[APEX1501] .DMR vs .SAD outputs

[Maya Struzak]

Hello,

I was wondering if anyone had more information on the difference between the .DMR and .SAD file outputs. I am modeling a single-subarea watershed (no routing) in APEX v1501 and have found evidence of discrepancies between .DMR and .SAD files. For instance, the WYLD reported in .DMR is 26mm on day 1, while .SAD reports ~0.5mm. Additionally, when trying to convert from ppm in DMR to kg/ha for the various nutrient and sediment yields under the assumption that ppm=mg/L, I also got different loads from what is reported in the .SAD file.

The .SAD-reported WYLD seems to agree with other output files (.DHY), and all files allegedly report WYLD in the same units (mm), according to the v1501 user manual. I couldn't find any information in the user manual nor the theoretical documentation on alternate calculation methods used DMR, either.

I would appreciate any insight into what might be driving this discrepancy and if .DMR is a reliable source for nutrient and sediment loading. 

Thanks!

Maya

[Luca Doro]

Dear Maya,

Thank you for pointing this out.

It seems that the DMR file was left behind after updating some units change in the source code. Specifically:

• WYLD calculation was correct but reporting the values in m3 instead of mm. I modified the header accordingly.
• Calculation for the remaining variables were off by an order of 10. I have modified the code correcting the conversion.

From my test, it seems that everything is working fine now. Here is a link to download a new executable (hopefully, this fully version is compatible with your input files), give it a try and let me know if everything looks good.

Luca

[Maya Struzak]

Hi Luca,

Thanks for getting back to me! I was able to try your executable, and the WYLD values from the .SAD and .DMR file now align. However, the soluble and sediment-bound N and P from .DMR and .SAD still do not. I used the following equations to convert from the .DMR concentrations to loads for my 5ha watershed, assuming ppm = mg/L:

Q[N or P] kg/ha = Q[N or P]ppm * WYLDm^3 * 10^(-3) / area in ha

Y t/ha = Yppm * WYLDm^3 * 10^(-6) / area in ha

Y[N or P] kg/ha = Y[N or P]ppm * Y t/ha * 10^(-3)

The sum of loadings calculated from the .DMR over the entire simulation relative to those reported in the .SAD file (as well as .DWS) are as follows:

Y (t/ha): DMR=547.6, SAD=547.6 

QN (kg/ha): DMR =1090.24, SAD=35.45

QP (kg/ha): DMR=4.43, SAD=4.43

YN (kg/ha): DMR=57.35, SAD=987.81

YP (kg/ha): DMR=10.31, SAD=175.7

I'm not sure if these inconsistencies are coming from a post-processing error of the .DMR values on my end, or if there might be something else I'm missing. I've also attached the corresponding .SAD and .DMR files if you'd like to take a look. 

Thanks! 
Maya

[Luca Doro]

Dear Maya,

I reviewed the code and the problem is that those outputs are not directly comparable. To give you an example, while QN in the SAD file indicates the fraction of mineral nitrogen transported with surface runoff, QN in the DMR output file is calculated differently. In fact, it is the ratio between the sum of mineral nitrogen lost with different flows and the sum of the water flows.

I feel that either the header is misleading or the calculation for the values in the DMR file should be done differently. I will discuss with the development team.

Let me know if you need more information about this.

Luca