Infiltration too low and runoff too high

[Stalled Mower]

All,

Please see attached plots of modelled and observed flows for the reach
I am trying to calibrate my model and I am getting far to little baseflow and too much runoff.
So far I have tried the following:
1. Reducing CN2/CNOP
2. Changing CNOP, PET Method, CNCOEF
3. Increasing SOL_K values
4. Adjusting ALL the GW parameters.

I've tested all these parameters to extremes with little or no effect. I simply can't find a way to increase infiltration and I can't see what is limiting it.
I think there is something not right with SW as it doesn't seem to fluctuate properly?

Any suggestions would be highly appreciated.

Phil

P.S. Water Balance is as follows:
AVE ANNUAL BASIN VALUES

PRECIP: 718.1
SURFACE RUNOFF Q: 126.14
LATERAL SOIL Q: 51.17
TILE Q: 61.99
GROUNDWATER (SHAL AQ) Q: 16.68
REVAP (SHAL AQ => SOIL/PLANTS): 0.49
DEEP AQ RECHARGE: 0.91
TOTAL AQ RECHARGE: 18.15
TOTAL WATER YLD: 254.3
PERCOLATION OUT OF SOIL: 16.84
ET: 401.3
PET: 491.6
TRANSMISSION LOSSES: 1.69

[Jim Almendinger]

This is dangerous, because I have not personally used the tile drainage routine in SWAT --

But at least one colleague has told me that with tile drainage, the lateral flow component was large and not really controllable.  In the end he concluded that the sum of tile Q + lateral Q was the "effective" flow due to tile drainage.  He didn't say that the groundwater and baseflow component was consequently limited -- but perhaps it was small in his watershed to begin with, in contrast to yours.  

I wonder, if you removed tile drainage from your watershed, would your model then be responsive to changing the groundwater parameters?  I realize this may not solve your problem -- but it might help us better understand what the model is doing.  Tile drainage is VERY important and we'd better be getting it about right.  I do know there are studies that have concluded that tile drainage has increased baseflow, so I presume there should be ways in the model to slow the delivery of water, via tiles, to the river such that it becomes part of the baseflow.  A subsurface tile that receives groundwater discharge is functioning like a new stream channel, and additional tile lines simply increase the channel network density.  They should contribute to baseflow like any channel, but the increased density should increase the baseflow recession rate, i.e., the alpha_bf.  At least, that would be my working hypothesis.  

There must be others who have used the tile drain routines in SWAT who have experience and opinions on the matter.  It would be great to get your advice on these things.  

Cheers,

-- Jim

---

From: "Stalled Mower" <sel...@yahoo.com>
To: swat...@googlegroups.com
Sent: Thursday, May 10, 2012 11:50:06 AM
Subject: [SWAT-user:3361] Infiltration too low and runoff too high

--
You received this message because you are subscribed to the Google Groups "SWAT-user" group.
To view this discussion on the web visit https://groups.google.com/d/msg/swatuser/-/cxbbmun8bkYJ.
To post to this group, send email to swat...@googlegroups.com.
To unsubscribe from this group, send email to swatuser+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/swatuser?hl=en.

---

Spam
Not spam
Forget previous vote

--

Dr. James E. Almendinger
St. Croix Watershed Research Station
Science Museum of Minnesota
16910 152nd St N
Marine on St. Croix, MN  55047
tel: 651-433-5953 ext 19

[Dave]

What version of SWAT are/were you using? And where is your watershed?

My working hypothesis is that some change to the SWAT algorithm(s) for draining/percolating/removing water from the soil profile will yield more realistic baseflow and runoff values in your watershed. Despite the presence of tile drains in your model watershed, water is building up in the soil profile, which increases the daily CN values and offsets the effects of your changes to the CN2/CNOP and other calibration parameters. I'm trying to understand the percolation routines myself. There seems to be some discrepancies I haven't resolved yet between the theoretical documentation, the Fortran code, and descriptions in published articles regarding vertical water flow between soil layers. (See my recent post re: Soil Water (Percolation and upward flow)).

Dave

[Timo Brussee]

Where is your watershed? And more importantly, how is the OBSERVED Q derived. With a Q-h relation equation?? Do you trust this Q-h equation? Because Q-h equations are generally not valid for a longer period than 5 years, sedimentation and erosion change channelshapes, anyway this could cause the Observed discharge to start at a higher baseflow than actually is the case.

Best,

Timo

Op donderdag 10 mei 2012 18:50:06 UTC+2 schreef Philip Selby:

[Jim Almendinger]

Reducing CNs by 10-25% has generally worked for me.  Occasionally I've reduced them even more (up to 30-35%) when I had reason to doubt the soils data, but that's beyond my usual comfort level. (And don't forget to re-write the changes to the txtinout folder, so SWAT sees the changes...).

You might try things without tile drainage to get a feel for the possible effects of changing CNs, just in case tile Q is complicating the issue, although I don't think it should. 

-- Jim

---

--
You received this message because you are subscribed to the Google Groups "SWAT-user" group.

To unsubscribe from this group and stop receiving emails from it, send an email to swatuser+u...@googlegroups.com.

To post to this group, send email to swat...@googlegroups.com.

Visit this group at https://groups.google.com/group/swatuser.
For more options, visit https://groups.google.com/d/optout.

---

Spam
Not spam
Forget previous vote