strength and temperature differences. This tech-
CONCLUSIONS OF PART II
nique warrants further field comparison with
Winter morning air temperature differences
rain settings.
observed in the Connecticut River Valley are com-
The influence of snow depth and cover on win-
parable with temperature differences observed
ter morning air temperature has been demon-
near Fairbanks, Alaska, by Benson and Bowling
strated. Lapse rate in the lowest 300 m diminishes
(1973) and in Hokkaido, by Maki and Harimaya
and lapse becomes less frequent as snow depth
(1988). Surface air temperature differences of 5 to
18C were commonly observed along slopes
increases. Inversion becomes stronger and more
frequent as snow cover increases. A surface-based
within 6 km of the river on the same days that less
than 2C temperature difference was observed
inversion is infrequent in the lowest 100 m of the
river basin when less than 20 cm of snow covers
along a 30-km path along the river. We attribute
the ground; a surface-based inversion diminishes
the largest differences to earlier formation of in-
river level temperatures when more than 20 cm of
version over small basins and diminished ex-
snow is present.
change of warmer tropospheric air across these
The diminution of winter morning tempera-
inversions through the night. A discussion of the
limited influence of cloudiness was provided in
tures in small and large basins results from dimi-
Hogan and Ferrick (1997).
nution of exchange of warmer tropospheric air
The moving probe method of temperature mea-
from above during night hours. The geometric
standard deviation of temperature difference, with
surement has produced results consistent with
respect to the lowest elevation, is least in basins
those obtained by a multiple thermometer and
and greatest on ridges and slopes. We propose
tethersonde array in Hokkaido. The inference of
temperature inversion or lapse structure in the
related to the geometric standard deviation of
lower 300 m of air over the Connecticut Valley
temperature differences.
from surface observations at two elevations has
yielded consistent results in describing inversion
29