INTRODUCTION
The work described in this poster is focused on the water
use of a sacaton grass (Sporobolus wrightii) flood plain and a mixed
grass/mesquite (Prosopis spp.) community. These two types of vegetation
communities are often associated with riparian areas across the Southwest
U.S. Numerous observations indicate that these plants are deep rooted; thus,
they have been thought to rely mainly on water taken up from the water table
when they exist in riparian corridors.
One of the purposes of this study is to improve the representation
of the evapotranspiration (ET) process within groundwater models. Current
basin aquifer modeling studies often rely upon coarse, empirical estimates
of riparian corridor ET to validate their modeling calibrations. Such estimates
of ET are often based on indirect measurements and methods developed elsewhere
and then applied (in this case) to the San Pedro Basin using local meteorological
information. In this study we employ micrometeorological techniques to measure
the fluxes of water and energy in both of these communities over several
seasons. |
METHODOLOGY
Two vegetation study areas nearby Lewis Springs on the
San Pedro River flood plain were chosen as the field sites for the study
(See Figure 1). One site was established in the mesquite-dominated upper
portion of the floodplain. The second was established in the sacaton grass-dominated
lower portion of the floodplain, between the mesquite area and the cottonwood/
willow forest gallery immediately adjacent to the river. 
To determine the amount of evaporation coming from these
two sites, meteorological towers were erected in 1996. These towers were
equipped with a set of standard meteorological instruments which measured
air temperature, relative humidity, incoming solar radiation, air pressure,
wind speed, wind direction, and precipitation. Additionally, the towers
were outfitted with instruments which measured the available energy (net
radiation minus ground heat flux) and Bowen ratio. With these instruments
the amount of energy that is consumed at the land surface by the evaporation
of water was determined over the course of several seasons.
In order to understand possible controls on evaporation,
we also monitored the state of the vadose zone soil moisture and the depth
to groundwater. Campbell Scientific Inc. (CS-615) water content reflectometers
were installed in a vertical transect (at .1, .25, .50, 1.0, and ~2.0 meters)
into the ground to measure the soil moisture under the vegetation. Piezometers
have also been installed to measure the fluctuations in the groundwater
table. |
 RESULTS
Figure 2 shows the cumulative water loss and incoming precipitation
over the two vegetation biomes from late February to December, 1997. The
mesquite grew leaves and became active after the last freeze of the spring.
From this point on to the first freeze of the autumn when the tree transpiration
immediately stopped, the mesquite transpired at nearly a constant rate (~1.9
mm/day). The total amount of evaporated water was nearly twice the amount
of incoming precipitation during this time frame. For the sacaton grassland,
significant transpiration activity did not occur until the onset of the
monsoon rains. |
To determine how precipitation influences the state of
the soil moisture, Figure 3 shows the evolution of the daily-average volumetric
soil moisture at three different depths into the soil. Data from deeper
depths (not shown) indicate no change in soil moisture throughout the entire
year. Remarkably, the only large changes in the soil moisture appear to
occur only within the top 25 cm for the grass site, and within the top 50
cm for the mesquite area. The increase in soil moisture only began to occur
after the onset of the heavier monsoon rains in early August (DOY 210).
The winter of 1997 was very dry and thus we suspect that little recharge
of moisture occurred in the vadose zone. The difference between the sites’
soil moisture status is largely due to the differences in the respective
soil types. The sacaton region has less-permeable silty soils while the
mesquite site is more sandy. |
DISCUSSION
Our observations clearly indicate a difference between
the water use characteristics of the sacaton grassland this site in this
year, the sacaton does not appear to be acting as a phreatophyte as its
water use is strongly linked to recent precipitation and growing season
length. Also, it is apparent that very little changes occur in the soil
moisture status of the vadose zone under the sacaton except in the near-surface.
Thus, indicating that the grassland must be relying upon shallow roots to
obtain moisture.
For the mesquite site, our results suggest that this biome
has a deeper source of water. The mesquite leafed out in May and stayed
active throughout the dry summer months of May, June, and July. The mesquite
evaporation rate exhibited little change with the addition of monsoon precipitation.
Most likely the slight increase in mesquite ET at the onset of the monsoon
rains is largely due to an additional contribution from bare soil and grass
evaporation rather than enhanced transpiration from the mesquite.
Since the soils are well-drained under the mesquite and
there were little changes seen in the dry (around 0.1 m3/ m3) and deeper
soil levels (at 1  and
2 meters), the mesquite at this site rely on a deeper source for moisture.
However, piezometer observations of the 10 meter-deep water table indicate
no diurnal fluctuations, which normally we would expect to occur under phreatophyte
withdrawal. Thus we suspect that the principal source of water for mesquites
lies in deep, vadose zone moisture.
Clearly, there are certain areas at some riparian sites
where mesquite grow in much more vigorous and denser stands than at our
site in the Upper San Pedro. Almost certainly, these areas are taking up
water directly from the near-surface aquifer. Figure 4 presents 1995 observations
of cumulative precipitation and ET over a very dense thicket of predominantly
mesquite, tamarisk, and some willow made in the Santa Cruz river valley
near Rio Rico, Arizona (Unland et al., 1998). At this site, the water table
was very near the surface and sometimes resulted in the formation of shallow
pools in low lying areas. Although the rainfall was much greater at this
site during this year, the strength of ET (yearly ET is 600 mm in excess
of precipitation) is vastly greater than our site. We suggest that this
is further evidence that the mesquite at our study site do not have free
access to groundwater. |
CONCLUSIONS
This research quantifies the evaporative demand of two
dominant riparian biomes. Preliminary results suggest that strength of the
vegetation water demand (for these two vegetation types) for the Upper San
Pedro are not as high as previously estimated. Most importantly, the vegetation
was much less of a groundwater user than previous estimates assumed. This
leads us to wonder if the dominant, and perhaps only significant, groundwater
consumptive use in the Upper San Pedro Basin is determined solely by the
cottonwood/willow gallery that stands adjacent to the river. |
ACKNOWLEDGMENTS
The authors wish to thankfully acknowledge support for this research.
This includes financial assistance from the EPA STAR Graduate Student Fellowship
Program, the USDA-ARS Global Change Research Program, NASA Grant W-18997,
and the Arizona Department of Water Resources. Additionally, we graciously
thank the US-ARMY Fort Huachuca Meteorological Support team, the US Bureau
of Land Management, and especially all the staff from the USDA-ARS located
in Tucson and Tombstone, Arizona. |
REFERENCES
Unland, Helene et. al., 1998: Evaporation from a riparian system in a
semi-arid environment. Hydrological Processes. In Press.
|
homepage