Soil Science Society of America Journal 64:1535-1543 (2000)
© 2000 Soil Science Society of America
DIVISION S-10-WETLAND SOILS
Stratigraphy and Hydrology of the Jackson-Frazier Wetland, Oregon
David V. D'Amorea,
Scott R. Stewartb,
J.Herbert Huddlestonb and
J.Reed Glasmannc
a Pacific Northwest Research Station, USDA Forest Service, 2770 Sherwood Lane, Suite 2A, Juneau, AK 99801 USA
b Dep. of Crop and Soil Science, Oregon State Univ., Corvallis, OR, 97331 USA
c Dep. of Geosciences, Oregon State Univ., Corvallis, OR 97331 USA
ddamore{at}fs.fed.us
 |
ABSTRACT
|
|---|
The relationship between wetland soils and hydrology can be better understood by linking soil geomorphological features to the measurement of groundwater depths in the soil. Soil stratigraphic analysis and long-term measurements of soil water levels in piezometers were used at the Jackson-Frazier wetland in western Oregon to investigate the interaction between local geomorphological history and the hydrology of the wetland. Morphological descriptions confirm the presence of a recent smectitic alluvial deposit (80180 cm) overlying Malpass clay (
35 cm thick), which overlies Irish Bend Silt. X-ray diffraction and isotope analysis support the conclusion of the presence of the Holocene alluvium and Irish Bend Silt, but are inconclusive regarding the Malpass clay. Piezometer data from 1992 to 1996 show that the smectitic alluvium controls saturation and drying of the wetland surface, and that a separate water table is present below the Malpass clay in the Irish Bend silt. The recent alluvium and Malpass clay act as an aquitard that restricts the vertical infiltration of surface water and helps restrict the groundwater table in the Irish Bend silt deposit. These stratigraphic relationships and associated hydrologic responses provide a means to identify wetlands and predict hydrologic conditions on similar wetland landscapes.
Abbreviations: PVC, polyvinyl chloride XRD, x-ray diffraction
 |
INTRODUCTION
|
|---|
WETLAND SOILS and wetland hydrology are intricately linked in the development and functioning of wetland ecosystems. Relationships between wetland soils and wetland hydrology are best understood when there are adequate data to fully characterize soil morphology, surface water hydrology, groundwater hydrology, and transmission or lack of transmission of water through the soil (Mausbach and Richardson, 1994).
Understanding hydrologic processes in relation to geomorphology and soils can provide indications of where wetlands may occur on the landscape, but determining these processes requires information on the geologic and hydrologic setting and their influence on the flow of water (Winter, 1988). Often, hard hydrologic data are unavailable for many wetland types, and inferences about wetland functions must be based on landscape features, soils, geomorphology, and vegetation patterns. Linking geomorphology, soil stratigraphy, and soil hydrology provides a basis for interpreting saturation patterns and wetland extent without intensive monitoring in similar areas.
Water movement, or impedance of movement, in the soil also plays a critical role in soil formation and the related soil physical and chemical transformations (Richardson et al., 1992). Studies that establish connections among soils, geomorphology, and hydrology can provide information about soil development as well as water movement. The recent hydromorphic soil study by Thompson et al. (1998) linked topography and soil geomorphology to water table dynamics to predict where saturated soil conditions occurred on the landscape. Steinwand and Fenton (1995) developed hydrologic relationships using soil stratigraphy and piezometer data for a glaciated landscape in Iowa. Richardson et al. (1992) used flownet analysis to illustrate water movement through soils on different landscape positions. All of these studies have illustrated the importance of the interactions of geomorphology, stratigraphy, and soils in the movement of water through the landscape. Establishing regional studies of water movement in specific landscapes provide information on water dynamics that can be applied to similar local landscapes, as well as similar soil geomorphic relationships in other regions.
In the Willamette Valley of western Oregon, wetlands are a common feature on broad, flat Late Pleistocene terraces and along the geomorphic boundary between the high terrace of the valley floor and the foothills of the Coast Range and Western Cascades. Complex soil stratigraphy related to interfingering of Late Pleistocene glacial flood deposits on older valley geomorphic surfaces (the Willamette Formation; Balster and Parsons, 1968, 1969) often results in abrupt textural differences within the solum that strongly influence pedohydrology (Boersma et al., 1972; Austin and Huddleston, 1999). In addition, deposition of Late Pleistocene lacustrine sediments within the Willamette Valley disrupted stream gradients along minor channels entering the valley from adjacent foothills, augmenting clay-rich, alluvial fan sedimentation. The Jackson-Frazier wetland represents one of these valley marginal, alluvial fan wetlands. The site represents a Holocene valley margin geomorphic surface underlain by clayey alluvium. The site has previously been used as pasture land, but has reverted to natural vegetation cover since 1980. The area is now a Benton County park and is a valuable wetland resource, both for its environmental qualities and for educational programs for the surrounding communities.
The objectives of this study were to characterize the soil geomorphic and stratigraphic relationships of the Jackson-Frazier wetland and to investigate the influence of these deposits on wetland hydrology.
 |
Materials and methods
|
|---|
Site Description
The Willamette Valley occupies a structuralerosional depression in western Oregon that lies between the Cascade Range to the east and the Coast Range to the west. The Jackson-Frazier wetland is an
64-ha area located at the western margin of the valley north of the city of Corvallis at an elevation of
66 m (Fig. 1)
. The overall configuration of the wetland is an alluvial fan with the Jackson and Frazier Creek source streams in the northwest corner dispersing water and sediment down the elevation gradient to outlet channels along the northeast (Frazier Creek ditch) and southeast edge (Stewart Slough) of the wetland. The subdued topography of the wetland and the presence of numerous beaver (Castor canadensis) dams below the confluence of Jackson and Frazier Creeks in the northwest corner of the wetland have created an area of complex overland flow and seasonal soil saturation. Vegetation includes forested, shrub-scrub, and prairie palustrine plant communities (Marshall, 1985). The area receives
1030 mm precipitation annually, most of which falls from October to March. The mean annual temperature of the area is 11.3°C.

View larger version (26K):
[in this window]
[in a new window]
|
Fig. 1 General location of Jackson-Frazier wetland and locations of monitoring sites, mineralogy sample sites, and stratigraphic transect
|
|
Soils and Stratigraphy
The predominant soil mapped (Knezevich, 1975) within the wetland is the Bashaw clay (very-fine, smectitic, mesic Xeric Endoaquert). The landscape associated with this soil consists of a backfilled, low-relief, broad swale that was eroded into the Calapooia surface (Balster and Parsons, 1968) by Jackson-Frazier Creeks. Neighboring soils on the Calapooia surface include the Dayton (fine, smectitic, mesic Typic Albaqualf) and Amity (fine-silty, mixed, superactive, mesic Argiaquic Xeric Argialboll) soils developed in Late Pleistocene lacustrine silts and clays of the Willamette Formation. The Dayton series has at least three major Late Pleistocene soil stratigraphic units, including the Greenback Member (surficial silt loam), the Malpass Member (clay), and the Irish Bend Member (silt loam to silty clay loam; Balster and Parsons, 1969). The erosive episode that developed the wetland swale scoured away the Greenback unit and may have altered some or all of the Malpass unit. The Bashaw soils are located on the Ingram surface (Balster and Parsons, 1968), which has alluvial deposits in place of the Greenback member.
The pedons selected for detailed soil description and sampling in this study lie within the Holocene alluvial fan that was deposited on the eroded Calapooia surface. Three soil pits were excavated to a depth of 2 m (Fig. 1; Sites 1, 2, and 3), subdivided into major horizons, and described (Soil Survey Division Staff, 1993). Although the soils at all three sites are classified as Bashaw, the vegetation at each site is different. Site 1 occurs at an interface between wet meadow and shrub-scrub cover types and is dominated by water parsley (Oenanthe sarmentosa K. Presl ex DC.). Site 2 lies in a shrub-scrub area and is dominated by slough sedge (Carex obnupta Bailey). Site 3 occurs in a forested area with a dominant overstory of Oregon ash (Fraxinus latifolia Benth.) and a dominant understory of reed canary grass (Phalaris arundinacea L.).
Five roughly north-south sampling transects were established across the wetland to observe soil stratigraphic relationships. Auger holes 2 m deep were hand-augered 50 m apart, but thick vegetation often made it necessary to lay out the transects in a zig-zag pattern. The location of each bore hole was determined from compass bearings and measured distance. The elevation of each site was determined with a transit and stadia rod from a reference benchmark. Soil color, texture, redoximorphic features, and horizon boundaries were recorded for subsequent stratigraphic characterization of the wetland (Soil Survey Division Staff, 1993).
Clay Mineralogy
Samples for clay mineral characterization were obtained from the 2Bt horizon at Site 1 and at depths of 50, 100, and 180 cm at four points along a northwest-southeast transect across the wetland (Fig. 1, Transect AA', Sites 48, 85, 109, and 129). Each sample was placed in a sealed plastic bag to retain field moisture content until analyzed. Clays were dispersed by mixing 5 to 10 g of moist soil with 150 mL of distilled water and 5 mL of 0.5% (w/v) Na-hexametaphosphate in 250-mL polyethylene bottles and shaking for 8 h. After dispersion, the clay fraction (<2 mm) was separated by centrifugation and was saturated with Mg by washing three times with 0.5 M MgCl2 followed by three rinses with distilled water. Oriented slides of the Mg-saturated clay were prepared for x-ray powder diffraction analysis (XRD) by a paste method (Theissen and Harward, 1962). The remaining clay was then K-saturated with 1 M KCl using a similar procedure and oriented slides were prepared for XRD. The slides were treated according to the method outlined by Glasmann and Simonson (1985), excluding the glycerol solvation and high temperature treatments. Slides were analyzed using a computer-automated Phillips XRG 3100 (Phillips, Eindhoven, the Netherlands) equipped with compensating slits and a focusing monochromator. Copper k
radiation was used (40 KV, 35 mA) and slides were step scanned from 2 to 34° 2
, using a step size of 0.02° 2
and a count time of 1 s. Semiquantitative interpretation of the clay mineralogy of each sample was facilitated by comparison of sample XRD spectra to computer-generated clay mixtures calculated using NEWMOD (R.C. Reynolds, Jr., 1985, Hanover, NH).
Isotope Analysis
The fine clay fraction (<0.2 µm) of several soil samples was separated by centrifugation after dispersing the clays as described above. The samples included soil from the Bss2, 2Bt, and 3BC horizons at Site 1 and a reference sample of the Malpass Member of the Willamette Formation, which comprises the restrictive clayey subsurface horizon of the Dayton series. Dayton soils do not occur within the Jackson-Frazier wetland, but the 2Bt horizon has the color and stratigraphic position of the Malpass clay and we wanted to determine if Malpass clay did extend into the wetland and influence subsurface hydrology.
Organic matter was removed from the <0.2-µm clay fraction using three alternating treatments with 5% (w/v) Na-hypochlorite and 20% (v/v) H2O2. Metal oxides were removed using dithionate-citrate-bicarbonate extraction as outlined by Jackson et al. (1986). The clays were then Mg-saturated as described above, except following the second rinse with distilled water, each sample was rinsed an additional three times with ethanol to facilitate complete removal of excess salt. Oriented slides of the Mg-saturated clays were prepared for XRD analysis, and the remaining clay was dried at 50°C and sent to an outside lab for O18 isotope analysis by mass spectrometry (analyses performed by Dr. F.J. Longstaffe, University of Western Ontario, London, ON, Canada). Reproducibility is ± 0.3
for
O (F.J. Longstaffe, 1996, personal communication).
Hydrology
Piezometers were installed in triplicate at each of the three detailed soil sampling sites (Fig. 1, Sites 1, 2, and 3) at depths of 25, 50, and 100 cm. A single 200-cm piezometer also was installed at each site. Piezometers were constructed from 1.9-cm o.d. schedule 200 polyvinyl chloride (PVC) pipe. Each piezometer was placed in a hole bored to the required depth. Then the hole was backfilled with fine sand to cover the openings in the PVC tube, and the remainder of the hole was backfilled with bentonite powder to seal the tube from surface leakage. The piezometric surface at each site was recorded weekly by measuring the depth from the ground surface to an indicator float.
 |
Results and discussion
|
|---|
Soil Stratigraphy
Detailed morphological descriptions at each of the three soil pits (Table 1)
indicate the presence of three distinct stratigraphic units in these wetland soils. The surface unit consists of black clay (dominantly 0 or 1 chroma) in which a Bss horizon has developed beneath the A horizon. The Bss horizon displays both slickensides and redoximorphic features as either soft Fe-Mn accumulations or small, hard Fe-Mn concretions (often both occur together). This surficial unit averages
100 cm thick across the wetland, but is as thick as 180 cm in the northwest corner and thins to
80 cm toward the southeast, suggesting proximal thickening of the alluvial fan deposits towards the Coast Range sediment source (Fig. 2)
. In spite of these thickness variations, the texture of Unit 1 remains clayey throughout its areal occurrence. The clay percentages remain consistent across the site with 50 to 55% clay in the surface horizon, and 60 to 66% clay in the Bss horizons.

View larger version (41K):
[in this window]
[in a new window]
|
Fig. 2 Southeast (A) to northwest (A') stratigraphic transect in Jackson-Frazier wetland developed with data from Sites 48, 85, 129, and 109
|
|
The second stratigraphic unit consists of a gray to olive brown silty clay (dominantly 2 or 3 chroma) 2Bt horizon that contains a diverse assemblage of redoximorphic features (Table 1). The boundary between this unit and the overlying Holocene alluvium is a gradual, wavy zone where the underlying unit has been partially reworked and enriched with material from the upper deposit. The second unit is distinguished by color changes from black (10YR 2/1) to dark grayish brown (2.5Y 4/2) at Site 1 and to light olive brown (2.5Y 5/3) at Site 2. The texture changes from clay in the overlying Bss to silty clay in the 2Bt horizons. The thickness of the 2Bt silty clay unit averages
35 cm across the wetland (Fig. 2). Unit 2 has the morphology and stratigraphic position of the Malpass clay of the Willamette Formation.
The third stratigraphic unit consists of olive brown (dominantly 2.5Y 4/3, 4/4, and 5/4) micaceous silty clay loam 3BC horizons that grade with depth to silt loam. This unit also contains a diverse assemblage of redoximorphic features similar to those found in overlying horizons. The color, texture, and micaceous mineralogy of this unit strongly suggest that it is related to the Irish Bend Member of the Willamette Formation. The Greenback member of the Willamette Formation (Balster and Parsons, 1969) is missing at this location, implying partial erosion of this younger lacustrine strata in the wetland, with deposition of Holocene clayey alluvium over Malpass clay (Fig. 2).
Clay Mineralogy
The clay mineral assemblage of Unit 1 (Holocene alluvium) is dominated by smectite, with minor amounts of dehydrated halloysite, vermiculite, and possibly interstratified smectite and kaolinite (Fig. 3)
. The character of XRD patterns of the Mg-glycol treated <2-mm clay from Unit 1 is nearly identical from site to site along transect AA'. This widespread lateral homogeneity in soil mineralogical composition is consistent with a point source for the sediments (i.e., Jackson-Frazier Creeks). Noticeably absent from the clay assemblage of the alluvium are micaceous clays, chlorite or chloritic intergrades, and more significant amounts of kaolinite. These minerals generally occur in paleosols that mantle foothill landscapes of western Oregon to the north of Jackson-Frazier wetland (Glasmann and Kling, 1980). Their exclusion in the sediment discharge of Jackson-Frazier Creeks suggests a process of preferential erosion of smectitic clays, derived from volcanic parent material, along low-order Coast Range streams.

View larger version (42K):
[in this window]
[in a new window]
|
Fig. 3 X-ray diffraction patterns from <2-µm, Mg-glycol treated clay at 50-, 100-, and 180-cm depths at auger sample Sites 109, 129, 48, and 85. 1 = Mg dismectite-glycol, 2 = di-vermiculite, 3 = illite, 4 = dehydrated halloysite, 5 = Mg-chlorite, 6 = kaolinite
|
|
The smectite-dominated clay mineral assemblage that characterizes the 50-cm depth across the wetland is also present in deeper samples (100-cm samples at Sites 129 and 109, Fig. 3); however, several samples show an increase in the mica component, which coincides with the stratigraphic break between Units 2 and 3 (weak indication of mica in the 100-cm samples 85 and 48, Fig. 3). The 100-cm sample from Site 48 is below the stratigraphic break between Units 2 and 3, while the sample from 85 is above the stratigraphic break between Units 2 and 3. The mica component increases with depth from Unit 2 to Unit 3 (180-cm samples at Sites 85 and 48 are in Unit 3; Fig. 3). Since the recent alluvium is non-micaceous, the increase in the mica component of the clay assemblage lower in the profile does not reflect pedogenic destruction of mica in Unit 1. Instead, the presence of mica probably indicates clay minerals derived from the Willamette Formation.
The Irish Bend Member of the Willamette Formation has a clay mineral assemblage characterized by the presence of smectite, mica, chlorite, and kaolinite (Glasmann and Kling, 1980). This unit was deposited by a series of catastrophic glacial floods from glacial lake Missoula and includes micaceous glacial flour scoured from loess deposits of the Channeled Scablands of Washington that were ultimately sourced from the Canadian Rockies and other metamorphic terranes. In contrast, Jackson-Frazier Creek drains an area underlain by deeply weathered sea-floor basalts and tufaceous sediments of Cascade origin (Orr et al., 1992), rock types that are deficient in mica. The micaceous character of the olive brown silts of Unit 3 is strong evidence that this unit represents the Irish Bend Member. Comparison of the XRD patterns of micaceous Unit 3 samples with patterns obtained from the clay fraction of the Irish Bend Member from the type locality at Irish Bend and other exposures (Fig. 4)
suggests many similarities and indicates that the Irish Bend member has a fairly homogeneous clay mineral assemblage across the southern Willamette Valley. The increased smectite component of the Irish Bend Member at Jackson Frazier wetland may reflect illuviation of Holocene smectitic alluvium and/or additional postdepositional pedogenic formation of smectite in the wetland environment.

View larger version (37K):
[in this window]
[in a new window]
|
Fig. 4 Irish Bend x-ray diffraction patterns: (a) sample from Jackson-Frazier Site 48 at 180 cm (upper pattern) compared with a sample from the Irish Bend type section at Irish Bend, OR (lower pattern), (b) sample from the Irish Bend type section at Irish Bend, Oregon (upper pattern), compared with a sample from the Calapooyia river near Turner, OR (lower pattern). 1 = Mg-dismectite-glycol, 2 = di-vermiculite, 3 = illite, 5 = Mg-chlorite, 6 = kaolinite, 7 = albite
|
|
In contrast to the somewhat micaceous and chloritic mineralogy of the Irish Bend silts in Unit 3, the mineralogy of Unit 2 clays is very similar to that of the Holocene alluvium in Unit 1 as shown in the 100-cm sample at Site 85 (Fig. 3), and in a sample from known Malpass clay in the Dayton soil at Corvallis airport (Fig. 5)
. Although Unit 2 has the color, texture, and stratigraphic position of the Malpass clay, there is very little mineralogical reason to distinguish it from the smectitic alluvium deposited by Jackson-Frazier Creeks. Comparison of Unit 2 clay mineralogy with examples of known Malpass clay (Fig. 5) suggests that Malpass mineralogy is not as distinct from the Holocene alluvial material as that of the Irish Bend member in the Jackson-Frazier sediments (Fig. 3). The origin of the Malpass clay is unclear, but it may have been deposited in depressional areas of the valley floor as overbank deposits from local streams. The period following deposition of the Irish Bend member was characterized by disrupted drainages due to filling of former stream courses by lacustrine silts. Widespread, frequent flooding of lowlands was probably an annual event, resulting in overbank clay deposition from a multitude of sources around the valley (Balster and Parsons, 1968; Parsons and Herriman, 1970). Such a model would help explain the mineralogical diversity of the Malpass clay and the weakly expressed micaceous character of the deposit, reflecting slight reworking of the underlying Irish Bend silts (Fig. 3).

View larger version (23K):
[in this window]
[in a new window]
|
Fig. 5 Comparison of Malpass x-ray diffraction patterns. Sample from presumed Malpass deposit at Jackson-Frazier wetland (upper pattern) and Malpass sample from Corvallis airport (lower pattern). 1 = Mg-dismectite-glycol, 2 = di-vermiculite, 3 = illite, 5 = Mg-chlorite, 6 = kaolinite
|
|
Further support for the Malpass-like character of Unit 2 clays in the Jackson Frazier wetland is found in the stable isotopic composition of the fine clay fraction (Table 2)
. Fine clay from the bulk soil matrix has a very uniform
18O composition, averaging
21.8
. By contrast, the fine clay fraction liberated from Fe-Mn soil nodules of the 2Bt and 3BC horizons (presumed Malpass and Irish Bend units) is depleted in 18O relative to the matrix clay (19.3 vs. 21.8
, Table 2). Similar contrasting isotopic compositions have been noted for fine clays from the soil matrix from recent smectitic alluvial soils of the Malabon (fine, mixed, superactive, mesic, Pachic Ultic Argixeroll) series on the Mary's River flood plain near Corvallis, OR, and Irish Bend silts from Woodburn soils (fine-silty, mixed, superactive, mesic, Aquultic Argixeroll) of the adjacent Calapooia terrace (Table 2, data provided by Jeff Schatz, Geosciences Dep., Oregon State Univ.). These observations suggest that the O isotopic composition of smectitic clays of the Irish Bend Member of the Willamette Formation (Woodburn soils, Table 2) is an average of 1.9 to 3.3
lighter (depleted of heavy isotopes) than that of smectitic clays (Malabon soils, Table 2) derived from Coast Range soils. The lower isotopic composition of Irish Bend smectite suggests the extra-valley formation of smectite from isotopically depleted water related to Late Pleistocene glaciation and geographic influences on the isotopic composition of precipitation (rainfall generally becomes depleted of heavy isotopes landward from the ocean). The presence of an isotopically light smectite component within redoximorphic nodules of Units 2 and 3 suggests that the nodules may preserve local "islands" of clay with Irish Bend-like isotopic composition due to a somewhat protected environment from weathering, whereas the surrounding matrix has acquired the isotopic character of clay equilibrated with local Willamette Valley meteoric water. This indicates that the matrix of both Units 2 and 3 have been influenced either by illuviation of locally formed smectitic clay or by the in situ genesis of smectite in equilibrium with local meteoric water. Doser et al. (1998) found the same type of meteoric water enrichment in fine clay deposits in Louisiana. The deposition of the clayey, Malpass-like Unit 2 and overlying Holocene clay strata within Jackson Frazier wetland has a profound impact on the hydrology of this wetland system.
View this table:
[in this window]
[in a new window]
|
Table 2 Oxygen isotope data for matrices, nodules, and known Malpass member and for Irish Bend/Holocene sediments for the <2-µm fraction in Willamette Valley, Oregon
|
|
Hydrology
Water levels in the piezometers and local weekly precipitation amounts at Site 1 are presented in Fig. 6
. Data from monitoring Sites 2 and 3 are similar to those from Site 1, but are not shown. Precipitation was below average during the 1992 to 1994 monitoring period. During 1994 to 1996, the Willamette Valley had record rainfall, including a 100-yr flood event in 1996. Water levels measured in piezometers placed in the Holocene alluvium at 25- and 50-cm depths indicate that the soil is saturated from the soil surface to a depth of 50 cm for almost the entire period from October to June in all years. The water level in the 25-cm piezometer fluctuated from 3 to 10 cm higher than that in the 50-cm piezometer from the point of initial wetting in November 1992 until February 1993 when the water levels at 25 and 50 cm showed the same degree of saturation. The same difference of 3 to 10 cm between the 25- and 50-cm piezometers occurred from November 1993 to February 1993 as well. The different response was probably due to the swelling of the surface clay and the decreased infiltration of water to the 50-cm piezometer due to rainfall and runoff from Jackson and Frazier Creeks after a prolonged, unsaturated period during the summers of 1992 and 1993. During August and September 1993, ponded water disappeared and the upper 50 cm of soil was dry. Though there are no data for the 1994 season, the data from 1995 and 1996 show that the abundant rainfall during the summer months maintained surface saturation for much longer periods of time than the 19921993 period when the soil surface completely dried out from July to November. Surface saturation increased from a minimum of 245 d during 19921993, to 297 d in 19941995, and 285 d in 19951996.

View larger version (34K):
[in this window]
[in a new window]
|
Fig. 6 Precipitation (bottom) and water table data (top) as observed in piezometers at Jackson-Frazier Site 1
|
|
The piezometers at the 100-cm depth were located in the lower portion of the Holocene alluvium, and the piezometers at 200-cm depth were located in the Irish Bend silt. The piezometric surfaces at these two depths show water levels rising toward the surface during the early part of the rainy season in November and December in all years (Fig. 6). The water in the 200-cm piezometer was close to 200 cm below the surface in early November 1992 after the summer drought, and climbed rapidly through late January 1993 to a level 40 cm below the soil surface in response to higher quantities of rainfall. The piezometric surface eventually corresponded with the soil surface in April 1993, before beginning to move downward through the profile in spring and summer. The pattern of an increase in water levels during the fall, and a water level drop in the spring was repeated each year. No data were recorded in the spring of 1994, but the measurements taken in the fall indicate that the water levels dropped sometime between April and October.
The piezometric surface in the 100-cm piezometers rose in a manner similar to that observed in the 200-cm piezometers, but it was below the piezometric surface of the 200 cm piezometers during 19921993. The patterns of water levels were similar for the piezometers at both 100 and 200 cm, and the piezometric surfaces were roughly equal from 1994 to 1996. The piezometric surfaces indicated that the soil was completely saturated in January each year when the piezometric surfaces of the 100- and 200-cm piezometers rose above the 50-cm depth. The pattern of rising and falling water levels in response to precipitation repeated during each yearly rainfall and drought cycle at 100- and 200-cm depths throughout the monitoring period.
Rainfall in the 1992 to 1994 monitoring period led to a quick response in the water levels at 50- and 100-cm depths in response to the infiltration of surface water downward into large cracks in the soil formed by the drying smectitic clay during the summers of 1992 and 1993. Water flowing over the surface moved directly down to the 50-cm level via the cracks. Once the cracks swelled shut, water movement through the fine pores in the slowly permeable clay was very slow, and the soil surface became saturated from the top downward as pores filled with water ponded on the soil surface. Once the pores within the Bss aquitard swelled shut due to the shrinkswell nature of the clay, the pressure of water rising from the 3BC silts was the dominant hydrologic factor at 100 cm. Rainfall throughout the year during the 1994 to 1996 seasons maintained surface saturation and prevented the clay from drying and cracking. The increased rainfall did not have as significant an impact on the subsurface water levels in the 100- and 200-cm piezometers as shown by the consistent drop during the summer months in all years, though to a lesser extent at 100-cm depth during 1995-1996.
Water levels in the 200-cm piezometers show a higher pressure potential than in the 100-cm piezometers ranging from 6 to 52 cm from January to June 1993, indicating the presence of a confined aquifer. Water in the silts was under pressure due to the overlying restrictive layer during the low rainfall years of 1992-1993 (Fig. 6). During the high-rainfall period of 1994 to 1996, the soil became completely saturated with water supplied by the lower silt deposit and the surface. During the 1994 to 1996 period when the surface did not dry, soil saturation was maintained at the 100-cm level. The presence of a confined aquifer was noted in similar soils of the Willamette Valley by Boersma et al. (1972).
The slowly permeable smectitic clay impedes water transmission downward in the profile, leaving an unsaturated zone between 50 and 100 cm for several months during periods of lower rainfall, such as during 1992-1993, and probably prevents additional detention and storage of water in the subsurface. Buffkin-Drost (1985) conducted a study to assess water inflow, outflow, and storage in the Jackson-Frazier wetland during 1984-1985. This study showed that the peak flow lag time between the input at Jackson-Frazier creek and output at Stewart Slough was longer at the onset of the rainy season in November than later in the rainy season in February, indicating that water was moving across the wetland more rapidly in February. This decrease in lag time indicates that groundwater detention storage decreased after the soils were completely saturated and surface runoff was the dominant flow pathway for incoming water. In our study, the piezometers at the 100-cm depth indicated that the subsurface (50100 cm) was partially unsaturated until late winter. Once the subsurface becomes saturated, the storage capacity of the wetland may be very limited. Also, the water confined below the clay aquitard may move laterally through the Irish Bend silt along the topographic gradient of the wetland.
 |
Conclusions
|
|---|
Two distinct stratigraphic units are present beneath the Jackson-Frazier wetland: a Holocene alluvium and the Irish Bend silt. There is also evidence for a third deposit that is similar to the Malpass clay of the Willamette formation. The alluvial fan shape, dark organic rich surface, and clayey subsurface confirm that the upper deposit is recent Holocene alluvial material from the Coast Range foothills. Mineralogical composition of the Holocene alluvium corresponds with the source material of the foothills as well. Morphological descriptions confirm the presence of the Irish Bend silt in the lower profiles of the Bashaw soils at the wetland. Comparison of known Irish Bend mineralogical samples with Jackson-Frazier samples also provides evidence for the presence of the Irish Bend deposit. The mineralogical evidence to support the presence of the Malpass clay material is unclear, but the deposit has the appropriate morphology and stratigraphic position, relative to the Irish Bend silt, of the Malpass clay.
Hydrology data for the 4-yr monitoring cycle reveal two important stratigraphic influences on water movement in the wetland: (i) Holocene smectitic (shrink-swell) clays are so slowly permeable that water is perched above them, and (ii) there is a separate water table in the Irish Bend silt, which is confined by the very slowly permeable Malpass-like deposit and Holocene clay strata. In drier years, water in the Irish Bend is under greater pressure than in the clay aquitard and acts as an artesian source. In wetter years, the water in the Irish Bend, Malpass, and lower Holocene alluvium is under the same pressure and the soil becomes continuously saturated from both above and below. The slowly permeable Holocene alluvium and the low relief of the wetland allow ponded water to spread laterally and decrease in velocity as it moves away from the wetland. The retention of surface water may help to attenuate peak flow conditions in Jackson-Frazier Creek and creates conditions conducive to growth of hydrophytic plants. The Holocene alluvium is an indicator of appropriate conditions for formation of wetland areas in similar geomorphic positions in the Willamette Valley. The Irish Bend silt indicates the presence of an independent aquifer below the Holocene alluvium, which helps maintain saturation of the lower portions of the soil profile.
Floodplain surfaces, such as the one described here, are not extensive in the Willamette Valley because of widespread suburban and agricultural development. The remaining intact floodplain wetland areas on similar geomorphic positions provide important examples of natural wetland conditions, which can serve as examples for restoration projects and are possible sites for wetland preservation. The interaction of the stratigraphic deposits and hydrology described here also provides an example of the influence of stratified sediments of variable permeability on the hydrology of floodplain wetlands for similar areas in other regions.
 |
ACKNOWLEDGMENTS
|
|---|
We would like to acknowledge the Natural Resources Conservation Service, National Wet Soil Monitoring Program for funding assistance with this project. We also wish to thank Will Austin (Oregon State University) for logistical support and the journal reviewers for offering many helpful comments that improved the manuscript.
Received for publication November 2, 1998.
 |
REFERENCES
|
|---|
- Austin W.E., Huddleston J.H. Viability of permanently installed platinum redox electrodes. Soil Sci. Soc. Am. J. 1999;63:1757-1762.[Abstract/Free Full Text]
- Balster, C.A., and R.B. Parsons. 1968. Geomorphology and soils, Willamette Valley, Oregon. Or. Agric. Exp. Stn. Spec. Rep. 265.
- Balster C.A., Parsons R.B. Late Pleistocene stratigraphy, southern Willamette Valley, Oregon. Northwest Sci. 1969;43:116-129.
- Boersma L., Simonson G.H., Watts D.G. Soil morphology and water table relations: I. Annual water table fluctuations. Soil Sci. Soc. Am. Proc. 1972;36:644-653.
- Buffkin-Drost, M. 1985. Preliminary investigation into the hydrology of Jackson-Frazier wetland. Rep. at the Dep. of Geosciences. Oregon State Univ., Corvallis.
- Doser L.S., Ferrell R.E., Jr., Longstaffe F.J., Walthall P.M. Fluid flow through clayey soils: stable isotope and mineralogical evidence. Clay Miner. 1998;33:43-49.[Abstract]
- Glasmann J.R., Kling G.F. Origin of soil materials in foothills of the Willamette Valley, Oregon. Soil Sci. Soc. Am. J. 1980;44:123-130.[Abstract/Free Full Text]
- Glasmann J.R., Simonson G.H. Alteration of basalt in soils of western Oregon. Soil Sci. Soc. Am. J. 1985;49:262-273.[Abstract/Free Full Text]
- Jackson M.L., Lim C.H., Zelazny L.W. Oxides, hydroxides, and aluminosilicates. In: Klute A., ed. Methods of soil analysis, 2nd ed Madison, WI: SSSA, 1986:101-150 SSSA Book Ser. 5..
- Marshall, J. 1985. A value assessment of the Jackson-Frazier wetland. M.S. thesis. Oregon State Univ., Corvallis.
- Mausbach M.J., Richardson J.L. Biogeochemical processes in hydric soil formation. Current topics in wetland biogeochemistry. 1994;1:68-127 Wetland Biogeochemistry Institute, Louisiana State Univ., Baton Rouge..
- Orr E.L., Orr W.N., Baldwin E.M. Geology of Oregon, 4th ed Dubuque, IA: Kendall Hunt Publ, 1992.
- Parsons R.B., Herriman R.C. Haploxerolls and Argixerolls developed in recent alluvium, southern Willamette Valley, Oregon. Soil Sci. 1970;100:299-309.
- Richardson J.L., Wilding L.P., Daniels R.B. Recharge and discharge of groundwater in aquic conditions illustrated with flownet analysis. Geoderma 1992;53:65-78.
- Knezevich, C.A. 1975. Soil survey of Benton County Area, Oregon. Soil Conservation Service. U.S. Gov. Print. Office, Washington, DC.
- Soil Survey Division Staff. 1993. Soil survey manual. Agric. Handb.18. U.S. Gov. Print. Office, Washington, DC.
- Steinwand A.L., Fenton T.E. Landscape evolution and shallow groundwater hydrology of a till landscape in central Iowa. Soil Sci. Soc. Am. J. 1995;59:1370-1377.[Abstract/Free Full Text]
- Thompson J.A., Bell J.C., Zanner C.W. Hydrology and hydric soil extent within a Mollisol catena in southeast Minnesota. Soil Sci. Soc. Am. J. 1998;62:1126-1133.[Abstract/Free Full Text]
- Theissen A.A., Harward M.E. A paste method for preparation of slides for clay mineral identification by x-ray diffraction. Soil Sci. Soc. Am. Proc. 1962;26:90-91.
- Winter T.C. A conceptual framework for assessing cumulative impacts on the hydrology of nontidal wetlands. Environ. Manage. 1988;12:605-620.
This article has been cited by other articles:

|
 |

|
 |
 
D. V. D'Amore, S. R. Stewart, and J. H. Huddleston
Saturation, Reduction, and the Formation of Iron-Manganese Concretions in the Jackson-Frazier Wetland, Oregon
Soil Sci. Soc. Am. J.,
May 1, 2004;
68(3):
1012 - 1022.
[Abstract]
[Full Text]
[PDF]
|
 |
|