In 2004, the UZ Flow Project joined a US Department of the Interior interdisciplinary study related to the evaluation of ecological habitat quality in the Mojave Desert. This work is supported by and carried out in collaboration with other investigators in the Science on the DOI Landscape in the Western Region (SDOILWR) program*. It complements work done under the topically related Recoverability and Vulnerability of Desert Ecosystems (RVDE) project of the Department of the Interior.
[*Our project’s role falls within Science Objective 2 (SO2) of the SDOILWR: Combined geospatial models in the Mojave Desert--Geomorphic process models of
deposit and soil texture, soil hydrology, and surficial geology. Within SO2 the work done by the UZ Flow Project is in Task 3: Three-dimensional Soil Moisture Model.]
Habitat suitability in the desert depends on soil moisture and its variability in several ways. Field water capacity is central to the assessment of water available
to plants after infiltration. The frequency and duration of dry-end plant-stress conditions is similarly important. The frequency and magnitude of runoff events, determined largely by precipitation
intensity relative to infiltration capacity, affects both the soil water distribution and processes of erosion. Soil moisture and its cyclicality also strongly influence the vulnerability of soil to compaction
and the recoverability of soil that has been compacted or disturbed in other ways that impair the soil's ability to support plant and animal life.
Soil moisture, hydraulic properties, structure, and texture are interrelated and are controlled by a wide variety of soil-modifying influences. These affect
vegetation species composition, cover, and productivity, as well as the rates of recovery of vegetation from disturbance. In particular, the soil texture or grain-size distribution is a useful predictor of other
physical characteristics. A model that evaluates the soil hydraulic properties in relation to soil texture and structure will be combined with a spatially-distributed
three-dimensional model of texture developed by SDOILWR collaborators from the Western Region Earth Surface Processes Team, and other scientists
investigating surface geology, soil texture, and climate. The three-dimensional texture model is created from geomorphic-process models for alluvial-fan environments coupled with other
process-driven texture models. Such a model that predicts texture as a function of depth and lateral extent is needed for evaluating ecological function.
The picture on the right shows a small excavation of a stream channel shortly after infiltration from a rainstorm. Wetter soil appears darker in color. The visible
pattern of water content results mostly from preferential drying on the open face; even slight contrasts in soil texture and other properties can have a great effect on how water is distributed within
the soil.
In the desert, soil water input is assumed to usually come from direct infiltration of precipitation. Runoff is generated on occasions (possibly rare) when the
precipitation rate exceeds the infiltration capacity. Typically within the Mojave National Preserve, runoff-generating areas have a silty, low-permeability vesicular layer near the soil surface
(Av horizon; example aggregate pictured below) that limits infiltration and strongly influences most aspects of soil-water behavior. After water has infiltrated, evapotranspiration and gravity-driven
downward flow are the main processes in the redistribution of soil water, probably exceeding the importance of horizontal flow and moisture-gradient-driven flow.
The measurements of soil moisture during infiltration and redistribution from the field tests, as well as from ongoing monitoring and other sources, serve to
(1) provide quantitative estimates of unsaturated hydraulic properties to be used in a dynamic soil moisture model, (2) evaluate the characteristic flow behavior of water in Mojave soils for
improved understanding of the basic desert soil-water processes, and (3) provide a set of quantitative observations to be used in testing a dynamic soil moisture model.
Measured inflow rates during the infiltration period of the experiments indicate infiltration capacity. Measured unsaturated-zone moisture conditions
during infiltration and 1-2 months of redistribution yield estimates of field water capacity and soil hydraulic properties.
Several studies indicate a likelihood that texture-based quantitative inference of soil hydraulic properties will be useful, though additional input and
modifications will be necessary for the more pronounced structure of Mojave surface soils. Nimmo and others (2002) showed that hydraulic properties of
sediments from a wash in the southwestern part of the Mojave Desert can be related to each other by simple scaling based primarily on particle-size distributions. Soil structure and compaction
also affect soil hydraulic properties. Winfield (2000) and Winfield and Nimmo (2005) explored effects of structure and geological processes on hydraulic properties
for Mojave soil samples and found that textural effects were dominant to a high degree. Nimmo and Akstin (1988) found that for a sandy soil, some types of structural disruption
strongly affected unsaturated hydraulic properties, though simple piston-driven compaction had no significant effect except near saturation.
Effective hydraulic properties will be taken to be uniform within each of the characterized soil layers. Evapotranspiration will be incorporated as a term expressing the rate of water loss from the root zone, quantified from empirical data. Other modes of hydraulic redistribution by plants would not be explicitly accounted for but to some
degree may be implicit within the field-measured soil hydraulic properties. As dictated by the property characterizations that will be available, the vertical resolution (centimeters) is much finer
than horizontal resolution (tens of meters). Modeling of thin layers may be important because phenomena such as infiltration rates and water availability to plants are sensitive to the layered
structure. If critical phenomena (e.g. runoff) vary significantly at sub-grid-block scales, this could be accounted for with a heterogeneity parameter based on the degree of variability of
properties (e.g. infiltration capacity) within a grid-block. The high spatial-resolution climate model being developed by another Task of SO2 will provide input to the soil-moisture model. To
avoid missing time- and intensity-sensitive effects such as runoff with the coarse (daily) time scales for climate information, we will evaluate and compensate for such effects with Monte Carlo
techniques or other statistical methods for simulating the likely duration and intensity of brief storms.
Objectives
To elucidate processes and relationships that influence habitat quality, we are measuring unsaturated hydraulic
properties and evaluating soil water behavior in field experiments. The knowledge gained will be combined with data related to soil moisture, geology, and climate to make robust models of soil
moisture that lead to regional ecological models. Our main goal is to develop a three-dimensional regional representation both of soil hydraulic properties and of soil moisture dynamics that will
supply quantitative information relevant to issues of field water capacity, the frequency and duration of dry-end plant-stress conditions, and the frequency and magnitude of runoff events.
Our approach has three main elements: (1) measurement, especially in the field, of soil water properties and behavior, (2) property transfer modeling to relate soil hydraulic properties to
textural and geologic characteristics, and (3) large-scale dynamic soil-moisture modeling, to be applied in connection with climate models to predict conditions essential to the evaluation of
habitat quality.
Measurements
The central activity of our soil-moisture measurements is to conduct field infiltration tests at three locations chosen to
represent different stages of soil development. The three field-test sites are on alluvial fans between the Providence Mountains and Kelso Wash, north of Kelso. At each site several types of
soil moisture probes installed at the surface and subsurface down to about 1 meter provide data on dynamically changing soil water in response to an artificial input of as much as 2000 liters
of water. Water supplied in these tests is ponded to a low level within an infiltration ring 1 meter in diameter for a period between two and three hours. This part of the test resembles conditions of flood
or extremely heavy rain. Simulation of such an extreme event is necessary so that the set of measurements before, during, and after the infiltration period will reflect the full range of soil moisture
conditions occurring in the desert. These experiments are similar to the flood-infiltration/instantaneous-profile experiments of Nimmo and others, 1999.
Property-transfer modeling
Results of the field soil-water experiments will be extended using textural and geologic information to
characterize soils over a large portion of the Mojave Desert. The purpose of developing this property-transfer model is to infer soil hydraulic properties and conditions from other
types of data obtained through various tasks of this Science on the DOI Landscape project as described above.
Dynamic Soil-Moisture Modeling
The model for predicting soil moisture for habitat evaluation will be implemented as a set of
1-dimensional models, one to each grid block of the domain. The hydraulic property and dynamic soil moisture models will be tested by comparing their results to field-measured data and
by evaluating their effectiveness in representing behavior known to occur, for example the cycling of water content within extremes typical of the Mojave study area.
| National Research Program | USGS | Water Resources | UZ Flow Homepage |