750 
energy balance equation because he realized the vagaries it involved. Ever since the intro 
duction of the use of infrared thermometry for vegetation (Fuchs and Tanner, 1966), scientists 
have been determined to try to find a way to use this simple measurement to exploit the physi 
cal relationship between surface temperature and evaporation. Jackson (1982) used infrared 
temperature in the Penman-Monteith equation (Monteith, 1965) for surface temperature. 
Huband and Monteith (1986) related infrared thermometer (IRT) measurements to aero 
dynamic surface temperature extrapolated from air temperature profiles, and even though they 
standardized the view angle relative to the sun, they encountered a systematic bias between 
the aerodynamic temperature that satisfies the big-leaf energy balance equation and the infra 
red temperature measurement. This is not surprising considering the myriad of factors that 
infrared temperature may depend on. 
Directional, infrared canopy temperature depends on many quantities, such as air temper 
ature, air vapor pressure, wind speed, canopy radiation balance, canopy architectural and 
spectral properties, canopy water vapor conductance (which depends on soil water availability 
and stomatal characteristics), soil heat and water characteristics and soil spectral properties, 
and the direction of view relative to the position of the sun. Although canopy temperature 
clearly is related to the partitioning of energy fluxes and thus evapotranspiration, numerous 
definitions exist for the term "canopy temperature." The various temperatures can be suffi 
ciently different for a given situation that infrared thermometer measurements may not be use 
ful in evaluating the canopy energy budget. 
This paper contains a discussion of directional temperature and emissivity of soil-vegeta 
tion systems and the implications for remote sensing within the framework of a comprehensive 
plant-environment model, Cupid (Norman and Campbell, 1983). The Cupid model, which has 
been compared with field measurements, can be used to explore the sensitivity of directional 
temperature and emissivity to various factors and assist in defining important quantities more 
precisely. 
2 - CUPID MODEL DESCRIPTION 
The model Cupid (Norman and Campbell, 1983) is used to explore the relations among the 
various definitions of canopy temperature. Cupid is a one-dimensional, layered model that 
includes atmospheric, canopy and soil exchanges of momentum, energy and mass. Radiative, 
convective and conductive processes are combined with soil and plant characteristics to pro 
duce an integrative model. Boundary conditions usually are specified some meters above the 
top of the canopy and below the bottom of the root zone so that conditions at the soil surface 
are an output of the model. Canopy architecture, leaf physiological and radiative properties, 
and soil conductive properties to heat and water are used in the combined radiative, convective 
and conductive equations to provide profiles of fluxes, concentrations and temperatures 
throughout the soil, canopy and atmosphere system. Cupid is a comprehensive model that has 
been used in numerous applications, including photosynthetic productivity, irrigation water 
balance, integrated pest management of insects and diseases, and bidirectional reflectance 
studies related to remote sensing (Norman, 1988). Presently the model also includes direc 
tional thermal radiance equations as well as bidirectional short wave and thermal reflection 
(Norman et al., 1985).