Skyscraper rooftop tracer concentration observations in Manhattan and comparisons with urban dispersion models

- Tracers were observed at rooftops (127 < z < 197 m) in the MID05 field study.
- Tracer releases took place near street level.
- Near-field ratios of rooftop to surface concentrations have a median of 0.02.
- Vertical spread is enhanced by large eddies around tall buildings.
- HPAC urban dispersion model underpredicts the rooftop concentrations and ratios.

This paper focuses on the observed and model-predicted rooftop concentrations on very tall buildings at distances less than a few hundred meters downwind of near-surface releases in built-up urban centers. These results are important when public health must be protected in populated urban areas with deliberate or accidental releases of toxic chemicals, or with significant traffic emissions. Observations of tracer concentrations taken at seven samplers on skyscraper rooftops (113 m < z < 197 m) during the Manhattan Midtown 2005 (MID05) field experiment are analyzed, with emphasis on the near-field (x < 100 m). To calculate the ratio of rooftop to surface concentrations, we pair each rooftop sampler with the closest street level sampler. Six tracer gases (SF6 and five perfluorocarbon tracers (PFTs)) were released near street level from several locations. In the near-field, the median ratio of observed rooftop to surface concentration is about 0.02, even very close to the source, although there is much scatter. The large recirculating eddies adjacent to the tall buildings may cause the relatively large vertical spread. It is noted that, at distances greater than a few hundred meters, the ratio approaches unity (although there is still significant scatter). The observed normalized rooftop and surface concentrations and rooftop to surface ratios are compared to the predicted concentrations and ratios by three urban dispersion options (Urban Dispersion Model (UDM), Urban Canopy (UC), and MicroSwiftSpray (MSS)) in the HPAC/SCIPUFF model. There is a general tendency towards an underprediction of the rooftop concentrations and a slightly smaller underprediction of the surface concentrations. The median ratio of rooftop to surface concentrations is underpredicted by most of the meteorology-urban module options, with much scatter for all options. These results underline the need to better parameterize the dispersion of plumes in the street canyons and recirculating eddies around tall buildings.