WMOD00570

• Title: Precipitation Enhancement Effects of Sorghum, Using a Daily Growth Model
• Date: 4/1/1982
• Weather Modification - Doc Type: Report

precipitation and yield. The rank correlation coefficient betwein this
14-day precipitation and yield was 0.46, significant at the I-pel-cent level
(the linear correlation coefficient was 0.35, significant at thel5-percent
level). While statistically significant, the correlation is only moderate,
having the same value as the stage 6 correlation coefficient andl'a substan.-
tially lower value than the correlation between yield and longer period
precipitation. It is likely that a different time period could be found that
would give a larger coefficient, but physical reasoning suggestslthat,
because of the interactions among precipitation, soil water, and evapotrans-
piration in the model, a simple statistical approach will not pr0vide a
satisfying description of the effects of timing. I

Before leaving the discussion of stage precipitation, it is interesting to
note that correlation coefficients (table 3) are significant at all five
sites only for stage 4 (floral initiation to end of leaf growth)] This
sugges-ts the possibil ity that the more the soil water is built Uw by precipi-
tation in this period of relatively low evapotranspiration, the ]ess the
plant will be affected by fluctuations in precipitation during s~bsequent
stages. An equally plausible explanation is that precipitation during this
period is important for plant vigor. I

Some effects of timing may be masked when data for all years arelconsidered
collectively, so plots of natural conditions for pairs of years ~Iere examined.
The pairs were specifically selected to illustrate the dominant factor of
timing, but the effects of amount may also be present.3/ The pl~ts, first
presented in figures 6a and 6b, show the daily progress of facto~'s that
influence plant growth and development. The vertical bars show/the amount of
precipitation; a trace event is depicted by a lip along the horitontal axis.
The solid line is soil water. The dot-dash line is water stress~ the dashed
line is temperature stress. The thin line rising to the right r~presents the
progress of grain filling.

Pairs of years were selected that have either reasonably similar precipitation
totals but different precipitation distributions and yields, or similar
yields but different precipitation totals and distributions. Examination of
Goodland 1961 and 1978 (figs. 6a and 6b) reveals that 1961 had mbre total
rain and growing-season rain than 1978. However, the yield was the same for
both years. The soil water in 1961 was consistently higher thanl1978 from
the middle of stage 3 to maturity. In 1978 a total of 26 mm of lain fell .in
three events in the 4 days prior to the onset of grain filling. The timing
of these events enabled 1978 to equal the yield of 1961 (3720 kgiha).

Oklahoma City 1955 and 1969 (figs. 7a and 7b) provide examples ofr good
timing and poor timing. Both total and growing-season rainfall \;Iere greatE~r
in 1955, yet the yields were essentially the same (about 4720 kg1ha). The
difference lies in the timing of the stage 6 rains. Both years 1eceived

I

3/ Questions about the effects of frequency (i.e., many small pr~cipitation
events versus a few large events) have been raised. These effects are
difficult to assess because of the complex interrelationships with amount ,and
timing. Therefore, within the context of this paper, no attempt!was made to
isolate the effects of frequency. :
I
I
I

21 i

;;

~

>>

,-