What aspects to consider when doing mobile monitoring?

Description

The most common installation of air quality monitoring and sensing devices is in fixed physical locations such as next to the window of a residential home, inside the home or in a neighbourhood location such as on a street pole. With the increasing miniaturization of the technology, it is now possible to mount Sensing Devices on vehicles such as bicycles or busses, or even carry them directly as wearables. These mobile-monitoring applications raise new challenges regarding data quality and usability, but open up a whole new range of investigations across multiple locations, and offering spatial continuity not available with fixed monitoring devices. These insights are particularly interesting for understanding the exposure of the individual as they move through a geographical area during their daily activities.

Why is this relevant?

Air pollution might have a very diverse spatio-temporal distribution, meaning that pollutant concentrations can change from one street to another, and from one minute to the next (e.g. when standing in a traffic light the concentrations are higher). Mobile monitoring offers the advantage of capturing spatial and temporal variability in environmental conditions, identifying pollution hotspots, and complementing stationary monitoring efforts, but it presents challenges in maintaining data quality assurance, achieving uniform spatial coverage, ensuring continuous data collection, coordinating logistical complexities, managing costs, interpreting complex data, and integrating diverse datasets for comprehensive analysis. Being aware of the pros and cons of mobile monitoring is crucial before selecting the method that best adapts to our needs, expertise and resources.

How can this be done?

By utilizing mobile platforms such as vehicles or drones equipped with sensors, mobile monitoring can rapidly cover large geographic areas and provide real-time data on air quality, noise levels, and other environmental parameters. This dynamic approach allows for targeted monitoring of pollution hotspots, identification of emission sources, and assessment of pollution dispersion patterns, which may not be feasible with static monitoring stations. Additionally, mobile monitoring enables flexibility in monitoring routes and sampling locations, allowing for adaptive monitoring strategies that respond to changing environmental conditions or emergent pollution events. Furthermore, mobile monitoring can complement static monitoring by providing high-resolution data in areas with limited stationary monitoring infrastructure, thereby enhancing the overall comprehensiveness and accuracy of environmental monitoring efforts.

However, before selecting mobile monitoring over static monitoring, one needs also to be aware of their challenges, including maintaining data quality assurance amidst sensor variability and environmental interference, achieving uniform spatial coverage, ensuring continuous or real-time data collection, coordinating logistical complexities such as route planning and equipment maintenance, managing the resource-intensive nature of campaigns, interpreting complex data due to spatial and temporal variability, and integrating diverse datasets for comprehensive analysis.

There are some crucial points to consider when selecting between mobile and fixed monitoring for air quality:

Spatial and temporal variability: assess the extent of spatial and temporal variability in air pollution levels across the monitoring area to determine whether mobile or fixed monitoring is better suited to capture these fluctuations effectively. With a mobile sensor device, you can cover large spatial areas, but as the sensor is moving you do not have continuous monitoring in any location, so you are blind on those locations until the device comes by again. This is, you miss the temporal information.

Coverage needed: evaluate the need for comprehensive coverage of different locations and pollution sources within the monitoring area to determine whether a fixed network of monitoring stations or mobile monitoring platforms would be more appropriate.

Data quality and interpretation: maintaining data quality in mobile monitoring can be challenging due to factors such as variability in sensor measurements, calibration drift during transportation, and environmental interference (e.g., vibrations, temperature fluctuations). Ensuring consistent data quality across different monitoring routes and platforms requires robust quality assurance procedures and calibration protocols. Moreover, analyzing and interpreting mobile monitoring data can be complex due to factors such as spatial heterogeneity, temporal variability, and potential confounding variables. Integrating mobile monitoring data with other datasets, such as static monitoring data or meteorological observations, may be necessary to provide context and facilitate accurate interpretation. With PM optical sensors this can be problematic as a particle under the effect of non-regular speed of the air flow may “bounce” in the sensor detection chamber and be counted multiple times. And in a more practical way, if you are following a polluting vehicle, you will measure high concentrations during all that time, which results in ‘skewed’ data that show a high level of pollution across a specific space that is not otherwise as highly polluted. Mobility of the sensor also requires an additional level of contextualization of the location of the data gathered to maintain good data quality and trust in the readings– for example, a wearable device will follow the wearer from indoors to outdoors to inside public transportation, over a short period of time. If the context is not well documented, the data will be very difficult to interpret in a usable way. See (Chap 7.3 Why is it important to document context and how does it help better understand collected data?).

Data integration with other information sources: mobile air quality sensors provide high-resolution data on localized pollution levels, whereas traditional stationary monitoring stations offer broader coverage over larger geographic areas. Integrating data from these sources requires methods to reconcile differences in spatial and temporal resolution to create cohesive datasets for analysis. Also, mobile devices might have higher sampling frequency compared to stationary monitoring. Standardizing data formats is crucial when integrating data from diverse sources.

Privacy and security: mobile air quality sensing devices may collect sensitive location-based information from individuals using wearable or smartphone-based monitoring devices. Ensuring privacy protection and data security is paramount to maintain public trust and compliance with data protection regulations.

We can split mobile monitoring in 2 families: vehicle mounted and wearables.

Vehicle mounted: often plugged onto vehicle’s power source these devices usually combine multiple environmental sensors such as air quality but also vibration, noise... and high precision GPS positioning to consider speed and canyon effect that may impact positioning precision. They often have their own data transmission system and come as plug and play systems. They are installed on top of vehicles which leads to specific issues such as vibrations, vandalism, electromagnetic perturbations and so on. Mobile monitoring is also done with moving reference instruments, for example mounting them on a large vehicle or in a truck one can move from one place to another (an example by the Environmental Agency in UK). As data is collected over time, the output map for a specific period of time (such as one day) will provide a representation of the air quality collected at different times of the day in one visualisation. This data can be complimented with more data from a network of fixed stations. Fixed stations providing measurements over time in a specific location can furthermore serve to provide a correlation / calibration service for the mobile stations when they come close to a reference / stationary equipment, thus improving the quality of the data.

Wearables: are the most popular type of mobile monitoring, used on bikes, at home or during specific activities such as the daily commute to work. Wearable devices offer promising opportunities for advancing air quality monitoring by providing individuals with real-time access to information about air pollution levels and potential health risks. Wearables enable personalized insights into air quality, empowering users to make informed decisions about outdoor activities, travel routes, and exposure reduction strategies. Moreover, wearable air quality sensing devices have the potential to contribute to Citizen Science initiatives by aggregating data from multiple users to create detailed maps of air pollution hotspots and trends. However, challenges such as sensor accuracy, data interpretation, and user adoption need to be addressed to fully harness the potential of wearable devices for air quality monitoring and public health protection.

Useful resources

• Assessment of PM2.5 Exposure during Cycle Trips in The Netherlands Using Low-Cost Sensors. Scientific paper showcasing the use of 500 dust (PM2.5) sensors mounted on bicycles to estimate typical PM2.5 levels to which cyclists are exposed in the province of Utrecht, the Netherlands, in the year 2020.

You might also be interested in….

• Why is it important to document context and how does it help better understand collected data?