Jump to content
TUFLOW Forum

lucy

Members
  • Content Count

    30
  • Joined

  • Last visited

Community Reputation

0 Neutral

About lucy

  • Rank
    Advanced Member

Profile Information

  • Gender
    Female
  1. Thank you so much Rachael! Exactly what I was after, champion
  2. Hi I am trying to present the overall peak floodplain at a number of ARI's in a township. 1. What is the best way to determine the critical storm? I have run various duration storms at each ARI. I have currently chosen the event which shows the maximum floodplain extent for the majority of locations. However, I know there could be some locations which peak in different duration events. 2. I am wondering if it possible to present a single flood map which is an envelope of peak flood depths from a combination of peak storms? This follows on from my first query. If I can't determine a single critical duration, I could present an envelope of multiple duration events. Any tips would be great. Thanks Lucy
  3. lucy

    PO Lines

    Hi I am using PO lines to extract surface hydrographs from the downstream end of a subcatchment within a larger model. To further understand the results, I am wanting to know a bit more about how the PO lines work. How does the length of the line influence the recorded flow? Does stationary water register in the hydrograph? Eg, if a PO line was over ponded water, would the hydrograph register this "flow" of the stationary water? Any pointers or more info on the PO lines and their interaction with the surface flow would be greatly appreciated. Thanks Lucy
  4. Hi Phillip I have previously used this tool but I can no longer download it. Is it possible to re upload it please? The page seems to have moved when I try to access it Thanks Lucy
  5. Hi I am using direct rainfall to develop a model for a stormwater mangement plan. Generally for this type of SMP you would use an ILSAX/DRAINS model to develop hydrographs for input into TUFLOW at pits. However, this town is very flat with minimal existing infrastructure. Firstly, creating a DRAINS model for this catchment is unlikely to produce an accurate hydrograph for the large catchments. Secondly, applying these hydrographs in TUFLOW is unlikely to produce an accurate floodplain due to the sparse number of pits. Thus I am required to justify the direct rainfall approach over a DRAINS approach. I have been asked to compare hydrographs from each model to compare peak flow, time to peak, volume etc. I have chosen a sub catchment, set up a DRAINS model and am comparing the outflow to a hydrograph from a surface hydrograph read from a PO line in TUFLOW. I am having trouble calibrating them. The TUFLOW hydrograph is slightly (10%) lower than the DRAINS peak and lags behind the DRAINS hydrograph. Also, the DRAINS hydrograph returns to 0 quite quickly but the TUFLOW hydrograph peaks and then plateaus when the simulation finishes without reaching 0. Thus the TUFLOW hydrograph is showing much higher volume compared to the DRAINS. I am running a 1 hour storm for 3 hours in a 100 year event. I am using a nested grid of 2m over the town and 8m elsewhere. There are no gauging stations or existing data to calibrate either model. There are a number of justifications we have come up with summarised below. If anyone has any other thoughts on the differences in these approaches, reasons they are different and things to consider I would appreciate any suggestions. catchment breakdown: DRAINS assumes clear distinction and all sub catchment flow reaching the outlet but due to the flat topography, in reality there may be inter catchment flowstorage: TUFLOW takes into account the actual overland flow path rather than a simplified direct overfland flow path as in DRAINSThanks Lucy
  6. Hi all The double precision model did the trick! With that being the only change the Peak Cumulative ME reduced down to 0.33%. Good to know it can make that much of a difference. The single precision model took 22 hours and the double precision took about 27 hours which is line with what Stephanie had outlined. I had already set the cell wet dry depth to 0.0002m thanks. I also changed the d value of the 2D type boundary between the nested grids. Visually this hasn't made a huge difference to the results however it took 31 hours to run and reduced the Peak Cumulative ME to 0.3%. However for that change I am going to leave d at default 0 due to the extra run time (I presume this is what caused that). Thanks everyone for your help Lucy
  7. Hi PHA Thanks so much for your response! Great to know I'm on the right path and there are some simple things to address which can help. I am currently running a double precision model and once that is finished I will also set the 2D boundary d value to 5. My timestep is already 0.5 seconds for the 2m grid and 2.5 seconds for the 8m grid so I will leave that as is. I'll also check the location of the 2D/2D boundary and move it away from flooding where possible. Also, good to know re the MB viewer. I had a play with SMS before but my company doesn't have it. I will have a play with Crayfish in QGIS though. I will let you know how these results go Lucy
  8. Hi I have a model of a rural town, model area approx 24km2. I am using direct rainfall on grid and a nested grid of 8m and 2m over the town centre. There is minimal stormwater infrastructure in the town. This is my first direct rainfall model and my first time using the nested grid module. When I used a single grid size of 4m my Peak Cumulative ME was 0.98%. However, with the nested grid described above, I am getting Peak Cumulative ME 87%. I am trying to work out what exactly is causing this and how I should address it. Some thoughts I have had from reviewing the forum etc are: Precision - when I used the global 4m grid size I was using double precision and I have changed to single. Would this make that much of a difference in the error? I am at elevations of around 40m so I thought it might be okBoundary location between the two grids - could this impact the model stability? Does it matter where you place this?d value in Type 2D boundary between grids - I have left this at 0 to create nodes at vertices along the boundaries. Could this be a problem?I have output the MB1 and MB2 .dat files but I cannot process them using TUFLOW to GIS (I get "can't find the specified timestep" error when I use the -max switch)I will try and run the double precision but it takes 20 hours to run as it is so I was hoping to define the problem a bit more. Does anyone have any other suggestions? Thanks, Lucy
  9. Hi Rachel Thanks for your response. So it breaks the fraction impervious into a cell by cell basis? For example; if I have a 400m2 block area with 80% impervious, and a 4m grid size, 20% of each 4m cell will infiltrate for each cell over the 400m2 block. Do I have that right? Thanks Lucy
  10. Hi I am running a model with GA infiltration. I have a materials layer which defines 4 material types (residential housing, rural housing, roads and undeveloped land). I am wondering how the fraction impervious is considered with infiltration? How does TUFLOW know where to apply the lack of infiltration for the impervious fraction over the entire material area defined? Hope this makes sense, any tips would be much appreciated. Thanks Lucy
  11. Hi I am trying to determine why additional nodes are appearing in my 1d_nwk_N check layer. Additional in that I have not specified a pit at this point. They are appearing "above" some of my pits ID of the pit name plus a ".1" on the end ANA = 1 Conn = SXL (same as the associated pit) but the check node at the actual pit location has a no Conn type Conn width of some value, I'm not sure where this is generated from though In the attached screenshot, the red dots are my pits and the green stars are the 1d_nwk_N check layer Can't find anything on these anywhere and not sure what impact they are having Any tips would be greatly appreciated Thanks Lucy
  12. Hi there I am trying to model an equalising siphon which goes under a lowered roadway. The theory is there will be two pits either side of the roadway connected to a "U" shape set up of three pipes, with the central pipe running under the roadway and the two side pipes connecting this to the pits on either end. Water should pool on one side of the roadway and "bubble up" from the pit on the other side. I have this set up as follows SXL connections for each of the two pits, W type with 0.9m width. US invert is set to 0.1 below the DEM surface SXZ 1d_nwk_bc points at each of these pits (do I need this as well as the SXL connection at the pit?!) I don't think they are working correctly. When I run the model, there is flow through the pipes but only in a very short interval. I'm not sure water is being picked up/carried correctly. One thing which doesn't look right is the mmq and 1d_nwk check layers The 1d_nwk_C layer has a line coming off vertically from where the pit is modelled The mmq layer puts a node on this line with 0 peak flow Also the 1d_nwk_N layer places a node at both ends of this line, with different bed level Not really sure what this line means, it is not a pipe I have modelled, but seems to be an additional polyline off the pit. It doesn't occur on any other of my pits. I am wondering if this is a clue as to why the siphons may not be working correctly I have attached an image of the left hand side of the siphon pipes which run right to left. The vertical line is the one appearing in the check layers. The large green dots are the mmq check and the small dots are the nwk_N check bed levels Am a bit lost on this one, any thoughts/tips at all would be greatly appreciated Lucy
  13. lucy

    Access Violation

    Thanks Phil! That would be super helpful
  14. lucy

    Access Violation

    Hi Phil I just worked it out - it was a projection issue, when i re projected both of them from scratch it worked fine so they must have been misaligned Thanks for your reply Lucy
×
×
  • Create New...