A modeling study of the impact of urban trees on ozone

Abstract

Modeling the effects of increased urban tree cover on ozone concentrations (July 13–15, 1995) from Washington, DC, to central Massachusetts reveals that urban trees generally reduce ozone concentrations in cities, but tend to increase average ozone concentrations in the overall modeling domain. During the daytime, average ozone reductions in urban areas (1 ppb) were greater than the average ozone increase (0.26 ppb) for the model domain. Interactions of the effects of trees on meteorology, dry deposition, volatile organic compound (VOC) emissions, and anthropogenic emissions demonstrate that trees can cause changes in dry deposition and meteorology, particularly air temperatures, wind fields, and boundary layer heights, which, in turn, affect ozone concentrations. Changes in urban tree species composition had no detectable effect on ozone concentrations. Increasing urban tree cover from 20 to 40% led to an average decrease in hourly ozone concentrations in urban areas during daylight hours of 1 ppb (2.4%) with a peak decrease of 2.4 ppb (4.1%). However, nighttime (20:00–1:00 EST) ozone concentrations increased due to reduced wind speeds and loss of NO_x scavenging of ozone from increased deposition of NO_x. Overall, 8-hour average ozone concentration in urban areas dropped by 0.5 ppb (1%) throughout the day.

Introduction

Vegetation in cities, particularly trees, can directly and indirectly affect urban air quality. Tree transpiration and tree canopies affect: (a) meteorology (air temperature, radiation absorption and heat storage, wind speed, relative humidity, turbulence, surface albedo, surface roughness and consequently the evolution of the mixing-layer height) (e.g., Heisler et al., 1995; Berman et al., 1997); (b) dry deposition of gases to the earth's surface (deposition velocity) (e.g., Baldocchi et al., 1987); (c) emission of volatile organic compounds (VOCs) that can contribute to the formation of ozone (O₃) and carbon monoxide (CO) (Brasseur and Chatfield, 1991); and (d) anthropogenic emissions through reduced energy use due to lower air temperature and shading of buildings. While lower pollutant emissions generally improve air quality, lower nitrogen oxide (NO_x) emissions, particularly ground-level emissions, may lead to a local increase in O₃ concentrations under certain conditions due to reduced NO_x scavenging of O₃ (Rao and Sistla, 1993; Rao and Mount, 1994). These four impacts of vegetation, which lead to changes in atmospheric chemistry and physics, all interact to affect ozone concentrations.

Research integrating the cumulative effects of urban vegetation on air quality, particularly ozone, is very limited. Cardelino and Chameides (1990) modeled vegetation effects on ozone concentrations in the Atlanta region using the OZIPM4 model. The study's primary focus was on the interaction of VOC emissions and altered air temperatures, and revealed that a 20% loss in the area's forest due to urbanization could have led to a 14% increase in O₃ concentrations for the modeled day. Although there were fewer trees to emit VOCs, an increase in Atlanta's air temperatures due to the urban heat island, which occurred concomitantly with tree loss, increased VOC emissions from the remaining trees and anthropogenic sources, and altered O₃ chemistry such that concentrations of O₃ increased.

A model simulation of California's South Coast Air Basin suggests that the air quality impacts of increased urban tree cover may be locally positive or negative with respect to ozone. The net basin-wide effect of increased urban vegetation is a decrease in ozone concentrations if the additional trees are low VOC emitters (Taha, 1996). This study used the Colorado State University Mesoscale Model (CSUMM) and Urban Airshed Model (UAM-IV), and accounted for vegetation effects on meteorology and emissions. The most significant meteorological impacts were temperature reductions, which altered chemical reaction rates, reduced temperature-dependent biogenic VOC emissions, and changed the depth of the mixed layer. The study also accounted for increased pollution deposition and possible increased VOC emissions due to increased vegetative cover.

McPherson et al. (1998) estimated the cost effectiveness of residential yards trees for improving air quality in Sacramento, CA, using Best Available Control Cost Technology cost analysis. The benefit–cost ratio over 30 yr for planting 500,000 trees was 0.5. However, the findings of this study are debatable as it omitted the quantification of the net effect of urban trees on air quality (Nowak et al., 1998a).

This paper details findings of a study that investigates the cumulative and interactive effects of increased urban tree cover on urban and regional ozone concentrations in the Northeastern United States. The study uses meteorological, emission, and air quality models and field data on urban vegetation to assess the impacts of altered meteorology, dry deposition, biogenic VOC emissions, and anthropogenic emissions, due to changes in urban tree cover, on ozone concentrations from July 13 (11:00 EST) to July 15 (14:00 EST), 1995. This time period was chosen due to the availability of model input data and used clean initial and boundary layer conditions to isolate the effects of urban tree cover on ozone concentrations. The purpose of these simulations were not to validate model performance but to compare model outputs against model inputs as related to urban tree cover.