Where particulates matter: High-rise construction emissions of PM2.5, PM10, BC, and UFP in Dutch cities

Abstract

Construction sites are major sources of particulate matter pollution to cities, with heavy diesel equipment producing ultrafine particulate matter (UFP), black carbon (BC), and contributing (alongside earth-moving activities) to emissions of fine (PM_2.5) and coarse (PM₁₀) dust. These air pollutants are hazardous to the health of those working or living within their plumes. Hence, in this study we characterize the short-term pollutant concentrations at three construction sites in the Netherlands. We measured concentrations of four pollutants at three early-stage high-rise construction sites in Rotterdam and Amsterdam. We find that among the pollutants measured, UFP number concentrations are most strongly correlated with construction activities, with construction site concentrations at high levels of activity ∼20,000 # cm⁻³, while concentrations during periods without construction activity are <5000 # cm⁻³. Although we can attribute concentration peaks to specific pieces of construction equipment, there were no strong correlations between UFPs and the measured concentrations of PM_2.5 and PM₁₀, as these latter particles have different sources in construction sites (e.g. earthmoving equipment) and in urban settings, as well as longer atmospheric lifetimes. At a relatively more isolated site with one major piece of diesel equipment, we do observe a strong correlation between UFP and BC (R = 0.62), suggesting that such measurements can be used to obtain emissions ratios. As new regulatory standards are being developed for health-hazardous UFP, knowledge of their number concentrations at and near construction sites will be crucial to developing source apportionments and dispersion modeling. We recommend increased measurement of UFP at construction sites, including the use of wearable monitors, to better understand exposure, and creating incentives to shift from diesel to electrical equipment wherever possible.