Nutrient Delivery Ratio Model - Surface/Subsurface Retention


Can you please help me with this... 

I am running the NDR model for nitrogen retention in Denmark, which is basically flat. An hydrologist here told me that for that reason almost 100% of all the nitrogen export happens through subsurface flow. On top of that, the soil here has different layers that have different effects on retention (due to oxigen availability). That means each area has a different subsurface retention, independent of LULC, which would be responsible for the retention only until the root zone. 

Ok, that said, the data I had access here is a subsurface retention map for 15km2 basins for the whole country. Seems like a very good work. I also have access to data on leaching through the root zone for different crops. So I have a retention map for subsurface flow that does not relate to LULC, no data for surface flow and data for what gets to the root zone for each crop.

But the problem is that there is no way to differentiate the subsurface retention in the model, there is only one overall retention value for it. And the NDR equations are different for the different flows, so I can't just use the subsurface as the surface and set the subsurface proportion to 0. 

Any ideas? 

The only solution I can think of is to use the leaching until the root zone as the load and then the subsurface retention as the surface retention? But would that work?



  • DougDoug Administrator, NatCap Staff
    Hey Raphael,

    Thanks for making a new post for this question, I believe I saw it at the end of a previous thread. Perhaps @Perrine or @Adrian can provide some guidance!

    Thanks for your patience
  • Hey Doug,

    I thought it would be better to have a new post. I am sorry if I broke some forum rule. Thanks for your help. 

  • PerrinePerrine Moderator, NatCap Staff
    Dear Raphael, 

    I think your idea is good, and it means you don't need the InVEST NDR model for these analyses. As you mention, if you already have the soil retention map, you essentially know the service provided by the environment . Based on your post, vegetation doesn't not seem to provide additional retention, so the most straightforward way to compute N concentration in the stream would be to multiply each pixel's amount of leaching by your soil retention in this area.

    Note that this model is still under development so any suggestion for improvements are welcome.



  • Dear Perrine,

    Thank you for your answer. I am putting more thought on how I am going to do this... If I just multiply each pixel won't I loose the effect of the accumulated nutrient that is not retained at each pixel and travels downstream?

    Do you see the model having the subsurface retention in the biophysical table (for each LULC) in a later version? Do you plan on releasing any updates in the next 2 months?

  • PerrinePerrine Moderator, NatCap Staff
    Hi Raphael,

    Sorry for the slow reply. Given your response, I think I misinterpreted what your "soil retention map" was representing. It doesn't provide directly the proportion of nutrient load that is retained by the soil, does it?

    If you need to use a LULC-dependent retention coefficient, could you use the current model with the assumption that the "eff_n" coefficients represent subsurface retention only? (And if you need to use both surface and subsurface models, you could use the model twice, and just sum up the two outputs to get total export).

    Re: next versions of the model, we are about to release a new version with a minor change in the way we compute loads.  We are not expecting to make further changes, unless we realize there is a high demand for some specific functionalities. 



Sign In or Register to comment.