Variation in Gas Exchange Characteristics Among Desert Plants
367
an annual basis (data not presented). The vast majority of storm events
result in less than 10 mm precipitation, and it may be questionable as to just
how useful this input is to increasing soil moisture at soil depths where
plants can effectively use the precipitation input. Sala and Lauenroth (1982)
examined the impact of small precipitation events (-5 mm) on plaht water
relations in a semi-arid glassland. They concluded that for shallow-rooted
grasses these small precipitation events could significantly improve gas exchange for 1-2 days following the storm.
Given the frequency distributions in Fig. 18.13, total annual precipitation
and the number of storm events should be tightly correlated. Figure 18.4
shows that not only is there indeed a significant linear relationship between
Date of first winter rains
15 Indio
10
5
0
en
15 Needles
....
CIl
Q)
>- 10
0
....
Q)
.0
E
5
~
Z
15 Phoenix
10
5
0 0 "k. <:>. "0
o/,
0t.,
&""
&"" v~
"' v. 01 // 7 7
U>/,
(9"..~
~~~1-7
o~
7
(9"..
(9"..
7
7
is'
7
7
Fig. 18.5. The frequency distribution of the date of the first winter rains for three sites
differing in the predictability of that winter precipitation. The first winter storm is defined
as the first storm after October 1 that was ;310 mm. Calculations are based on data from
the US Weather Bureau records (NOAA) during the period 1929-1979. The three figures
are arranged in a west-to-east transect from top to bottom
367
an annual basis (data not presented). The vast majority of storm events
result in less than 10 mm precipitation, and it may be questionable as to just
how useful this input is to increasing soil moisture at soil depths where
plants can effectively use the precipitation input. Sala and Lauenroth (1982)
examined the impact of small precipitation events (-5 mm) on plaht water
relations in a semi-arid glassland. They concluded that for shallow-rooted
grasses these small precipitation events could significantly improve gas exchange for 1-2 days following the storm.
Given the frequency distributions in Fig. 18.13, total annual precipitation
and the number of storm events should be tightly correlated. Figure 18.4
shows that not only is there indeed a significant linear relationship between
Date of first winter rains
15 Indio
10
5
0
en
15 Needles
....
CIl
Q)
>- 10
0
....
Q)
.0
E
5
~
Z
15 Phoenix
10
5
0 0 "k. <:>. "0
o/,
0t.,
&""
&"" v~
"' v. 01 // 7 7
U>/,
(9"..~
~~~1-7
o~
7
(9"..
(9"..
7
7
is'
7
7
Fig. 18.5. The frequency distribution of the date of the first winter rains for three sites
differing in the predictability of that winter precipitation. The first winter storm is defined
as the first storm after October 1 that was ;310 mm. Calculations are based on data from
the US Weather Bureau records (NOAA) during the period 1929-1979. The three figures
are arranged in a west-to-east transect from top to bottom