Until about 1970, urban heat island (UHI) studies were generally limited to analyses of surface temperatures from fixed-site networks with long term data sets or from data obtained during short term mobile traverses. The helicopter and fixed wing aircraft studies of Oke and East (1971) in Montreal, McElroy (1971) in St. Louis, and Bornstein (1968) in NYC were the first to provide insight into the vertical structure of the polluted nocturnal urban boundary layer. They showed UHIs that extended upward through a depth of several 100 m meters, with magnitudes that decreased almost linearly with height. The urban boundary layer was shown to be plume shape when viewed along the flow axis, but dome shaped when viewed along its cross-wind axis and during calm conditions.

UHI layers were shown to be capped by one or more weak and shallow elevated temperature inversions and by a “reverse heat island” layer of about 100 m, in which a negative UHI existed. Various explanations have been put forth as to the origin of urban elevated inversions, e.g., they are remnants of the surface based rural inversion as it is advected over a city or they form from radiative cooling from the top of the urban pollution layer. Various explanations have also been put forth as to the origin of the “reverse heat island” layer aloft, e.g., radiative flux divergence from the top of the urban pollution layer or UHI-induced convergence, which produces rural downward motion and subsidence warming over surrounding rural areas.

The presentation thus presents a summary of these classic observational studies, along with a history of the modeling attempts (from 1-D analytical to 3-D numerical) that tried to explain these observed features of the urban boundary layer. It finishes with suggestions for required future observational and modeling studies that could answer some still-existing questions concerning the vertical structure of PBL UHIs.