I managed to grant permission from the local authorities (Junta de Andalucia) here in my city Almeria to place two low-cost monitors among the reference instruments in order to determine whether the two consumer products could offer fair readings to the users.

After filing some paperwork and the help of the responsible people in my city, they granted me access to the reference Monitor ES1393A (Mediterráneo) that complies with all EU standards. It is a quite powerful station as it is equipped with many expensive instruments (figure 1) which measure many types of pollutants, such as Nitrogen Oxide (NO), Nitrogen Dioxide (NO2), ground-level Ozone (O3), Carbon Monoxide (CO), Sulphur Dioxide (SO2), PM10, PM2.5 and many more.

The portable consumer-grade monitors that I placed in the reference AQ station were the Atmotube Pro and the Flow 2. Of course, I placed them outside the station and I attached them to the PM2.5 beta attenuation monitor air inlet (as you can see in figure 2) because I wanted to ensure the greatest PM2.5 correlation possible.

A beta attenuation monitor (BAM) is a universally used air monitoring technique that employs the absorption of beta radiation by solid particles extracted from airflow. This technique allows for the precise detection of PM2.5 without humidity affecting the results (a common problem for the laser scattering sensors). However, there are some drawbacks, for instance, it doesn’t provide data in real-time and it takes only a few measurements in an hour.

Figure 1: Air Pollution Monitors

Figure 2: BAM air inlet and Portable AQ Monitors

The portable monitors took measurements continuously for three and a half days with an interval of 1 minute between each measurement. The reference monitors took measurements every 10 minutes, so the data from the reference monitor are not as dense as from the Atmotube and Flow but are much more accurate and even without the same density we will be able to get a good idea of what is going on.

Before presenting the results from this comparison, I would like to mention that my measurements cannot be taken as a “de facto”. In order to determine the real correlation of a monitor or a sensor, you need at least 3 units and various days to months of data. Maybe some other units from both parties, PlumeLabs and Atmotube, may give different results. The climate also affects the performance of the sensors, so a different geographical location can give us different results.

Nitrogen Dioxide

Let’s begin with the comparison of the NO2 gas sensor inside the Flow 2 and the reference monitor Teledyne Model 200E. The official measurements were in μg/m3 so I had to convert them into parts per billion (ppb). Converting the original values to ppb is not an easy task as you have to know the molecular weight of the pollutants (NO2=46.0055g/mol) and most importantly the temperature and atmospheric pressure at the time each value was taken, the problem was that they didn’t provide me with that data and I didn’t have them either. As a result, I looked them up and the average values were 15°C/59°F and 1019hPa, so to simplify my workflow I used those numbers in the formula for all the conversions. I know, I know, the data from the reference monitor are not 100% correct now but they are very close so they will do the trick.

As you can see in the graph above there are quite many similarities in the trajectory of the lines. I was quite surprised and happy to be honest. You can clearly see how the low-cost sensor keeps getting exited from the pollutant(s) for a longer time but it was able to follow the official measurements. Keep that in mind, the official monitor Teledyne Model 200E worths around 8,000€ and Flow worths 159€.

Particulate Matter

Let’s continue our comparison with the Particulate Matter sensors from Atmotube Pro and Flow 2. Firstly, let’s see the PM10 measurements as the reference monitor Met One Instrument BAM 1020 (~18,000€) only provided the PM10 concentrations. The air inlet for the PM10 was about a meter away from where the 2 portable AQ monitors were placed and this is one of the reasons we can see that the concentrations from the BAM 1020 were much higher than the low-cost sensors.

Atmotube was able to follow the same trajectory as the BAM 1020 and when I multiply by x3.0 the output concentrations of the Atmotube the trend was much more similar (see graph below). On the other hand, Flow 2 was not able to keep up with the reference monitor. Flow’s (Yellow) line does not follow the reference monitor and I was unable to find any correlation.

As the BAM PM2.5 measurements are being analysed in a lab right now in order to ensure accuracy by the state I will compare the result in a different post. They gave me an estimation of about a month.

Conclusion

It gives you a level of confidence to know that the little AQ monitor that you carry with you is able to reveal the truth about the air you breathe. Although the results are not perfect they can provide some general guidelines to the users.

Stay tuned for the final PM2.5 comparison, will Flow 2 have better PM2.5 measurements and will Atmotube Pro keep up?