Clarification on Using Field Soil Nutrient Data and NoData Outputs in NDR

[Aatreyee]

Dear Community,

I am currently working with the Nutrient Delivery Ratio (NDR) model and have the following queries:

1. I have access to field-level soil data that includes nitrogen and phosphorus concentrations at both surface and subsurface levels. Can these measurements be directly used as nutrient load inputs in the NDR model? If so, how should they be integrated or processed to fit the model requirements?
2. While running the model using global datasets over a region characterized by rugged terrain, I observed that the model correctly returns NoData values along the stream channels. However, it is also returning NoData values in areas adjacent to the streams, which I expected to have valid outputs. What could be the possible reasons for this behavior?
   I am attaching an image below to help illustrate the issue more clearly.

Any insights or suggestions would be greatly appreciated.

[Aatreyee]

the NoData values appear to follow a spatial pattern—they are predominantly concentrated on north-facing slopes. Could this indicate a topography-related issue in model behavior or input sensitivity?

[megannissel]

Hi @Aatreyee,

Are you running NDR with the D8 Flow Direction Algorithm? There’s a known issue with NoData values when using D8; this could be the cause of the behavior you’re seeing. This bug should be fixed in the upcoming 3.16.1 release in a couple of weeks!

[jesseG]

Although, this looks like MFD since D8 streams are only 1 pixel wide. @Aatreyee , are there any holes in any of your input layers in those same locations? If so, that may lead to the NoData values in the results.

-Jesse

[Aatreyee]

(post deleted by author)

[Aatreyee]

Hello Jesse,
I’m using the D8 algorithm, and none of my inputs have holes, i have checked it

[Aatreyee]

And can I get some insights for my first question also?