scholarly journals Closing the gap on lower cost air quality monitoring: machine learning calibration models to improve low-cost sensor performance

2017 ◽  
Author(s):  
Naomi Zimmerman ◽  
Albert A. Presto ◽  
Sriniwasa P. N. Kumar ◽  
Jason Gu ◽  
Aliaksei Hauryliuk ◽  
...  
Keyword(s):  
Air Quality ◽  
Relative Error ◽  
Low Cost ◽  
Personal Exposure ◽  
Minimum Data ◽  
Calibration Models ◽  
The Us ◽  
Forest Models ◽  
Random Forest Models ◽  
Us Epa

Abstract. Low-cost sensing strategies hold the promise of denser air quality monitoring networks, which could significantly improve our understanding of personal air pollution exposure. Additionally, low-cost air quality sensors could be deployed to areas where limited monitoring exists. However, low-cost sensors are frequently sensitive to environmental conditions and pollutant cross-sensitivities, which have historically been poorly addressed by laboratory calibrations, limiting their utility for monitoring. In this study, we investigated different calibration models for the Real-time Affordable Multi-Pollutant (RAMP) sensor package, which measures CO, NO2, O3, and CO2. We explored three methods: 1) laboratory univariate linear regression, 2) empirical multivariate linear regression and 3) machine-learning based calibration models using random forests (RF). Calibration models were developed for 19 RAMP monitors using training and testing windows spanning August 2016 through February 2017 in Pittsburgh, PA. The random forest models matched (CO) or significantly outperformed (NO2, CO2, O3) the other calibration models, and their accuracy and precision was robust over time for testing windows of up to 16 weeks. Following calibration, average mean absolute error on the testing dataset from the random forest models was 38 ppb for CO (14 % relative error), 10 ppm for CO2 (2 % relative error), 3.5 ppb for NO2 (29 % relative error) and 3.4 ppb for O3 (15 % relative error), and Pearson r versus the reference monitors exceeded 0.8 for most units. Model performance is explored in detail, including a quantification of model variable importance, accuracy across different concentration ranges, and performance in a range of monitoring contexts including the National Ambient Air Quality Standards (NAAQS), and the US EPA Air Sensors Guidebook recommendations of minimum data quality for personal exposure measurement. A key strength of the RF approach is that it accounts for pollutant cross sensitivities. This highlights the importance of developing multipollutant sensor packages (as opposed to single pollutant monitors); we determined this is especially critical for NO2 and CO2. The evaluation reveals that only the RF-calibrated sensors meet the US EPA Air Sensors Guidebook recommendations of minimum data quality for personal exposure measurement. We also demonstrate that the RF model calibrated sensors could detect differences in NO2 concentrations between a near-road site and a suburban site less than 1.5 km away. From this study, we conclude that combining RF models with the RAMP monitors appears to be a very promising approach to address the poor performance that has plagued low cost air quality sensors.

scholarly journals A machine learning calibration model using random forests to improve sensor performance for lower-cost air quality monitoring

2018 ◽  
Vol 11 (1) ◽  
pp. 291-313 ◽  
Author(s):  
Naomi Zimmerman ◽  
Albert A. Presto ◽  
Sriniwasa P. N. Kumar ◽  
Jason Gu ◽  
Aliaksei Hauryliuk ◽  
...  
Keyword(s):  
Air Quality ◽  
Relative Error ◽  
Low Cost ◽  
Personal Exposure ◽  
Minimum Data ◽  
Calibration Models ◽  
The Us ◽  
Forest Models ◽  
Random Forest Models ◽  
Us Epa

Abstract. Low-cost sensing strategies hold the promise of denser air quality monitoring networks, which could significantly improve our understanding of personal air pollution exposure. Additionally, low-cost air quality sensors could be deployed to areas where limited monitoring exists. However, low-cost sensors are frequently sensitive to environmental conditions and pollutant cross-sensitivities, which have historically been poorly addressed by laboratory calibrations, limiting their utility for monitoring. In this study, we investigated different calibration models for the Real-time Affordable Multi-Pollutant (RAMP) sensor package, which measures CO, NO2, O3, and CO2. We explored three methods: (1) laboratory univariate linear regression, (2) empirical multiple linear regression, and (3) machine-learning-based calibration models using random forests (RF). Calibration models were developed for 16–19 RAMP monitors (varied by pollutant) using training and testing windows spanning August 2016 through February 2017 in Pittsburgh, PA, US. The random forest models matched (CO) or significantly outperformed (NO2, CO2, O3) the other calibration models, and their accuracy and precision were robust over time for testing windows of up to 16 weeks. Following calibration, average mean absolute error on the testing data set from the random forest models was 38 ppb for CO (14 % relative error), 10 ppm for CO2 (2 % relative error), 3.5 ppb for NO2 (29 % relative error), and 3.4 ppb for O3 (15 % relative error), and Pearson r versus the reference monitors exceeded 0.8 for most units. Model performance is explored in detail, including a quantification of model variable importance, accuracy across different concentration ranges, and performance in a range of monitoring contexts including the National Ambient Air Quality Standards (NAAQS) and the US EPA Air Sensors Guidebook recommendations of minimum data quality for personal exposure measurement. A key strength of the RF approach is that it accounts for pollutant cross-sensitivities. This highlights the importance of developing multipollutant sensor packages (as opposed to single-pollutant monitors); we determined this is especially critical for NO2 and CO2. The evaluation reveals that only the RF-calibrated sensors meet the US EPA Air Sensors Guidebook recommendations of minimum data quality for personal exposure measurement. We also demonstrate that the RF-model-calibrated sensors could detect differences in NO2 concentrations between a near-road site and a suburban site less than 1.5 km away. From this study, we conclude that combining RF models with carefully controlled state-of-the-art multipollutant sensor packages as in the RAMP monitors appears to be a very promising approach to address the poor performance that has plagued low-cost air quality sensors.

scholarly journals How emissions, climate, and land use change will impact mid-century air quality over the United States: a focus on effects at National Parks

2014 ◽  
Vol 14 (19) ◽  
pp. 26495-26543 ◽  
Author(s):  
M. Val Martin ◽  
C. L. Heald ◽  
J.-F. Lamarque ◽  
S. Tilmes ◽  
L. K. Emmons ◽  
...  
Keyword(s):  
United States ◽  
Land Use ◽  
Air Quality ◽  
National Parks ◽  
Land Use Changes ◽  
The United States ◽  
Western Us ◽  
The Us ◽  
Us Epa ◽  
Surface O3

Abstract. We use a global coupled chemistry-climate-land model (CESM) to assess the integrated effect of climate, emissions and land use changes on annual surface O3 and PM2.5 on the United States with a focus on National Parks (NPs) and wilderness areas, using the RCP4.5 and RCP8.5 projections. We show that, when stringent domestic emission controls are applied, air quality is predicted to improve across the US, except surface O3 over the western and central US under RCP8.5 conditions, where rising background ozone counteracts domestic emissions reductions. Under the RCP4.5, surface O3 is substantially reduced (about 5 ppb), with daily maximum 8 h averages below the primary US EPA NAAQS of 75 ppb (and even 65 ppb) in all the NPs. PM2.5 is significantly reduced in both scenarios (4 μg m−3; ~50%), with levels below the annual US EPA NAAQS of 12 μg m−3 across all the NPs; visibility is also improved (10–15 deciviews; >75 km in visibility range), although some parks over the western US (40–74% of total sites in the US) may not reach the 2050 target to restore visibility to natural conditions by 2064. We estimate that climate-driven increases in fire activity may dominate summertime PM2.5 over the western US, potentially offsetting the large PM2.5 reductions from domestic emission controls, and keeping visibility at present-day levels in many parks. Our study suggests that air quality in 2050 will be primarily controlled by anthropogenic emission patterns. However, climate and land use changes alone may lead to a substantial increase in surface O3 (2–3 ppb) with important consequences for O3 air quality and ecosystem degradation at the US NPs. Our study illustrates the need to consider the effects of changes in climate, vegetation, and fires in future air quality management and planning and emission policy making.

scholarly journals City Wide Participatory Sensing of Air Quality

2021 ◽  
Vol 9 ◽  
Author(s):  
Andrew Rebeiro-Hargrave ◽  
Pak Lun Fung ◽  
Samu Varjonen ◽  
Andres Huertas ◽  
Salla Sillanpää ◽  
...  
Keyword(s):  
Air Pollution ◽  
Air Quality ◽  
Low Cost ◽  
Age Groups ◽  
Personal Exposure ◽  
Urban Environments ◽  
Sensor Data ◽  
Automated System ◽  
Participatory Sensing ◽  
Good Correspondence

Air pollution is a contributor to approximately one in every nine deaths annually. Air quality monitoring is being carried out extensively in urban environments. Currently, however, city air quality stations are expensive to maintain resulting in sparse coverage and data is not readily available to citizens. This can be resolved by city-wide participatory sensing of air quality fluctuations using low-cost sensors. We introduce new concepts for participatory sensing: a voluntary community-based monitoring data forum for stakeholders to manage air pollution interventions; an automated system (cyber-physical system) for monitoring outdoor air quality and indoor air quality; programmable platform for calibration and generating virtual sensors using data from low-cost sensors and city monitoring stations. To test our concepts, we developed a low-cost sensor to measure particulate matter (PM2.5), nitrogen dioxide (NO2), carbon monoxide (CO), and ozone (O3) with GPS. We validated our approach in Helsinki, Finland, with participants carrying the sensor for 3 months during six data campaigns between 2019 and 2021. We demonstrate good correspondence between the calibrated low-cost sensor data and city’s monitoring station measurements. Data analysis of their personal exposure was made available to the participants and stored as historical data for later use. Combining the location of low cost sensor data with participants public profile, we generate proxy concentrations for black carbon and lung deposition of particles between districts, by age groups and by the weekday.

scholarly journals Stationary and portable multipollutant monitors for high-spatiotemporal-resolution air quality studies including online calibration

2021 ◽  
Vol 14 (2) ◽  
pp. 995-1013
Author(s):  
Colby Buehler ◽  
Fulizi Xiong ◽  
Misti Levy Zamora ◽  
Kate M. Skog ◽  
Joseph Kohrman-Glaser ◽  
...  
Keyword(s):  
Air Quality ◽  
Low Cost ◽  
Personal Exposure ◽  
Atmospheric Pollutants ◽  
Sensor Data ◽  
Concentration Data ◽  
Spatiotemporal Resolution ◽  
Online Calibration ◽  
High Spatiotemporal Resolution ◽  
Wide Range

Abstract. The distribution and dynamics of atmospheric pollutants are spatiotemporally heterogeneous due to variability in emissions, transport, chemistry, and deposition. To understand these processes at high spatiotemporal resolution and their implications for air quality and personal exposure, we present custom, low-cost air quality monitors that measure concentrations of contaminants relevant to human health and climate, including gases (e.g., O3, NO, NO2, CO, CO2, CH4, and SO2) and size-resolved (0.3–10 µm) particulate matter. The devices transmit sensor data and location via cellular communications and are capable of providing concentration data down to second-level temporal resolution. We produce two models: one designed for stationary (or mobile platform) operation and a wearable, portable model for directly measuring personal exposure in the breathing zone. To address persistent problems with sensor drift and environmental sensitivities (e.g., relative humidity and temperature), we present the first online calibration system designed specifically for low-cost air quality sensors to calibrate zero and span concentrations at hourly to weekly intervals. Monitors are tested and validated in a number of environments across multiple outdoor and indoor sites in New Haven, CT; Baltimore, MD; and New York City. The evaluated pollutants (O3, NO2, NO, CO, CO2, and PM2.5) performed well against reference instrumentation (e.g., r=0.66–0.98) in urban field evaluations with fast e-folding response times (≤ 1 min), making them suitable for both large-scale network deployments and smaller-scale targeted experiments at a wide range of temporal resolutions. We also provide a discussion of best practices on monitor design, construction, systematic testing, and deployment.

MOPGA/Improving Air Quality in West Africa: Low-cost sensors as a solution to improve the understanding of spatial and temporal variability in urban air pollution

2021 ◽  
Author(s):  
Julien Bahino ◽  
Michael Giordano ◽  
Véronique Yoboué ◽  
Arsène Ochou ◽  
Corinne Galy-Lacaux ◽  
...  
Keyword(s):  
Air Pollution ◽  
Air Quality ◽  
West Africa ◽  
Low Cost ◽  
West African ◽  
Passive Samplers ◽  
Quality Monitoring ◽  
Spatial And Temporal Variability ◽  
Air Quality Monitoring ◽  
Calibration Models

<p>This study was carried out within the framework of the Improving Air Quality in West Africa<strong> </strong>(IAQWA) project funded by the Make Our Planet Great Again (MOPGA) program. In recent years, West African countries have experienced an economic upturn driven by GDP growth of nearly 3.7% in 2019 (AfDB, 2020). This economic boom is mainly felt in the cities where it promotes the construction of highway infrastructure, real estate development, and industry. All these activities are sources of air pollution. Unfortunately, there is almost no air quality monitoring in these cities partly due to the high cost of monitoring instruments. Low-cost air quality monitoring instruments can help improve the spatial and temporal resolution of measurements at relatively lower cost. However, the installation of these instruments in West African environments characterized by high relative humidity requires their calibration through collocation with reference instruments. The IAQWA project aims to improve our understanding of air pollutants such as fine particulate matter mass (PM<sub>2.5</sub>), ozone (O<sub>3</sub>), nitrogen oxides (NOx), sulfur dioxide (SO<sub>2</sub>), and carbon monoxide (CO) in Abidjan and Accra, two major West African capitals, through the deployment of Real-time Affordable Multi-Pollutant (RAMP) monitors.</p><p>Since February 2020, five RAMPs have been installed and are operating continuously at various sites in Abidjan and Lamto in Cote d'Ivoire, and four RAMPs have been operating in Accra, Ghana. Some of the RAMPs have been collocated with PM and/or NOx reference instruments. At other sites the RAMPs have been collocated with INDAAF passive samplers and passive aerosol collectors. These collocations have allowed for the development of calibration models for these low-cost sensors. The performance of these calibration models is presented here along with the diurnal and seasonal variations of air pollution at the different sites in Abidjan and Accra. These results will eventually be used to improve our understanding of the drivers of air pollution in these major West African cities, which is essential to choosing sustainable development pathways in the future.</p>

scholarly journals An investigation into seasonal and regional aerosol characteristics in East Asia using model-predicted and remotely-sensed aerosol properties

2008 ◽  
Vol 8 (22) ◽  
pp. 6627-6654 ◽  
Author(s):  
C. H. Song ◽  
M. E. Park ◽  
K. H. Lee ◽  
H. J. Ahn ◽  
Y. Lee ◽  
...  
Keyword(s):  
Air Quality ◽  
East Asia ◽  
Chemical Species ◽  
Temporal Variations ◽  
Aerosol Formation ◽  
Model Simulations ◽  
Aerosol Retrieval ◽  
Modis Aod ◽  
The Us ◽  
Us Epa

Abstract. In this study, the spatio-temporal and seasonal distributions of EOS/Terra Moderate Resolution Imaging Spectroradiometer (MODIS)-derived aerosol optical depth (AOD) over East Asia were analyzed in conjunction with US EPA Models-3/CMAQ v4.3 modeling. In this study, two MODIS AOD products (τMODIS: τM-BAER and τNASA) retrieved through a modified Bremen Aerosol Retrieval (M-BAER) algorithm and NASA collection 5 (C005) algorithm were compared with the AOD (τCMAQ) that was calculated from the US EPA Models-3/CMAQ model simulations. In general, the CMAQ-predicted AOD values captured the spatial and temporal variations of the two MODIS AOD products over East Asia reasonably well. Since τMODIS cannot provide information on the aerosol chemical composition in the atmosphere, different aerosol formation characteristics in different regions and different seasons in East Asia cannot be described or identified by τMODIS itself. Therefore, the seasonally and regionally varying aerosol formation and distribution characteristics were investigated by the US EPA Models-3/CMAQ v4.3 model simulations. The contribution of each particulate chemical species to τMODIS and τCMAQ showed strong spatial, temporal and seasonal variations. For example, during the summer episode, τMODIS and τCMAQ were mainly raised due to high concentrations of (NH4)2SO4 over Chinese urban and industrial centers and secondary organic aerosols (SOAs) over the southern parts of China, whereas during the late fall and winter episodes, τMODIS and τCMAQ were higher due largely to high levels of NH4NO3 formed over the urban and industrial centers, as well as in areas with high NH3 emissions. τCMAQ was in general larger than τMODIS during the year, except for spring. The high biases (τCMAQ>τMODIS) may be due to the excessive formation of both (NH4)2SO4 (summer episode) and NH4NO3 (fall and winter episodes) over China, possibly from the use of overestimated values for NH3 emissions in the CMAQ modeling. According to CMAQ modeling, particulate NH4NO3 made a 14% (summer) to 54% (winter) contribution to σext and τCMAQ. Therefore, the importance of NH4NO3 in estimating τ should not be ignored, particularly in studies of the East Asian air quality. In addition, the accuracy of τM-BAER and τNASA was evaluated by a comparison with the AOD (τAERONET) from the AERONET sites in East Asia. Both τM-BAER and τNASA showed a strong correlation with τAERONET around the 1:1 line (R=0.79), indicating promising potential for the application of both the M-BAER and NASA aerosol retrieval algorithms to satellite-based air quality monitoring studies in East Asia.

scholarly journals GeoAir—A Novel Portable, GPS-Enabled, Low-Cost Air-Pollution Sensor: Design Strategies to Facilitate Citizen Science Research and Geospatial Assessments of Personal Exposure

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3761
Author(s):  
Yoo Min Park ◽  
Sinan Sousan ◽  
Dillon Streuber ◽  
Kai Zhao
Keyword(s):  
Air Pollution ◽  
Air Quality ◽  
Low Cost ◽  
Rapid Evolution ◽  
Personal Exposure ◽  
Quality Data ◽  
Portable Devices ◽  
Design Strategies ◽  
Laboratory Test Results ◽  
Voc Sensor

The rapid evolution of air sensor technologies has offered enormous opportunities for community-engaged research by enabling citizens to monitor the air quality at any time and location. However, many low-cost portable sensors do not provide sufficient accuracy or are designed only for technically capable individuals by requiring pairing with smartphone applications or other devices to view/store air quality data and collect location data. This paper describes important design considerations for portable devices to ensure effective citizen engagement and reliable data collection for the geospatial analysis of personal exposure. It proposes a new, standalone, portable air monitor, GeoAir, which integrates a particulate matter (PM) sensor, volatile organic compound (VOC) sensor, humidity and temperature sensor, LTE-M and GPS module, Wi-Fi, long-lasting battery, and display screen. The preliminary laboratory test results demonstrate that the PM sensor shows strong performance when compared to a reference instrument. The VOC sensor presents reasonable accuracy, while further assessments with other types of VOC are needed. The field deployment and geo-visualization of the field data illustrate that GeoAir collects fine-grained, georeferenced air pollution data. GeoAir can be used by all citizens regardless of their technical proficiency and is widely applicable in many fields, including environmental justice and health disparity research.

scholarly journals IMPACT OF DUST PARTICLES ON AIR QUALITY AND ASSOCIATED HUMAN HEALTH IN ILORIN NIGERIA

2020 ◽  
Vol 4 (3) ◽  
pp. 481-488
Author(s):  
Mukhtar Balarabe ◽  
Bello Saadu
Keyword(s):  
Air Quality ◽  
Human Health ◽  
Atmospheric Condition ◽  
Dust Particles ◽  
High Concentration ◽  
Total Suspended Particles ◽  
The Us ◽  
Us Epa ◽  
Study Sites ◽  
The Impact

To improve our understanding of the impact of desert dust on human health, there is need to constantly monitor and examined the dust related phenomena. Therefore, twenty 20 year’s (1998–2018) data of visibility for Ilorin Nigeria were used to estimate the concentrations of the Total Suspended Particles (TSP) and Particulate Matter PM10 as usually used to monitor air quality on international level. The results established the threshold for daily concentration of TSP (254) and PM10 (186) μgm−3 at the study sites. It also identified months (November-March) of the following year with the greatest number of days having low air quality (high concentration of TSP and PM10). These months are responsible for 47% of the annual air pollution and number of days above the US EPA-NAAQSTSP, US EPA-NAAQS PM10 as well as the 24-hour EU-LVAQ regulations, respectively. Furthermore, some considerable numbers of days were found to experienced hazardous atmospheric condition for the total number of days, Harmattan and summer respectively. The concentrations of PM10 (0-54 μgm−3) showed absence of good air quality throughout the period of study. Even though, there were significant number of days associated with moderate air quality most of which occurs during summer. Consequence of which can lead to increased respiratory symptoms and aggravation of lung diseases. It was also observed that, the concentrations of TSP and PM10 start of build up in the atmosphere by October, reaching peak in December and January before it decline by April and remain low with almost uniform values until September.

Sign in / Sign up

Export Citation Format

Share Document