How to increase total flow/surface runoff during calibration?

[Rosie Records, CA Cooperative Fish & Wildlife Research Unit]

Hi all,

I am doing manual daily calibration for surface runoff. My simulated
surface runoff and my total flow are *very* low compared to my
measured data (say 10-20 times too low). I am increasing CN,
decreasing SOL_AWC, and increasing CH_K(2) based on seepage studies
showing groundwater flow to streams in this area. I have also tried
changing the ET method since I saw Dr. Srinivasan recommended this.
All of this slightly increases the flow but not by much.

The precipitation and the ET look about right in the output.std file.

I would really appreciate other SWAT expert's advice on how to boost
my flow levels.

Thank you, all!!!

Rosie

[Jim Almendinger]

Rosie --
I'd do some back-of-envelope calculations to help diagnose the problem.  You should be able to get a fairly good water balance from the model  if it is fully "warmed up" -- namely, on an annual average basis, does P - ET = water yield (otherwise called "mean annual runoff", or "runoff depth")?  (Here, annual water yield = total annual flow volume divided by the whole basin area.)  And, check your monitoring data -- how does your stream's annual water yield compare to the regional water yield (P - ET)?  If if the stream's water yield is much larger, then it indicates that your stream may be capturing groundwater flow from a larger area than its surficial watershed.  Or, could you be getting meltwater from retreating snow or ice fields (if this is at altitude, at your Crater Lake site)?

If P and ET are fairly close to known values, and yet P - ET still greatly exceeds your water yield, then you're losing water someplace in the model.  Either it is filling up some huge storage component in your model that is not yet at equilibrium (like, a huge reservoir where PVOL >> initial volume), or you're still filling up the shallow aquifer (GW Qmin >> 0).  Or, if you've got Ponds, Wetlands, or Reservoirs with K > 0, then you're losing water to seepage, which is lost in the current releases of SWAT.  And, I presume that you're not allowing any loss to deep seepage. 

Increasing CN is the right first step to try -- but I don't think you can make up a 10-20x difference in flow that way.  Something seems fundamentally out of balance here.   

I bet you've already covered most of these possibilities... 

-- Jim

---

From: "Rosie Records, CA Cooperative Fish & Wildlife Research Unit" <cac...@gmail.com>
To: "ArcSWAT" <arc...@googlegroups.com>
Sent: Tuesday, November 30, 2010 2:11:13 PM
Subject: [ArcSWAT:2783] How to increase total flow/surface runoff during calibration?

--
You received this message because you are subscribed to the Google Groups "ArcSWAT" group.
To post to this group, send email to arc...@googlegroups.com.
To unsubscribe from this group, send email to arcswat+u...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/arcswat?hl=en.



--
BEGIN-ANTISPAM-VOTING-LINKS
------------------------------------------------------

NOTE: This message was trained as non-spam.  If this is wrong,
please correct the training as soon as possible.

Teach CanIt if this mail (ID 01DBwbrGl) is spam:
Spam:        https://canit.smm.org/canit/b.php?i=01DBwbrGl&m=73068b5aaa06&t=20101130&c=s
Not spam:    https://canit.smm.org/canit/b.php?i=01DBwbrGl&m=73068b5aaa06&t=20101130&c=n
Forget vote: https://canit.smm.org/canit/b.php?i=01DBwbrGl&m=73068b5aaa06&t=20101130&c=f
------------------------------------------------------
END-ANTISPAM-VOTING-LINKS



--
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

[Felix]

Hi Rosie,
One mistake I once had
-make shure that you didn't check the "print log flow" (or similar),
because then your runoff seems to be very low just because it is
printed logarithmic.. :-)

Felix


On 30 Nov., 21:11, "Rosie Records, CA Cooperative Fish & Wildlife

[Rosie Records, CA Cooperative Fish & Wildlife Research Unit]

Felix:

thank you so much!!! I did indeed have the "print log flow" box
checked (it's amazing how something so small can mess with your
results so badly). Now my predicted flows are too high...back to
calibration! Jim: thank you for the water budget directions. My
problem has changed, but I'm sure the advice will still be very
useful.

Rosie

[congratulations]

Hi

I have exactly the same problem with you, observed flow rate values
are 10 times greater than the predicted values.
However, I noticed that print log flow box is not checked, what do you
think?


On 1 Aralık 2010, 19:57, "Rosie Records, CA Cooperative Fish &

[Rosie Records, CA Cooperative Fish & Wildlife Research Unit]

Hi,

have you tried searching "Increase flow" on this discussion group? Is
your flow 10 times higher across the board, or is it just for peak
flows?

Jim's recommendations for how to check your water balance are very
useful. In general, my experience is that the PET calculation method
can have a sizeable effect on your water yield...I would start there,
and also see if changing your routing method gives you better
results. If snowmelt is important in your area, changing SFTMP,
SMTMP, or other snowmelt parameters could also help. I found the
calibration section of one of Jim's papers very useful to refer to
when understanding how to adjust total flows:
http://www.smm.org/static/scwrs/tapwaters_sunrise.pdf

Decreasing CN2 while increasing SOL_AWC can reduce your peak flows
especially, while decreasing ESCO allows the model to evaporate from
deeper in the soil profile and seems to reduce total flow. Depending
on what the groundwater hydrology is of the region you're working in,
increasing RCHRG_DP will also decrease your total flow; I've found
that increasing REVAPMN seems to do the same thing, but am still
trying to understand how the subsurface flow parameters work together
so have been using them cautiously...Willem might be able to assist
you here.

Good luck!

Rosie

> > > > I am doing manual dailycalibrationfor surface runoff.  My simulated
> > > > surface runoff and mytotalflow are *very* low compared to my

[Akash kp]

hi mam i am doing project in arc swat, my arc swat run was successful, in output simulated values are higher than gauge recorded value i want to calibrate runoff so plese if possible send me any procedure pdf..thank u.

[ammara talib]

Hi all,

My simulated flow is extremely low compared to observed. So I followed up on Jim suggestion. Prec-ET is not equal to water yield or runoff (SURQ). 

 The difference between "water yield or runoff (SURQ)" and "Prec-ET" is from -160 mm to 115 mm. So for some years runoff is more and for some years Prec-ET is higher than runoff. Any suggestions? I don't think tweaking parameters will help to increase the simulated flow.  

[Sudhir Singh]

Hello Rosei,

I have the same problem as you faced. My observed discharge values are very much as compared to simulated flow out. Please suggest me the appropriate procedure.

Regards

Sudhir