We learned this week that an engineer (who shall remain unnamed) presented a paper at the DSS meeting in Maryland about water age modeling in which he claimed that mixing CANNOT decrease water age. 

We respectfully disagree.

The best way to understand how mixing can influence water age is to consider the differences between a completely mixed and a totally unmixed tank.

Most operators assume that a perfectly mixed tank will have a mean water age that is the product of its average daily turn-over divided by its average daily capacity. For example: a 1 MG tank with an average turn-over of 250,000 gallons would have, by this calculation, an average water age of 4 days. The actual mathematics of this dilution is a lot more complicated, but let’s keep with this simple assumption for now.

But what is the water age in an unmixed tank?  Imagine the same tank again but this time it is thermally stratified, with 50% of the water floating on top of the other 50% of the water with a strong thermocline separating the two. The top 500,000 gallons does NOT mix with the water below.  You withdraw 250,000 gallons during the day and refill it with 250,000 gallons of fresh water at night.

What’s the water age?

Well, in the lower half of the tank, you have withdrawn 250,000 gallons from the layer with a total of 500,000 gallons. So the lower layer has a water age of 2 days.  But the upper layer grows older by one full day every day.  At the end of a week the lower half of the tank still has a water age of 2 days, but the upper half of the tank has a water age of 7 days.  The average of these two is 4.5 days. But at the end of the second week, the lower half of the tank has a water age of 2 days (because 50% of the volume is being recycled every day) but the upper half of the tank now has a water age of 14 days. The average water age in the tank is now 8 days.

At one standpipe we worked on in Washington State, the upper 80% of water thermally stratified in the Spring and remained isolated *all summer long*.  This slug of bad water rode up and down on top of the fresh, colder water at the bottom of the tank. The water at the top of the tank had lost all disinfectant residual and the tank as a whole probably had a water age in excess of 100 days.

We don’t know all the details of some of the water age models that are presented by engineers. But we suspect that many of these models *assume completely mixed tanks* as part of their calculations. This assumption is only correct if you have active mixers installed and operating. If you do not – you really can’t say how old the water is in your distribution system.

We appreciate your comments and questions.

Sincerely,

Peter S. Fiske, Ph.D.

CEO, PAX Water Technologies