Jump to content

Direct Rainfall modelling

Recommended Posts

Hi there,

I’ve been running some 'direct rainfall' urban flood models, and calculating residual depths once the catchment has fully drained. Basically I’m coming up with average residual depths of around 30mm over a catchment, for different catchment gradients and land-uses.

I do believe this initial loss due to the depressions in the DEM is exaggerated, due to grid or ALS resolution, and also discontinuities or blockages of sheet-flow paths due to the DEM generation process. I’m just in the process of thinking through the cleanest way to account for this initial loss due to depressions in the DEM.

I initially thought of a process such as subtracting the 30mm from the ARR type IL values in the TUFLOW Materials File(.tmf). This however starts to get messy (negative ILs), is hard to audit, and may require rainfall data to be adjusted.

The best method I can think of is to run an initial simulation for say 6 hours, applying 30mm rainfall over the first hour to fill depressions, then waiting until the catchment drains. This would be with no IL or CL in the tmf file. Then, create a flood surface at t=6 to use as part of a restart or initial condition for the real flood simulation. This way, adjustments of the ‘standard’ IL and CL values in the tmf are not needed, and these should still be comparable to test book values such as in ARR.

It’s interesting that this initial loss component due to depressions in the DEM will vary depending on the grid interpolation method (eg, TIN, IDW), and even cell size (larger cells apply more smoothing and possibly contain less depressions). Instead of trying to reduce the IL values in the tmf file, I think the second method could work well for any DEM scenario.

Just wanted to ask if anyone had looked into all this in detail. Are there any other techniques being used?



Share this post

Link to post
Share on other sites


Felix Taaffe from our office investigated this issue in detail for his undergraduate thesis, including the effects of varying grid resolution. The attached paper was presented by Mark Babister on his behalf at IAHR in Brisbane last year.

As you identify, the DEM cannot represent the small-scale drainage features in the upper catchment, leading to artificial extra storage being created. I think rather than applying negative loss values, methods to modify the DEM either by smoothing, filling or pre-wetting are probably the way to go. If filling or pre-wetting pre-wetting, you might need to consider whether there are legitimate trapped depression storages in your catchment, and whether it is an appropriate antecedent condition to have them brim-full before your design storm burst.




Share this post

Link to post
Share on other sites

We undertook some detailed analysis of these issues for some urban catchments we did in Melbourne, assessing both loss rates and runoff coefficients against more traditional hydrological modelling approaches linked to flood models. We used a pre-priming method described by Paul in one of these cases by applying 10mm to the catchment a number of hours before the design storm to remove depressions that were not directly connected to the drainage system. Th 10mm was based on an analysis of the volume remaining across the catchment after the storm had finished (i,e, the bits that would never drain). Interestingly the results showed virtually no difference to the case when this was not done (less than 1cm of difference in levels along the main flow lines at 2 SD). The flows and flood depths generated by direct rainfall approach were broadly consistent with those derived from distributing a equivalent volume from a hydrological model flow across all pits in each small subcatchments. Overall catchment size was about 6 square km and subcatchments were about 2-4 ha.

We tested with a very large storm event and good pluviograph and level data. The most salient point was that if we had been only doing a major drainage study, the runoff coefficients needed to be greater (or loss rates smaller) in the direct rainfall approach and the distributed hydrological modelling approach to achieve the same flood levels as the broadscale catchment approach (flow input along the main drainage lines only). I think this is likely due to the influence of minor surface features causing retardation, the efficiency of the pipe network and the flow path length in the distributed models.

I think that the method you chose really depends on your catchment and the level of detail required - our results showed that for these urban areas, the difference in pre-filling these storages was negligible. Your results may vary!

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...