116 Setting overflow permits – how complicated can it be

A new government guidance document “Draft information and guidance on storm overflows” for England has been published for consultation.  This sets out to provide clarification of the requirements of existing documents including:

• The Water Industry Act.
• The Storm Overflows Discharge Reduction Plan (SODRP)
• The revised Version 2 of the Storm Overflow Assessment Framework.  (This is supposedly out for consultation but I cannot locate a copy.  However I did see a presentation on its content at the UDG conference a couple of weeks ago.)

The new guidance also briefly mentions the new statutory requirement for sewerage utilities to develop Drainage and Wastewater Management Plans (obviously something I am very interested in) but does not seem to be aware that detailed DWMP guidance is being drafted that will also cover assessing system performance for storm overflows.

It is telling that additional guidance is required to explain existing requirements.  They are obviously not clear enough in themselves.

For now I will just look at what seems to be new in the “Draft information and guidance on storm overflows”

Targets

The targets for the operation of storm overflows from sewerage systems in England are set out in the Storm Overflows Discharge Reduction Plan.  This sets out three metrics:

1. No more than 10 spills a year on average
2. No local adverse ecological impact
3. Significantly reduce harmful pathogens in designated bathing waters

The SODRP puts the 10 spills target last, suggesting that it is the least important, but I have put it first as for most overflows it is probably the most restrictive.

The targets are at first sight very clear and need little further explanation or embellishment, but the new guidance adds lots of words to these simple targets without greatly clarifying what is required.

Keep it simple stupid

I have argued before, that performance targets should be set based on the outcome that we want to achieve and leave how to achieve it to the sewerage utilities.

So now that the driving target for overflow operation is to have no more than 10 spills a year, I think that how the sewerage system is designed, built and operated is largely the business of the sewerage utility, provided that they meet that target.  Unfortunately the new guidance document doesn’t adopt this principle but goes for a belt and braces approach by adding design details to back up the simple performance targets.

So what is new

I picked up a few things in the new guidance document that I thought were new, or at least that I was previously unaware of.

Best value

We already knew that the SODRP does not require a test of value for money for reductions in overflow spill frequency.  The targets are absolute and must be met whatever the cost.

However, the new guidance points out that, although it was a statutory requirement to publish the SODRP, the contents of it are not legally binding.  They only have to be met because the sewerage utilities have committed to meet them.

Targets can be imposed on an individual overflow by being defined in the discharge permit, but I am not clear whether the conditions set in the permit need to take cost benefit into account.

Interesting, so is there an option to withdraw that commitment to meet the absolute targets and come up with a best value plan?

Assessing performance

The guidance states that compliance with the 10 spills requirement will be assessed based on the average over 10 years of measured data.  That opens up a whole range of questions. 

How many years data are required to give an initial indication that there are problems or can the sewerage utilities put off doing anything until there is 10 years of data.  The guidance says that the Environment Agency are working on a framework to give early assessment of performance.  It will be interesting to see what that says as the devil will be in the detail.

Incidentally there was an interesting paper by Paul Martin at the recent UDG conference that pointed out that to be certain of not exceeding 10 spills a year over 10 years we probably needed to design for 8.  Will the EA guidance take this uncertainty into account?  Or have we just moved the goalposts again.

The new guidance actually states that it would be better to design an overflow using a representative 25 years of data and demonstrate that all 10 year periods comply.  It is bad enough having to run large complex models for 10 years.  We cannot afford the time or cost of increasing the computational load.  If anything we need to simplify things by designing for a worst case single year.  Any ideas on how we would define this?

The guidance states that compliance with spill frequency targets should take account of future climate change affecting rainfall.  It is good to see it stated explicitly.

It also states that there is no de-minimus spill volume that can be ignored as being unlikely to have any impact.  A spill of even a few litres of storm sewage still counts as a spill.  I think that we need an injection of common sense here.

Belt and braces

The guidance then puts an extra layer of security on meeting the 10 spills target by defining required pass forward flows at overflows and storm tanks.  Why is the regulator defining this; it should be up to the sewerage undertaker to decide on the combination of pass forward flows and storage that it wishes to use.

The guidance restates that for storm overflows at storm tank the Flow to Full Treatment (FFT) should be 3 * foul flow + Imax.  Where Imax is the maximum infiltration and inflow on dry days following a dry day.

The guidance also addresses the required pass forward flow at the inlet to the treatment works.  This has traditionally been defined as Formula A to match the minimum requirement for network overflows.  However this can give spill frequencies of 40 spills a year.

The guidance now gives an alternative definition that can be used instead of Formula A.

6 domestic foul flow + 3 trade effluent + Imax

It is not clear what to do when (inevitably) this gives a different answer to Formula A.  Is it a requirement to use the larger one or can the sewerage utility choose which to use.

The guidance also sets out the required volume of storm tanks at a treatment works.  This has traditionally been defined as 68 l per head of population or 2 hours at the difference between pass forward flow at the works inlet and the FFT.  The guidance confirms these rules, but doesn’t seem to have considered the change in the definition of pass forward flow.  The flow difference would be 3 * domestic foul flow for 2 hours.  That is a quarter of the daily flow per person or about 35 l per head of population. 

So which volume is required?

My argument would be that all of this is unnecessary.  The required performance is no more than 10 spills per year.  It is not necessary to say how that is achieved.  What if the new definition of required pass forward flow gives only 5 spills a year?  Is it still a requirement?

Conclusions

Most of the new guidance document says nothing new and nothing that isn’t said elsewhere.  So, is it needed.

The one new thing that it does introduce, a new definition of required pass forward flow, is confused and is in any case not necessary as the performance target is for no more than 10 spills a year.

The regulators are wanting to use both belt and braces, but in the process have forgotten to put on their trousers.