When do CONA participants light their wood fires?

This initial analysis is going to focus on seven homes from which we were collecting data over 12 days in August. As part of the sign up agreement we promised not to share private data with anyone else. For that reason we will not be identifying which homes are which in this, or any similar analysis.

Why are we interested in fire-lighting? To understand air quality in the town it matters a lot to know WHEN smoke is being emitted from people’s chimneys, and from how many. One of the things we know very little about is how much people vary the time they light their fire, or whether we are all creatures of habit and have the same fire-lighting routine each night. Does it depend on the weather, the design of our homes, our lifestyles?

We know when our volunteers lit their fires because they had a thermocouple – a device which can measure high temperatures – fixed to the flue of their woodburner. When a fire is lit the temperature recorded by the thermocouple jumps up tens of degrees in just a few minutes. Nothing else would cause this to happen. The thermocouple records the time and we determine to within about ten minutes when the fire was lit. This is also true when the fire is ‘re-lit’ or boosted after it has been dying away. You can see an example above from one of our homes where there was a big jump in flue temperature at 7:30pm. You can also see the room temperature (dashed orange line) warming up by 8 degrees over the next 2 hours. Earlier in the afternoon there had been a slight rise in room temperature but the flue was the same temperature indicating that this was probably due to the sun warming the home, not the fire. After 9:30pm the room, and the flue, are cooling down again as the fire is left to gradually die away.

Looking at a bar chart of the time (to the nearest hour) when a fire was lit for all homes pooled together you can see that the most popular time for lighting a fire was between 6 and 7 pm on both weekdays and weekends. However, some fires were also being lit at 7 - 8am and 11 am – noon. Also it can be seen that we have records of at least one fire being lit for every hour of the day, except 2 – 3 am and 4 – 5 am.

Looking at the seven homes individually there was quite a range of different patterns. Firstly, there was more fire-lighting on weekends than on weekdays (by about 20% on average). During the weekdays two of our homes lit their fires only once per day. They both clearly had a steady routine as one regularly lit their fire between 7pm and 8pm while the other one lit their fire between 3:45pm and 5pm. In contrast only two of our homes lit their fires regularly in the morning between 6:30am and 8am. These two homes also lit their fires twice, and sometimes three times in the evening. Another home had a different routine, lighting their fire in the middle of the day and then again after 9:20pm. Much of these differences between homes will be due to whether the occupants work and how many people are at home. It may also depend on how warm they like their home to be. But the data does also hint that some homes are requiring more heating than others. We may look at that in more depth in the future.

During this early part of the study in August, there were three distinct phases of weather – cool nights until the 12^th/13^th, followed by 6 warm nights warm nights (one cold night in the middle) and then a series of cold nights. Over the short sample being considered here there does not seem to be any obvious change in fire-lighting behaviour in response to these weather changes. But as we collect longer datasets like this into the future it may be possible to detect how heating habits change through the winter.

In upcoming analysis we plan to look at how warm and cold each got, why and how they compare to each other.