Skip Navigation

USGS Home
Contact USGS
Search USGS

Surface Heat Flux

The surface heat flux was calculated using techniques described by Weller et al (1995), and the flux was introduced into the top sigma layer at each grid location. These techniques produce bulk estimates of latent, sensible and longwave radiation, given sea surface temperature, air temperature, insolation, relative humidity, barometric pressure and wind. The limited availability of these data led us to use air temperature, barometric pressure and wind data at the Boston Buoy, relative humidity at Logan Airport and insolation at Woods Hole to represent bay-wide conditions (Figure 2.3). The sea surface temperature used in the heat flux estimates was initially determined from observations at the Boston Buoy. This proved problematic, because Cape Cod Bay is significantly warmer than the Boston Buoy site during most of the summer, which should result in reduced heat flux (all else being equal), but there was no feedback mechanism to allow this. The result was that model temperatures in Cape Cod Bay became too warm. To address this problem, later runs used model-generated sea surface temperature fields to calculate surface heat flux. Using the modeled sea surface temperature field allows spatial variability in sea-surface temperatures to feed back into the heat flux routines. Under upwelling conditions, for example, there can be 10 or more difference between surface water temperatures in the Bay, which can in turn result in heat flux differences of 200 W m or more. Allowing for spatial differences in the surface heat flux due to varying water temperatures led to a significantly improved correspondence between modeled and observed water temperatures in Cape Cod Bay. Hourly estimates of heat flux obtained from the bulk formulae were averaged over 4 hours to reduce the size of the model input file.