Phosphorus overestimation

[Sepideh Kheirkhah]

Hi everyone,

Does anyone know what could cause unreasonably high Phosphorus load result? I am using SWAT model for a large watershed with potholes (applied with fraction) which is calibrated satisfactorily for streamflow. The sediment load in output file is also close to the observations. However, TP load (kg) is about 1000-10000 times higher than measured values across the basin. 

I analyzed sensitivity of the results to parameters related to P processes and routing, but the simulated changes were negligible in that level of high P load predictions. Does anyone have an idea of what might lead to this overestimation? Or which files in my txtInOut should I check for P input or initialization in soil or river? Thanks.

Regards,

[Natalja C.]

Hello,

Which form of P is giving you trouble? If you say that the sediments all reasonable, then I suspect it will be the soluble P. And this can come primarily from point sources or unrealistically parametrized reservoirs. My troubleshooting algorithm would be then:

- Determine the P form.

- Determine the P source (which basin or reach does the issue originate).

- If soluble P - then look at potential solP contributors, identify the one and correct the issue.

- if it is particulate P - then look at the sediment transport, soil and bank/bed erosion, settling parameters. 

- last, but not least, groundwater and lateral flow contributions. There are GW_P and LAT_P (or something along those lines) parameters, which should be 0 by default, but maybe you have different unrealistic values.

Hope this helps!

Best,

Natalja

[Ashwin Ghimire]

Hi,

Did you figure it out? I am having the same issue as yours. 1000 to 100000 times higher.