Non-linear probabilistic calibration of low-cost environmental air pollution sensor networks for neighborhood level spatiotemporal exposure assessment.

BACKGROUND: Low-cost sensor networks for monitoring air pollution are an effective tool for expanding spatial resolution beyond the capabilities of existing state and federal reference monitoring stations. However, low-cost sensor data commonly exhibit non-linear biases with respect to environmental conditions that cannot be captured by linear models, therefore requiring extensive lab calibration. Further, these calibration models traditionally produce point estimates or uniform variance predictions which limits their downstream in exposure assessment. OBJECTIVE: Build direct field-calibration models using probabilistic gradient boosted decision trees (GBDT) that eliminate the need for resource-intensive lab calibration and that can be used to conduct probabilistic exposure assessments on the neighborhood level. METHODS: Using data from Plantower A003 particulate matter (PM) sensors deployed in Baltimore, MD from November 2018 through November 2019, a fully probabilistic NGBoost GBDT was trained on raw data from sensors co-located with a federal reference monitoring station and compared against linear regression trained on lab calibrated sensor data. The NGBoost predictions were then used in a Monte Carlo interpolation process to generate high spatial resolution probabilistic exposure gradients across Baltimore. RESULTS: We demonstrate that direct field-calibration of the raw PM2.5 sensor data using a probabilistic GBDT has improved point and distribution accuracies compared to the linear model, particularly at reference measurements exceeding 25 µg/m3, and also on monitors not included in the training set. SIGNIFICANCE: We provide a framework for utilizing the GBDT to conduct probabilistic spatial assessments of human exposure with inverse distance weighting that predicts the probability of a given location exceeding an exposure threshold and provides percentiles of exposure. These probabilistic spatial exposure assessments can be scaled by time and space with minimal modifications. Here, we used the probabilistic exposure assessment methodology to create high quality spatial-temporal PM2.5 maps on the neighborhood-scale in Baltimore, MD. IMPACT STATEMENT: We demonstrate how the use of open-source probabilistic machine learning models for in-place sensor calibration outperforms traditional linear models and does not require an initial laboratory calibration step. Further, these probabilistic models can create uniquely probabilistic spatial exposure assessments following a Monte Carlo interpolation process.

Evaluating the Performance of Using Low-Cost Sensors to Calibrate for Cross-Sensitivities in a Multipollutant Network.

As part of our low-cost sensor network, we colocated multipollutant monitors containing sensors for particulate matter, carbon monoxide, ozone, nitrogen dioxide, and nitrogen monoxide at a reference field site in Baltimore, MD, for 1 year. The first 6 months were used for training multiple regression models, and the second 6 months were used to evaluate the models. The models produced accurate hourly concentrations for all sensors except ozone, which likely requires nonlinear methods to capture peak summer concentrations. The models for all five pollutants produced high Pearson correlation coefficients (r > 0.85), and the hourly averaged calibrated sensor and reference concentrations from the evaluation period were within 3-12%. Each sensor required a distinct set of predictors to achieve the lowest possible root-mean-square error (RMSE). All five sensors responded to environmental factors, and three sensors exhibited cross-sensitives to another air pollutant. We compared the RMSE from models (NO2, O3, and NO) that used colocated regulatory instruments and colocated sensors as predictors to address the cross-sensitivities to another gas, and the corresponding model RMSEs for the three gas models were all within 0.5 ppb. This indicates that low-cost sensor networks can yield useable data if the monitoring package is designed to comeasure key predictors. This is key for the utilization of low-cost sensors by diverse audiences since this does not require continual access to regulatory grade instruments.

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

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.

Statistical field calibration of a low-cost PM2.5 monitoring network in Baltimore.

Low-cost air pollution monitors are increasingly being deployed to enrich knowledge about ambient air-pollution at high spatial and temporal resolutions. However, unlike regulatory-grade (FEM or FRM) instruments, universal quality standards for low-cost sensors are yet to be established and their data quality varies widely. This mandates thorough evaluation and calibration before any responsible use of such data. This study presents evaluation and field-calibration of the PM2.5 data from a network of low-cost monitors currently operating in Baltimore, MD, which has only one regulatory PM2.5 monitoring site within city limits. Co-location analysis at this regulatory site in Oldtown, Baltimore revealed high variability and significant overestimation of PM2.5 levels by the raw data from these monitors. Universal laboratory corrections reduced the bias in the data, but only partially mitigated the high variability. Eight months of field co-location data at Oldtown were used to develop a gain-offset calibration model, recast as a multiple linear regression. The statistical model offered substantial improvement in prediction quality over the raw or lab-corrected data. The results were robust to the choice of the low-cost monitor used for field-calibration, as well as to different seasonal choices of training period. The raw, lab-corrected and statistically-calibrated data were evaluated for a period of two months following the training period. The statistical model had the highest agreement with the reference data, producing a 24-hour average root-mean-square-error (RMSE) of around 2 µg m -3. To assess transferability of the calibration equations to other monitors in the network, a cross-site evaluation was conducted at a second co-location site in suburban Essex, MD. The statistically calibrated data once again produced the lowest RMSE. The calibrated PM2.5 readings from the monitors in the low-cost network provided insights into the intra-urban spatiotemporal variations of PM2.5 in Baltimore.

Compact, Automated, Inexpensive, and Field-Deployable Vacuum-Outlet Gas Chromatograph for Trace-Concentration Gas-Phase Organic Compounds.

The identification and quantification of gas-phase organic compounds, such as volatile organic compounds (VOCs), frequently use gas chromatography (GC), which typically requires high-purity compressed gases. We have developed a new instrument for trace-concentration measurements of VOCs and intermediate-volatility compounds of up to 14 carbon atoms in a fully automated (computer-free), independent, low-cost, compact GC-based system for the quantitative analysis of complex mixtures without the need for compressed, high-purity gases or expensive detectors. Through adsorptive analyte preconcentration, vacuum GC, photoionization detectors, and need-based water-vapor control, we enable sensitive and selective measurements with picogram-level limits of detection (i.e., under 15 ppt in a 4 L sample for most compounds). We validate performance against a commercial pressurized GC, including resolving challenging isomers of similar volatility, such as ethylbenzene and  m/ p-xylene. We employ vacuum GC across the whole column with filtered air as a carrier gas, producing long-term system stability and performance over a wide range of analytes. Through theory and experiments, we present variations in analyte diffusivities in the mobile phase, analyte elution temperatures, optimal linear velocities, and separation-plate heights with vacuum GC in air at different pressures, and we optimize our instrument to exploit these differences. At 2-6 psia, the molecular diffusion coefficients are 6.4-2.1 times larger and the elution temperatures are 39-92 °C lower than with pressurized GC with helium (at 30 psig) depending on the molecular structure, and we find a wide range of optimal linear velocities (up to 60 cm s-1) that are faster with broader tolerances than with pressurized-N2 GC.

Field and Laboratory Evaluations of the Low-Cost Plantower Particulate Matter Sensor.

Due to the rapid development of low-cost air-quality sensors, a rigorous scientific evaluation has not been conducted for many available sensors. We evaluated three Plantower PMS A003 sensors when exposed to eight particulate matter (PM) sources (i.e., incense, oleic acid, NaCl, talcum powder, cooking emissions, and monodispersed polystyrene latex spheres under controlled laboratory conditions and also residential air and ambient outdoor air in Baltimore, MD). The PM2.5 sensors exhibited a high degree of precision and R2 values greater than 0.86 for all sources, but the accuracy ranged from 13 to >90% compared with reference instruments. The sensors were most accurate for PM with diameters below 1 µm, and they poorly measured PM in the 2.5-5 µm range. The accuracy of the sensors was dependent on relative humidity (RH), with decreases in accuracy at RH > 50%. The sensors were able to produce meaningful data at low and high temperatures and when in motion, as it would be if utilized for outdoor or personal monitoring applications. It was most accurate in environments with polydispersed particle sources and may not be useful in specialized environments or experiments with narrow distributions of PM or aerosols with a large proportion of coarse PM.