Jump to content
Richard Hughes

Cross scetion representation

Recommended Posts

I am currently reviewing a TUFLOW model, however, the cross sections have been entered in a way that i am not familiar with. The surveyed cross sections have been entered as CS lines, type XZ. The CS lines (cross scetions) have been placed at the centre of 1d_nwk sections. Upstream and downstream inverts have been manually entered for each 1d_nwk section in the attributes table. Am i right in thinking that the channel 'shape' described by a CS line is applied to the whole 1d_nwk section ( albeit with elevation changes to meet the specified invert levels). If this is correct, how does ESTRY interpolate between channel shape, if a channel shape is applied to a whole 1d_nwk section. If consecutive cross sections vary greatly in shape, will there be a sudden channel shape change at the 1d_nwk section break?



Share this post

Link to post
Share on other sites

Hi Richard

By CS lines I'm presuming you're referring to cross-sections being referenced via a 1d_xs layer (ie. the 1d_tab format). 1d_xs lines can be either snapped to the channel ends or lie midway along the length of the channel. In the latter, which is the case I think you're referring to, you're correct in assuming that the cross-section shape applies to the entire length of the channel. This is also the case for end cross-sections except that the cross-section shape applied to the length of the channel is an average (of the processed data) of the two end cross-sections. In any 1D modelling, for each 1D segment/element/channel/Q-point/unit the cross-section properties are kept constant for the length of the element or for half the length u/s and d/s (this is the nature of the 1D solution).

If using midway 1d_xs lines, the u/s and d/s inverts have to be entered, or if using -99999 the u/s and d/s inverts are the same and are set to the bed of the cross-section (ie. there is no slope on the channel). If the inverts have been entered, as in your case, the cross-section is shifted up or down to match the manually entered inverts. Searching for the channel in the .eof file is a good way of reviewing the final hydraulic properties for the channel.

Yes, if consecutive cross-sections vary greatly in shape (or more importantly the conveyance at similar water levels is substantially different) then there will be a sudden change in 1D conveyance that can lead to a poor representation and numerical problems (for any 1D scheme). This situation rarely occurs in reality except where there is a man-made constriction or structure, where the creek/river is down to bedrock, or sometimes in the urban environment at transitions/junctions of man-made channels. In many of these cases it is better to use a structure to represent the sudden change in shape as the flow will not be predominantly one-dimensional and the 1D equations are not representative due to the rapid expansion/contraction of flow in the horizontal and/or vertical planes. This applies irrespective of whether end-cross-sections or midway cross-sections are used.

The above issue can be a common cause for the poor performance, mass errors and stability of any 1D scheme, especially if cross-sections are "blindly" extracted from a DEM that has very poor accuracy within the creek/river banks (ie. DEMs from most Lidar or other aerial survey techniques).

The 1d_hydprop_check layer is useful as a first pass review to label or thematically map the conveyance and other hydraulic properties at the top of the cross-section. Another very good way of identifying locations where this might be a problem is by viewing the 1D velocities (use _mmV.mif and/or _TS.mif layers or over time using 1D WLL output that hasn't used the "Read MI WLL Points" command to set the elevations). Where you have sudden changes in velocity is an indaction that there is a sudden change in conveyance and possibly cross-section shape.

I hope this helps!



Share this post

Link to post
Share on other sites

On this topic, what is the best way to review the shape of cross sections and the lengths of reach over which they have been applied ?

Share this post

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...