I compared the usle output from Invest with a single-flow soil-erosion calculation in ArcGIS. The difference is obvious. There are some parcels they have almost the same amount, but usually the InVest parcels are higher (sometimes more than twice as high)
The LS factor for the single-flow calculation is based on the work from Mitasova et al., 1996.
I used the same rasta-files for both calculations.
The catchment has 30km², the rastersolution is 1m x 1m, the area is kind of steep but the K-factors are relatively low. the usle from the the single-flow method seems more realistic to me.
Is the difference caused from different flow-methods?
overall the ls-raster from InVest seem realistc, but threr are a view “hot spots” where the value goes up to 5000! this is causing major inaccuracies.