VZJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
Related Collections
Right arrow Watershed and Landscape Processes
Published in Vadose Zone Journal 3:1007-1016 (2004)
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

ORIGINAL RESEARCH

Regolith Water in Zero-Order Chaparral and Perennial Grass Watersheds Four Decades after Vegetation Conversion

Tanja N. Williamsona,*, Brent D. Newmanb, Robert C. Grahamc and Peter J. Shoused

a Dep. of Geosciences, Univ. of the Pacific, Stockton, CA 95211
b Los Alamos National Lab., MS J495 EES-2: Earth and Environmental Sciences Division, Los Alamos, NM 87545
c Dep. of Environmental Science, Univ. of California, Riverside, CA 92521
d USDA-ARS, U.S. Salinity Lab., 450 W. Big Springs Rd., Riverside, CA 92507
* Corresponding author (twilliam{at}pacific.edu)

Received 14 August 2003.



   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
In 1960, areas of chaparral were converted to perennial grass after a fire burned most of the San Dimas Experimental Forest in southern California. This conversion provided an opportunity to compare regolith moisture patterns of zero-order watersheds under native chaparral with those under nonnative veldt grass (Ehrharta calycina Sm.). We collected data as a function of vegetation type and watershed element to test the hypothesis that conversion from chaparral to grass altered water distribution in the vadose zone as a result of changes in the physical environment, including rooting depth and soil horizonation. Patterns in vadose zone water distribution during the dry season, including soil water potential and residual flux, were significantly different in converted areas, reflecting the different rooting habits of the two vegetation types. In chaparral areas, there was no significant change in soil water potential between the surface and the 150-cm depth; soil water potential was consistently below –1.5 MPa, reflecting the extensive root system. In grass areas, soil water potential was most negative close to the surface, where grass roots were most abundant. Plant available water was present below the 100-cm depth, suggesting that recharge to groundwater may occur under grass in average or wetter years. Under both vegetation types, the largest differences in residual water fluxes were near the soil–weathered rock contact. However, there was a significant relation between minor differences in fluxes and soil horizon boundaries, confirming the effects of vegetation conversion on soil properties and vadose zone soil water.

Abbreviations: SDEF, San Dimas Experimental Forest


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE SAN DIMAS EXPERIMENTAL FOREST (SDEF, Fig. 1a) is a 7000-ha area on the southern flank of the San Gabriel Mountains. It is representative of this part of southern California where dry-ravel (downward movement of dry sediment due to the steepness of slopes) equals or exceeds surface-water induced erosion (Anderson et al., 1959; Kraebel and Sinclair, 1940; Sinclair, 1953; Wohlgemuth, 1985). The consequently thin soils overlie weathered bedrock that is generally encountered at depths <60 cm. In this mediterranean environment, the native vegetation is chaparral, an evergreen, sclerophyllous, summer-dormant vegetation community (Hanes, 1974). Some chaparral species have roots that extend >8 m deep, where moisture held in weathered rock is accessed to survive long dry periods (Hellmers et al., 1955).



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 1. (a) Location map of the San Dimas Experimental Forest (SDEF) and the study area. (b) Schematic of a zero-order watershed showing watershed positions. The boundary between the soil and weathered rock indicates the irregularity of this contact and that it does not always follow surface topography.

 
Chaparral is a fire-prone community found on steep slopes where hillslope erosion and mass wasting are a concern. Fire rehabilitation in chaparral areas commonly includes seeding of nonnative grass species to decrease postfire erosion (Rice and Foggin, 1971; Riggan et al., 1985; Taskey et al., 1989; Wohlgemuth, 1985). Research in the chaparral environment has investigated the potential for increasing groundwater storage and base flow by introducing vegetation with lower evapotranspiration rates and interception amounts (Hibbert et al., 1982; Hopkins, 1958).

After a fire in 1960 burned more than 96% of the SDEF, large areas were rehabilitated using a combination of manual vegetation removal, herbicide application, and seeding (Dunn et al., 1988). Where chaparral was converted to annual grasses, interception was reported to decrease by 50%, potentially reducing the net loss due to evaporation from the canopy (Corbett and Crouse, 1968; McMillan and Burgy, 1960). Conversion of chaparral to annual and perennial grasses produced residual soil water that led to increased base-flow duration and runoff rates from first-order watersheds, as well as a threefold increase in rates of soil slumping (Corbett and Green, 1965; Hill and Rice, 1963; Rice and Foggin, 1971; Rowe, 1963; Rowe and Reimann, 1961). One rehabilitation method involved high density hand-seeding of perennial grasses in watersheds that were also stabilized by planting barley (Hordeum vulgare L.) along contours in 0.6-m intervals (Corbett and Green, 1965; Rice et al., 1965). Veldt grass was only 15% of the original mixture, but is now the predominant vegetation in areas of this high density perennial grasses and barley treatment.

Watershed and vadose zone hydrology is structured by interactions with vegetation that affect processes such as surface runoff, infiltration, and regulation of subsurface water (Horton, 1932). The effect of plant cover on precipitation interception and energy dissipation was established by Schumm and Lichty (1965) in their examination of effective precipitation and erosion. Other effects on hydrology stem from the influence of vegetation on soil structure, including root distribution and incorporation of organic matter in soil, influencing soil and weathered rock water patterns (Howard, 1981). Together, soil and the underlying weathered rock create the regolith. Regolith water distribution is further affected by evapotranspiration through plant uptake and insulation of the soil surface by vegetation canopy and litter (Howard, 1981; Selby, 1982). Vegetation is usually linked to regional climate or geology, however, making it difficult to separate effects that are solely due to plant type.

Understanding the long-term effects of vegetation conversion on water availability requires evaluation of water in the vadose zone. The relation between regolith water distribution, including water potential patterns and residual flux, and watershed behavior can best be understood in zero-order watersheds. Low-order streams are important in quantitative analysis because of their similarity in different drainage systems (Horton, 1945). Subsurface flow is a dominant process in water movement at this scale. Within the zone of rooting, residual flux (the net downward movement of water) reflects regolith moisture patterns that result from evapotranspiration (Phillips, 1994). Below the depth of rooting, residual flux is equated to groundwater recharge. Newman et al. (1997) compared residual fluxes under piñon (Pinus edulis Engelm.)–juniper [Juniperus monosperma (Engelm.) Sarg.] woodlands and ponderosa pine (Pinus ponderosa C. Lawson) forests and showed that evapotranspiration demand was principally supplied by the upper 10 cm of soil for both communities, but variability was greater under the piñon–juniper woodland. Differences in residual flux between up-slope, mid-slope, and down-slope positions in a Banksia spp. woodland were ascribed to differences in vegetation density (Sharma et al., 1991). This type of lateral variability in residual water flux is frequently enhanced by effects of soil horizonation and soil hydraulic properties (Johnston, 1987; McCord et al., 1997; Newman et al., 1997; Sharma et al., 1991). In areas of vegetation conversion, residual flux within the rooting zone should reflect resultant differences in evapotranspiration and soil properties.

To understand the decadal effects of vegetation conversion on watershed and vadose zone hydrology, the physical environment created by vegetation must be linked to regolith water distribution. The SDEF provides an opportunity to examine the effects of vegetation conversion on zero-order watershed hydrology. Our objective was to identify differences between the regolith water patterns in zero-order watersheds with native chaparral and those that underwent conversion to nonnative grasses in 1960. Our hypothesis was that conversion from chaparral to perennial grass resulted in higher water flux in the subsurface as a consequence of changes in horizonation and rooting depth, providing sufficient water to increase streamflow and soil slumping in areas converted to grass vegetation.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Environmental Setting
On the SDEF, Typic Xerorthents are the most prevalent soils (Ryan, 1991), but Typic Haploxeralfs are also common (Williamson and Graham, 1998). Bedrock, a mixture of highly weathered banded gneiss and granitics, is generally encountered at <60 cm depth (Nourse, 1998). The regional structure has a NNW to NNE dip of 30 to 58° (Nourse, 1998). Krammes (1969) reported that the water retention curves of the soil and weathered bedrock are similar. The mediterranean climate produces cool, wet winters and hot, dry summers with an annual temperature range of about 38 to –4°C. Annual precipitation varies from 292 to 1224 mm, with a mean of 678 mm (Dunn et al., 1988); most precipitation occurs as rain between November and April. On the SDEF, the 1- to 3-m-tall, dense canopy, chaparral community includes chamise (Adenostoma fasciculatum Hooke & Arn.), scrub oak (Quercus dumosa Nutt.), hoary-leaf ceanothus (Ceanothus crassifolius Torr.), black sage (Salvia mellifera Greene), bigberry manzanita (Artcostaphylos glauca), California buckwheat (Eriogonum fasciculatum Benth.), and yorbasanta (Eriodictyon spp.).

As a result of vegetation conversion after the 1960 fire, perennial veldt grass is now the dominant vegetation type in some areas of the SDEF. Veldt grass, a southern Africa prairie grass, has been used internationally for erosion control (Mulroy et al., 1992) because of its adaptability to mountainous regions with sandy soils and a mediterranean climate (Tothill, 1962). Veldt grass commonly occurs in association with other sclerophyllous vegetation types, including the heath of South Australia (Tothill, 1962). On the basis of soil water potential data, Specht (1957) deduced that veldt grass has evapotranspiration rates approximately equal to those of heath vegetation. Veldt grass roots grow as deep as 2.4 m, and new roots are initiated at the surface following rain events (Tothill, 1962).

Six watersheds, all <700 m2, were selected, three with chaparral vegetation and three with veldt grass. Watersheds range in elevation from 830 to 920 m, and average slope ranges from 33 to 52% (15–23°). All have easterly aspects. Mean surface and subsurface characteristics for watersheds of each vegetation type are summarized in Table 1 . To capture spatial variability within each watershed, three watershed elements were sampled for vadose zone water distribution (Fig. 1b):


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of surface and soil characteristics for watersheds.{dagger}

 
Vadose Zone Water
Vadose zone water potential was measured in six watersheds at the end of the 1997 summer season, the driest time of year in the SDEF. Rainfall for this year was 711 mm, close to the long-term average. The channel, side-slope, and summit elements were sampled in three watersheds of each vegetation type during September 1997. Because of a rainstorm in the middle of sampling, additional samples of the upper 20 cm (the approximate depth of wetting from the September storm) were measured in November 1997 after a dry period. Samples were collected in known increments (usually 10 cm) to the depth allowable by hand auger, including the soil and underlying weathered rock.

Regolith water potential was compared rather than water content to relate to plant usage and minimize differences due to regolith properties. Water potential was determined for each depth increment sampled and was calculated from regolith water activity as determined by a chilled-mirror dew point technique on a Decagon Devices, Inc. (Pullman, WA) Aqualab CX-2 (Rawlins and Campbell, 1986). For soils sampled immediately after the September storm, data reported for the upper 20 cm are from the November sampling. Vadose zone water potential was compared by vegetation type and by watershed element. Standard error is reported for all data. Differences in means were tested using a two-tailed t test for samples with unequal variance. Statements of significance indicate a p ≤ 0.05.

Residual Flux
Residual flux is regolith water that is in excess of evapotranspiration needs and thus is conducted downward through the rooting zone. Residual flux was analyzed using natural Cl– as a tracer (Allison et al., 1994; Phillips, 1994; Stone, 1984). Regolith samples were allowed to air dry and then a 50-g split was mixed with 50 mL of deionized–distilled water. The mixture was allowed to equilibrate for 3 d, during which it was periodically stirred. The leachate was centrifuged at 11000 rpm for 22 min (based on a Stoke's Law calculation) and filtered through a disposable Gelman 0.2-µm filter. Chloride concentrations were measured on an Alpkem RFA/2 320 ion colorimeter (OI Analytical, College Station, TX). Water content was determined gravimetrically (Gardner, 1986), and bulk density values were determined using 5.4-cm-diam. cores collected during soil description (Table 1). These data were used to calculate the Cl– concentration in regolith water from each depth increment.

Meteoric Cl– in precipitation and dry deposition moves into soil with infiltrating water and serves as a natural tracer of water movement in soil and weathered rock (Allison et al., 1994). Water Cl– concentrations in the regolith increase as evapotranspiration depletes available moisture. Minimal Cl– is taken up by plants during evapotranspiration, thus allowing a calculation of regolith water age using a mass-balance approach. All Cl– samples were taken within the known rooting zone of SDEF vegetation, so residual flux, not recharge, will be discussed (Phillips, 1994). Stone's (1984) cumulative Cl–– cumulative water method was used to derive residual fluxes. One of the assumptions of this method, downward piston flow, is violated in the SDEF, where lateral subsurface flow is common. This lateral subsurface flow should be present throughout the study area so should not cause anomalous effects between sites. Changes in residual flux rates were identified using cumulative Cl––cumulative water depth profiles (Stone, 1984). Linear segments on these plots indicate zones of constant flux. Residual moisture flux (R) was determined as

[1]
where P is the mean annual precipitation (678 mm), Clp is the [Cl–]precipitation (0.48 mg L–1), and Clsw is the [Cl–]soil water (Allison et al., 1994; Phillips, 1994). The mean Clp value was calculated from NADP (1998) data for a site 1 km from the study area watersheds and is a best estimate of both dry and wet deposition.

Soil water age (A) was determined for the base of each flux zone using the relation

[2]
where

[3]
and {theta}v is the volumetric water content, {Delta}z is the sampling interval, and d is the depth at the base of the flux zone (Newman et al., 1997).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Vadose Zone Moisture
Vadose zone water distribution was compared by vegetation type (for three watershed elements in each of three watersheds; n = 9 for each vegetation type) and watershed element (one replicate from each watershed; n = 3). Analyses included regolith water potential (Fig. 2) and regolith water Cl– for residual flux (Fig. 3 and 4) .



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 2. Depth profiles of regolith water potential. Three watersheds of each vegetation type were sampled at each position. Mean data are reported for 10-cm depth increments. (a) Vegetation means derived from all profiles under each vegetation type (n = 9). Note the significant changes at depths 70 and 100 cm under grass vegetation and that water potential is significantly different between the two vegetation types below 100 cm. (b) Position means for chaparral watersheds (n = 3). (c) Position means for grass watersheds (n = 3). Standard error is shown for each depth increment.

 


View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Depth profiles of regolith water Cl– concentrations. Water Cl– was measured on the same samples used in the water potential analysis. (a) Vegetation means derived from all profiles under each vegetation type (n = 9). The same significant changes are present under grass vegetation at 70 and 100 cm. The high spike at 145 cm for chaparral is due to a single depth increment at one site and suggests local water uptake by a nearby root. (b) Position means for chaparral watersheds (n = 3). Note that the Cl– concentrations from the summit element are significantly lower within the upper 50 cm. (c) Position means for grass watersheds (n = 3). Standard error is shown for each depth increment.

 


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 4. Cumulative Cl– vs. cumulative water plots from one watershed of each vegetation type. Chloride (Clswi) and water ({theta}v) accumulation with depth were used to identify portions of the regolith with different fluxes. The calculated moisture flux (m yr–1) is reported next to the appropriate straight-line segment (e.g., 0.0031 for the A and Bw horizons of the chaparral channel). Inflection points in the cumulative curves were compared with soil horizon boundaries based on the known depth of each sample point. Soil horizons and boundary depths (cm) are indicated. Arrows indicate the transition between input and leaching zones for each profile (averaged in Table 3). Note that minor inflection points frequently coincide with horizon boundaries.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Residual water zone characteristics.{dagger}

 
Regolith water potential (Fig. 2, Table 2) was compared instead of water content to relate results to plant usage and to minimize differences due to regolith properties. Under chaparral, water potential shows minor fluctuations with depth, but there is no significant change between the surface and 150-cm depth (Fig. 2a). Under grass, the depth profile can be separated into three significantly different segments (p < 0.001). Above 70 cm, regolith water potential is the most variable, and there is no difference between data from chaparral and those from grass watersheds. Between 70 cm and 100 cm under grass, there is an increase in regolith water potential toward less negative values (indicating higher water availability). Below 100 cm under grass, there is little variability in regolith water potential with depth, and there is plant-available water (i.e., water potential > –1.5 MPa). This is the depth range where the chaparral and grass curves are significantly different (p < 0.01).


View this table:
[in this window]
[in a new window]
  		Table 2. Mean regolith water potential and Cl– values.{dagger}

 
A difference due to vegetation type is evident from the relative position of the chaparral and grass Cl– profiles (Fig. 3a, Table 2); however, the high variability in regolith water Cl– concentrations for individual depth increments makes this difference statistically nonsignificant. Patterns evident in the regolith water Cl– curves for each vegetation type are similar to those observed with water potential data. In all chaparral plots, Cl– concentrations fluctuate throughout the depth sampled. Under grass, the same three significantly different zones are evident in the Cl– depth profiles (i.e., variable from 0–70 cm, transitional from 70–100 cm, and approximately constant below 100 cm).

Comparison by watershed element is not as clear. No differences due to watershed element are evident in the grass areas for either water potential (Fig. 2c) or Cl– data (Fig. 3c). In contrast, there is a difference among watershed elements in the upper 50 cm under chaparral. Regolith water potential is more negative at the summit element, indicating the least available water of the three elements (Fig. 2b), but the variability makes this difference statistically nonsignificant. In contrast, Cl– concentrations are significantly higher at the side-slope and channel relative to the summit throughout most of the upper 50 cm, suggesting higher flux at the summit (Fig. 3b).

Residual Flux
Chloride data were used to assess water movement in the subsurface. Because of the significant differences seen between vegetation types in the water potential profiles and the significantly different zones seen in both water potential and Cl– depth profiles under grass, this portion of the analysis focused on comparisons between vegetation types. Cumulative Cl––cumulative water plots were used to identify changes in flux with depth (Fig. 4); linear segments were interpreted as regolith zones with constant flux (Newman et al., 1997; Stone, 1984). For both vegetation types, two zones were identified based on inflections in these cumulative plots (Table 3, Fig. 5) . Data were averaged from three elements in three watersheds of each vegetation type (n = 9), so the depth of the transition between the upper and lower zones varies.



View larger version (41K):
[in this window]
[in a new window]
  		Fig. 5. (a) Histogram of regolith water flux in the input (n = 9) and leaching (n = 12) zones under chaparral and grass. Note that the 0.01 to 0.015 interval includes an observation from both the chaparral input and the chaparral leaching zones. (b) Histogram of regolith water age at the deepest part of the input (n = 9 for chaparral and grass) and leaching zones (n = 7 for chaparral and n = 9 for grass). The transition between flux rates was interpreted as the base of the input zone. The base of the leaching zone is the deepest interval sampled (as allowable by hand auger).

 
Chloride concentrations in the upper zone are high, indicating that downward movement of water (residual flux) is relatively slow. In the lower zone, Cl– concentrations are less, indicating leaching of regolith water. For both vegetation types, residual water flux is higher at depth, but this increase from the upper zone is only significant under grass. The upper zone flux is barely differentiable between chaparral and grass (p = 0.18). There is no statistically significant difference in the lower zone flux because of high variability under both vegetation types.

Regolith water age was calculated for the lowest depth increment in each zone (Table 3). For one depth increment, the deepest sample (140 cm) from a chaparral summit element, the Cl– analysis estimated regolith water to be >60 yr old. However, at 14 of the 18 sites, all water sampled was <10 yr old (Fig. 5).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Vadose Zone Water and Rooting Characteristics
Analysis of both water potential and water Cl– data showed similar implications for vadose zone water movement (Fig. 2a and 3a). For both data sets, the depth profiles show a change in the relative position of the curves for the two vegetation types within the transitional 70- to 100-cm depth zone. Above the transitional zone, water depletion is greater under grass, and below this zone, water depletion is greater under chaparral. Variability below the 70-cm depth is less under grass where the Cl– concentrations are significantly lower, indicating a higher residual flux. Water potential and water Cl– data significantly correlate under grass (R = –0.91), but not under chaparral (R = –0.06). This difference in correlation between the two vegetation types is significant (p < 0.01). Differences in subsurface water patterns between the two vegetation types reflect the different rooting characteristics.

Global analysis of grasslands has shown that 75% of roots are concentrated within 20 to 30 cm of the surface (Jackson et al., 1996; Sims and Coupland, 1979). The fibrous grass root system in the SDEF consists of a high number of roots that are evenly distributed near the surface. Water potential and residual flux data indicate that evapotranspiration loss under grass is highest in the upper 5 to 10 cm and decreases with depth (Fig. 2, 3, and 4). This corresponds to the average A horizon thickness in these grass watersheds, where the very fine roots are most abundant (Table 1). Under grass, vadose zone water is relatively abundant below the 100-cm depth. Sims and Coupland (1979) reported only 9% of grass roots below the 90-cm depth. Tothill (1962) reported veldt grass roots to 2.5 m, but during soil description in the SDEF grass watersheds, roots were rarely observed below the 90-cm depth (Williamson, 1999). Apparently, sufficient moisture is available in the upper 1 m of soil at these sites to sustain grass growth. Hill and Rice (1963) saw 9 to 24% percolation under perennial grass during rain years that produced 0 to 8% percolation from chaparral soils. At the end of the 1997 rain year, during which total rainfall was close to the 70-yr average, plant available water was present below 100 cm, suggesting that recharge to groundwater may occur under grass in average or wetter years.

Chaparral species have elaborate root systems that access water throughout the soil and weathered rock (Hellmers et al., 1955). Roots are heterogeneously distributed, resulting in variable regolith water usage. This heterogeneous root distribution is reflected by the irregularity of the soil water potential (Fig. 2) and soil water Cl– (Fig. 3) depth plots to >100 cm. The chaparral is efficient at obtaining water throughout the depths sampled.

Vadose Zone Water and Watershed Element
Comparison of depth profiles from the channel, side-slope, and summit suggests a relation between vadose zone water distribution and watershed element. Chloride data from chaparral watersheds show significantly higher fluxes at the summit relative to the channel and side-slope (Fig. 3b). Variability in regolith water distribution due to slope position has been noted elsewhere. McCord et al. (1997) showed that differences in Cl– concentration due to slope position could be explained by the effects of surface slope on subsurface flow paths. In an analysis of recharge variability, Sharma et al. (1991) related differences in recharge among slope positions to differences in vegetation density. Surface cover in chaparral watersheds in the SDEF was lower and/or more variable at the summit (71 ± 34%) relative to the side-slope (94 ± 8%) and channel (74 ± 12%) (Williamson, 1999). Lower surface cover suggests lower transpiration rates and higher residual flux.

Residual Flux and Vegetation Type
For both vegetation types, two zones were identified based on differences in cumulative Cl––cumulative water plots (Table 3). We designated the upper zone as the "input zone" because this is where Cl– enters the system. Chloride concentrations are high in this zone and downward movement of water (residual flux) is relatively slow since most water in this zone is used for evapotranspiration. We designated the lower zone, where Cl– and water have moved downwards, as the "leaching zone." This water has moved beyond the largest concentration of roots, so the effects of evapotranspiration are minimized. Chloride concentrations are less and residual flux is increased as regolith water is available for leaching, and ultimately, recharge to groundwater. In all but one case, the transition between these zones is within or below the first horizon with weathered rock (Fig. 4). Data were averaged from three elements in three watersheds of each vegetation type (n = 9), so the depth of the transition between the two zones varies. For both vegetation types, residual water flux is higher in the leaching zone; however, this increase from the input zone is only significant under grass. The input zone flux is barely differentiable between chaparral and grass (p = 0.18). Under both vegetation types, infiltration capacity (measured by cylinder infiltrometer) at the side-slope and summit elements exceeds the highest recorded rainfall rates for the SDEF (Reimann and Hamilton, 1959; Williamson, 1999), suggesting that differences in vadose zone water are not attributable to the capacity for water movement into the soil. The differences in rooting habit between the two vegetation types do not produce a statistically significant difference in the leaching zone flux because of the high variability under both vegetation types. However, water potential and Cl– data indicate that the concentration of grass roots near the surface results in increased evapotranspiration near the surface and a significantly higher flux in the lower, leaching zone, relative to chaparral.

The Cl– analysis showed no significant differences in the age of regolith water in the depths sampled; neither by vegetation type nor by zone. In the leaching zone, variability in water age is lower under grass relative to chaparral because of the paucity of roots in this zone. Most importantly, except for one 10-cm increment, all of the regolith sampled showed water ages <40 yr, indicating that most of the regolith water sampled entered these watersheds after the vegetation conversion. This confirms that differences in vadose zone hydrology between the two vegetation types reflect the alteration of the regolith in watersheds that were converted from chaparral to grass.

Calculation of fluxes differentiated the input and leaching zones based on the largest changes in slope of the cumulative Cl––cumulative water curves. However, minor differences in flux are visible from inflection points in several cumulative plots (Fig. 4). Inflection points in the cumulative curves were compared with soil horizons described in the field (Williamson, 1999). In several instances, the depth of horizon boundaries and the depth of cumulative curve inflection points coincide, approximating a 1:1 relation (Fig. 6) . Statistical analysis (Snedecor and Cochran, 1967, p. 432–436,) shows that neither regression has a slope that is significantly different from one. Horizon boundaries signify differences in soil physical characteristics, including soil texture, porosity, and organic matter content. These soil physical characteristics affect how water moves between and through horizons. In addition to hydraulic conductivity and water retention, soil horizons also affect, and are affected by, root distribution. The combination of soil physical characteristics and root distribution causes changes in flux at horizon boundaries; it is noteworthy how well the cumulative Cl– and cumulative water content method can identify horizon-related changes in hydrologic behavior.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6. Depth of soil horizon boundaries vs. transitions in residual flux rate. Depths of inflection points in the cumulative plots (Fig. 4), signifying minor differences in flux, were compared with depths of soil horizon boundaries described in the field. Data are included from all nine profiles for each vegetation type. The chaparral data have a higher coefficient of determination. However, correlation for the 0- to 53-cm depth (the depth range for which data are available for grass) is similar to that for grass. Neither regression equation has a slope that is statistically differentiable from one.

 
In addition to indicating horizon effects on vadose zone hydrology, the correlations also show that the vegetation conversion has affected water dynamics in the regolith. Description of soil physical characteristics showed changes in A horizon thickness and physical properties as a result of the vegetation conversion (Williamson et al., 2004). Correlation of differences in flux to soil horizons indicates that the vegetation conversion affected water dynamics in the regolith. In contrast, surface covers of approximately 90% in both chaparral and grass areas suggests little difference in interception by the canopies, and there is no difference in infiltration rates between the two vegetation environments (Williamson et al., 2004). Consequently, changes in watershed behavior, including increased base flow, runoff rates, and soil slumping, are most likely the result of increased leaching of regolith water below grass vegetation.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
This research on regolith water distribution illustrated that vadose zone water availability under grass increases (i.e., potential becomes less negative and residual flux increases) between 70 and 100 cm, resulting in the possibility for annual recharge below this depth. The concentration of grass roots near the surface in converted watersheds and consequent evapotranspiration result in low vadose zone water availability in the soil. Significantly more water is accessible in the weathered rock below the 70-cm depth, where the frequency of grass roots decreases. In chaparral watersheds, water is accessed throughout the soil and weathered rock, resulting in variable amounts of water throughout the depth sampled, even during the driest time of year. Under both vegetation types, the distribution of vadose zone water was significantly related to soil and weathered rock horizonation. In summary, these data show that the vegetation conversion from chaparral to grass altered the vadose zone hydrology of zero-order watersheds in the SDEF. The consequences of this change were seen by the increased rates of soil slumping (Rice and Foggin, 1971) and increased base flow from converted first-order watersheds (Rowe, 1963) at the SDEF following the conversion. Thus, modeling of watershed behavior and vadose zone hydrology must not overlook the effects of changing vegetation.


   ACKNOWLEDGMENTS
 
Research supported by the Cooperative State Research Service, USDA (Agreement no. 93384208793).


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Vadose Zone JHome page
B. D. Newman and R. C. Graham
Species-Level Impacts on Chaparral Root Zone Hydrology
Vadose Zone J., August 13, 2008; 7(3): 1110 - 1118.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
J. E. Schoonover, B. G. Lockaby, and B. S. Helms
Impacts of Land Cover on Stream Hydrology in the West Georgia Piedmont, USA
J. Environ. Qual., October 27, 2006; 35(6): 2123 - 2131.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
Y. A. Wood, T. Meixner, P. J. Shouse, and E. B. Allen
Altered Ecohydrologic Response Drives Native Shrub Loss under Conditions of Elevated Nitrogen Deposition
J. Environ. Qual., January 3, 2006; 35(1): 76 - 92.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Williamson, T. N.
Right arrow Articles by Shouse, P. J.
Related Collections
Right arrow Watershed and Landscape Processes

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome