We use cookies to make your experience of our website better. You can delete or block cookies, but some parts of this website won't work without them. By using this website you accept our use of cookies.

Reservoir flows and small timesteps - unrealistic results

Posted 05/03/2019 17:51 by pcross2

Last post 06/03/2019 11:12

5 replies

I have a 1D model that has a series of reservoirs at its upstream extent. I wish to route my inflows through these reservoirs to assess their attenuation, however I am getting an odd attenuated /stepped profile from the resrvoir (stage and flow) when run with timesteps of 2 seconds or below.

I've tried several variations of the model considering:

  • reservoir type (reservoir / online pond)
  • outflow structure (weir, spill, orifice)
  • inflow connection (hydrograph directly connected, connection via routing reach, connection via real channel)
  • changes to the geometry profile within the sotrage area relationship

None of the above make much differerence and the model gives the strange stepped profile for small timesteps (the largest and smoothest profile comes from a 4 second timestep). I've also rerun a different routing model that was previously run with a 900 second timestep at a much smaller timestep and this has resulted in the model giving very unrealistic results.

The obvious answer here is to run the model at a larger timestep, however I wish to link it to TUFLOW and will probably require a 0.5 or 1 second timestep to do so. I'm also wondering if there is a general limit on timestep with respect to reservoir units.

Thanks in advance



Error E1721 - Parse error in rule

Posted 08/02/2019 15:55 by Siripen Songprasit

Last post 11/02/2019 11:10

2 replies

Hi, I'm trying to fix the above error. I've checked everything and the node names and the letter case are correct, but still can't find what's wrong with the rule. 

The 'Open' rule is 

IF (LEVEL(STAN1_1105u).GT.LEVEL(STAN1_1105d))

Details of the error are below.

Datafile line 2119:
*** error E1721 *** at label: STAN1_1105u

Parse error in rule Open See above for details.

head/flow operators must have corresponding
bracket characters surrounding hydraulic label
error in subroutine parser

Any help or advice would be much appreciated.



Don’t change default top slot parameters for CIRCULAR, SPRUNG, FULLARCH CONDUITS (v4.4 and earlier)

Posted 05/11/2018 09:40 by Konrad

No replies yet


It has recently come to our attention that there is a bug which incorrectly calculates the free surface width within a conduit top slot for curved-topped, i.e. CIRCULAR, SPRUNGARCH and FULLARCH conduits. (unless the default values are used). This in turn can lead to mass balance issues, and sharp changes in flow along the length of such culverts.

The advice is therefore, when using the top slot option for CIRCULAR, SPRUNGARCH and FULLARCH conduits, to leave the height values as default – to do this, simply leave the Top Slot Height and Top slot dh value fields blank (or equivalently, set to zero), in either/both the network (.dat) file and the simulation (.ief) file.

There may also be an issue with calculating the transition depths in the arched (sprung and full) conduits, which I'm still looking into. In the meantime, do check mass continuity specifically at these conduit types.

This will be fixed in v4.5



Energy Junction - Warning Message W2065

Posted 29/01/2018 16:30 by Jennifer

Last post 06/02/2018 10:38

1 reply

Hi there,

I'm having some trouble getting my model to run and am hoping someone can help. The steady state run fails with the following message: 

Model time 1.0000 hrs:
*** warning W2065 *** at label: 14.5_DS_A

bad energy junction configuration at 14.5_DS_A
check junction is attached to channel sections everywhere


My model includes a culvert crossing which has two different sized pipes located at different invert levels (approx. 0.5m height difference). I've tried many set-ups  (including using an open junction) and different ways of connecting up the junction, but I still can't get the model to run. The error message seems to suggest it should be only river sections that are connected to the junction? I currently have an energy juction which is connected to an upstream river section, two culvert inlet nodes, and a spill.

Any suggestions anyone has would be greatly appreaciated!



1D Unsteady Model Crashes when Bridge is added

Posted 16/01/2018 16:28 by Niamh O'Malley

Last post 17/01/2018 12:12

3 replies

I have built a river model that includes 2 existing bridges to represent the existing scenario and the model build process was relatively straightforward. I then had to divert a section of the river, remove one bridge and add 2 new bridges to test the impact of a building a new roadway. I can divert the river no problem but when I add in the bridges the model crashes when I run the 1D in the Unsteady state. I used the same procedure as I did for the existing model build for adding the proposed bridges, which have a far greater hydraulic capacity than the existing. The model keeps telling me the water level is higher than the channel boundary which doesn't make sense as the when I run it without the bridges the water levels stay within the banks. I have tried adding more interpolated sections where it suggests and even added more either side of the bridge just to be sure. I also increased the dflood value to 10m. I keep getting the following error:

Calculated water level is higher than the channel boundary

Highest point on section = 62.061 m
(after adding vertical wall of height 10.000 m)

Calculated water level = 145.615 m

Try adding more section data
or increasing the wall height dflood

Model time 2.6183 hrs:
*** error E1023 *** at label: New_20_BR

ERROR in unit type US BPR Bridge
which is attached to the following labels:-
stopped in error


It is as if the bridge is not allowing any flow to pass through it. Any suggestions would be greatly appreciated




Running window in v4.3

Posted 07/11/2017 10:05 by Monica Macias Jimenez

Last post 07/11/2017 11:09

1 reply

Hi all,

I am new using v4.3. I have found that the running window does not give me a pop window when the model crashes as it used to do before. Not in 1D2D linking models at least. It is quite hard to know when the model fail while doing other things. I need to keep looking until I see the output window has given me an error. I am not sure if I am missing a preference button in which I can set to be notified when the model crashes though. Any possibility to have that?

As an observation I would like to point out that I find less useful to have the TuFLOW window within the FM one because to track the model run is more complicated.




Reopen simulation progress window in V4.3

Posted 07/11/2017 09:48 by N Chapman

Last post 07/11/2017 11:13

1 reply



I'm using FM V4.3.


I have closed the simulation progress window, but left the model running in the background.


is there a way to reopen the window to check on the model progress?




FloodCloud & ReFH2 units

Posted 02/10/2017 11:24 by Marcus Shepherd

Last post 02/10/2017 13:09

1 reply

Hi all,


Is it possible to use ReFH2 in FloodCloud? I don't have a local ReFH2 license on my machine, and am likely to need this for an upcoming project. If it's possible to run this in the Cloud, it would potentially be a good saving.






Error in unit type river

Posted 14/09/2017 11:45 by Sarah

Last post 14/09/2017 11:52

1 reply


I'm adapting a model to include a new FAS, which was working fine until I added a flow control structure using an orifice and weir structure I now seem to get the "ERROR E1023 in unit type river at cross section a couple of nodes downstream. I've tried to manipulate the ICs to solve the problem but it won't seem to budge. Any ideas?

Many thanks!


Error in rules

Posted 12/08/2017 02:27 by Chulani

Last post 14/08/2017 10:59

1 reply

When I try to run the model after inserting vertical sluice gate, error occured as mentioned below,

Rules 2 and 1 are valid simultaneously for
logical operation of sluice connected to label Galamuna_U
This is not permitted.

Computation error in subroutine rules.
Stopped in rules.


How can I sort it out? 

1D Mass Balance

Posted 28/07/2017 10:15 by Konrad

Last post 25/07/2019 08:43

5 replies

Thought I'd try and generate a discussion on the new 1D Mass Balance output tool in v4.3. Pls see the following article which appeared in this week's newsletter for how it works, and what is uses.


Would be interested to hear others' thoughts ... one question I've heard repeated amongst a small audience to date is the obvious "what's an acceptable value?" Discuss...


Save settings for printing cross sections

Posted 12/06/2017 14:22 by Thomas

Last post 01/08/2017 12:11

2 replies

Hello everybody!

Another question concerning the visualisation/print of 1D cross section data with Flood Modeller Free:
is there a chance to save all the preferred settings (colours, marks...)
after changing them in the „Edit chart…“ window, to use them again later?

Thanks in advance!

Cross-Section Data dxf export

Posted 08/06/2017 08:28 by Thomas

Last post 26/07/2017 13:57

2 replies

Hello everybody!


Having built up a 1D model with Flood Modeller Free (Version: 4.1.5816.33282) I encounter the following problem:

Double-ckicking on a node symbol of a river section I get the data window for this profile,
hitting the „Plot…“ button I get a drawing of the cross section to which I can add results, such as maximum stage.
Using the right mouse button i get the chance to export this drawing to AutoCAD.
But whe i try to click on Export -> Autocad File (*.dxf) I get no window to choose the location
of the file or anything. Is this feature not implemented in the free version or do I make any mistake?

Many thanks for your replies!


ReFH2 and storm duration optimisation

Posted 18/05/2017 10:23 by miguel piedra

Last post 18/05/2017 13:04

2 replies

Hello, does anyone know how to run the storm duration optimisation tool when using a ReFH2 unit? I have tried but it seems that FM does not accept ReFH2 units for optimisation...



Lateral inflow to unit

Posted 14/03/2017 10:26 by P McAteer

Last post 15/03/2017 09:31

4 replies

Good morning - I have an overflow reservoir linked to a 1D channel via a spill. When running the 1D - 2D model, I receive this warning:

'Sending lateral inflow to unit S6 with no d/s unit - sending to upstream unit S5 instead'.

Both S5 and S6 are in the main channel. The warning occurs when rising river levels meet the invert of the spill. Immediately downstream of S6 is a junction with two outlets, the 1D channel (s7) and the spill linked to the reservoir. As there appears to be nothing wrong with the 1D model, should I assume that the problem lies with the link between the spill and the downstream reservoir? Altering initial water levels in the reservoir appears to make no difference.



Sensitivity analysis

Posted 15/02/2017 11:38 by NP

Last post 13/06/2019 13:11

1 reply


I have used flow hydrographs and NCBDY normal depth for my upstream and downstream boundaries in my model. I have been asked to carry out sensitivity analysis on the boundary conditions. How do I do this please, presumably varying inflows for u/s boundary, what about d/s boundary?

Also, what is the  minimum parameters apart from boundary conditions I need to carry out sensitivity tests on?

Thanks in advance



can see or export outflow at last node ?

Posted 06/12/2016 06:11 by teerawat pechsringam

Last post 06/12/2016 10:00

1 reply

I want to see outflow at lastnode or Export outflow to .csv, .text, .xls file node-to-node.

Who know please answer my question.



1D Diagnostics

Posted 02/12/2016 14:24 by Sarah

Last post 02/12/2016 14:35

3 replies

I'm looking for the 1D Diagnostics tool on FM and can't see it in the bottom right of the screen or under the layout tab. Where can I add it to my interface?



[New to Flood Modeller] Not sure how to fix error 1020 by looking the previous thread "https://www.floodmodeller.com/shared/forums/flood-modeller/1d/running-1d-models/error-1020/"

Posted 03/10/2016 13:26 by Ken_Li

Last post 03/10/2016 17:23

3 replies

Hi there,

I am new to floodmodeller and doing my final year flood modelling project at uni.

Sorry that I didn't realise thre was another forum thread regarding the same issue before I posted my thread on 10/2/2016 2:42 PM...

In that thread "https://www.floodmodeller.com/shared/forums/flood-modeller/1d/running-1d-models/error-1020/", your team suggests to...

  1. Incrementally build the model and create initial conditions as you progress with the model build
  2.  Look to increase the minimum flow of the model inflows to improve stability.

I have generated the initial conditon through steady (direct) simulation and have used the .zzs file as the initial condition during unsteady simulation (fixed timestep). So I am not sure what you mean by "Incrementally build the model and create initial conditions".... And about "increase the minimum flow", could you please specify where I can do this because I don't really get it as I am really new to this Flood Modeller? 


Here is the error I got. I have overcome the error occurred in node 76 (beginning) by running steady simulation, in order to input the initial condition for the unsteady run. I have tried amending the river section data (x,y values) in 190, but it doesnt help at all.....

*** error E1020 *** at label: 190

Water level exceeded maximum section data level dflood ( 3.00m)
Check model, extend section data or increase dflood
Check model, extend section data or increase dflood
stopped in ondam


Sorry for writing such long and hope to hear from your team soon!!laughing


Many thanks



1D model - bridge surcharge

Posted 06/09/2016 09:46 by envirologic

Last post 06/09/2016 09:58

1 reply


I am running a 1D model.  When a USBPR bridge in the model surcharges the upgradient water levels appear to be severely overestimated.  Is there any way of specifying the distance between top of orifice opening and top of bridge wall.

I would like to propose a mitigation measure of a flat deck bridge also but not sure how to input this.  This would allow water to flow over the top of the bridge during surcharge, and presumably limit the upgradient flood levels.  I've tried a culvert but similar problem.




Posted 16/07/2016 10:42 by Ian Corder

Last post 18/07/2016 08:20

1 reply

I'm working on the calibration of a simple model, and want to adjust Manning's N throughout the network to see what effect it has. Is there a simpler way of doing this than going through every cross-section in turn and manually changing the figures at each point in the cross-section?



Model losing water?

Posted 02/07/2016 12:28 by Ian Corder

Last post 18/07/2016 08:26

4 replies

Hi there,

I'm running scenarios on a small, fairly simple network in FloodModeller Free. The shape of output hydrographs downstream is consistent with the upstream inputs, (lagging slightly, as expected), but there seems to be a constant loss of volume as it goes downstream! After 20 nodes (2km) the signal has lost about 2 cumecs (8% of the initial flow). Any ideas where I might be going wrong? I am getting some errors such as 'U09307   - NO SPILL ATTACHED', but not anywhere near the nodes that the loss is occuring.

Thanks a lot,



Automatic Priesmann slot

Posted 28/06/2016 11:11 by ketaki

Last post 28/06/2016 14:10

1 reply


What difference does selecting this option during simulation make in the results? 

My models run smoothly when this option is selected and donot run when its not. So just wanted to know if this option just drives the model to run or affects the results the as well.




Posted 10/05/2016 16:19 by Brookr

Last post 10/05/2016 17:52

1 reply

 Good afternoon.

This is a fairly general question around the use of Theta (Advanced Paramaters).

I am currently working on a third party model that appears to unstable towards the downstream. I have tried various things to rectify the problem but to no avail. I have found that if I increase theta the poor convergance is removed.

It is set to 0.7 as default, however if we increase this value 0.85 or greater that seems to work.

Within the help file it states that a higher value will make the model run more stably, if this is the case why not have it set to 1 all the time? It obviously has some draw back/ limitation.

I was hoping you might be able to explain these limitations.

Regards Richard Brook

Sign in to create or reply to forum posts.

Privacy & Terms