SWMM 5 subcatchment too small??? 1

[hansbenford]

I am running a model with a year of rain data. It rained 55.74 inches in that year.

My subcatchment is 0.468 acres of 100% impervious area. For whatever reason, the rain turns out to be 55.621 inches which i am assuming comes from the change that SWMM does from volume input to intensity which is what it uses for calcs.

No big deal there. The runoff is in the 55 inch range as well as expected.

It travels to a storage unit which is my "infiltration trench". I have two outlets from the trench: 1) an infiltration rating curve starting from the bottom of the trench. 2)a rating curve at the overflow level.
If i sum the total flows from these two, they come out right around 55 inches over the watershed.

So far so good....

When I change the size of the subctachment to more appropriately fit the trench, the total of the two flows decreases dramatically. I am decreasing the area of the subcatchment a lot (0.0213 acres).

In both cases I am converting the flows from CFS to inches over the respective watershed.

For the smaller area I am still seeing 55.621 inches of rain running off, but the sum of the infiltration and overflow at the trench turns out to be 46.18 inches.

I have no explanation and have tried to find the cause to no avail. Am I at some threshold in SWMM? Is it not capable of modeling such a small area?

 

[tsgrue]

.

I've used SWMM5 for areas smaller than that. The size shouldn't be a problem.

What are your time steps and your simulation surface runoff and flow routing percent errors?

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[hansbenford]

Runoff Quantity continuity error of -0.21%

No Flow Routing continuity error

Routing every 5 seconds

Wet weather every 5 minutes

Reporting every 15 minutes

Then I sum the 15 minute data.

The inflow to the trench is where the runoff volume of 55.621 drops to 46.@#$.

 

[tsgrue]

.

I expected higher continuity errors, especially with those time steps. I suggest the following:

- run a single storm event at time intervals of 2 seconds and see what results you get
- run 2 weeks of wet weather at time intervals of 2 seconds and see what results you get
- post the inp fule and I'll look it over

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[RWF7437]

Suggested reading:

https://www.forester.net/sw_0607_voodoo.html

 

[hansbenford]

Alright.

So ran a storm and a couple weeks of rain data using 2 second intervals and not seeing nay improvement.

I look at the precipitation and runoff in the status report and it matches with the rain.

Then when i take the total flow into the trench (cfs) and sum those flows and convert them to inches over the watershed, I am still losing significant volume.

Here are the inputs for the two runs.

Thanks for any insight.

 

[hansbenford]

couple weeks

 

[tsgrue]

.

You need to post the precipitation data as well.

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[tsgrue]

.

I also noted that your infiltration rating curve had a 'blip' in it (outflow decreasing at 4.7ft to 4.8ft head increase)...

ft cfs
4.6 0.0036476
4.7 0.0038595
4.8 0.0037601
4.9 0.0040821

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[hansbenford]

Storm Data

 

[hansbenford]

Couple weeks

 

[hansbenford]

As far as that curve goes, it was something I changed but must not have saved. I just erased points 4.7, 4.8 and 4.9.

I don't think that will solve all of my problems........... but who knows.

 

[tsgrue]

.

Can you post the prec data?

.



tsgrue: site engineering, stormwater
management, landscape design, ecosystem
rehabilitation, mathematical simulation

http://hhwq.blogspot.com

 

[hansbenford]

The rain data is posted above. The indivdual storm and couple weeks of wet weather. The entire year of data is not posted.

 

[ForensicMeteorology]

hansbenford:

I looked over the rainfall data you are using. Is this point rainfall/basin average or other? Time steps seem to be 15 min?

If you are looking for individual storm event rainfall with smaller time steps try this site:

http://www.ncdc.noaa.gov/nexradinv/

I can provide you further guidance if you find it may be useful.