SWMM-USERS Digest - 5 Jun 2008 to 6 Jun 2008 (#2008-101)

0 views

Skip to first unread message

Automatic digest processor

unread,

Jun 7, 2008, 12:00:32 AM6/7/08

to Recipients of SWMM-USERS digests

There are 13 messages totalling 907 lines in this issue.

Topics of the day:

1. Slope property of the subcatchements (8)
2. Preissman slot (4)
3. System storage

**********************************************************
* To sign off, email to: list...@listserv.uoguelph.ca *
* In the body of the message type: signoff swmm-users *
**********************************************************

----------------------------------------------------------------------

Date: Fri, 6 Jun 2008 06:25:00 -0300
From: Stephanie Piers de Raveschoot <stephan...@EPFL.CH>
Subject: Slope property of the subcatchements

Hello,

I was wondering: how can it be done if I have not been able to obtain the
slopes for every subcatchment in my real-case study ? Different slope values
will mean different peak flows, at different times, etc. Is there any way of
minimizing the impact of lack of correct slope values ?

Thank you,

Stephanie.

------------------------------

Date: Fri, 6 Jun 2008 05:56:23 -0400
From: Robert E Dickinson <robert.d...@GMAIL.COM>
Subject: Re: Slope property of the subcatchements

Stephanie,

You should contact Rob James and try to get the best estimate of the slope
in your subcatchments. It is a key hydrologic parameter and you should try
to expend further effort in getting a less uncertain estimate of your
slope. Why can't you get the slope for every subcatchment? At a minimum
you should be able to use the mean of the known subcatchment slopes and use
a sensitivity analysis to find out how sensitive your assumption of the mean
slope is in your model.

Regards,
Robert Dickinson

On Fri, Jun 6, 2008 at 5:25 AM, Stephanie Piers de Raveschoot
<stephan...@epfl.ch> wrote:

> Hello,
>
> I was wondering: how can it be done if I have not been able to obtain the
> slopes for every subcatchment in my real-case study ? Different slope
> values
> will mean different peak flows, at different times, etc. Is there any way
> of
> minimizing the impact of lack of correct slope values ?
>
> Thank you,
>
> Stephanie.
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************
>

--
Best and Kind Regards,
Robert Edgar Dickinson
MWH Soft Inc.

------------------------------

Date: Fri, 6 Jun 2008 08:03:34 -0400
From: Richard Landon Stanford <rl...@CMS.MAIL.VIRGINIA.EDU>
Subject: Re: Slope property of the subcatchements

Stephanie -

Are you (or rather, is your site) in Switzerland? If so - you could try to
find the slope using topographic maps or DEM files. Such places as
http://www.cartographic.com/xq/asp/switzerland/topographic/maps/navmode.R/rid.247/tid.1/sid.0/cid.0/vid.0/oid.0/qx/hub/index.asp
and others have them for sale, although the scale might not be large enough
for your purposes.

Good luck,
Rick

On Fri, 6 Jun 2008 06:25:00 -0300
Stephanie Piers de Raveschoot <stephan...@EPFL.CH> wrote:
> Hello,
>
> I was wondering: how can it be done if I have not been able to obtain the
> slopes for every subcatchment in my real-case study ? Different slope
>values
> will mean different peak flows, at different times, etc. Is there any way
>of
> minimizing the impact of lack of correct slope values ?
>
> Thank you,
>
> Stephanie.
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************

------------------------------

Date: Fri, 6 Jun 2008 14:25:05 +0200
From: Gabriele Pasteris <paster...@GMAIL.COM>
Subject: Re: Slope property of the subcatchements

Hi,

I'm in the same case of carence in catchment slope data. Any suggestions
will be very appreciated.

Best Regards.

G. Pasteris
gabriele...@a3.epfl.ch

On 06/06/2008, Robert E Dickinson <robert.d...@gmail.com> wrote:
>
> Stephanie,
>
> You should contact Rob James and try to get the best estimate of the slope
> in your subcatchments. It is a key hydrologic parameter and you should try
> to expend further effort in getting a less uncertain estimate of your
> slope. Why can't you get the slope for every subcatchment? At a minimum
> you should be able to use the mean of the known subcatchment slopes and use
> a sensitivity analysis to find out how sensitive your assumption of the
> mean
> slope is in your model.
>
> Regards,
> Robert Dickinson
>
> On Fri, Jun 6, 2008 at 5:25 AM, Stephanie Piers de Raveschoot
> <stephan...@epfl.ch> wrote:
>
> > Hello,
> >
> > I was wondering: how can it be done if I have not been able to obtain the
> > slopes for every subcatchment in my real-case study ? Different slope
> > values
> > will mean different peak flows, at different times, etc. Is there any way
> > of
> > minimizing the impact of lack of correct slope values ?
> >
> > Thank you,
> >
> > Stephanie.
> >
> > **********************************************************
> > * To sign off, email to: list...@listserv.uoguelph.ca *
> > * In the body of the message type: signoff swmm-users *
> > **********************************************************
> >
>
>
>
> --
> Best and Kind Regards,
> Robert Edgar Dickinson
> MWH Soft Inc.
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************
>

------------------------------

Date: Fri, 6 Jun 2008 07:26:44 -0500
From: "Gheith, Hazem" <HGh...@BRWNCALD.COM>
Subject: Re: Slope property of the subcatchements

Stephanie,

Same, if your issue is not having enough terrain elevation data, I used
the world wide Digital Terrain Elevation Data (DTED) few years ago to do
a surface water analysis in mountainous areas in Sinai, Egypt. The data
was prepared and on sale by the US Department of Defense (DoD) and comes
in different resolution, from 1 km x 1 km cells down to 30 m x 30 m (and
even less than that, but less resolution might not be available for
public use yet.) You can find it and purchase it through the internet.

Thank You
Hazem

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Richard Landon Stanford
Sent: Friday, June 06, 2008 8:04 AM
To: SWMM-...@listserv.uoguelph.ca
Subject: Re: [SWMM-USERS] Slope property of the subcatchements

Stephanie -

Are you (or rather, is your site) in Switzerland? If so - you could try
to find the slope using topographic maps or DEM files. Such places as
http://www.cartographic.com/xq/asp/switzerland/topographic/maps/navmode.
R/rid.247/tid.1/sid.0/cid.0/vid.0/oid.0/qx/hub/index.asp
and others have them for sale, although the scale might not be large
enough for your purposes.

Good luck,
Rick

On Fri, 6 Jun 2008 06:25:00 -0300
Stephanie Piers de Raveschoot <stephan...@EPFL.CH> wrote:
> Hello,
>
> I was wondering: how can it be done if I have not been able to obtain
>the slopes for every subcatchment in my real-case study ? Different
>slope values will mean different peak flows, at different times, etc.
>Is there any way of minimizing the impact of lack of correct slope
>values ?
>
> Thank you,
>
> Stephanie.
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************

------------------------------

Date: Fri, 6 Jun 2008 09:54:48 -0400
From: "Heineman, Mitchell" <Heine...@CDM.COM>
Subject: Re: Slope property of the subcatchements

Stephanie:

I often simply use the slope of the corresponding collection system as
the catchment slope. While this sounds like a gross simplification, it
is often not worth the additional effort of refinement.

SWMM effectively aggregates slope (S), Manning's N, and width (W) into a
single parameter based on the non-linear reservoir runoff equation. This
works out to K = W/N S^0.5. Thus a factor of four error in slope is
equivalent to a factor of two error in W or N. While we can strive to
measure N, we can only estimate W, and it is best identified through
calibration. It is unlikely that you know W within a factor of 2 a
priori, so the model is thus relatively insensitive to mild errors in S.

Even if we know the true S and N, we still need to adjust W to account
for the degree of aggregation in the catchment. W represents both the
actual surface width and the undiscretized portions of the collection
system network within the catchment (i.e. street gutters, omitted small
pipes, etc). Unless you have a highly discretized model where you
explicitly represent ALL channelized flow, the W parameter must be
adjusted downwards to account for those processes which are not directly
represented in the model. As W is thus inevitably a key calibration
parameter, and its impact is proportional with the square root of the
slope, I don't worry about limited precision of my slope measurements.

Regards,
Mitchell Heineman, P.E., BCEE
Principal Engineer | Collector Systems and Surface Water Subdiscipline
Leader
CDM
50 Hampshire Street
Cambridge MA 02139
www.dynsystem.com/netstorm/cdmuswmm.htm

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Stephanie Piers de Raveschoot
Sent: Friday, June 06, 2008 5:25 AM
To: SWMM-...@listserv.uoguelph.ca
Subject: [SWMM-USERS] Slope property of the subcatchements

Hello,

I was wondering: how can it be done if I have not been able to obtain
the slopes for every subcatchment in my real-case study ? Different
slope values will mean different peak flows, at different times, etc. Is
there any way of minimizing the impact of lack of correct slope values ?

Thank you,

Stephanie.

------------------------------

Date: Fri, 6 Jun 2008 16:22:29 +0200
From: Laurent Courty <lco...@CABINET-MERLIN.FR>
Subject: Re: Slope property of the subcatchements

Richard Landon Stanford a écrit :
> Stephanie -
>
> Are you (or rather, is your site) in Switzerland? If so - you could try to
> find the slope using topographic maps or DEM files. Such places as
> http://www.cartographic.com/xq/asp/switzerland/topographic/maps/navmode.R/rid.247/tid.1/sid.0/cid.0/vid.0/oid.0/qx/hub/index.asp
>
> and others have them for sale, although the scale might not be large enough
> for your purposes.
>
> Good luck,
> Rick
>
>
> On Fri, 6 Jun 2008 06:25:00 -0300
> Stephanie Piers de Raveschoot <stephan...@EPFL.CH> wrote:
>> Hello,
>>
>> I was wondering: how can it be done if I have not been able to obtain the
>> slopes for every subcatchment in my real-case study ? Different slope
>> values
>> will mean different peak flows, at different times, etc. Is there any way
>> of
>> minimizing the impact of lack of correct slope values ?
>>
>> Thank you,
>>
>> Stephanie.
>>

Hello,

Maybe the free SRTM data can do the job ?

http://www2.jpl.nasa.gov/srtm/

Regards,

--
Laurent Courty

------------------------------

Date: Fri, 6 Jun 2008 16:30:06 +0200
From: Gabriele Freni <fr...@IDRA.UNIPA.IT>
Subject: Preissman slot

Dear SWMMers,
I'm playing a little with SWMM5 and some experimental data we have
collected in laboratory studies about pipe pressurization.
Is there any simple way to change Preissman slot properties in SWMM5
(i.e. dimensions)?
I suppose that the answer should be "No, unless you modify the source
code and recompile the executable" so here is my second question:
Where exactly I have to look at for modifying such properties?
The idea is to evaluate "how bad/good is Preissman approximation in
case of rapid pipe surcharge".
Thank you for your always valuable help.
Regards,
Gabriele
________________________________
Gabriele Freni
Ph.D. in Hydraulic Engineering
Dipartimento di Ingegneria Idraulica ed Applicazioni Ambientali
Università di Palermo
Viale delle Scienze
90128 Palermo
tel: +39 0916657730
fax: +39 0916657749
E-mail: fr...@idra.unipa.it
Alt E-mail: gabrie...@gmail.com

------------------------------

Date: Fri, 6 Jun 2008 10:42:40 -0400
From: Robert E Dickinson <robert.d...@GMAIL.COM>
Subject: Re: Preissman slot

Hi Gabrielle,

SWMM 5 uses approximately the same surcharged algorithm as in SWMM 3&4 but
with some iteration changes. The features of the slot or transition zone i=
f
you prefer are calculated in dynwave.c and link.c and applied in the node
solution function of dynwave.c If you want to experiment with the
transitory shape and the denominator of the surcharged solution you need to
look at those two c section - if you search for dqdh and sumdqdh you will
find most of the code used in the surcharge algorith.

I will send you other information offline that may help you with your
search.

Regards,
Robert Dickinson

On Fri, Jun 6, 2008 at 10:30 AM, Gabriele Freni <fr...@idra.unipa.it> wrote=
:

> Dear SWMMers,
> I'm playing a little with SWMM5 and some experimental data we have
> collected in laboratory studies about pipe pressurization.
> Is there any simple way to change Preissman slot properties in SWMM5
> (i.e. dimensions)?
> I suppose that the answer should be "No, unless you modify the source
> code and recompile the executable" so here is my second question:
> Where exactly I have to look at for modifying such properties?
> The idea is to evaluate "how bad/good is Preissman approximation in
> case of rapid pipe surcharge".
> Thank you for your always valuable help.
> Regards,
> Gabriele
> ________________________________
> Gabriele Freni
> Ph.D. in Hydraulic Engineering
> Dipartimento di Ingegneria Idraulica ed Applicazioni Ambientali
> Universit=E0 di Palermo
> Viale delle Scienze
> 90128 Palermo
> tel: +39 0916657730
> fax: +39 0916657749
> E-mail: fr...@idra.unipa.it
> Alt E-mail: gabrie...@gmail.com
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************
>

--=20
Best and Kind Regards,
Robert Edgar Dickinson
MWH Soft Inc.

------------------------------

Date: Fri, 6 Jun 2008 10:51:44 -0400
From: William James <bi...@COMPUTATIONALHYDRAULICS.COM>
Subject: Re: Preissman slot

Gabriele and others,
You are probably aware of the slew of recent papers (e.g. by Steve Wright
and Karen Ridgeway, and their collaborators) on this and related topics in
the monographs of Proceedings of the annual stormwater and water quality
modelling conferences (formerly SWMM User-group meetings) held every 3rd
week of February in Toronto, and if so I humbly post the following link to
them:
http://www.computationalhydraulics.com/Publications/Books/index.html
However my main point is to urge you and other SWMMers to publish your
findings in those future proceedings. The topic is of great importance and
interest - I believe that Karen found that SWMM5 (unlike other more
expensive programs) can be made to reproduce certain rapidly varied flow
transients.
Thanks and sincerely,
William James
____________________________________________
Plan to attend the International Conference on Stormwater and Urban Water
Systems Modeling, Toronto, February 19-20, 2009.

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Gabriele Freni
Sent: Friday, June 06, 2008 10:30 AM
To: SWMM-...@listserv.uoguelph.ca
Subject: [SWMM-USERS] Preissman slot

Dear SWMMers,
I'm playing a little with SWMM5 and some experimental data we have collected
in laboratory studies about pipe pressurization.
Is there any simple way to change Preissman slot properties in SWMM5 (i.e.
dimensions)?
I suppose that the answer should be "No, unless you modify the source code
and recompile the executable" so here is my second question:
Where exactly I have to look at for modifying such properties?
The idea is to evaluate "how bad/good is Preissman approximation in case of
rapid pipe surcharge".
Thank you for your always valuable help.
Regards,
Gabriele
________________________________
Gabriele Freni
Ph.D. in Hydraulic Engineering
Dipartimento di Ingegneria Idraulica ed Applicazioni Ambientali Università
di Palermo Viale delle Scienze
90128 Palermo
tel: +39 0916657730
fax: +39 0916657749
E-mail: fr...@idra.unipa.it
Alt E-mail: gabrie...@gmail.com

------------------------------

Date: Fri, 6 Jun 2008 17:22:33 +0200
From: Gabriele Freni <fr...@IDRA.UNIPA.IT>
Subject: Re: Preissman slot

Thank you, Robert and William.
I'm always amazed of fast and accurate replies that we can receive
from the LIST.
I can preliminary confirm that SWMM5 results are really promising. I'm
only struggling with some stupid oscillations that take part in the
transition zone. They do not take any serious inconvenient (apart the
need to cut that few points from the graphs) but I wonder if some
modifications to the code can help.
I will keep you informed.
Best Regards,
Gabriele

________________________________
Gabriele Freni
Ph.D. in Hydraulic Engineering
Dipartimento di Ingegneria Idraulica ed Applicazioni Ambientali
Università di Palermo
Viale delle Scienze
90128 Palermo
tel: +39 0916657730
fax: +39 0916657749
E-mail: fr...@idra.unipa.it
Alt E-mail: gabrie...@gmail.com

On 6 Jun 2008, at 16:30, Gabriele Freni wrote:

> Dear SWMMers,
> I'm playing a little with SWMM5 and some experimental data we have
> collected in laboratory studies about pipe pressurization.
> Is there any simple way to change Preissman slot properties in SWMM5
> (i.e. dimensions)?
> I suppose that the answer should be "No, unless you modify the source
> code and recompile the executable" so here is my second question:
> Where exactly I have to look at for modifying such properties?
> The idea is to evaluate "how bad/good is Preissman approximation in
> case of rapid pipe surcharge".
> Thank you for your always valuable help.
> Regards,
> Gabriele
> ________________________________
> Gabriele Freni
> Ph.D. in Hydraulic Engineering
> Dipartimento di Ingegneria Idraulica ed Applicazioni Ambientali
> Università di Palermo
> Viale delle Scienze
> 90128 Palermo
> tel: +39 0916657730
> fax: +39 0916657749
> E-mail: fr...@idra.unipa.it
> Alt E-mail: gabrie...@gmail.com
>
> **********************************************************
> * To sign off, email to: list...@listserv.uoguelph.ca *
> * In the body of the message type: signoff swmm-users *
> **********************************************************
>

------------------------------

Date: Fri, 6 Jun 2008 11:50:09 -0400
From: Rob James <r...@COMPUTATIONALHYDRAULICS.COM>
Subject: Re: Slope property of the subcatchements

Thanks Mitch for a great post. We quite agree with you that W is the most
uncertain of the three parameters that make up WCON, and that its
calibration is important.

I'd like to make a minor point about adjusting Width downward to account for
attenuation from gutters and omitted small pipes. A smaller width will also
reduce the total volume of contribution from the pervious surfaces, as the
water spends longer sitting on an infiltrating surface. In reality, this
water is spending time on an impervious surface (gutter or pipe). Luckily
the time spent in the gutter and pipe is relatively short compared to the
time spent in surface flow, so Width may not need to be adjusted that much.

However if you are using the average maximum length of overland flow (AMLOF)
method of determining width (Width=Area/AMLOF), you might not want to
include the full distance to the outlet (if much of the flow path takes the
form of gutters/pipes) in drawing the flow path(s). If you do, there may be
a need to reduce infiltration to achieve reasonable runoff volume and
increase slope to reduce attenuation (celerity in the gutters and pipes is
considerably faster than that in sheet flow). The importance of this depends
on the amount of pervious area.

In a 100% rural arid watershed we had good luck with using just the average
maximum length of overland "sheet" flow (i.e. the distance from the
subcatchment boundary to the natural channels), even though the channels
themselves were not explicitly modeled. This model included a groundwater
component to achieve the long term response from rainfall.

More information about estimating slope can be found in this topic:

www.chi-kb.com/forums/thread-view.asp?tid=3118

Sincerely,

Rob James
Computational Hydraulics Int.
www.computationalhydraulics.com
Tel. 519-767-0197

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Heineman, Mitchell
Sent: Friday, June 06, 2008 9:55 AM
To: SWMM-...@listserv.uoguelph.ca
Subject: Re: [SWMM-USERS] Slope property of the subcatchements

Stephanie:

I often simply use the slope of the corresponding collection system as
the catchment slope. While this sounds like a gross simplification, it
is often not worth the additional effort of refinement.

Regards,
Mitchell Heineman, P.E., BCEE
Principal Engineer | Collector Systems and Surface Water Subdiscipline
Leader
CDM
50 Hampshire Street
Cambridge MA 02139
www.dynsystem.com/netstorm/cdmuswmm.htm

Hello,

I was wondering: how can it be done if I have not been able to obtain
the slopes for every subcatchment in my real-case study ? Different
slope values will mean different peak flows, at different times, etc. Is
there any way of minimizing the impact of lack of correct slope values ?

Thank you,

Stephanie.

------------------------------

Date: Fri, 6 Jun 2008 16:58:23 -0400
From: Rob James <r...@COMPUTATIONALHYDRAULICS.COM>
Subject: Re: System storage

Thanks Bob for the correction - the node minimum surface area is added to
the allocated link surface area rather than replacing it (for change in head
calculations in non-surcharged junctions).

For the sake of completeness, one final interesting but perhaps academic
point. Non-ponded junctions do not add to the reported system storage. In
other words, this minimum surface area does not contribute to the volume
reported in the continuity calculations or the output file's System Storage
time series.

Even though it is not reported, flow is attenuated through non-surcharged
junctions due to this minimum surface area. So the statement "A non-ponding
junction between two 10 ft deep lengths of open channel does not add 125
cubic feet (12.566ft2 x 10ft) of storage capacity to the system" is true as
far as reporting goes, but the system will behave as if there was this
storage capacity (i.e. be dampened).

I gather this is not significant, but you can take from this that adding
junctions along a channel or pipe will change (lower) the local HGL
somewhat. In an urban setting, a non-surcharged junction representing a
fully benched manhole may perhaps add more attenuation than exists in
reality (although in a typical urban system, this default minimum surface
area at the nodes is probably < 2% of the total surface area). In a natural
channel, adding junctions along a reach in order to get the head detail will
also have this effect, although typically the default minimum surface area
is usually relatively small compared to the surface area in the adjoining
natural channels.

Sincerely,

Rob James
Computational Hydraulics Int.
www.computationalhydraulics.com
Tel. 519-767-0197

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Robert E Dickinson
Sent: Thursday, June 05, 2008 6:53 AM
To: SWMM-...@listserv.uoguelph.ca
Subject: Re: [SWMM-USERS] System storage

Hi,

I hope I can clarify this issue with a small example of the internal
workings of SWMM 5:

Step 1: Calculate the node surface area at the beginning of an iteration:

If the node is a storage node then the storage area for the
depth is used as newArea in the program;
If the node is a non storage area then the area is 0.0

before Xnode[i].newSurfArea 0.000000

Step 2: Compare newArea to the minimum user defined area of a node

If the value of newArea is less than the minimum nodal area (in
this case 10 feet square),
The value of newArea is set to the default node area

after Xnode[i].newSurfArea 10.000000

Step 3: Add in the node surface area contribution from the connecting
links

In this case the links are dry so the link contribution is only
0.09 square feet

area in node continuity 10.090000

The node area in this one iteration increased from 0 to 10 to 10.09 square
feet in on iteration. All nodes have a minimum surface area as defined by
the user.

Best and Kind Regards,
Robert Edgar Dickinson
MWH Soft Inc.

On Wed, Jun 4, 2008 at 4:11 PM, Shyamprasad, Sangameswaran
<Shyamp...@cdm.com> wrote:
Rishab,
Every node in the network by default has a storage area of 12.56 sq feet
and that is the reason you have a value for total storage volume in the
system.
-Shyam

CDM
Cambridge, MA 02139
phone/fax: (617) 452-6139

-----Original Message-----
From: SWMM-USERS [mailto:SWMM-...@listserv.uoguelph.ca] On Behalf Of
Rishab Mahajan
Sent: Wednesday, June 04, 2008 3:13 PM
To: SWMM-...@listserv.uoguelph.ca
Subject: [SWMM-USERS] System storage

I had a query regarding the system storage. It is sum of the total nodal
storage volume. But even if the network does not have any stoarge node
in it, it shows value, which is quite surprising.

Thanks

------------------------------

End of SWMM-USERS Digest - 5 Jun 2008 to 6 Jun 2008 (#2008-101)
***************************************************************

Reply all

Reply to author

Forward

0 new messages