Ideas Lab

In GIS visualiser, it would be useful if Interpolate and Replicate sections could be correctly positioned between the upstream and downstream River or Culvert units.

At present, the first Interpolate or Replicate unit is located halfway along the line between the upstream and downstream units. The second Interpolate or Replicate is then positioned halfway between the first and downstream unit etc.

I think it would be a very useful thing if we could have interpolate units between the various types of culvert sections instead of having to use replicates. This would be particularly useful when modelling culverts that have steep gradients and require lots of replicates to be calculated and manually input.

ISIS results files from large models can be very large (100Mb+ for .zzn files) and results from multiple runs can quickly use up large amounts of storage. If files are compressed manually they take up much less space (compression to about 10-30% of original size using a variety of freely available zip software) but can no longer be easily viewed or analysed.

If ISIS incorporated an option to compress the results to a format which allowed the plotting of long sections, use of tabular CSV etc. then it would massively reduce data storage requirements for large projects.

TUFLOW has recently released a double-precision version, which greatly stabilises (and increases the accuracy of) rainfall modelling. It would be great if this was able to be linked with ISIS river models, but unless ISIS is also working in double-precision the data transferred between the programs is clearly incompatible.
Pre-existing ISIS-TUFLOW models could then be used for pluvial modelling without the need to consider conversion to ESTRY or filling in the river up to peak levels. I think this is going to be an increasing area of work, so it would be good to have the capability in place now as the practice emerges.

(OK, it had to come out again!)
Labyrinth Weirs (and similar structures, e.g. horseshoe and oblique weirs) have markedly different behaviour in modular and drowned flow, specifically due to their effective breadth. There are several ways to model at present (different models for flood & low flows; QH control; sluice that models the extra breadth [& closes when drowned], ...), but none are 100% satisfactory. How about a dedicated Labyrinth Weir unit in ISIS?

At the moment, if you are viewing cross sections and scroll through a model, the panel markers and Manning's values are removed every time you cross a structure.

I think it would be helpful if panel markers and Manning's values re-appeared when you got to the next river section unit, rather than having to be re-selected each time.

Could we have some automation to ordering the initial conditions in the data file. During model build and stabilisation, we often revisit units in the data file. This results in a re-ordering of the initial conditions in the data file. When post-processing results, the order at which these are in is vital, especially if the model is large and you are only interested in particular reaches. Thanks. TJ

I am currently developing a real time flood forecast model. It would be useful to be able to update levels using telemetered level data in online storage areas (as can already be done for flows).

Does anyone else agree this would be a useful ISIS development?

Judging from the amount of people who have discovered the formatting error in the Culvert Inlet input values table in the Help file within one day or so of release, I realise many people must regularly refer to this table for the input values.
How useful would it be to have a "Culvert Inlet wizard" that takes you through the culvert types and enters the values for you? Should minimise transcription errors and make input easier.
konrad

It would be useful if you could print out a zoomed area of the IXY so that you can show complicated areas in more detail. Also, if your model is very large, the IXY file is not clear even on an A0 size sheet, so it would be useful to have the 'print a zoomed area' facility to produce your schematic on two or more sheets to go into a report and it will still be possible to see all the units and connections. It is possible to print out a zoomed area by specifying the print limits using those shown on the visualiser screen (bottom left to top right corners), but it prints right to the edge of the sheet over the border and legend and is too messy to go in reports.

A paper on this subject was presented at the ISIS seminar earlier in the year but it is not on the list of features on the horizon.

Randomly varying parameters within a tolerance range and then running a number of simulations is a way of providing a distribution of possible outcomes. This should not be too difficult to implement as a number of .dat files could be produced to run in batch mode (as a starter).

It would be nice to be able to view all of the output data sets simultaneously.

A further advantage of running this type of simulation is that it may encourage modellers to think about the confidence limits of all of the data, both input and output. We often see sensitivity analysis presented in reports, but it is rarely put to any use.

I suspect that it will also provide a good test of model stability.

Rather than saving bitmaps of model run graphs, how about using a more efficient modern format, such as jpg, tif, gif, png

I know server space is fairly cheap, but emailed bmps take too long to download.

and another thought - please could you save these by default for all model runs, so they are an opt-out feature, not opt-in?
thanks,
Ian.

I've just noticed that some of the curious zeros in brackets at the end of the headings for ideas viewed in the main ideas lab are infact now ones!
And clicking on the heading reveals that this means someone has commented upon the idea.

Some of these comments are really quite halpful to see, being responces from the ISIS team for the most part. One such comment is in responce to the multiple language support idea, for example, and is asking which languages people would like to see. A worthwhile question that it would be good to have responces to!

It would seem to me then, that'd it'd be helpful if these comments were a little more visable in some form. Alternatively, (or perhaps additionally) it would be usefull if at least the ISIS development teams responces were visible on the main page; allowing better access to both the follow-up questions they may have and their ideas for implementation.

Further to PHA's suggestion "Show comment units in IXY, GXY", it may be useful to have an option to display a short field associated with internal .dat file comments - i.e. one that would be visible alongside the Label1, Unit and Sub Unit columns - perhaps a comment title with a limited character length.

I often use the comment units to split up blocks of hydraulic units by reach and enter a short description of what is represented by each block in the comments. Being able to quickly see something like "River XYZ Right Bank Spills" in the .dat file window could be handy on occasion.

Might it be possible to have something along the lines of an optional "Start New Reach" and "End New Reach" field entry command which would lock the position of the comments to associated nodes in the .dat file? If the commments fields were locked to georeferenced units at the start and end of each reach, it may be possible to display them in the visualisers as PHA suggested. Some type of positional locking system for comments may also ensure that further useful model audit information, e.g. survey details etc, is preserved in a useable location within the .dat file.

Prior to using ISIS Professional the software package I used for flood risk modelling enabled me to swiftly import surveyed cross sections directly from a csv or txt file. I could also build up a ground model from OS NextMap data (again in csv or txt format). It would be really useful if cross section and ground model data could be imported directly into the model from an Excel or txt file.

Presently I am having problems converting this data as I believe MapInfo + an add-on is required to convert the files to a format that ISIS will read.

If anyone has a workaround for this I would be grateful. We use a number of surveyors who generally provide us with data in XYZ format as a CSV file as this is how we've asked for it in the past.

It may be useful if the X-Y series plot included an option to add spot flow gaugings for rating calibration (currently the Add Series option is unavailable in X-Y plots).
P.S. Please enable X-Y plots from the visualiser.

The "Go to network chart" button on the GIS Visualiser is really useful if you have to keep switching back and forth from the main network chart. It would be useful if this feature was also available for the Scematic Visualiser (IXY viewer).

For some larger models, or those whose node ordering is not very helpful (like one exported from Infoworks for exapmle!) it can be that you spend most of you time working form the IXY or GXY viewer. The downside of this however is that one does not necessarily see the comments associated with a locale in the model. (If the comment is by the channel sections and you are looking at the bridge unit, you wouldn't necessarily see the comment even if you looked at the main data file window.)

To overcome this, the addition of comment icons to the IXY and GXY viewers would be very helpful.

To implement this would probably mean that comments will need a node lable equivilent (given the way a GXY works, but this could be auto generated by ISIS when each comment is created (and would probably require a utility to convert old comments so they work with this).

It may also be good if this was an optional thing per comment; some ISIS files come with a comment containing the survey data for each secion, and these might not need or want georeferencing to clutter up the viewer!

When using ISIS free, it would be helpful if a prompt was produced when trying to do something which will prevent you from saving or running the file (so when creating new units and surpassing the 200 limit, or when pressing ok to changes in a unit when editing a file with more than 200 nodes).

Such a dialogue box should have the option to carry on regardless (but you're now well aware that you can't save; useful for changing bank markers to view results for example) or to undo the last change (i.e. node addition) so that you can save what you'd already done.

It's a pain to start with 195 nodes, not realise and add another 20 and then find you can't do anything with it and the last half hour is wasted.

When modelling historical (or indeed realtime) events, recorded flows may drop below what ISIS is capable of modelling. This requires the setting of minimum flows in the inflow data.
Currently the flow data has to be altered externally to ISIS. It would be much easier if there was an additional option within QTBDYs to impose a minimum flow (like there is in FEHBDYs for exapmple).

There may also be other instances where similar functionality would be useful, such as HTBDYs...