In order to in provide a high resolution of air pollution sensing: (i) we deployed our reference ground sensor (FROG) since mid-August 2019 in Houston’s East End with TFA as base station, (ii) we performed multiple drone missions in that same neighborhood and are currently analyzing the data collected using our custom-built drones, and (iii) we implemented a mobile app that provides the community with real-time collected data of both ground sensors and drones.
Our objective is to perform measurements of VOC pollutants using high-spatial-resolution drone-based sensing in addition to high-accuracy ground sensing. This will allow us to provide the community with unprecedented spatial and temporal characterization of VOC concentrations. Furthermore, ground-based VOC data will be used to identify the existence of a gas leak and therefore determine when to launch drone missions.
In terms of ground sensing, and over the past year, we collected more than 9 months of ground concentrations of VOC pollutants at a 10min temporal resolution using our FROG high-accuracy sensor. Compared to governmental collected data (TCEQ stations) which is published at an hourly scale, our design allows us to have a better temporal resolution in terms of continuous pollution characterization. Our ongoing analysis of the collected ground data shows that the peak values of VOC concentrations can be very difficult to identify with just an hourly resolution as some peaks occur and then disappear within the same hour.
In terms of drone-based sensing, we collected more than 10,000 aerial samples in different locations and times around Houston’s East End and within the campus of Rice University. The collected data allowed us to study the performance of aerial sensing, which can be highly impacted by drone propellers and vibrations. Our evaluation results show that the dynamic airflow caused by drone propellers affects first the temperature and humidity levels of the ambient air, which then affect the measurement quality of VOC gas sensors. Based on that, we show that the quality of drone-based VOC measurements can be inferred with up to 88% accuracy based on humidity and temperature data.
In addition to collecting and processing high-resolution air pollution data, we implemented a mobile app that provides the community with real-time collected data, most up-to date drone missions’ collected measurements and regulatory thresholds that are associated with the measured pollutants.
Publications:
Data Set – Baseline BTX air pollution levels in the TFA area.