExpl2040, models.chosen = 'all', metric.binary = 'all', metric.filter = 'all', build.clamping.mask = F)

myBiomodEMProj <- BIOMOD_EnsembleForecasting(bm.em = myBiomodEM, proj.name = 'try', new.env = myExpl2040, models.chosen = 'all', metric.binary = 'all', metric.filter = 'all')

[HeleneBlt]

Hello there !

If you don't have all the predictions for your variables, you have two options :

• You can assume the anthropogenic variables don't change and you add the current anthropogenic variables to the future climatic variables (this is what you're already doing)
• You build a model with the climatic variables only and you project it with the future climatic variables.

Be careful with the result of Biomod_EnsembleForcasting : be sure which EM you're looking at. For example, EMmean and EMcv will not have the same range because they don't represent the same thing.

If it doesn't help, please don't hesitate to let us know more about your case so we can help you. Hélène

[ShreePoudel0]

Hello again. Is it better to stack new expl variable and use it directly in EnsembleForecasting with new.env or we can altogether start the process from biomodFormatting using new future explanatory variable. Thank You

On Mon, May 6, 2024 at 5:22 PM HBlancheteau @.***> wrote:

Hello there !

If you don't have all the predictions for your variables, you have two options :

• You can assume the anthropogenic variables don't change and you add the current anthropogenic variables to the future climatic variables (this is what you're already doing)
• You build a model with the climatic variables only and you project it with the future climatic variables.

Be careful with the result of Biomod_EnsembleForcasting : be sure which EM you're looking at. For example, EMmean and EMcv will not have the same range because they don't represent the same thing.

If it doesn't help, please don't hesitate to let us know more about your case so we can help you. Hélène

— Reply to this email directly, view it on GitHub https://github.com/biomodhub/biomod2/issues/451#issuecomment-2095810174, or unsubscribe https://github.com/notifications/unsubscribe-auth/BBPZRFVR2BFCB5BKVV6PZ7DZA5TO7AVCNFSM6AAAAABHHPBWBSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJVHAYTAMJXGQ . You are receiving this because you authored the thread.Message ID: @.***>

[HeleneBlt]

Hello Shree,

If you stack your new expl variables and use them directly in BIOMOD_EnsembleForecasting, the projection will be created with your Ensemble Model created with your initial expl variables. (That's why you can add a new variable and you have to keep the same names.) If you restart the process from BIOMOD_Formating, you will create a new model based on these new variables. So it depends on what you want.

Hope it's clearer, Hélène