Hourly atmospheric longwave radiation (L down arrow) flux densities ob
served at an alpine tundra site in southern British Columbia were comp
ared to eight different L down arrow models that utilized screen-level
temperature and/or vapor pressure input data. Models tested were the
Stefan-Boltzmann equation using an effective atmospheric emissivity of
.70, and the models formulated by Brunt, LeDrew, Swinbank, Idso, Idso
-Jackson, Brutsaert, and Berdahl-Martin. All models were tested on bot
h cloudless and cloudy sky conditions. Most models performed well, wit
h the exception of the LeDrew model. Overall, the Idso-Jackson model w
as superior. The results demonstrated that L down arrow could be relia
bly estimated from air temperature alone, and challenged the accepted
notion that L down arrow models do not work effectively in mountainous
environments. Measured terrestrial longwave radiation (L up arrow) va
ried according to the dictates of the Stefan-Boltzmann Law. Comparison
s between measured L up arrow and surface (skin) temperature indicated
that blackbody assumptions of surface emissivity were appropriate.