Frog Twaddle

My partner and I recently moved back east to be closer to family and we settled on buying a home in upstate New York. The climate here is significantly different than the milder winters and summers of Portland, Oregon that we had become accustomed to. The landscape, flora, and fauna are also different. This variation should be no surprise given the geologic history of the area. The homes are very different too.

We found a home built in 1922 in the craftsman style. We love it. This is our first home that is older than we are — significantly so.

It’s also worth noting that we both work from home and so we spend a significant number of our hours in our new abode. While we’ve worked from home for the past decade, this is the first time we’ve done it in a century-old house. That made me curious about the air we were breathing and so we decided to do some environmental investigation.

Data Collection

When we first moved in, we discovered the previous owners had repaired the air intake for the furnace. Unfortunately, they had used PVC glue which has a strong odor. After the first week in the house, the odor had not noticeably diminished and so we put some sensors in place that could at least detect the presence of VOCs. We placed three Eve Room sensors which measure and record some basic indoor air quality information: VOCs, temperature, and humidity.

There are several things I like about the Eve Sensors. First, they are small and inconspicuous in the spaces where we placed them. Second, they can run from several weeks or months on a single charge which means that we don’t need to keep them plugged in (I hate cords). Third, they have an application that syncs with the devices and allows me to download the individual measurements to a csv file. Fourth, and finally, they integrate with HomeKit which means they were easy to add to our smart home hub and apps.

After the sensors had run in the house for a week, were happy to learn that VOCs were not elevated in the house. The house did still smell bad though so we hired an environmental engineer to assess our home.

We did a little research and found a company out of Rochester, NY called Environmental Engineering that specializes in home environment remediation. They came out and set up a variety of sensors for VOCs and CO2 as well as took air samples to check for any mold or other biological incursions. A few weeks later we had a thorough report of our air quality and an hour long consultation with a remediation expert. Thankfully, there was nothing of immediate concern. During our call though, we inquired about a spike in radon that had showed up during our initial home inspection. The house passed the inspection but the spike seemed out of place. The environmental engineer leading the consult recommended we purchase a radon detector that we could leave in place for a few months to really gauge our exposure risk. That led to us putting an Air Things sensor into our basement.

We went with the Air Things Wave which measures radon as well as temperature, pressure, humidity, VOCs, and CO2. I was a little bummed to learn that we could not directly connect it to our home but nevertheless, we were able to export the measurements over time as a CSV file. It took a month before it would start to register radon (as per design) and we needed three months of measurements.

After the first month of measurements, we got our first radon reading and we were in the green and well under the 2.7 pCi/L level the EPA and WHO recommend. What we learned though, is that radon levels can vary pretty wildly. So we waited as we collected the three months of data needed to see where we really stood.

Data Types

Radon

The journey that resulted in this post really started with our concern for the levels of radon in our new old-home. While we waited for enough data to accumulate so we could make an informed decision about what next steps we should take, I collected and read several studies and papers published in reputable journals, the EPA, and the World Health Organization. I learned that radon is everywhere, which levels are dangerous, how is gets into our homes, and what remediation strategies are available. The short of it is, radon is the number two cause of lung cancer, second only to smoking. So knowing what radon levels are present in our home is definitely useful. While I knew our risk was likely low (based on the initial readings we were getting) I was eager to see what the full dataset would say.

Humidity

In addition to the radon, we had also taken an interest in humidity. Homes with basements, particularly older homes, can easily have environments above 60% sustained humidity which will lead to molds and mildew growth. Happily all our sensors were collecting this data as well.

Others

It should be obvious at this point that I was becoming interested in fully understanding our air quality. I was going down the rabbit hole.

Data Analysis

After more than three months of collecting data it was finally time to do some analysis. For this stage of the game I relied on my science degree and my computer programming skills. Since radon was our primary concern, I started to dive into those values first.

As you can see in the table below, we had collected almost 3,500 readings. We averaged 1.7 pCi/L over that period.

	radon pCi/L
count	3423
mean	1.710870581
std	0.679612862
min	0.68
25%	1.24
50%	1.57
75%	1.97
max	5.65

That max value of 5.65 had me a little nervous so I also took a look at a few selected quantiles.

Quantile	radon pCi/L
.99	4.1
.95	3.05
.90	2.54
.98	3.78

From this, I am reading that max as a definite outlier. I’ll not worry too much about it since it is well below the 8 pCi/L value the WHO sets as their absolute max. (To be clear, they do say that anything above 4 is actionable though.)

I also wanted to know how the radon related to the other measurements I took. So I built a correlative heat map comparing all the values I had from just the Air Things sensor. You can see in the chart below that most things aren’t really related BUT there is something going on between humidity and radon.

Hmmmm…. interesting. I dove in a little deeper to the relationship between these two features.

It’s pretty clear in that last chart that radon and humidity are linked. Of course, humidity is not a source of radon (or vice-versa). But it is possible that they share a common cause.

After tracking the humidity, we decided to put in a dedicated dehumidifier for the basement and, as I write this, our unit is waiting downstairs for installation this coming Tuesday. I’m guessing that we’ll see the correlation between radon and humidity broken once the humidity is maintained at a steady 45%. I am curious if there are any knock-on effects that we’ll see in the radon but if so… I’m sure it won’t be large. My working theory is that the gaps in the foundation walls and flooring are letting both radon and excess moisture into the basement. We’re in the process of repairing those gaps now but it will be another few months before we’re finished. Once it's done, I’d expect the dehumidifier to run less frequently. And, I hope, the radon levels will stabilize and decrease. Of course, if they don’t we’ll likely install a radon abatement system for peace of mind. But in the meantime, I’m happy to have this data which gives us a clue about what’s happening with our environment.

Side Note

I also did a comparison of the three Eve humidity sensors just to see how each floor of the house related to one another. You can see just how closely the humidity levels on each floor compare in the two graphs below.

What’s Next

As I mentioned above, the dehumidifier is being installed this Tuesday. We also have a call out to Environmental Engineering to review the data and make a decision about radon abatement.

I’ll write a follow up entry in a couple months once we have some additional post dehumidifier installation. And possibly again after the radon abatement is in place.

Additional Info

For those who would like to dig into the analysis further, I've posted my Jupyter notebook on GitHub here: Air Quality Analysis