J3.3 LAND-COVER INFLUENCES ON BELOW-CANOPY TEMPERATURES IN AND NEAR BALTIMORE, MD Gordon Heisler 1 *, Baohua Tao 2 , Jeffrey Walton 1 , Richard Grant 3 , Richard Pouyat 1 , Ian Yesilonis 1 , David Nowak 1 , and Kenneth Belt 1 1 USDA Forest Service, Syracuse, NY; 2 SUNY Environmental Science and Forestry; 3 Purdue University 1. INTRODUCTION As a contribution to the Baltimore Ecosystem Study (BES), a U.S. National Science Foundation Long Term Ecological Research (LTER) site, weather variables are being measured continuously at five locations near Baltimore, MD. Data also are available from two National Weather Service Automated Surface Observing System (ASOS) stations (Fig. 1). Measurements include temperature at a height of 1.5 m at all stations. One object of the analysis of these measurements is to develop an empirical model of below-canopy air temperature differences. Such a model is important for evaluating urban structural and vegetation influences on air temperature for studies related to human thermal comfort, carbon cycling, soil and stream temperatures, ozone formation, and interaction with effects of UV radiation. An anticipated application of the temperature- differences model is for mapping predicted temperatures across Baltimore based on 30- by 30-m imagery from the National Land Cover Database (NLCD). The mapping will be for times of special interest, such as early evening, when temperature differences usually are greatest, and midafternoon, when temperatures usually reach a maximum. At the five non-ASOS sites, measurements have been continuous since June 2003. These sites include a lawn area with nearby trees near a large apartment complex (Apartments, 1 in Fig. 1); a residential area with heavy tree cover but few buildings (Residential under trees, 2); a residential area with some trees and large lawn areas (Residential open, 3); a woodlot next to a large cultivated field (Woods, 4); and a large open pasture, (Rural open, 5). The ASOS sites are in downtown Baltimore (Downtown, 6), and at the Baltimore/Washington International (BWI) Airport (Airport, 7) Temperature differences on an hourly basis between each site and the Downtown ASOS site are related empirically by regression analysis to upwind tree, impervious, and water land cover from the NLCD 2001 (Homer et al. 2004). Additional predictor variables for temperature difference are atmospheric stability, vapor pressure deficit, antecedent precipitation, sky view and transmitted direct-beam solar radiation estimated from hemispherical photographs at each site, and topography. The initial analysis is for May through September 2004. In this paper we report preliminary results in the development of predictor variables. 7 6 5 4 1 3 2 Figure 1. Land use in Baltimore and vicinity and location of seven weather stations; from NLCD 2001 (http://www.mrlc.gov). 2. METHODS The planned analysis is similar to that carried out in Heisler and Wang (1998). The temperature differences (∆T) for each hour between the generally warmest site-- the Downtown site--and each of the other sites form the dependent variable. A major challenge is created by the range of the scales of influences on temperature and the correlation between potential predictor variables. _________________________________________ * Corresponding author address: Gordon Heisler, USDA Forest Service, 5 Moon Library, 1 Forestry Drive, Syracuse, NY 13210; email: gheisler@fs.fed.us 2.1 Sites The sites are separated by as much as 19 km (Fig. 1). Land uses around the sites according to the NLCD