Flow Based Sub Samples

There will be times when it is convenient to confine statistical analysis and cumulative frequency curves to a particular range of flows. It may be necessary to exclude periods of zero flow when assessing compliance with guidelines, or periods of baseflow and stormflow may need to be addressed separately. This can be done by specifying a flow-based sub-sample at any point.

The Flow-Based Sub-Sample Bounds... command on the Context-sensitive menu of a node or link opens the Select Flow Thresholds box where you can select upper and lower flow thresholds at that point. The thresholds modify the analysis at that point, but not the underlying simulation. They apply to the inflow to the node or link, and influence the outflow from the same point, but do not affect other locations in the catchment.

Both upper and lower flow thresholds may be set simultaneously, but in most cases only one will be required. To eliminate periods of zero flow from the analysis, tick the lower threshold box and set it to zero. Clear the check box to include periods of zero flow. To analyse only baseflow or only stormflow, tick the upper threshold box (for baseflow) or the lower threshold box (for stormflow) and enter an appropriate threshold value. The thumbnail graph may help in choosing the threshold. If extreme data values compress the thumbnail plot into one corner, as in the example shown, open a full Cumulative Frequency Graph of flow using the Flow-Based Sub-Sample submenu item, and enlarge the scale. Ensure that Auto Run Models is turned on (File menu), so that the graph is preserved as the threshold is changed.

To view statistics or graphs of a sub-sample, use the Flow-Based Sub-Sample submenu item of the Statistics or Cumulative Frequency Graphs command at that point. Remember that the underlying simulation has not been changed. Statistics and graphs of the full sample can be viewed simultaneously, if desired, using the All Data submenu item.

The examples below show how the flow-based sub-sample can be used. The first example shows a cumulative frequency graph for our example pond using all data. The outflow neatly meets a condition (contrived for this example) that in 90% of time-steps the TSS concentration does not exceed 12 mg/L. It’s not a hard condition to meet - in 80% of all timesteps there is no outflow anyway.

In the second example, zero inflows to the pond have been excluded from the analysis. Now only non-zero flows upstream of the treatment node (or prior to its construction) are considered. The flow range which applies is noted in the graph heading. Evaporation from the pond has further reduced the time-steps with non-zero flow, but the specified water quality condition is no longer met.

As can be seen, the conditions under which guidelines are applied can make a substantial difference to the outcomes achieved. Care is needed to ensure that guidelines and standards are interpreted correctly.

Remember that the flow-based sub-sample is always defined by flow, as its name implies, and always at the inlet to a node or link (except for a source node, which has no inflow). If you need to take the sub-sample at the outlet of a treatment node for a particular analysis, think of it as the inlet to the downstream link.