HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Porro, I. Search for Related Content PubMed PubMed Citation Articles by Porro, I. Agricola Articles by Porro, I. Related Collections Municipal Waste Water Quality

TECHNICAL REPORT
WASTE MANAGEMENT

Hydrologic Behavior of Two Engineered Barriers following Extreme Wetting

Applied Geosciences Dep., Idaho National Engineering and Environmental Lab., P.O. Box 1625, 2525 North Fremont Avenue, Idaho Falls, ID 83415-2107

Corresponding author (ixp{at}inel.gov)

Received for publication May 9, 2000.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Many engineered barriers are expected to function for hundreds of years or longer. Over the course of time, it is likely that some barriers will experience infiltration to the point of breakthrough. This study compares the recovery from breakthrough of two storage–evapotranspiration type engineered barriers. Replicates of test plots comprising thick soil and capillary–biobarrier covers were wetted to breakthrough in 1997. Test plots were kept cleared of vegetation to maximize hydrologic stress during recovery. Following cessation of drainage resulting from the wetting irrigations, water storage levels in all plots were at elevated levels compared with pre-irrigation levels. As a result, infiltration of melting snow during the subsequent spring overloaded the storage capacity and produced drainage in all plots. Relatively rapid melting of accumulated snowfall produced the most significant infiltration events each year during the study. Capillary barriers yielded less total drainage than thick soil barriers. By limiting drainage, capillary barriers increased water storage in the upper portions of the test plots, which led to increased evaporation from the capillary barrier plots compared with thick soil plots. Increased evaporation in the capillary barrier plots allowed more water to infiltrate in the second season following the wetting tests without triggering drainage. All thick soil plots again yielded drainage in the second season. Within two years of intentionally induced breakthrough, evaporation alone (without transpiration) restored the capability of the capillary barrier covers to function as intended, although water storage in these covers remained at elevated levels.

Abbreviations: CFA, Central Facilities Area • FY, fiscal year • TDR, time-domain reflectometry

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
WORKER and public health and safety concerns associated with remediation of contaminated sites, the lack of proven remediation technologies, and economic considerations have led to the consideration of containment and isolation as viable alternatives to managing buried wastes and contaminated soils. The most serious problems encountered in shallow land burial of wastes are related to water management (Jacobs et al., 1980; Fischer, 1986; Nativ, 1991). The movement of water through buried wastes is recognized as a major mechanism for transporting wastes beyond their original burial boundaries. As a result, a major objective of many containment and isolation projects is to minimize the movement of water into the waste (O'Donnell et al., 1994; Daniel and Gross, 1995). Engineered surface barriers are often used to satisfy this objective.

Most engineered surface covers are made up of several layers that can be categorized as follows: (i) surface layer, (ii) protection layer, (iii) drainage layer, (iv) barrier layer, and (v) gas collection or foundation layer (Daniel, 1994). Design of these covers generally follows the technical guidance for hazardous waste landfills issued by the U.S. Environmental Protection Agency (USEPA, 1989). However, not all layers are needed in every barrier. The environment in which the engineered barrier will be employed and the performance objectives of the barrier will often determine the specific design required. Additionally, not all layers function effectively over the long term. The traditional surface barriers incorporate a compacted soil layer designed to restrict water movement by virtue of its low permeability. There is increasing evidence that the permeability of these layers increases over time (particularly in arid environments) due to a variety of factors (e.g., desiccation cracking, freeze–thaw cycles, erosion, subsidence, root and animal intrusion, and initial flaws in construction) (Suter et al., 1993; Daniel, 1994; Melchior et al., 1994; Omidi et al., 1996). The limitations of currently required designs are driving the development of regulatory guidance that will allow the evaluation and use of new alternative materials and designs (Skahn, 1997).

Surface covers that do not use compacted soil layers and/or other impermeable materials to limit water flow have also been proposed. Anderson et al. (1993) describe a water storage–evapotranspiration cover for use in semiarid regions where potential evapotranspiration greatly exceeds precipitation. Their cover consists of a layer of soil sufficiently thick to store precipitation that falls while plants are dormant. Silt loam soils, in particular, have been shown to have superior water storage characteristics compared with soils of other textures (Petersen et al., 1995). Transpiration by plants established on the soil surface is used to recycle the stored water back into the atmosphere during the growing season, thereby preventing its movement downward into and through the waste. The water storage–evapotranspiration type covers are attractive in that mechanisms that increase the hydraulic conductivity of compacted soil covers and thereby significantly compromise their integrity may have little effect on the integrity of these covers because these covers do not rely on low hydraulic conductivity to function effectively. Including a capillary barrier below the water storage zone can provide additional protection against downward water flow. Capillary barriers refer to the interface between an upper fine-textured soil and a lower coarse-textured material. Under unsaturated conditions, water in the small pores of the fine-textured soil is held at high tensions and will not flow into the large pores of the coarse textured soil where the water tension is low. In order for water to flow out of the soil and into the coarse textured material it must be at sufficiently low tension. Tension decreases as the soil approaches saturation. Once breakthrough occurs, water will drain into the coarse material at a rate controlled largely by the hydraulic conductivity of the overlying soil. The restriction of water movement by a coarse layer in a soil has been recognized for a long time (Alway and McDole, 1917; Nelson and Baver, 1941; Robins, 1959; Eagleman and Jamison, 1961; Miller and Bunger, 1963). More recently, the flow-restrictive properties of capillary barriers are being exploited in designs of engineered surface barriers. Capillary barriers as components of engineered surface covers are the focus of several studies (Nyhan et al., 1990; Gee et al., 1993; Benson et al., 1994; Limbach et al., 1994; Dwyer, 1997; Ward and Gee, 1997; Warren et al., 1997; Khire et al., 1999). These studies notwithstanding, field data that are needed to properly evaluate capillary barrier performance and establish regulatory acceptance are limited (Daniel and Gross, 1995).

Two storage–evapotranspiration type surface covers are undergoing field tests at the Idaho National Engineering and Environmental Laboratory (Bhatt and Porro, 1998). One cover design consists of a thick soil (silt loam) layer with a plant cover. The other design incorporates a capillary–biobarrier within the thick soil (silt loam) layer. The cover(s) chosen for use at this site will be expected to function for hundreds, if not thousands, of years. Over the course of time, it is likely that the cover will be subjected to an extreme rainfall or flooding event that wets the cover to the point of breakthrough. In a worst-case scenario, this could happen when the cover is devoid of vegetation (e.g., in the aftermath of a fire). It is of interest to know how these covers recover from such a condition.

Few studies have focused on this aspect of cover behavior. Campbell et al. (1990), working at the Hanford Site in Washington state, observed continuous drainage from two capillary barrier lysimeters for a year following intentional wetting to breakthrough. The lysimeters were covered with plastic during drainage. A second wetting, still under plastic cover, resulted in a "very slow rate" of drainage for another year. The lysimeters were then uncovered and exposed to ambient conditions. Gee et al. (1993) reported the total annual drainage for the next three years as 0, 0, and 2 cm, respectively. Drainage during the last year was attributed to record snowfall. Stormont and Anderson (1999) conducted infiltration tests in the laboratory on soil columns of sand over gravel. They repeatedly wetted the columns to breakthrough. After each infiltration was terminated, columns were allowed to recover under laboratory conditions until the suction head near the interface increased above the breakthrough head of the gravel to restore the barrier. The breakthrough head did not change substantially after eight cycles of breakthrough and recovery.

Other studies, while not intentionally wetting test plots to breakthrough, have provided the opportunity to observe recovery as a result of exposure to natural conditions. Warren et al. (1997) reported the results from 45 mo of field-testing various covers, including two capillary barrier covers and one soil-only cover, in northern Utah. Breakthrough drainage occurred during each year of the study in response to natural precipitation. More than 90% of all drainage occurred during the months of February through May. This was associated mostly with snowmelt and rain events in late winter and early spring months. Drainage occurred each year despite the fact that annual precipitation totals and snowfall amounts were at or below long-term averages during much of the study, and despite the fact that the covers were vegetated and sloped to enhance evapotranspiration and runoff. Although the capillary barrier covers allowed less drainage than the soil-only cover, annual total amounts were not significantly different. Nyhan et al. (1990) reported the results from a 3-yr field study at Los Alamos, NM that compared the water balance of a capillary barrier to a conventional (crushed tuff backfill) cover. All breakthroughs occurred as a result of snowmelt. Breakthrough in the conventional plot occurred in the first and third winters of the study and lasted about a week and from December through April, respectively. Breakthrough in the capillary barrier plot occurred only after the third winter and lasted from early March through April. Breakthrough did not occur in either type of cover following the intervening winter despite a maximum record spring precipitation. All of the test plots were vegetated in this study and the capillary barrier was sloped at 5% to enhance its diversion capacity.

The objective of this study is to compare the recovery from breakthrough of two types of surface covers under natural weather conditions. Runoff and evapotranspiration were eliminated to maximize the hydrologic stress on the covers during recovery.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Test Facility
This study is being conducted on the cold desert rangeland of the upper Snake River Plain in southeastern Idaho. The test facility is a concrete structure consisting of five cells (plots) on either side of an enclosed access trench. Each cell has four walls and a floor and measures 3.05 m wide by 3.05 m long by 3.05 m deep. The top of each cell is open to the atmosphere. Because of shallow soil at this location, soil is backfilled up around the facility so that cell tops are slightly above grade level. Each cell has two floor drains that empty into separate sumps in the access trench. One drain drains a 0.10-m-wide trough that runs around the bottom perimeter of the cell. The other drain drains the remaining central portion of the cell. Each drained area is sloped toward its drain.

The access trench is approximately 26.2 m long by 3.0 m wide by 3.8 m deep and serves primarily as a protected area for housing the data acquisition system and those instruments (e.g., tensiometers) that penetrate the cell walls. A separate room at the south end of the access trench houses the data acquisition computer and serves as an office area. The access trench is supplied with 115-V electrical service and a telephone line allowing remote access to real time and stored data. A heat pump mounted on the south end of the roof of the access trench minimizes temperature variations and prevents freezing within the access trench.

Figure 1 illustrates the two engineered barrier designs being tested. One cover design consists of a uniform layer of silt loam soil. This soil is covered with a 0.15-m-thick layer of mixed silt loam soil (75% by volume) and gravel (25% by volume) designed to increase the cover's resistance to wind erosion. The other cover design consists of a similar soil–gravel surface layer underlain by 1.45 m of silt loam soil. Beneath this soil is a 0.15-m-thick layer of gravel and a 0.76-m layer of cobbles. A sharp interface between the gravel and the overlying soil is maintained by a high conductivity geotextile. The interface forms a capillary barrier that impedes downward water flow during unsaturated conditions. The cobble layer is intended to minimize biointrusion beyond this depth. A second layer of silt loam soil underlies the cobble layer. The silt loam soil in all test plots comprises 22% sand, 53% silt, and 25% clay, with a porosity of 47.3. Specific properties of the pea-size gravel and softball-size cobbles were not determined. Both types of covers are designed to exploit the transpiration capabilities of plants to extract water that infiltrates into the covers. However, all test plots were maintained devoid of vegetation during the testing period described in this article. The absence of vegetation allows evaluation of the behavior of the barriers under the most extreme hydrologic conditions that are likely to occur. Four replicates of each design were built. The thick soil barrier design test plots are labeled S1 to S4. The capillary–biobarrier design test plots are labeled B1 to B4. Two additional test plots (labeled D5 and I5) were built with the thick soil barrier design for supplemental soil sampling and hydraulic property testing purposes.

View larger version (56K): [in this window] [in a new window]   Fig. 1. Schematic of the cover designs being tested: (top) thick soil, (bottom) thick soil with capillary–biobarrier

View larger version (26K): [in this window] [in a new window]   Fig. 2. Water applied during FY-97 irrigations

All of the test plots subsided as a result of the irrigations. Subsidence was measured at 25 locations on each plot using a 75- by 75-cm grid. A pipe was laid across each plot with its ends resting on the concrete edges of the plot to serve as a grid point locator and to establish a vertical reference point. Averages of the inner 3 by 3 grid subsidence measurements were used for comparisons. Measurements made along the outer edges of the plots are not included in the averages because subsidence along the edges appeared to be affected by the edges and not representative of overall plot subsidence.

Barrier Performance Assessment
The function of the engineered barriers undergoing evaluation is to minimize the downward movement of water through the barriers. Rather than providing just an impermeable barrier to stop downward water movement, these barriers are designed to rely on several natural, interactive processes to function effectively. Determining the engineered barrier's water balance has been identified as one of the most effective ways to evaluate the barrier's overall performance (Nyhan et al., 1990; Ward and Gee, 1997). The water balance of the engineered barrier test plots quantitatively accounts for all water affecting the test plots. The water balance equation is represented as:

[1]

where S is the change in soil water storage, P is precipitation, I is irrigation, R is surface runoff, D is drainage, and ET is evapotranspiration. Changes in soil water storage in the following water balance evaluation are based on water contents determined from the measurement of bulk soil dielectric constants using time-domain reflectometry (TDR). Three TDR probes were installed at depths of 20, 40, 60, 80, 100, 120, 140, 155, 255, and 295 cm in each test plot. Three additional probes were installed at depths of 180, 205, and 230 cm in each thick soil test plot. The TDR probes were monitored hourly in FY-97 and once every 3 h in FY-98 and FY-99. The water balance calculations were made on a fiscal year basis rather than a calendar year basis to avoid soil-freezing effects on TDR readings. The change in water stored in each test plot, S, at any given time was calculated as:

[2]

where L is the depth of the test plot, is the volumetric water content, z is the depth interval of the water content measurement, and t is time. Moisture contents in the 0.91-cm-thick capillary–biobarrier were not determined because TDR probes were not installed in that zone. This is of little consequence in the following discussion because the water holding capacity of the barrier zone is minimal. Precipitation at the test facility is measured using a tipping bucket rain gauge equipped with a heater. This rain gauge is located on the surface of the backfill surrounding the test plots. Irrigation was measured by flow meters. Surface runoff is zero because the test cell sidewalls extend above the surface of the test plots. Drainage is measured by both tipping buckets installed at the outlet of the plot drains (to measure volume of drainage) and pressure transducers installed on the floor of the drainage sumps (to measure water level in the sump). Evapotranspiration is determined by solving Eq. [1] for ET with measured values substituted for all other components. During FY-97, FY-98, and FY-99 the ET term represents evaporation only because all plot surfaces were kept cleared of vegetation, thereby eliminating transpiration by plants.

Thermocouples were used to measure soil temperatures during the study. Two thermocouples were installed at depths of 5, 20, 40, 60, 80, 155, and 270 cm in each test plot. Five additional thermocouples were installed at depths of 60 and 270 cm in each plot.

Cables from all instruments buried in a test plot were routed to a plot cable tower. Each tower is centrally located within a plot and serves to minimize cable intrusions within the plot as well as provide a centralized path for cables exiting the plot. The tower is a PVC tube that is installed vertically within the plot and does not extend to the plot surface, thereby minimizing surface penetrations and possible artificially created pathways for water flow into the plot. Instrument cables enter the tower through sealed penetrations and are routed through a sealed penetration in the floor of the plot to the data acquisition system in the access trench. Readings from all of the above instruments are made automatically by the data acquisition system.

	RESULTS AND DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Precipitation
Cumulative precipitation measured at the test facility during FY-97, FY-98, and FY-99 is shown in Fig. 3 along with the long-term average precipitation measured 11 km northeast of the test facility at the Central Facilities Area (CFA). The CFA is the nearest location to the test facility where meteorological data have been collected for a long time (since 1950). Total precipitation for each of the three periods was close to the long-term average. The patterns of precipitation for each period, however, differed from the long-term average and from each other. In FY-97 precipitation in December and the first few days of January was significantly greater than the long-term average. This was followed by more than four months of below-average precipitation, which brought the total back to normal. Another period of rainfall in August raised the level above normal where it remained until the end of FY-97. In FY-98 precipitation in November and December was less than the long-term average and the cumulative amount remained below average until May. Precipitation remained near normal throughout early summer, fell slightly below normal toward the end of summer, and recovered to match the long-term average at the end of FY-98. Precipitation in FY-99 followed the long-term average curve closer than in the previous two years through spring when it increased above the long-term average. Minimal precipitation after 6 June kept the yearly total below the long-term average.

View larger version (19K): [in this window] [in a new window]   Fig. 3. Cumulative precipitation during FY-97, FY-98, and FY-99 and long-term average cumulative precipitation

View larger version (19K): [in this window] [in a new window]   Fig. 4. Depths of soil freezing during FY-97, FY-98, and FY-99 in a representative test plot

View larger version (26K): [in this window] [in a new window]   Fig. 5. Snow depths at the Central Facilities Area (CFA) and the test facility during FY-97, FY-98, and FY-99 and long-term average snow depth

View larger version (20K): [in this window] [in a new window]   Fig. 6. Average time-domain reflectometry (TDR)-measured water content profiles in B plots (a) and S plots (b)

View larger version (22K): [in this window] [in a new window]   Fig. 7. Average cumulative drainage as a function of time

Since all the test plots are located near one another, they all received essentially equal inputs of water due to the winter precipitation. Equal inputs, however, resulted in about twice the total amount of drainage from the thick soil test plots (average from 1 Mar. 1998 to 30 Sept. 1998 = 11.3 cm) as from the capillary–biobarrier test plots (average from 1 Mar. 1998 to 30 Sept. 1998 = 6.0 cm). Even Plots S1 and S4 yielded more drainage than their corresponding capillary barrier plots (B1 and B4) despite the fact that irrigations on both S1 and S4 were terminated prior to initiation of drainage. Plots S4 and B4 received comparable amounts of water during irrigation, whereas Plot S1 received considerably less water than Plot B1.

Figure 7 (FY-98 drainage) also shows that drainage from the capillary–biobarrier plots slowed down more abruptly than drainage from the thick soil test plots (note the sharper knee in the curve for the capillary–biobarrier plots compared with the thick soil plots). This is consistent with observations made of drainage immediately following the irrigation. Also, drainage from the capillary–biobarrier plots again continued for a longer time than drainage from the thick soil plots. Table 4 clearly shows that all of the thick soil plots ceased draining in July (with only a small recurrence of drainage in Plot S2 in September), whereas three of the four capillary–biobarrier plots continued to drain into December (Plot B3 drained into November). It warrants repeating, however, that the capillary–biobarrier plots generated less total drainage for the year than the thick soil plots even though they drained for a longer time.

The drainage curve for the thick soil covers again exhibits a noticeable inflection, this time at the beginning of June. This inflection most likely is due to a 10-d period beginning 28 May that totaled 49.5 mm of rainfall. The largest single-day total during this period was 21.8 mm on 29 May. As in the previous year, this rainfall increased the drainage that still was occurring as a result of the March thaw. The capillary–biobarrier covers, on the other hand, prohibited this rainfall from contributing to drainage.

Evaporation
Calculated evaporation is shown in Table 2. The ET term represents just evaporation, E, because test plots were kept cleared of vegetation during FY-97 to FY-99. The negative value for evaporation for Test Plot S1 in FY-97 stands out as an anomaly. This probably results from the accumulated imprecision of measurements when small values of ET are determined by summation. This value is excluded from further discussions. On average, evaporation from the capillary–biobarrier test plots is greater than evaporation from the thick soil test plots in all three years (16.35 vs. 8.27 cm, 21.95 vs. 13.79 cm, and 20.92 vs. 13.60 cm in FY-97, FY-98, and FY-99, respectively). This observation is consistent with the other factors of the water balance equation discussed above. The capillary–biobarrier test plots have allowed less drainage than the thick soil test plots even when equal amounts of infiltration have occurred. Water that is redistributed downward and contributes to drainage in the thick soil plots is held in the upper portions of the capillary–biobarrier plots by the capillary barrier. Figure 8 illustrates the higher moisture contents found in the upper portions of the capillary barrier plots compared with the thick soil plots. The closer water is to the surface, the more it is subject to the evaporative forces of the atmosphere. Consequently, greater evaporation is observed from the capillary–biobarrier test plots than from the thick soil test plots.

View larger version (15K): [in this window] [in a new window]   Fig. 8. Average soil moisture profiles on 1 Oct. 1997

Evaporation as a percentage of precipitation (precipitation plus irrigation in FY-97) differed significantly between FY-97 and the next two years. In the capillary–biobarrier test plots, evaporation represented 90.6 and 99.7% of precipitation in FY-98 and FY-99, respectively, but only 22.6% of precipitation (plus irrigation) in FY-97. In the thick soil test plots, evaporation represented 56.9 and 64.8% of precipitation in FY-98 and FY-99, respectively, but only 9.6% of precipitation (plus irrigation) in FY-97. The lower percentages in FY-97 result from the fact that much of the water that infiltrated ended up filling empty storage capacity in the plots, particularly in the lower reaches of the plots where the water is out of reach of evaporative forces.

Subsidence
As noted above, all of the test plots subsided as a result of the irrigations. Thick soil test plots subsided an average of 16.1 cm whereas capillary–biobarrier test plots subsided an average of 7.3 cm. Silt loam soils are reasonably strong and incompressible when dry, but they are highly susceptible to collapse when wetted (Mitchell, 1976; Rodriguez et al., 1988). Collapse resulting from saturation is likely when the density of the soil is sufficiently low to give a void space larger than that needed to hold the liquid limit water content (Gibbs and Bara, 1967). The test plot soils were compacted to 82% of maximum, as per design specifications, during installation. The level of compaction was designed to strike a balance between the minimization of collapse and the creation of a zone suitable for plant root penetration when the plots are vegetated. Evidently, these soils are still susceptible to collapse at this level of compaction.

The difference in construction between the two types of test plots is the inclusion of the gravel and cobble layers in the capillary barrier plots. Greater subsidence in the thick soil plots seems to indicate that settlement took place in the silt loam and not in the gravel and cobble layers, although this has not been confirmed. Assuming settlement occurred over the full length of the silt loam (including the gravel–soil mixture at the surface), the percentage of settlement is 3.48% in the capillary barrier plots versus 5.37% in the thick soil plots. Assuming settlement only occurred above the capillary barrier, the percentage of settlement is 4.56%. If the amount of settlement is proportional to the depth of soil, the percentages between the two designs should be equal. The results above indicate that the occurrence of settlement above the capillary barrier but not below it fits this assumption better than settlement over the full length of soil. This, also, has not been confirmed. The planned excavation of selected plots when all experiments are completed will allow us to determine soil properties and compare them to initial properties. This should shed some light on how the settlement occurred.

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 
Test plot monitoring and water balance analyses revealed significant differences between the performance of the thick soil and capillary–biobarrier test plots both before and after the irrigation to breakthrough. In the year prior to the irrigations, none of the test plots produced drainage as a result of exposure to ambient conditions. However, differences in the performance of the two barrier designs were already beginning to be revealed in the internal distribution of water within the plots. The capillary barrier was effective in prohibiting the downward migration of water.

Winter weather conditions at the site caused numerous precipitation events to accumulate in the form of snow on the soil surface. Combined with soil freezing, this effectively decoupled infiltration from individual precipitation events during this period. The relatively rapid thawing of accumulated snow and soils at the onset of warm weather resulted in the most significant infiltration events observed during testing (with the exception of the irrigations).

Water storage levels in all test plots were at elevated levels compared with pre-test levels following cessation of drainage resulting from the irrigations. As a result, infiltration of melting snow during the subsequent spring overloaded the storage capacity and produced drainage in all plots. Drainage from capillary barriers slowed down faster than drainage from thick soil barriers, and, although drainage from capillary barriers continued for a longer time than drainage from thick soil barriers, the capillary barriers yielded less total drainage.

By limiting drainage, the capillary–biobarriers increased water storage in the upper portions of the test plots compared with the thick soil barriers. This led to increased recycling of water to the atmosphere through evaporation. Increased evaporation in the capillary barrier test plots reduced the amount of water stored in the plots thus providing room for all of the infiltrating water during the FY-99 spring thaw (the second season following the irrigations) without triggering drainage. Drainage occurred in all of the thick soil test plots in FY-99. During current testing, evaporation alone (without transpiration) at this site had the ability to restore the functioning capability of the capillary barrier cover. Continued monitoring of the test plots will determine to what degree both barrier types are restored to their intended functioning and the time scale required to do so.

Users of silt loam soil for engineered barriers should be aware of its structural characteristics. The silt loam soil used to construct the barriers collapsed during the irrigations as evidenced by surface subsidence. Subsidence occurred in spite of some compaction of soils during test plot installation.

	ACKNOWLEDGMENTS

 
The author gratefully acknowledges project management support provided by Tom Honeycutt and Rajiv Bhatt, and Tim Kaser's efforts in maintaining the data acquisition system. Work supported by the U.S. Department of Energy, Assistant Secretary for Environmental Management, under DOE Idaho Operations Office Contract DE-AC07-99ID13727.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS AND DISCUSSION CONCLUSIONS REFERENCES

 

• Alway, F.J., and G.R. McDole. 1917. Relation of water-retaining capacity of a soil to its hygroscopic coefficient. J. Agric. Res. 9:27–71.
• Anderson, J.E., R.S. Nowak, T.D. Ratzlaff, and O.D. Markham. 1993. Managing soil moisture on waste burial sites in arid regions. J. Environ. Qual. 22:62–69.[Abstract/Free Full Text]
• Benson, C.H., P.J. Bosscher, D.T. Lane, and R.J. Pliska. 1994. Monitoring system for hydrologic evaluation of landfill covers. Geotech. Testing J. 17:138–149.
• Bhatt, R.N., and I. Porro. 1998. Evaluation of engineered barriers at the Idaho National Engineering and Environmental Laboratory. In Proc. of the WM `98 Conf., Tucson, AZ. 1–5 Mar. 1998. WM Symposia, Tucson, AZ.
• Campbell, M.D., G.W. Gee, M.J. Kanyid, and M.L. Rockhold. 1990. Field lysimeter test facility: Second year (FY 1989) test results. PNL-7209. Pacific Northwest Lab., Richland, WA.
• Daniel, D.E. 1994. Surface barriers: Problems, solutions and future needs. p. 441–487. In Proc. of the 33rd Hanford Symp. on Health and the Environ., Richland, WA. 7–11 Nov. 1994. Battelle Press, Columbus, OH.
• Daniel, D.E., and B.A. Gross. 1995. Caps. p. 119–140. In R.R. Rumer and J.K. Mitchell (ed.) Assessment of barrier containment technologies. Int. Containment Technol. Workshop, Baltimore, MD. 29–31 Aug. 1995. U.S. Dep. Of Energy, Washington, DC.
• Dwyer, S.F. 1997. Large-scale field study of landfill covers at Sandia National Laboratories. p. 87–107. In T.D. Reynolds and R.C. Morris (ed.) Landfill capping in the semi-arid West: Problems, perspectives, and solutions. Conf. Proc., Grand Teton National Park, WY. 21–22 May 1997. Environ. Sci. and Res. Foundation, Idaho Falls, ID.
• Eagleman, J.R., and V.C. Jamison. 1961. The influence of soil textural stratification and compaction on moisture flow. Res. Bull. 784. Missouri Agric. Exp. Stn., Columbia, MO.
• Fischer, J.N. 1986. Hydrogeologic factors in the selection of shallow land burial for the disposal of low-level radioactive waste. Circ. 973. U.S. Geol. Surv., Reston, VA.
• Gee, G.W., D.G. Felmy, J.C. Ritter, M.D. Campbell, J.L. Downs, M.J. Fayer, R.R. Kirkham, and S.O. Link. 1993. Field lysimeter test facility Status Rep. IV: FY 1993. PNL-8911. Pacific Northwest Lab., Richland, WA.
• Gibbs, H.J., and J.P. Bara. 1967. Stability problems of collapsing soils. J. Soil Mech. Found. Div. A.S.C.E. 93(SM4):577–594.
• Jacobs, D.G., J.S. Epler, and R.R. Rose. 1980. Identification of technical problems encountered in the shallow land burial of low-level radioactive wastes. ORNL/SUB-80/13619/1. Oak Ridge Natl. Lab., Oak Ridge, TN.
• Khire, M.V., C.H. Benson, and P.J. Bosscher. 1999. Field data from a capillary barrier and model predictions with UNSAT-H. J. Geotech. Geoenviron. Eng. 125(6):518–527.
• Limbach, W.E., T.D. Ratzlaff, J.E. Anderson, T.D. Reynolds, and J.W. Laundre. 1994. Design and implementation of the protective cap/biobarrier experiment at the Idaho National Engineering Laboratory. p. 359–377. In Proc. of the 33rd Hanford Symp. on Health and the Environ., Richland, WA. 7–11 Nov. 1994. Battelle Press, Columbus, OH.
• Magnuson, S.O. 1993. A simulation study of moisture movement in proposed barriers for the Subsurface Disposal Area, INEL, EGG-WM-10974. EG&G Idaho, Inc., Idaho Falls, ID.
• Melchior, S., K. Berger, B. Vielhaber, and G. Miehlich. 1994. Multilayered landfill covers: Field data on the water balance and liner performance. p. 411–425. In Proc. of the 33rd Hanford Symp. on Health and the Environ., Richland, WA. 7–11 Nov. 1994. Battelle Press, Columbus, OH.
• Miller, D.E., and W.C. Bunger. 1963. Moisture retention by soil with coarse layers in the profile. Soil Sci. Soc. Am. Proc. 27:586–589.
• Mitchell, J.K. 1976. Fundamentals of soil behavior. John Wiley & Sons, New York.
• Nativ, R. 1991. Radioactive waste isolation in arid zones. J. Arid Environ. 20:129–140.
• Nelson, W.R., and L.D. Baver. 1941. Movement of water through soils in relation to the nature of the pores. Soil Sci. Soc. Am. Proc. 5:69–76.
• Nyhan, J.W., T.E. Hakonson, and B.J. Drennon. 1990. A water balance study of two landfill cover designs for semiarid regions. J. Environ. Qual. 19:281–288.[Abstract/Free Full Text]
• O'Donnell, E., R.W. Ridky, and R.K. Schulz. 1994. Control of water infiltration into near-surface, low-level waste-disposal units in humid regions. p. 295–324. In Proc. of the 33rd Hanford Symp. on Health and the Environ., Richland, WA. 7–11 Nov. 1994. Battelle Press, Columbus, OH.
• Omidi, G.H., J.C. Thomas, and K.W. Brown. 1996. Effect of desiccation cracking on the hydraulic conductivity of a compacted clay liner. Water Air Soil Pollut. 89:91–103.
• Petersen, K.L., S.O. Link, and G.W. Gee. 1995. Hanford site long-term surface barrier development program: Fiscal Year 1994 highlights. PNL-10605. Pacific Northwest Lab., Richland, WA.
• Robins, J.S. 1959. Moisture movement and profile characteristics in relation to field capacity. Int. Comm. Irrig. Drainage 8.509–8.521.
• Rodriguez, A.R., H. del Castillo, and G.F. Sowers. 1988. Soil mechanics in highway engineering. Trans Tech Publ., Clausthal–Zellerfeld, Germany.
• Skahn, K. 1997. EPA's performance, monitoring, and maintenance requirements for landfill caps. p. 19–25. In T.D. Reynolds and R.C. Morris (ed.) Landfill capping in the semi-arid West: Problems, perspectives, and solutions. Conf. Proc., Grand Teton National Park, WY. 21–22 May 1997. Environ. Sci. and Res. Foundation, Idaho Falls, ID.
• Stormont, J.C., and C.E. Anderson. 1999. Capillary barrier effect from underlying coarser soil layer. J. Geotech. Geoenviron. Eng. 125(8):641–648.
• Suter, G.W., R.J. Luxmoore, and E.D. Smith. 1993. Compacted soil barriers at abandoned landfill sites are likely to fail in the long term. J. Environ. Qual. 22:217–226.[Abstract/Free Full Text]
• USEPA. 1989. Technical guidance document: Final covers on hazardous waste landfills and surface impoundments. USEPA Rep. 530-SW-89-047. Office of Solid Waste and Emergency Response, USEPA, Washington, DC.
• Ward, A.L., and G.W. Gee. 1997. Performance evaluation of a field-scale surface barrier. J. Environ. Qual. 26:694–705.[Abstract/Free Full Text]
• Warren, R.W., T.E. Hakonson, and K.V. Bostick. 1997. The hydrologic evaluation of four cover designs for hazardous waste landfills. p. 181–197. In T.D. Reynolds and R.C. Morris (ed.) Landfill capping in the semi-arid West: Problems, perspectives, and solutions. Conf. Proc., Grand Teton National Park, WY. 21–22 May, 1997. Environ. Sci. and Res. Foundation, Idaho Falls, ID.

This article has been cited by other articles:

				J. W. Nyhan A Seven-Year Water Balance Study of an Evapotranspiration Landfill Cover Varying in Slope for Semiarid Regions Vadose Zone J., June 9, 2005; 4(3): 466 - 480. [Abstract] [Full Text] [PDF]

				R. M. Lehman, K. E. Baker, and E. D. Mattson Distribution of Microorganisms and their Activities in Capillary Barriers: Implications for Modeling of Hydrologic Transport through Capillary Barriers Vadose Zone J., February 1, 2004; 3(1): 134 - 142. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (3) Citing Articles via Google Scholar Google Scholar Articles by Porro, I. Search for Related Content PubMed PubMed Citation Articles by Porro, I. Agricola Articles by Porro, I. Related Collections Municipal Waste Water Quality