Microscale Urban Heat Island and Air Quality Assessment Through Multi-sensor E-Bike Monitoring: Integrating High-Density Observations with Geospatial Analytics

Urban air quality and heat island effect are major concerns for human health and thermal comfort, particularly under the increasing trend of extreme heat events. But traditional networks of stationary monitoring stations provide limited spatial representativeness of intra-urban microclimate variability at the scales where citizens experience thermal stress and are exposed to air pollution.

To address this observational gap, we have developed a mobile monitoring system integrating a number of compact sensors mounted on an electrical bicycle platform. This setup enables high-density geolocated sampling along pedestrian areas, walkways and bike lanes -where citizens are truly exposed to the urban environment- while minimizing perturbations to the environmental conditions. It also presents advantages over other mobile platforms, such as UAVs or electric vehicles, in terms of operation permissions, accessibility to pedestrian areas, and time endurance.

The system simultaneously records: (i) meteorological variables (air temperature, relative humidity, wind speed and direction) (ii) radiative components including, sun/shade discrimination, shortwave hemispherical irradiance (sunlight), directional radiometric temperature (from six directions), enabling estimation of the standard Wet Bulb Globe Temperature (WBGT) for outdoor thermal comfort assessment; and (iii) environmental pollution indicators (suspended black carbon, PM2.5, PM10 and ambient noise). The data is geolocated by GNSS and recorded by a datalogger every second, providing approximately 3m spatial resolution at 10km/h cycling speed.

The mobile monitoring system has been tested during a summer heat wave in Valencia, Spain, performing mobile transects at solar noon and after sunset to capture differential cooling dynamics across urban morphologies. The 20 km routes were designed to pass through different types of neighbourhoods and densities of green-blue spaces. 

The geolocated measurements are integrated within a Geographic Information System (GIS) framework together with several layers of geospatial city information, like distribution of buildings, pavement types, urban green and blue spaces, individual tree inventory, vegetation density indices from satellite imagery, and building height (DEM). This set of multisource data enables advanced geospatial analytics combining spatial statistical and deep-learning to: (i) generate informative maps of thermal comfort and air quality at high spatial resolution; (ii)  identify urban hot and cold spots; and (iii) quantitatively evaluate the effectiveness of different nature-based solutions (NbS) for UHI mitigation.

This novel mobile monitoring approach, delivering unprecedented spatial density of CUHI observations combined with multi-source geospatial data, provides a scalable methodology for microscale air quality mapping and evaluating urban planning strategies and nature-based cooling interventions.