raff-k
commented
1 year ago the length of the rain event which triggered the landslide might be different from case to case. I suggest to take a length which covers the entire rain event. The impelemted function based on Melillo et al. 2015 - and was also validated with this data (or data of Rossi et al. 2017). Anyway, there are a couple of hyperparameters or thresholds to "kick-out" irrelevant precipitation or to account for dry periods depending on the season. I set the defaults to the Melillo-publication. Maybe if your study area is different, you should adapt these values. - But this should explain your "full duration"-issue. However, the sub-events should show rain events of smaller length. And, if you use dates via parameter date.of.failure a subsetting is performed to this failure data. But you can also input already subsetted data.
Hello, I'd like to ask you some questions about the event function. In this function, all I need to input is the rainfall data and the rainfall date, but the function does not say how long before the landslide? The critical rainfall and duration of rainfall in the output of the function are mostly the full duration of the rainfall data I input, which is not the rainfall event that I want to see cause landslide. Looking forward to your reply