6B.6 IMPACT OF SKY CONDITIONS ON ERYTHEMAL UV-B EXPOSURE UNDER TREE CANOPIES Gordon M. Heisler USDA Forest Service, Syracuse, New York Richard H. Grant Purdue University, W. Lafayette, Indiana Wei Gao Natural Resource Ecology Laboratory, Colorado State University, Fort Collins, Colorado 1. INTRODUCTION Solar ultraviolet (UV) radiation at the Earth's surface has many implications for human health. There appear to be complex relationships between skin cancers and sun exposure (Heisler and Grant 2000a, Heisler and Grant 2000b, Vitasa, et al. 1990). These health effects may have been exacerbated by depletion of the stratospheric ozone layer and attendant increases in UV-B (320-280 nm) radiation (de Gruijl 1995), because average UV-B irradiance has increased about 4 to 6% in mid-latitudes since the 1970’s (Madronich, et al. 1998). However, changes in habits of recreation, dress, and the increased value placed on a “tan” are evidently largely to blame for most of the increases in skin cancer rates in recent decades (Heisler and Grant 2000a). The health impact of solar UV depends on the environmental conditions as well as human habits. Researchers have made significant improvements in the prediction of open-environment exposure of surfaces at various slopes (Grant 1998, McKenzie, et al. 1997, Parisi and Wong 1994). However, most people spend much of their outdoor time in environments in which there is significant obstruction to sky and direct sun by buildings and trees. Prior studies have shown that the irradiance at pedestrian heights is significantly influenced by canopies of trees and buildings (Grant 1997, Grant and Heisler 1996, Grant, et al. 2002, Heisler, et al. 2003a). Tree cover in most urban areas is substantial; in a survey study of the tree cover of many cities, Nowak and coworkers (1996) indicate that tree cover in residential areas ranges from 48% in forested climates, to 27% in grassland climates, and 11% in desert climates. Human exposure to UV-B radiation and the effects of shading objects have often been described for clear sky conditions. Although survey measurements showed that cloud cover influences the relative exposure to UV- B radiation in shaded environments (Moise and Aynsley 1999); it is well known that cloud cover does not greatly reduce the UV irradiance until the cover is opaque and nearly overcast (Barton and Paltridge 1979, Frederick and Steele 1995, Grant and Heisler 2000). However, we expect the type, thickness, and fractional cover of clouds to cause variation in UV-B irradiance beneath trees; especially because the irradiance under partly cloudy skies is affected by scattering off the sides of clouds (Mims and Frederick 1994). One approach to evaluating cloud cover effects on UV-B exposure of people under tree canopies in different neighborhoods is to model the relative irradiance beneath tree canopies compared to above canopy irradiance and then to use measured or modeled UV-B irradiance above the canopy to approximate the irradiance under the canopy. The above-canopy UV-B irradiance can be derived from monitoring networks such as that of the United States Department of Agriculture UV-B Radiation Monitoring Program (Bigelow, et al. 1998) or from other UV-B measurements (Heisler, et al. 2003b). A recent model of UV-B irradiance above canopy, including cloud effects is also available (Madronich, et al. 1998), and in this paper we use this model and a model of radiation transmission through tree canopies to estimate UV-B exposures under tree canopies in the Baltimore, MD area under variable cloud conditions and tree cover. 2. METHODS The model developed to predict UV-B irradiance below vegetation canopies is an adaptation of a 3-D relative irradiance model (Gao 1997, Gao, et al. 2002, Grant, et al. 2002). This model assesses the UV-B irradiance at a point below a canopy given sky conditions and a canopy consisting of a finite number of ellipsoidal crowns that act as discrete scattering volumes. The foliage in the crowns is characterized by a single foliage density. In our modeling for this analysis, the probability of penetration of sky diffuse radiation was computed using separate hemispherical sky radiance distribution models for cloud-free skies (Grant, et al. 1997a), partly cloudy skies (Grant, et al. 1997b), and overcast skies (Grant and Heisler 1997). To approximate the relative irradiance (I r ) under a canopy of essentially infinite extent, we modeled an 11 X 11 array of regularly spaced tree crowns and averaged irradiance over a grid below the central four crowns. We carried out this analysis for solar zenith angles of 15°, 30°, 45°, and 60° and the National Weather Service cloud cover classes of CLeaR (<1 octa), FEW (1-2 octas), SCaTtered (3-4 octas), BroKeN (5-7 Octas), and OVerCast (8 Octas). We interpreted