When I analyse SDR raster results there are some white squares with no data and also sediment export doesn’t seem to respond equal in all river basin. I have noticed that only the middle of the basin as attributed values…
I don’t know what is happening, I tried to change TFA value from 1000 to 100. TFA1000 results in projected streams closer to reality compared to TFA100, my pixel size is 100m. However, the change didn’t changed much.
I also don’t understand the meaning of some values like Borselli K Parameter, MAXIMUM sdr value, Borcelli IC0 parameter and Maximum L value . I am using default values but I dont know in what extent changing them will change my results.
Here is a file with my data and results with logfile. I have two results (2018 and future) but checking one will be enough since I only change land use map.
Looking at your results, I don’t see any regions of missing outputs that aren’t supposed to be there. The only areas of your basin that are not included in the output are areas that InVEST’s routing determines that it does not drain to a stream. For example you can see in the intermediate/sdr.tif output that there are a few missing regions in the southeast corner of your study area:
I think the user’s guide chapter has a much better discussion and description of how these values are used than what I’ll be able to describe here. Could you take a look at the chapter and let us know if you have any specific questions?
Previous scholars have outlined detailed explanation. You should understand that when estimating sediment deposition, the InVEST-SDR model assumes that no sediment is exported to the stream, that’s why in all stream networks you have seen "no data” pixels.
Actually, the SDR model is designed to quantify the amount of sediment exported to the stream. What it does not do is quantify erosion within stream channels, which is why there is no data for the stream pixels.