One of the really unique aspects of our business is that we’re automatically harvesting large amounts of data on swimming pools, regularly and reliably. One of these data points is water temperature. Using a large enough group of pools to be statistically meaningful I’ve looked at water temperature with some surprising results. For each of the graphs below – click on the image for a higher resolution version if you want a closer look.
About the data
The temperature readings are taken from our customers pools, all are based in Sydney and there’s currently a significant skew towards homes in the eastern half of metropolitan Sydney (closer to our head office).
Many of our customers pools have heaters, in some cases we have data about their use but no attempt has been made to allow for this in the analysis below. In all cases we don’t have visibility of the thermostat temperature set on the heater where used.
First up, let’s look at some high level data based on the averages of our readings.
This chart shows in black the average water temperatures of all pools we monitor, in green and red are the maximum and minimum daily air temperatures (taken from the Beureau of Metereologies data for Sydney) on the same day. From the same source is daily rainfall data.
Next I’ve taken the average of the maximum and minimum temperatures, then smoothed both the air and water temperatures by taking a 5 day rolling average for each value.
This makes the data a little cleaner and easier to see patterns in. A close correlation between rainfall (itself coupled with lack of sunshine and lower temperatures) and water temperature is evident. Not surprisingly, a cool clear day when the sun is out reduces water temperature much less than rainy days with no sun.
Very clearly we can see where pool temperatures typically reside in Sydney, dropping into the low teens through winter and climbing to the mid twenties in summer.
Personally 28 degrees is a really nice water temperature, 30 degrees is great but if I’m already hot and it’s a warm day it can be a little too close to bath water. On the flip side, 25 degrees or below I find uncomfortable for lounging about in the pool unless it’s a really hot day. Let’s take this as the desired water temperature range.
So if you have a pool in Sydney you need a way of increasing the water temperature if you want to extend the swimming season beyond the summer months. Let’s look at what’s working in the pools we manage.
The above chart shows the deviation above and below the group average of the pools we monitor. You would expect to see a bell curve on this data with lots of pools clustered around the mid point and then tailing off for those above and below average. There is a notably longer tail for pools with above average temperatures from customers that choose to heat their pools.
What makes a warm pool?
On the basis that we have an average water temperature of circa 25 degrees in summer but we’d all like our pool to be closer to 28 degrees, I had a closer look at the pools that were averaging these higher temperatures to find out how.
It’s important to stress that whilst most of the data in this post is presented on data sets which are large enough to be statistically reliable – this one is not. We don’t have a meaningful number of pools above average. There are other data points which follow that support some of these values though and they correlate closely with my anecdotal experience.
For the above data – some of the pools had only one component (e.g only solar heating) whilst others had a mix (a cover plus gas heating). All of the pools had at least one of solar heating, gas / heat pump heating or a cover.
The prevalence of covers is very notable here as they’re relatively uncommon (circa 10%) when looking across all pools we manage. Solar heating a little less so but it’s definitely over represented in the warmer pools against average.
What makes a cold pool?
Not surprisingly a lack of sun. No graph for this one and again it’s based on a small data set but observations from the data of the coldest pools are:
- They were consistently subject to heavy shading from trees or buildings.
- Two had mesh covers (used for safety or leaves) which are suspended over the pool (vs bubble covers that float on the water).
- A number had wet edges (also called infinity edges) or part of the pool wall was formed in glass.
- One was a fibreglass above ground pool.
A closer look at pool temperatures
Given the relatively small sample set of pools on the warmer and colder ends of the spectrum, I plotted a few examples of different pool configurations for a closer look:
This chart shows that covers and solar heating are very much up there in water temperature averages with pools heated via gas or heat pumps.
Pool construction effect on temperature
I was expecting to see a correlation in water temperature between pools which are submerged entirely in ground vs those either partially of completely above ground. I’ve not been able to find evidence of this in our data set. I’ve a number of ideas or theories as to why:
- It’s very common for swimming pools to be partially above ground in Sydney being a hilly and undulating city. To reduce construction costs if the property is on a slope excavation for the pool is regularly a mix of cut and fill.
- I’m reliant on site photos for identification of above ground vs in-ground or partially exposed. It’s not always clear when a pool is partially exposed.
- Taking the daily average of maximum and minimum ambient air temperatures across a full year in Sydney gives an average value of 19.05 degrees. I can’t find reliable values for average annual soil temperature at 1-2 m depths but multiple sources suggest it’s in the region of 19-20 degrees. The average of all daily water temperature readings we have available comes out at 19.14 degrees.
- In summary – the average ambient air temperature (affecting above ground pools) probably isn’t that much different to the average ground temperature (affecting in-ground pools) – hence not much difference in the average water temperature.
We really don’t yet have enough data to make reliable conclusions on this one but I expect that other factors such as shading, evaporation, wind, covers and heating will have more significant affects on swimming pool water temperature than the construction attributes.
Relationship between heating method and pool temperature
An unexpected insight from the data was that covers and solar heating are so prevalent in the pools our data showed to be warmer on average. Across the three types of pool heaters available – gas, heat pump and solar, the later has the lowest capacity in its ability to generate heat. Solar is the most energy efficient and cost effective to run, followed by heat pumps then gas. Gas on the other hand has the capacity to change the temperature of the pool most rapidly of the three options.
What our current data set suggests however is that people with gas or heat pumps simply aren’t using them as much or at all as those with solar heating. You can heat your pool if you buy a gas heater but you probably won’t! Given the significant cost of operating pool heating this is understandable. I suspect solar heaters and covers both get used a lot more by their owners.
If you want a warm pool:
- Make sure it gets sun on it
- Use a pool cover (more details in our post on the pros and cons of swimming pool covers)
- Install solar heating