This forum is shutting down! Please post new discussions at community.naturalcapitalproject.org

seasonal water yield model

MattiasMattias Member
edited August 2018 in Freshwater Models
I downloaded map of Hydrological soil groups from this dataset: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5952866/ as input for the seasonal water yield model.
After cutting the global map for my study area, I found out that there are some gaps in the map, probably due to permanently snow covered areas. Do these gaps cause a crash in the model or the the model overcome the issue in some way?
in case, how can I fix it?
thanks in advance
Mattias
Post edited by Mattias on

Comments

  • swolnyswolny Member, NatCap Staff
    Hi @Mattias -

    Anywhere there is NoData in one of the input rasters, there will be NoData in the output maps, so if you leave the gaps, you will have missing output data there.

    I've seen these kinds of gaps before, in snow areas or bodies of water usually. What we often do is fill in the missing areas with the dominant value from around the gap. This assumes that the soil properties are similar, but in the absence of data otherwise, we generally find it to be a good way to deal with it. The other option would be to do some research and see if you can find out what type of soil is likely to be in those areas, and manually fill in the value appropriately. 

    ~ Stacie
  • Hi Stacie,

    thanks for your answer.
    Since I want to use the outcome of SWY model as input for nutrient retention model, NoData will be a problem?
    Among the the outputs of SWY, which raster should be used as nutrient runoff proxy as input for NDR model? and should I normalize it before application in NDR model as described in the online manual, or the model routines will do it automatically?
    Matt
  • swolnyswolny Member, NatCap Staff
    Hi Matt -

    If you have NoData in the SWY output that you're using in NDR, then you will have NoData in NDR. It will also affect the calculations in the model, because water is routed down the landscape until it reaches a stream, and if it encounters a NoData before it reaches a stream, the flow pattern will be incorrect and you'll get inaccurate results.

    NDR is an annual average model, so I would recommend using annual runoff, which is the QF.tif result from SWY. And you do not need to normalize the runoff proxy raster.

    ~ Stacie


Sign In or Register to comment.