Difference in sediment load estimation with two SWAT versions

[Jeewanthi Sirisena]

Dear All, 

I have checked the sediment load (SED_OUT from .rch file) at my basin outlet for simulations done by two SWAT versions: Version 635 and 670. I found significant difference as shown in attached figure. Does anybody come cross with this kind of differences or any idea on reasons behind this? 

Note: These simulations results are based on Default SWAT parameter.

Thanks. 

Jeewa

[Natalja C.]

Hello,

There are two explanations:

1. The output.rch file changed in rev670, so the column for sediment output is different than of the rev635 - are you absolutely sure you are comparing the same thing?

2. It is expected that there might be different outputs between the versions. To help resolve this I usually turn to the revision history. Now, however, it was not updated yet, so I cannot say what has changed or how, to identify the source of differences in the outputs.

Best,

Natalja

[Jeewanthi Sirisena]

Hi Natalija, 

Thanks for the reply. 

I have checked the results of two versions again. I'm checking the SED_OUT (tons) column in output.rch (Column 11). In both versions, Column 11 is SED_OUTtons. Please correct if I'm wrong. 

I understand that there might be difference in outputs between two versions. But changes in order of magnitude is the problem. 

Thanks. 

Jeewa

[Natalja C.]

I just checked my outputs (difference between rev. 664 and 670), the difference does not exceed 16% (yearly average loads).

In my case the columns for SED_OUT start from space 100 (rev. 664) and space 101 (rev 670). Not sure if the difference in total output can be that large (like I mentioned, I have no idea what procedures/formulas changed between the versions). And are you sure that you have hundreds of "Million tons" of sediment load? That seems to be a bit much :)

Best,

Natalja

[Jeewanthi Sirisena]

Hi, 

Thanks for your reply. As per your calculation, there is a little difference in Rev.664 and 670, and this might be significant difference in Rev.635 and 670 (or 664). 

In terms of the numbers (Hundreds of million tons), those are possible. Not much. 

Jeewa  

[Jeewanthi Sirisena]

Hi Natalja, 

I'm back again for discussing the above subject (Difference in sediment load estimation with two SWAT versions)

I found that very low sediment simulation (Big difference in two versions) is occurred only in the Main Canal (River). All the tributaries have almost similar results for two versions. 

Also I found document mentioned about the SWAT Revisions (Attached herewith). Could you please guide me on this regard?

Thanks. 

Jeewa