assumes that the bottom heat flux is constant throughout the exercise. One example of the
difference between having measured initial soil temperature profile data and letting the
program calculate one is seen in Figure 4.4 below for the Yuma, AZ, data. The "yes" case
is the same as the model results shown in Figure 4.3. Based on the user-specified number
of layers, position and number of initial soil temperature and moisture measurements, and
the total thickness, the model divided the profile into 15 nodes. For the two "no" cases,
everything was retained from the "yes" scenario except the measured initial soil
temperatures. In one "no" run, the number of nodes was forced to be the same as in the
"yes" case; in the other "no" run, the model determined the required number. As can be
seen, essentially no difference is discerned between the three runs except at the start of
the simulation.
Yuma, AZ 1993
yes, 15
no, 15
no, 12
330
325
320
315
310
305
300
295
290
285
280
73
74
75
76
77
78
79
80
81
82
83
Day of the Year
Figure 4.4 Effect on the results of having initial soil temperature profile data (yes) versus
letting the model compute one (no). The number in the legend indicates the number of
nodes.
We used the Yuma, AZ, data to perform a sensitivity study on the model. The variables
tested were number of layers, albedo, and emissivity. We let the model calculate the
thermal conductivity and diffusivity. We found no dependence in the results on the
number of layers. The results showed little sensitivity to the emissivity. The albedo does
affect the surface temperature since the larger the albedo, the less shortwave radiation is
absorbed by the surface. The results are seen in Figure 4.5.
38
Previous Page
