Grantee Research Project Results
Final Report: The Hawaii Island Volcanic Smog Sensor Network (HI-Vog): Tracking airquality and community engagement near a major emissions hotspot
EPA Grant Number: R836183Title: The Hawaii Island Volcanic Smog Sensor Network (HI-Vog): Tracking airquality and community engagement near a major emissions hotspot
Investigators: Kroll, Jesse H. , Heald, Colette L.
Institution: Massachusetts Institute of Technology , The Kohala Center
EPA Project Officer: Callan, Richard
Project Period: May 1, 2016 through April 30, 2019 (Extended to April 30, 2022)
Project Amount: $750,000
RFA: Air Pollution Monitoring for Communities (2014) RFA Text | Recipients Lists
Research Category: Air , Air Quality and Air Toxics , Endocrine Disruptors , Environmental Engineering , Environmental Justice , Watersheds
Objective:
Air quality on the Island of Hawai‘i (“the Big Island”) has traditionally been poor due to high emissions of sulfur dioxide (SO2) from Kīlauea Volcano. The resulting “volcanic smog” (“vog”), a mixture of SO2 and fine particulate matter (PM), can have negative impacts on human health as well as agriculture, and consequently is a major concern of local communities. Because of the high variability of the volcanic plume, community members’ exposures to vog is not straightforward to estimate. The region thus served as a unique test case for the use and assessment of distributed air quality (AQ) networks based on portable low-cost sensors. This project entails the development and deployment of a state-of-the-art community-based AQ sensor network across Hawai‘i Island, for the measurement of SO2 and particulate matter (PM) levels with high spatial and temporal resolution. The network aimed to provide improved measurements of air quality and vog exposures across the island, as well as to assess the utility of AQ sensor networks as educational resources and as tools for atmospheric chemistry research.
Summary/Accomplishments (Outputs/Outcomes):
Initial efforts focused on developing, testing, and calibrating sensor nodes for the measurement of SO2 and PM. In 2017 and 2018, proof-of-concept sensors for each were built and deployed in Hawai‘i. A major effort involved calibration of these sensors, in order to convert sensor output to pollutant concentrations. This was done via co-location, in which the low-cost sensors were located next to high-fidelity air quality monitoring instrumentation, providing accurate concentrations against which the sensor output could be compared. For SO2 we found that a non-parametric method (“k-nearest neighbors”, or kNN) can correct for the influence of temperature on sensor output, which is especially important at low concentrations. Little sign of drift of degradation was observed over the 6-month period. Such results provide confidence in the reliability of these sensors for air quality measurements (particularly when disseminated to the public in real time).
Additionally, this work highlighted an important practical consideration in sensor calibration, which is that the calibration data (“training set”) from such co-location studies may not be fully representative of atmospheric conditions. This limits the utility of most commonly-used non-parametric approaches (e.g., kNN), which cannot extrapolate to conditions beyond those of the training set. To address this, we developed a new hybrid approach, which utilizes the kNN calibration at low SO2 levels and the linear calibration at high levels. This approach was used for all subsequent SO2 measurements in this project.
In May 2018, Kīlauea Volcano entered a new eruptive phase, with the Lower East Rift Zone (LERZ) eruption emitting SO2 at levels that were ~10 times higher than they had been previously. In response we designed and built 30 new sensor nodes to measure SO2 and PM, calibrated them by co-location with regulatory-grade monitors, and deployed around the island (primarily at local schools) within 3 weeks of the beginning of the eruption. This enabled the collection of spatially- and temporally-resolved one-minute data for both pollutants throughout the remainder of the eruption. From these measurements we were able to quantify the exposures of the population to both pollutants with high granularity; we found that only a small fraction of the population was exposed to high levels of SO2, but a much larger fraction was exposed to high levels of PM. Further, the sensor measurements of SO2 and PM allowed us to estimate the rate by which SO2 oxidized in the atmosphere to form H2SO4 (the main component of PM during the eruption); this provides important information of the evolution of those two atmospheric components. To our knowledge this was the first time that low-cost sensors had been used to track the rate of an atmospheric chemical reaction.
Eruptive activity ceased in early August 2018, leading to extremely clean AQ conditions throughout the island. We thus designed, built, and deployed a new sensor network aimed at meteorological parameters (temperature, relative humidity, pressure, irradiation, wind speed, wind direction, and precipitation). The goal was to complement the existing AQ network, by providing local information on the transport of volcanic pollution around the island, both in the past (while we were monitoring volcanic AQ) and in the future, as well as to provide a new real-time dataset on environmental and atmospheric parameters for use by teachers and students in the participating schools.
After a 2.5-year hiatus, Kīlauea began erupting in December 20, through May 2021, and then again starting in September 2021. By January 2022 we had installed a new network of 17 more-robust nodes, providing SO2, size-resolved PM, and meteorological information. This is still active; a real-time map can be viewed.
Conclusions:
In addition to sensor deployment and data analysis, a major component of this project centered on education and outreach. We carried out a range of educational outreach activities, centered around teacher curriculum development related to the sensor data, as well as more generally to air quality, weather, climate, and kilo (traditional Hawaiian observation of the environment). These were organized by our community partner (The Kohala Center, TKC), and involved both MIT and TKC team members. Activities included a public talk, teacher workshops, a professional development program for teachers, classroom visits and activities, a student symposium, and virtual classes during the COVID-19 pandemic. Overall, educators successfully designed new learning materials, and feedback from program participants through surveys and questionnaires identified three main themes: i) teachers valued the sensor networks, though many needed additional technical support and training to integrate data and develop data-centered educational modules, ii) inclusion of cultural perspectives of vog and weather deepened appreciation and interest in local environmental phenomena, and iii) the program fostered valuable networks to connect participants with local educational resources and scientific institutions.
Journal Articles on this Report : 3 Displayed | Download in RIS Format
Other project views: | All 11 publications | 3 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Hagan DH, Isaacman-VanWertz G, Franklin JP, Wallace LMM, Kocar BD, Heald CL, Kroll JH. Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments. Atmospheric Measurement Technniques 2018;11(1):315-328. |
R836183 (2017) R836183 (2018) R836183 (2019) R836183 (2020) R836183 (Final) |
Exit Exit |
|
Hagan DH, Kroll JH. Assessing the accuracy of low-cost optical particle sensors using a physics-based approach. ATMOSPHERIC MEASUREMENT TECHNIQUES 2020;13(11):6343-6355. |
R836183 (2020) R836183 (Final) |
Exit Exit |
|
Crawford B, Hagan DH, Grossman I, Cole E, Holland L, Heald CL, Kroll JH. Mapping pollution exposure and chemistry during an extreme air quality event (the 2018 Kīlauea eruption) using a low-cost sensor network. Proceedings of the National Academy of Sciences. 2021;118(27). |
R836183 (Final) |
Exit |
Supplemental Keywords:
Volcanic emissions, air pollution, vog, sulfur dioxide, sulfate, particulate matter, low-cost sensors, sensor networks, community engagement, science curriculumProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.
Project Research Results
- 2020 Progress Report
- 2019 Progress Report
- 2018 Progress Report
- 2017 Progress Report
- 2016 Progress Report
- Original Abstract
3 journal articles for this project