Date of Award
Master of Science (MS)
Yeo H. Lim
In November 2022, Allegheny County residents raised concerns about increased air pollution, especially in the Clairton region area. Due to the recent air inversion, the Clairton region was regarded as a highly polluted area because of the concentrated hydrogen sulfide smell invading the area and the exceedance of the daily PM2.5 average. Consequently, Allegheny County Health Department installed low-cost sensors (PurpleAir) in Mon Valley regions to increase the air monitoring rate and verify the efficiency of federal Reference Methods in place. Low-cost sensors are a relatively new technology adopted in air monitoring. Therefore, this project seeks to assess the effectiveness of these sensors relative to the reference method (FRM) while considering only PM2.5 data for the two methods. According to the EPA, establishing a relationship between the sensor and FRM data requires plotting the variables on a correlation graph and generating an equation of a line with a coefficient of determination (R2). The coefficient of determination is given a value of 1 and the closer the value of R2 generated from the plot is to 1, the more likely the sensor behaves like FRM. According to the analysis of the data collected from Allegheny County Health Department using Microsoft Excel and Minitab, R2 = 0.52, 0.69, and 0.12 for three different sensors. Then, a correction was made to remove the effect of relative humidity consequently, R2 = 0.63, 0.67, and 0.09. Also, during the analysis, it was observed that the sensors did not generate data for a certain number of days, and as a result, the corresponding number of days was removed from the FRM data to enable correlation. For better visualization, an air quality index (AQI) was generated for the sensor data with corresponding AQI categories, and a heat map was developed for the sensor within the study area. Finally, analysis shows that the first-second sensor established moderate relationships with the reference instrument and the third sensor did not establish any relationship. The sensor was able to detect the presence of pollution but could not generate accurate concentration. This work could not establish why sensor three has a low R2 but an assumption was made to conclude that the poor performance of the third sensor is either due to the fact that the sensor is bad or early degradation occurred and probably the effect of an unforeseen environmental factor. Therefore, more work needs to be done in the future so that the sensors can be reassessed during other seasons to compare their performance.
Tijani, Issa Adegbola, "Assessing The Effectiveness Of Low-Cost Sensors As Probable Alternatives For Conventional Air Monitoring Devices" (2023). Theses and Dissertations. 5346.