SDR model and landslide risk

Hello,

I am hoping to model landslide risk alongside using the SDR model. The user guide indicates that this is possible to do in post-processing ‘The additional soil loss can be added in post-processing.’. I was wondering if there is any more advice, or examples, on analysis needed to add to the model results to show landslide risk?

Many thanks

Hi @MegC -

None of our InVEST models will help with landslides, for that kind of analysis you’ll need to find a different model or method. So maybe I should rephrase that in the User Guide, since you can’t post-process SDR to get landslide information. @RafaSchmitt did landslide modeling for a project in Nepal, I don’t think that has been published yet, but perhaps he can give some advice.

~ Stacie

1 Like

Hi guys,
I was also hoping to use the outputs of the SDR model to estimate landslide risks, perhaps using the new output maps of avoided erosion in a given pixel or total sediment eroded. I came here on the community group to see if anyone else has also used this approach and found these messages from 2021. I believed that using this amount of erosion produced and the amount of rain in a given pixel, plus the history of landslide in the region, it would be possible to correlate to develop an Index of landslide risk. Could be even something using a RandomForest approach, for example, using the landslide history with the amount of sediment plus rain to create a landslide-risk index. Or even to produce a rough threshold using GIS of areas with high sediment erosion and areas with above x% of heavy rain. Something like that.

@swolny Stacey, did you mean that there is not any output ready in the model for landslides, but would be possible to interpret the outputs to produce such an index? or do you believe that the model output should not be used for such purpose due to how it was developed?

@RafaSchmitt did you eventually use a landslide analysis with a SDR output? can you provide further guidance?

Thanks,
Stella

Hi @stellamanes ,

I’m chiming in here to reiterate Stacie’s point that “None of our InVEST models will help with landslides, for that kind of analysis you’ll need to find a different model or method.” SDR is simply not intended nor designed to address mass wasting questions.

-Jesse