| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
a Dep. of Land, Air and Water Resources, Univ. of California, One Shields Ave., Davis, CA, 95616
b Tree Division Services, Parks and Recreation, Stockton, CA 95202
c Kelley & Associates Environmental Sciences, 216 F St. #51, Davis, CA 95616
* Corresponding author (atogeen{at}ucdavis.edu).
 |
ABSTRACT |
|---|
 TOPNOTES ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: MPWT, maximum perched water thickness • RMF, redoximorphic feature
|
INTRODUCTION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
In the Mediterranean climate of California, vernal pools function as wetlands during the winter and spring before completely desiccating by late spring or early summer. Extremes in the seasonal variability of the hydrologic status range from saturation during the winter–spring period to desiccation (water potential less than –1.5 MPa) during summer and fall. In spite of their ephemeral nature, they often meet standards of the jurisdictional wetland definition for hydrology and vegetation characteristics; however, hydric soil features can often be challenging to identify (Environmental Laboratory, 1987).
The presence of hydric soil indicators is critical to the delineation of jurisdictional wetlands on these landscapes. A variety of hydric soil indicators may apply to vernal pool wetlands in northern California: (i) in vernal pools, a depleted matrix at least 5 cm thick within 15 cm of the soil surface (hydric soil indicator F9), (ii) in closed depressions, 5% or more distinct or prominent redox concentrations as masses or pore linings in a 5-cm-thick layer within 15 cm of the soil surface (hydric soil indicator F8), (iii) in the case of red parent materials, a 10-cm layer 7.5YR or redder with chroma of 3 or less and 2% or more redox depletions or concentrations within 30 cm of the surface (test hydric soil indicator TF2), or (iv) instances where redox dark surfaces are observed, such as a matrix color value of 3 or less and chroma of 2 or less and the presence of redox concentrations in a 10-cm layer within 30 cm of the surface (hydric soil indicator F6) (Hurt and Vasilas, 2006).
Hydric soil indicators can be difficult to apply to many vernal pools in California because soil properties change across short distances (<1 m) within the complex microtopography (Nikiforoff, 1941; Weitkamp et al., 1996). In many instances, seasonally submerged soils are the only portions of the landscape that meet hydric soil criteria, often representing just 3 to 5% of the landscape. There are also several specific problems that contribute to difficulties in recognizing hydric soil indicators in vernal pool landscapes: (i) high organic matter content associated with mineral soils can obscure redoximorphic features in the upper soil horizons, (ii) in young volcanic soil materials, the naturally low chroma of the parent material hinders identification of redox-depletion zones, (iii) high Fe content can mask redoximorphic features, (iv) it may be difficult to determine whether redoximorphic features are contemporary or relict features associated with wetter periods during the late Pleistocene, (v) mixing of soil materials due to cattle grazing, small mammal disturbance, and shrink–swell forces often obliterates redoximorphic features, and (vi) due to the strong seasonality in moisture, redoximorphic features may change as the pools wet and dry, making identification of redoximorphic features difficult during certain times of the year.
The goal of this study was to quantify how soil properties and hydric soil indicators vary in response to hydrologic conditions and landscape position on selected landscapes in northern California. Our primary objective was to investigate soil stratigraphic relationships across microtoposequences (summit, rim, and basin) in four contrasting vernal pool catenas. Information from this study may be used to further guide vernal pool conservation and construction or enhancement (expansion of existing vernal pools), a common mitigation practice in northern California.
|
MATERIALS AND METHODS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
5% of the total land surface of the Central Valley of California (Holland, 1998). Four vernal pool catenas were studied at two general locations in Butte County, California. Catenas were named based on the soil series that the summit position most closely resembled. The Toomes pool was examined at the City of Chico's Doe Mill Preserve (39°43'30'' N and 121°46'45'' W). This 6.1-ha preserve protects vernal pool habitat for the endangered Butte County meadowfoam (Limnanthes floccosa Howell ssp. californica Arroyo). The Toomes catena is underlain by consolidated lahar and is located on an intermediate fan terrace with rolling topography and undulating microrelief. Elevation ranges from 79 to 82 m. Soils formed from andesitic Pliocene lahars and pumaceous tuff of the Tuscan Formation and associated alluvium along the east side of the Sacramento Valley (Saucedo and Wagner, 1992). Three additional study sites (Anita, Redding, and Tuscan) were examined at Wurlitzer Ranch Preserve (39°51'30'' N and 121°57'30'' W). The ranch is located on an intermediate fan terrace with rolling topography and undulating microrelief. Elevation ranges from 56 to 64 m. The Anita and Tuscan soils formed from the Modesto and Riverbank aged formations, which are composed dominantly of andesitic, late- to middle-Pleistocene alluvium derived from Tuscan Formation lahars (Saucedo and Wagner, 1992). The Redding soils formed from mixed, mid-Pleistocene alluvium of sedimentary origin from the Coast Ranges. Sierra Nevada alluvium is mixed with Coast Range alluvium in this region because the Sacramento River has meandered across the east–west extent of the valley through time.
The climate in the study area is Mediterranean with an average of 660 mm of precipitation occurring from late fall to mid-spring. Annual precipitation was 1166 and 762 mm for the 1994–1995 and 1995–1996 water years, respectively. The average annual air temperature is 16.7°C, with average monthly temperatures of 7.8°C in January and 27.8°C in July. Vegetation at the sites are grassland communities dominated by non-native plants, interspersed with vernal pool communities dominated by endemic plants (Hobson, 1998).
Field Methods
A summit-to-basin transect was examined at each catena by excavating a 5- to 10-m-long trench to a depth of 1.5 m with a backhoe during the dry season. Soil profiles in the summit, rim (defined as the high-water level), and basin positions were described. Soils were classified according to Soil Survey Staff (2004) and bulk samples collected for each genetic horizon (Soil Survey Laboratory Staff, 1992). Redoximorphic features were described according to Vepraskas (1992).
Laboratory Methods
Bulk density was measured on intact clods using the paraffin-clod method (Singer, 1986). Bulk soil samples were air dried, gently ground, and sieved to remove coarse fragments (>2 mm). Soil pH was analyzed in 1:1 soil/water suspensions after a 30-min equilibration. Particle-size distribution was determined by the pipette method after organic matter removal with H2O2, Fe removal with dithionite–citrate, carbonate removal with pH 5 NaOAc on affected horizons, and Si removal with 0.1 mol L–1 NaOH on affected horizons (Soil Survey Laboratory Staff, 1992).
Soils were extracted with dithionate–citrate to determine the quantity of organically bound, amorphous, and various crystalline oxyhydroxides of Fe and Mn, using one 16-h extraction at 23°C (Holmgren, 1967). Ammonium oxalate was used to selectively extract amorphous and organically bound Fe and Mn by one 4-h extraction at pH 3 in the dark (Parfitt, 1989). Extractable Fe and Mn concentrations in all extracts were measured by inductively coupled plasma spectrometry. All values represent the means of triplicate analyses.
Exchangeable cations (Ca2+, Mg2+, K+, and Na+) were determined by displacement with 1 mol L–1 NH4OAc and cation-exchange capacity by saturation with 1 mol L–1 NH4OAc (pH = 7), followed by displacement with 10% acidified NaCl (Soil Survey Laboratory Staff, 1992). Displaced NH4+ was measured on a conductimetric N analyzer (Carlson, 1978). Carbonate concentrations were determined by the evolved gas method (Williams, 1948). Total C and N were determined by dry combustion of powdered samples (C/N analyzer, Carlo-Erba, Milan, Italy). Organic C was calculated as the difference between total C and the C released from carbonates.
|
RESULTS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
).
|
|
|
|
|
The physical properties of the soils were uniform throughout the catena. Particle-size distribution did not change appreciably with depth or across the catena. Clay content of the thin basin profile was similar to that of the Bt horizons in the summit and rim locations, suggesting that erosion of surface layers could be the cause of the undulating microtopography. Bulk density was high throughout the catena and ranged from 1.59 to 1.98 Mg m–3. Bulk density was slightly higher in the summit position than the rim and basin positions (Table 2).
Tuscan Catena
The Tuscan catena formed in andesitic alluvium and has a duripan that perches water. The general horizon stratigraphy was similar across the catena (Ap–AB–Bt1–Bt2–Btkqm–Btq or Btkq). In the basin position, however, AB and Bt2 horizons were not present (Fig. 2
). Despite the differences in the degree of development, only slight differences in soil morphology were observed in the Tuscan catena. These differences appear to result from soil erosion by water and deflation by wind at the basin landscape position. Alternatively, upland positions may reflect landscape aggradation via deposition of dust. Soil thickness above the duripan decreases from 66 cm at the summit, to 50 cm at the rim, and 22 cm at the basin. Strong coarse prismatic structure overlies the duripan at all landscape positions. Platy structure is present in A horizons of the basin and rim, probably due to the seasonally submerged conditions. The reddest hues (5YR and 2.5YR) were encountered in all Bt horizons overlying the duripan (Table 3, Fig. 2).
|
Redox depletions (chroma of 2 or less) were absent in all Tuscan profiles. Iron and Mn concentrations (masses, coatings, or nodules) were most abundant at the basin and rim locations. At all landscape positions, the abundance of Fe concentrations was lowest in the Bt horizons and highest in A horizons. Iron concentrations also were abundant in the duripan, but are assumed to be relict features due to the impermeable nature of the duripan and because Fe concentrations within and below the duripan had redder hues (2.5 YR) than overlying horizons (7.5 YR; Table 3). Manganese concentrations were common throughout most of the profile at each landscape position except in the Bt2 horizon of the summit, which lacked Mn concentrations (Fig. 2).
Redding Catena
Soil morphologic features were similar across the Redding catena. The catena formed in mixed sedimentary alluvium and has a duripan present across the landscape. The horizon sequence was similar across the catena (Ap–Bt–Btqm–2BC or 2Bt) with the exception that an AB horizon was present at the summit and rim position. Soil structure was often platy in A horizons and angular blocky in all horizons above the duripan. The duripan at the rim and basin positions displayed coarse angular blocky or platy structure, while at the summit the duripan was massive. Soil color was also similar across the catena, with 10YR hues in A horizons and 7.5 YR or redder hues in subsoil horizons. Bulk density was high (>1.78 Mg m–3) throughout the catena and increased with depth. As was the case with the Tuscan catena, bulk density values were highest in the horizons above the duripan (2.36–2.50 Mg m–3) and lowest in Ap horizons (1.78–1.86 Mg m–3; Table 4, Fig. 3
).
|
Redox depletions (chroma of 2 or less) were absent at all landscape positions even though this site was one of two sites that were submerged the longest. Manganese concentrations (mangans and nodules) were present in all profiles but only in the duripan, suggesting that Mn concentrations are relict features when present in these cemented horizons. Iron concentrations (masses and nodules) were present at all landscape positions in and below the duripan. At rim and basin positions, redoximorphic Fe concentrations within and below the duripan are believed to be relict features because they had redder hues (5YR) than those above the pan (7.5YR; Table 4). Contemporary Fe concentrations were present in all horizons overlying the duripan at rim and basin landscape positions. At the summit position, Fe concentrations were only present in the clay-rich Bt2 horizon and within and below the duripan (Table 4, Fig. 3).
Anita Catena
The Anita catena, formed in andesitic alluvium, has a duripan that extends across the catena. The horizon sequence across the catena was very similar (Ap–A–Btss–Btkqm). Slickensides were present in clay-rich horizons overlying the duripan. Redoximorphic features indicated that aquic conditions were present in the basin and rim positions (Fig. 4
). At the summit position, the soil matrix chroma was too high, excluding aquic conditions at the suborder level.
|
Physical properties were relatively uniform across the catena. The trend in particle-size distribution across the catena was similar to the other catenas where it appears that subsurface horizons have been exposed at the basin position by erosion and sediment export. Clay content in Ap horizons was higher at rim and basin profiles (29–31%) and similar to the A horizon (32% at 6–16-cm depth) in the summit. Clay content of all Btss horizons was similar across the catena, ranging from 42 to 56%. Bulk density was high (1.87–2.47 Mg m–3) throughout the catena and increased with depth to the duripan (Table 5).
The Anita catena was the only site with visible redox depletions (chroma of 2 or less). Throughout the catena, common to many depletions were present in all horizons above the duripan with the exception of the Btss2 at the summit location, which may have served as an impermeable layer. At each landscape position, common to many Fe concentrations (masses and nodules) were found throughout the profile. Manganese concentrations (mangans and nodules) were common within the rim and basin profiles but few were present at the summit (Fig. 4).
Soil Chemical Properties
Toomes Catena
The measured chemical properties were relatively uniform from the Toomes catena. Soil pH increased slightly with depth at each landscape position. The average pH was 6.2, moderately to slightly acid. Differences in base saturation were small among horizons and profiles along the catena. Base saturation ranged from 60 to 75% and followed a similar trend to that of pH, increasing with depth at the rim and summit positions. Base saturation decreased with depth at the basin position. Organic C for each profile along the catena decreased from the summit to the basin, ranging from 7.4 kg m–2 at the summit to 3.6 kg m–2 in the basin (Table 2).
The ratio of oxalate-extractable Mn to dithionite–citrate-extractable Mn (Mno/Mnd), a measure of the degree of manganese oxide crystallinity, was high throughout the catena, ranging from 0.78 to 0.90. These ratios suggest that the amount of poorly crystalline extractable Mn is high, although organically bound Mn may also contribute to these high ratios in surface horizons. In contrast, the oxalate-extractable Fe to dithionite–citrate-extractable Fe ratio (Feo/Fed) was low throughout the catena and decreased with depth: A horizons = 0.27 to 0.22 and Bt/Cr horizons = 0.14 to 0.16. The low Feo/Fed ratio indicates that most of the extractable Fe is in crystalline forms and the amount of poorly crystalline Fe is similar at each landscape position (Table 2).
Tuscan Catena
Differences in the chemical properties of the soils in the Tuscan catena were most evident between the horizons above and below the duripan at all landscape positions. The pH of horizons above the duripan ranged from 5.7 to 6.2, compared with values of 7.3 to 8.2 in horizons below the duripan. Base saturation followed a similar trend, with values ranging from 59 to 81% in horizons above the duripan to values of 81 to100% below the duripan. The presence of pedogenic carbonates within and below the duripan indicates that leaching was limited in the lower portion of the profile (Fig. 2). The trend in organic C within the profiles differed from the other sites and was highest at the summit (15.8 kg m–2), decreased to 8.7 kg m–2 at the rim, and increased to 14.5 kg m–2 at the basin (Table 3).
The Mno/Mnd ratio (0.57–1.0) was high throughout the catena but slightly lower in the summit. Patterns in extractable Fe (Fed and Feo) were similar across the catena, showing sharp decreases in Fed and Feo in horizons below the duripan. At the summit position, the Feo/Fed ratio was low (0.11–0.24) and generally decreased with depth. The Feo/Fed ratios were higher in practically all horizons at the rim and basin positions, ranging from 0.31 to 0.44, and did not decrease sharply below the duripan. Patterns in extractable Fe indicate that the basin and rim profiles contain slightly more poorly crystalline Fe minerals than the summit position.
Redding Catena
As was the case with the Tuscan catena, differences among chemical properties in the Redding catena were most evident between horizons above and below the duripan. The pH was the lowest among all the catenas studied. The pH increased with depth, and the magnitude of the increase was greater when associated with the duripan. Base saturation also increased with depth, displaying a sharp increase to 100% in horizons below the duripan. These sharp changes in pH and base saturation below the duripan are probably a signature of relict conditions where these horizons are excluded from the contemporary hydrologic regime and active leaching by lateral flow above the clay-rich Bt and duripan. Organic C in horizons above the duripan ranged from 14.7 kg m–2 at the summit to 13.6 and 13.3 kg m–2 at the rim and basin, respectively (Table 4).
As was the case with the Toomes and Tuscan sites, the Redding catena displayed high Mno/Mnd ratios (0.75–1.0) in horizons above the duripan; however, the ratio did not change in horizons below the duripan (1.0). At all landscape positions, Fed decreased with depth, displaying sharp decreases in horizons below the duripan. The Feo/Fed ratio was highest in horizons above the duripan except for the rim position, where Fed was very low in horizons underlying the duripan. The Feo/Fed ratio was particularly high in A horizons of the rim and basin, 0.60 and 0.34, respectively, where prolonged saturation and high organic matter content coincide to facilitate greater soil reduction. These values correspond with the distribution of RMFs (Fig. 3).
Anita Catena
The chemical properties of the Anita catena followed trends similar to those of the other catenas. The pH increased with depth, displaying sharp increases in horizons below the duripan. Base saturation followed similar trends, with values as high as 100% in horizons below the duripan and values ranging from 45 to 88% in horizons above the duripan. Organic C was low and similar to the Toomes catena, decreasing from 5.8 kg m–2 at the summit to 3.6 kg m–2 in the basin (summit > rim > basin; Table 5).
The Mno/Mnd ratio (0.50–1.0) was high throughout the catena and tended to decrease in horizons below the duripan. Extractable Mn did not change appreciably with profile depth across the catena; however, Mnd was higher at the basin profile (1.4–1.5 g kg–1) than the summit and rim profiles (1.2–1.3 g kg–1). As was the case with the other sites with duripans, sharp decreases in Feo/Fed occurred in horizons below the duripan. Extractable Fe (Fed and Feo) in horizons above the duripan decreased with depth across the catena with the exception of Fed at the summit profile, which increased with depth. The Feo/Fed ratio was highest in A horizons, especially at the rim and basin positions.
|
DISCUSSION |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Bulk density was high, often exceedingly high (>2 Mg m–3), at all sites and did not show trends related to the duration of standing water. Platy soil structure was common in all soil profiles. High bulk density and platy structure are probably a result of repeated wetting and drying (hydroconsolidation). Wetting and drying is emphasized instead of the impacts of standing water because these features were also present at summit locations that are not submerged by water. Studies have shown that hydroconsolidation can occur in soils with a wide range of clay contents (5–40%) and the degree of compaction depends on clay mineralogy (Assallay et al., 1998; Bull, 1964). Soils dominated by kaolinite mineralogy are more susceptible to hydroconsolidation at higher clay contents (e.g., >30%) compared with other active clays such as smectites and vermiculites, which resist hydroconsolidation at lower clay contents (Assallay et al., 1998). Clay mineralogy in the Toomes, Anita, and Tuscan catenas indicates that the soils are conducive to hydroconsolidation. X-ray diffraction showed that the clay fraction in all profiles at the Toomes catena and the basin position at the Tuscan catena are dominantly kaolinite. Clay fractions at rim and summit positions at the Tuscan catena and rim and basin positions at Anita are dominated by kaolinite and smectite. The summit position at the Anita catena is dominated by smectite, with lesser amounts of kaolinite. The Redding soils were not analyzed (Hobson, 1998).
Historical cattle grazing, especially when soils were wet, also has contributed to platy structure and high bulk density. For example, in a woodland setting in central California, the surface bulk density of soils in cattle concentration areas was 0.37 to 0.47 Mg m–3 higher than areas that had not been grazed for 6 and 26 yr (Tate et al., 2004).
High bulk densities at all sites reflect a reduction in porosity, which restricts O2 diffusion, increases CO2 accumulation, and promotes anaerobic conditions (Linn and Doran, 1984; Conlin and van den Driessche, 2000; Shestak and Busse, 2005). The reduced pore space may help facilitate the development of RMFs in vernal pool soils. Micromorphological techniques have demonstrated that compaction causes reduced macroporosity, dispersed organic material in peds, and the presence of faint RMFs (Sweeney et al., 1992). Thus, in addition to high organic C content, compaction may also play a role in the abundance of redox concentrations in the A horizons of most catenas (Fig. 1–4
). Moreover, the platy soil structure that is common in the surface horizons of most catenas may limit O2 diffusion into the soil.
The duration of standing water in each catena loosely corresponded with the amount and type of RMFs. The Tuscan and Toomes pools had standing water for the shortest amount of time and contained redder hues, reflecting more oxidized conditions. The absence of Fe depletions and concentrations at the Toomes site corresponds with the duration of standing water, which was the shortest of all sites. At the Tuscan site, the duration of standing water lasted 26 d longer in 1994–1995 and 11 d longer in 1995–1996 relative to the Toomes catena. Since the Tuscan site contained RMFs at the soil surface, this extended period of saturation appears to reflect a threshold at which RMFs can form in this region. The differences in RMFs could also be attributed to soil organic C levels, which are higher at the Tuscan catena, particularly in AB and Bt horizons. With the exception of the entire Toomes catena and the Redding summit position, the abundance of Fe concentrations was common to many in A horizons, which corresponded with the highest organic C contents and the highest Feo/Fed ratios.
The absence of RMFs at the Toomes catena is puzzling. Although the duration of standing water was the shortest at this site, standing water was present at temperatures conducive to microbial activity. In the 1994–1995 water year, the average air temperature was 10.4°C in the months when standing water was present, ranging from 5.8°C in December to 14°C in April. In the 1995–1996 water year, the average air temperature was 11.4°C during the months when standing water was present and ranged from a low of 8.2°C in January to 15.4°C in April (Table 1; California Irrigation Management Information System, www.cimis.water.ca.gov/cimis/welcome.jsp, verified 28 Dec. 2007).
The duration of standing water was also less telling at the other catena sites. The duration of ponded water was greatest at the Redding and Anita catenas, but the expression of RMFs was very different. Low-chroma (=2) redox depletions were present only at the Anita site. The distribution and abundance of RMFs were similar at the Tuscan and Redding catenas (especially rim and basin positions) even though the duration of standing water was different. It is possible that the duration of saturation did not correspond with the duration of standing water. For example, clay content was highest at the Anita and Tuscan catenas and may have extended the duration of saturation or slowed O2 diffusion, allowing the Anita catena to remain anoxic for the longest duration and the Tuscan catena to remain anoxic for a period similar to the Redding catena.
The dominant factors that dictate the presence of redoximorphic features in vernal pool landscapes are duration of standing water, landscape position, horizon stratigraphy, hydrologic flow path, soil temperature, and abundance of labile C (Vepraskas, 2001). The presence of redox features at summit and rim locations appears to be controlled by the proximity of episaturation to surface horizons where organic C content is high. This is controlled by the degree of relief and the height of ponded water, but also by the horizon stratigraphy.
The thickness of the soil above the duripan or lithic contact is similar for the summit positions of each catena, ranging from 48 to 66 cm (Fig. 1–4); however, abrupt clay increases in horizons overlying these water-restrictive layers appear to control the distribution and abundance of near-surface RMFs. We refer to the thickness of soil overlying a hydraulically restrictive layer (bedrock, duripan, or sharp clay increase) as the maximum perched water thickness (MPWT). The MPWT is greatest at summit positions in the Toomes and Redding catenas (56 and 40 cm, respectively), where RMFs were not described at or near the surface. The MPWT was not as thick in the Anita and Tuscan summit locations (16 and 30 cm, respectively), and as a result, both profiles had RMFs at or near the soil surface (Fig. 1 and 4). In addition, the difference between the summit elevation and maximum water table height was 15 cm or less for the Tuscan and Anita sites (Table 1). The high bulk densities (absence of macropores) and loamy or finer surface textures present at the Anita and Tuscan catenas suggest that capillary rise could maintain saturation at summit positions while standing water was present (Tables 3 and 5). At the Toomes catena, the capillary fringe should also have extended well into the summit profile because the distance between the summit and maximum height of the water table was only 17 cm.
The absence of RMFs at the summit profile in the Redding catena suggests that capillary rise does not extend into the topsoil. The difference between summit elevation and maximum water table height was greater (25 cm) than for other summit profiles. Furthermore, it is possible that the thickness of the capillary fringe is much less at this site because the sand content is higher (Tables 1 and 4).
The MPWT is less for all rim positions than summit positions, presumably due to truncation of the soil profile by erosion. The MPWT in the rim positions of Redding, Anita, and Tuscan pools is less (26, 16, and 12 cm, respectively), and for these soils it appears as if the water table reaches the surface horizons long enough to develop RMFs that satisfy hydric soil criteria (Table 1, Fig. 2–4).
Bioturbation by gophers may also explain the absence or reduced abundance of RMFs in upslope positions. Gopher activity is extensive in these landscapes and occurs preferentially on the drier summit and rim positions (Arkley and Brown, 1954; Cox, 1990); this may destroy evidence of hydric soil features. The shorter duration of standing water at the Toomes catena may be more hospitable to burrowing organisms, which may explain the absence of RMFs near the soil surface.
The redox potential (Eh) and pH were monitored for the 1994 to 1996 water years at the Anita and Tuscan catenas (Hobson, 1998). The Eh and pH values were sufficient to reduce NO3, Mn, and sometimes Fe. In general, Eh values remained low for the longest duration in the micro-low positions. Of the three depths monitored (5 and 15 cm and above the duripan), the 5-cm depth tended to have the lowest Eh, probably due to higher organic C contents (Hobson, 1998; Hobson and Dahlgren, 1998). The abundance of RMFs generally corresponded with reported Eh values in that RMFs at the Tuscan and Anita catenas were greatest in the A horizons of basin and rim positions (Fig. 1 and 4).
Soil Chemical Properties
The pH and base saturation above the duripan were much lower than below the duripan at all three sites with duripans. The process of ferrolysis may be responsible for the sharp changes in pH and base saturation with depth. Ferrolysis involves the cyclic reduction and oxidation of Fe and Mn. The seasonal nature of vernal pools creates repetitive cycles of anaerobic and aerobic conditions. During periods of reduction, soluble Fe+2 and soluble Mn+2 are formed, displacing base cations from the exchange complex. These base cations are leached from the soil via lateral flow (Rains et al., 2006) or accumulate at the depth of leaching. Subsequent oxidation of exchangeable Fe and Mn produces protons via hydrolytic reactions and the precipitation of Fe oxyhydroxides, which acidify the soil and contribute to low base saturation by displacing base cations. The processes of ferrolysis are outlined by the following equations (Miller, 1983):
 | [1] |
 | [2] |
 | [3] |
). A pH fluctuation of nearly 3 units was measured for the 5-cm depth between the oxidized and reduced states in the Anita pool during the course of a seasonal wetting and drying cycle (Fig. 5
).
|
Extractable Fe concentration was a more informative indicator of hydromorphic processes. The Toomes catena showed little difference in extractable Fe with depth and landscape position. The short duration of standing water and minimal evidence of pedogenesis may explain the absence of Fe concentrations and similarities in extractable Fe with depth and across the catena. Notably, there was a high amount of Fed at this site, similar to that of the Tuscan catena. The abundance of crystalline Fe at these sites suggests that the shorter duration of standing water may favor crystalline Fe over poorly crystalline Fe. No clear trend in profile-weighted means (using the soil thickness above the duripan) of poorly crystalline Mn and Fe oxyhydroxides was observable at Toomes. Crystalline Fe profile-weighted means were highest at the summit and lowest at the basin (Fig. 6 ).
|
Implications for Hydric Soils
Designation of hydric soil status is based on the presence and abundance of redox depletions or concentrations within the upper portion of the soil profile (Hurt and Carlisle, 2001; Vepraskas, 2001; Hurt and Vasilas, 2006). Hydric soils were identified in the basin positions of each catena. All rim positions contained hydric soils except in the Toomes catena. All soils at summit positions were not hydric except the Tuscan catena. None of the vernal pool soils had features that matched the criteria for F9 (vernal pools hydric soil indicator). Indicators F8 (redox depressions) and TF2 (test indicator for red parent materials) were most commonly applied (Hurt and Vasilas, 2006).
In the Toomes catena, the duration of saturation and the presence of hydrophytic vegetation suggests that hydric soils should be present in lower landscape positions. The basin profile is a hydric soil meeting the F8 criteria. The profiles at rim and summit positions did not meet F8 because RMFs were only present below 15 cm. Rim and summit profiles were very close to meeting the hydric soil criteria of redox dark surface (F6) and test indicator red parent material (TF2) (Hurt and Vasilas, 2006). The chroma for each profile was one chip higher than allowed for F6. In addition, summit and rim profiles had prominent Mn coatings only on gravels, not the fine-earth fraction.
The Tuscan catena contained hydric soils at summit, rim, and basin positions when considering the test indicator for red parent materials (TF2). The rim and basin profiles also met F8, with distinct or prominent redox concentrations as soft masses in a 5-cm layer within 15 cm of the surface. The basin position also met F6. The RMFs in the summit position only satisfied TF2 (Hurt and Vasilas, 2006).
In the Redding catena, red hues and an abundance of prominent Fe masses at the rim and basin positions met the hydric soil criteria of TF2 and F8. The summit was not hydric because RMFs were not present within the upper 30 cm of the soil (Hurt and Vasilas, 2006).
Hydric soils were present at the basin and rim locations in the Anita catena. Both profiles contained features that satisfied the criteria for F8, but not F9 because low-chroma depletions were not abundant enough (Table 5). The summit position was not hydric. It did not fit into F8 because it is an upland. In addition, the chroma was one chip too high for F6 and hues were not red enough for TF2 (Hurt and Vasilas, 2006).
Interestingly, the summit profile in the Anita catena had the strongest evidence of prolonged saturation near the surface but failed to meet hydric soil criteria. The Anita summit profile had common, prominent Fe depletions and concentrations throughout the upper 30 cm of soil but was not red enough for TF2, the chroma was too high for F6, and it didn't satisfy the landscape requirements for F8. The MPWT was not as thick in the Anita and Tuscan summit locations (16 and 30 cm, respectively) and, as a result, both profiles had RMFs at or near the soil surface (Fig. 1 and 4) that satisfy or come close to satisfying hydric soil indicators.
In some instances, the classification of soils did not adequately reflect the hydrologic conditions. For example, in the Toomes catena, the summit profile met criteria for the oxyaquic subgroup because standing water maintained saturated conditions within 100 cm (either through capillary rise or perched water on the bedrock) for >20 consecutive days or 30 cumulative days (Table 1). Since the basin and rim profiles of this catena are shallow to bedrock, the lithic subgroup keys out ahead of oxyaquic and takes precedence in Soil Taxonomy. Thus, the concept of wetness in these lower landscapes is completely hidden by the taxonomic names (Table 2).
A similar scenario occurs in the Tuscan catena, where all three soils display redox concentrations in the surface layers yet aquic conditions are recognized only in the basin soil at the subgroup level (Table 3). This is the case because the basin is an Inceptisol, and only redox concentrations are needed for an aquic subgroup for that order. The upland profiles are Alfisols. Redox depletions are required for an aquic subgroup for Alfisols. Furthermore, from a genetic standpoint, the basin soil is close to being classified as an Alfisol, in which case the hydromorphic features would have been concealed by the taxonomic name.
|
CONCLUSIONS |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Hydric soils were identified in the basin and rim positions of each catena except the Toomes site, where the basin contained the only hydric soil. The Tuscan catena was the only site with a hydric soil at the summit position. The summit profile at the Anita catena was very close to satisfying hydric soil criteria. These two summit positions were the wettest and second driest sites in terms of duration of standing water. While the duration of standing water did not appear to have a strong impact on the presence of near-surface RMFs at these sites, the duration of saturation, which probably lasts longer, appears to influence their distribution and abundance. The duration of near-surface saturation extends beyond the duration of standing water and is governed by water supply and intrinsic properties of the soil that limit deep percolation, such as clay content, degree of compaction, and horizon stratigraphy (e.g., thickness of the MPWT). The persistence of anaerobic conditions near the surface is probably due to low porosity and shallow depth to water-restrictive horizons. Hydric soils were identified in settings where anaerobic conditions coincided with horizons with high organic matter to form RMFs near the surface.
Chemical properties of the soils were surprisingly similar across the microtopography in each catena. The greatest differences in chemical properties were observed within individual profiles that had duripans (Anita, Tuscan, and Redding), where sharp changes in pH, base saturation, Mno/Mnd, and Feo/Fed were present between horizons above and below the duripan (Tables 3–5). Profiles at the Toomes site, which were weakly developed soils, displayed similar pH, base saturation, and extractable Fe and Mn with depth and across the catena, which corresponded with the uniformity in morphology.
Similarities in biogeochemical, physical, and morphological properties across each catena coincide with the findings of a recent study that identified a hydraulic connectivity across large extents of vernal pool landscapes (Rains et al., 2006). Thus hydric soil indicators may need to be revised for the micro-upland positions (summits and some rims) of vernal pool landscapes to ensure that these landscapes are not unnaturally dissected into well-drained (summits and some rims) and poorly drained (basins and some rims) areas for land-use purposes. This is especially important in northern California, where identification of RMFs is often difficult. One possible solution is to expand the depth criteria for hydric soil indicators in vernal pools (F9) and redox depressions (F8) to greater depths for all micro-high positions that are associated with hydric soils in adjacent micro-lows. None of the soils met F9, thus the hydric soil criteria for vernal pools may need to be revised. Changing the criteria for F9 to include redox concentrations as an alternative to depletions may be appropriate. This change would eliminate the confusion of using other hydric soil indicators such as F6, F8, and TF2, which were not specifically established for vernal pools.
|
NOTES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Received for publication April 2, 2007.
|
REFERENCES |
|---|
|
TOPNOTESABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
