|
|
INVESTIGATING
LEACHATE MIGRATION
CASE STUDY:
FIELD ANALYSIS AND MODELLING OF A MUNICIPAL LANDFILL IN NORTHERN ENGLAND
TANYA SPLAJT
Centre
for Waste and Pollution Research
Geography
Department, The University of Hull
Hull,
UK, HU6 7RZ
Telephone:
(44) (1482) 465-371
Fax:
(44) (1482) 466-340
Email
Address: T.D.Splajt @ geo.hull.ac.uk
ABSTRACT
Contamination of water
resources by landfill leachate is a growing problem. Waste management companies
often struggle with the challenge of containing and controlling leachate
migration. This case study of a
landfill site in Northern England is an example of post-management problems
that can occur after leachate-containment structures are implemented within a
landfill site. The case study site was identified as a regional threat in 1992.
Leachate containment structures built in recent years have improved water
quality conditions however problem areas within the site continue to exist.
This investigation reviews the geologic and hydrologic site conditions. The
SEEP/w hydro-geologic model was used to construct a finite element,
two-dimensional simulation of the problem area on site. Conclusions were formed
by integrating data analysis findings and SEEP/w sensitivity analysis. The
analysis found that high water table levels on site are causing the leachate
management structures to under-perform. Leachate from the landfill is migrating
through and bypassing containment walls and re-circulation systems resulting in
local and downstream surface and groundwater contamination. Landfill leachate
migration is a long-standing problem that is expected to increase with the
growth of urban populations and resulting landfills. This study points to the
importance of understanding the hydro-geologic composition of a landfill and
recognising the geo-physical limitations of engineered pollution control
systems. These are integral issues that will determine the effectiveness of
leachate management strategies used in municipal landfill sites.
Keywords: leachate
migration, municipal landfill site management, hydrology, water pollution
control, hydro-geology
INTRODUCTION
Monitoring leachate movement in groundwater is a challenging problem. (Hudak & Loaiciga, 1992) The threat of migrating leachate originating from landfill sites is an important water quality and waste management issue. This is especially the case in the UK where 70% of controlled waste goes into landfill sites as an integral part of the national approach to waste management. (DOE, 1995, p.8) Further research is needed in areas of landfill hydraulics to more effectively predict and control leachate migration. (Pickens, 1976)
This research paper is a
case study analysis of landfill leachate management using a landfill site in
Northern England. For the purpose of this paper, the landfill will be named
LANDFILL SITE A. The project reviews background geologic and hydrologic
conditions of Landfill Site A and uses the SEEP/w hydro-geologic model to
simulate a problem area on site that is a likely source of migrating leachate
contamination. The project's assessment was based on published leachate
migration studies conducted by Dykes (1994), Haitjema (1991), Kjeldsen (1992),
Sykes (1982) and Pickens (1976) The purpose behind this investigation was to
better understand leachate migration patterns and associated hazards facing
Landfill Site A.
THE GEO-PHYSICAL ASSESSMENT OF LANDFILL SITE A
SITE
LOCATION AND HISTORY
Landfill Site A was opening
in 1983 and annually accepts 22,000 tones of bailed domestic, commercial and
industrial waste. Historically the site was an air force landing strip, later
used for agricultural purposes. It is surrounded by farmland, bordering with a
railway line, an industrial zone and a water treatment plant.
The site was identified as
a source of local leachate migration in 1992. Since then, several engineered
structures have been constructed on site to monitor and control leachate from
migrating into surrounding soils and waters. Prior to construction, leachate
migrated from the northern and eastern perimeters of the site into downstream
surface water regimes and groundwater sources.
From 1994 through to 1996
toe drains, bore holes and subsurface trench wall (constructed of geofin and
tyres) containing leachate-collecting pumps and drains were built along the
northern and eastern perimeter of Phase 1 to maintain acceptable leachate
levels. (Figure 4) A bentonite cut-off wall was constructed encircling Phases
1, 2 and 3 to laterally contain leachate from migrating off site. The wall is approximately 6m tall, ending 2m
below the depth of the buried waste. It is 0.60 cm thick having a hydraulic
conductivity of 8.64 x 10-4
m/d. (CL Associates, 1995)
REGIONAL
HYDROLOGY AND GEOLOGY
Landfill Site A contributes
to the local surface water catchment through ditches and sewage outlets. A
network of small streams drains into two popular beaches on the North Sea.
These beaches are potentially threatened by upstream drainage as they are
bathing beaches meeting EC Bathing Water Directives. Information about the
regional rainfall is shown in Tables 1.
The Site's topography
ranges from 7-16 m AOD due to landfilling operation. There are approximately
18m of highly complex Quaternary Till with discontinuous sequences of clay
containing gravel and sand lenses. The Till has a hydraulic gradient of 0.0002.
Regional trends confirm that Till erodes easily thought to account for regional
trends of geologic collapse. (Frostick, 1998) The regional aquifer is a chalk
aquifer with a hydraulic gradient of 0.004. Clay bands confine the chalk groundwater
levels.
Three large gravel and sand
lenses are believed to exist on site. Their exact location and size are
unconfirmed however piezometric sampling on site found constant levels of
hydraulic head throughout the drift indicating that the lenses are in hydraulic
continuity with each other. The piezometric head with the drift varies between
7m AOD in summer to 8.5m AOD in winter. (Entec, 1996) The presence of gravel and sand lenses is of extreme hydrologic
importance. These materials have a hydraulic conductivity that is much higher
then those of the surrounding Quaternary Till and clay. Leachate may be
migrating through these easily permeable patches into surrounding groundwater
flow paths. (Entec, 1996)
CURRENT
CONDITIONS
The water quality around
Landfill Site A has been closely monitored since 1992. In recent years leachate
migrated southeast in the direction of the regional groundwater flow and
locally in the direction of the greatest hydraulic gradient. Water quality
monitoring (Figure 1) has indicated that the cut-off wall on most parts is an
effective hydraulic barrier isolating the present landfill from the surrounding
hydraulic regime. The need for caution is at bore hole 20 where leachate
migration continues to occur. (Figure 2)
Figure 3 describes the
hydrological conditions on site at the time of this investigation (1998)
indicating a regional increase in water table levels of 3 m (from 7.5 to 10.5 m
AOD). A perched water table also formed
on site, along with overland flow from daily landfilling operations.
BORE
HOLE 20
Bore hole 20 (Figure 4) is located along the northern perimeter of Phase 1, on the outer side of the cut-off wall. It is located within a highly permeable sand and gravel lens with a depth of 7.25m. It was drilled to monitor leachate that may be migrating under or through the cutoff wall. Figure 2 shows that leachate concentrations at bore hole 20 were similar to other nearby bore holes monitored during 1996. A cause for concern arose in December 1996 when leachate concentrations intensified and remained high through 1997 and 1998.
MODELLING USING SEEP/w
SEEP/w is an established
engineering, hydrological model. Its behaviour and predictive capabilities are
known for soils in temperate regions. SEEP/w is based on Darcy's Law for
saturated flow and Richard's Equation for unsaturated conditions. The input
parameters define the model-produced output. This modelling investigation used
boundary conditions that simulated precipitation input, leachate pumping and
nodal flux.
SEEP/w was used to
construct a 2-dimensional model to explore leachate migration patterns that may
be occurring around the area of bore hole 20. (Figures 5 and 6) The model was
run under steady-state and saturated flow conditions. (SEEP/w Manual, 1991)
MODELLING
Four model scenarios were constructed to investigate the sensitivity of engineered structures and geologic conditions around bore hole 20. (Table 2) The cross section model was constructed with four assumptions in mind:
1. The
model's geologic layout was based on 3 bore hole samples taken near bore hole
20 and does not account for the geologic complexity of the area.
2. Field
data (e.g. bore hole and water samples) contain some degree of error and
approximation that increases when added to the model.
3. Input
variables (e.g. hydraulic conductivity and volumetric moisture content) are
theoretical values and may be different from field conditions.
4. The
model scale is 1cm=1m. Some grid features (e.g. the distance from the cutoff
wall to the geofin wall) are not to scale with realistic distances around bore
hole 20.
CALIBRATION, VERIFICATION, VALIDATION and SENSITIVITY ANALYSIS
SEEP/w cannot be
meaningfully calibrated because only direct field measurements are used to
define input data. Model verification was
done by graphing calculations of pressure head verses the number of iterations,
x-y gradient, x-y velocity and volumetric moisture content. The process of
validation compared field monitored water table levels in Landfill Site A to
water table levels in SEEP/w calculations. Sensitivity analysis was conducted
by testing the sensitivity of model parameters in Scenario 1 and then by adding
conditions to the model that simulated a cutoff wall, geofin wall and leachate
re-circulation pump.
SCENARIO
1
The first scenario
increased precipitation input in three intervals simulating low to high
recharge rates. Groundwater flow, overland flow and water table levels
increased when recharge was raised from 7 to 10 (m/s). Flow through the sand
lens toward the ditch increased when recharge was high. Groundwater flow was in
the direction of the greatest hydraulic gradient.
SCENARIO 2
Scenario 2 took conditions
of Scenario 1 adding a bentonite cut-off wall. Low and high recharge conditions
were tested. Figure 5 shows that when recharge rates were high, the water table
rose from 7.5 to 10 m AOD, horizontal flow in the sand lens began flowing
through and beneath the cutoff wall and surface water flow toward the toe drain
intensified. The presence of the cutoff wall was found to decrease the
hydraulic flux through the refuse and sand lens when compared to Scenario 1.
SCENARIO 3
Scenario 3 simulated the
presence of a geofin wall without a pumping system. This scenario simulated the
conductive role that a geofin wall may play if the leachate re-circulation
pumps are not working or if collection pipes within the geofin wall are damaged
or not in use. The geofin wall created a direct lateral drain for percolating
moisture when recharge rates were high. (Figure 6) Changing the location of the
geofin wall with in the sand lens decreased and reversed groundwater flow.
SCENARIO 4
In the final version of
modelling, a pumping node was added to Scenario 3 to simulate leachate pumps
within the model. The sensitivity of a pump suction strength, depth, location
and quantity of were tested. Optimal hydraulic results were produced when the
pump was located in the centre of the geofin wall at 7.0 m AOD depth, suction
strength was high (at 10 m/s) and recharge conditions were low. (Figure 7).
Pump suction strength was a
sensitive parameter impacting water table levels and water flow directions. An
optimal water table level of 7.5-m AOD was produced when pump suction strength
was set equal to recharge rate. Pump quantity, depth and locations in the sand
lens were sensitive parameters. Adding an additional pump into the sand lens
decreased fluxes in the sand lens, cutoff wall and ditch. Flow found
alternative flow paths when the pump depth was at 8.5 or 6.5 m AOD. When the
pump and geofin wall were moved closer to the cutoff wall at a depth of 7.5 m
AOD, flow was redirected to laterally escape through the geofin wall.
CONCLUSIONS
The case study of leachate
migration at Landfill Site A pointed to several issues of concern:
LATERAL
SUBSURFACE FLOW
Lateral subsurface flow
away from the site is likely to be occurring. The unconfirmed location and size
of three sand-gravel lenses within the clay sequence suggest this. Channels of
contaminated groundwater may exist within the site flowing in the direction of
the regional groundwater gradient. Modelling results showed that the sand lens
was a focal flow path under most scenario conditions. The hydraulic
conductivity and volumetric soil moisture content assigned as input parameters
determined the rate of model-predicted flow. The site requires a spatially
representative geologic survey that investigates the hydro-geologic variations
in compaction and depth of buried refuse and areas where soil removal and
replacement operations have occurred. These factors were not considered in
previous geologic assessments and may be important factors impacting current
paths of leachate migration.
LEACHATE
MIGRATION AROUND AND THROUGH THE CUTOFF WALL
Leachate is likely to be
migrating around and through the cutoff wall. The quality assurance tests
conducted in 1995 by CL Associates show that the cut-off wall would optimally
contain inter-landfill leachate levels of up to 7.5m AOD. The cut-off wall
cannot be expected to contain leachate from migrating off site if inter-site
water levels are higher then 7.5 m AOD. The regional water table level has also
increased in recent years. If water
table levels inside the wall are higher then those outside the wall, a
hydraulic gradient is created, resulting in leachate migration across and
around the cutoff wall. Bore hole data shows that aside from bore hole 20, the
cutoff wall is an effective hydraulic barrier within the sand lens. Modelling
confirmed this by acceptable water table and flow conditions when recharge was
low.
CUTOFF
WALL EROSION
The pressure gradient
within the sand-gravel lens in the area of bore hole 20 is likely to be much
higher then that of upper and lower clay layers. (Domenic & Schwartz, 1990,
p.80) Figure 2 shows that leachate migration was minimal at bore hole 20 from
its construction in February 1996 until December 1996. Following this period,
leachate migration levels have increased and fluctuated. The stronger gradient
conditions within the sand lens may have weakened the bentonite clay wall at
the time of construction, deteriorating its low hydraulic conductivity with
time. This hypothesis is supported by DOE (1995, p.172) which states that
"heterogeneous conditions within a landfill will impose forces on a vertical
structure... Lateral forces will deflect the structure out of vertical alignment
and may crush or shear it." SEEP/w was not able to simulate cutoff wall erosion
however migration through the cutoff wall occurred when recharge was high or
when the hydraulic conductivity of the wall was lowered. At least one more bore
hole is needed on the inner side of the cutoff wall across from bore hole 20 to
calculate the migration gradient across the wall at this point on site.
OVERLAND
FLOW
Overland flow from surface
landfill operations was observed after heavy storm events during field
investigations. Modelling Scenarios 3 and 4 produced a perched water table and
surface overland flow when recharge rates were high. The perched water table
formed above the cut-off wall cap, which had a lower hydraulic conductivity
then, the surrounding clay. Surface overland flow and toe drain water quality
needs to be monitored after heavy rain events. Other conditions that intensify
surface overland flow include surface soil compaction, the hydraulic
conductivity of capped areas and the height and direction of surface waste
piling.
EFFECTIVE
HYDRAULIC BARRIERS
Modelling found that the
cutoff wall, geofin wall and leachate pump (Scenarios 2-4) reduced flux and
hydraulic head in the sand lens and refuse areas when recharge was low. The
geofin wall and re-circulation pump (Scenario 4) decreased water table levels
under both low and high recharge conditions while the cutoff wall was most
effective when under rates were low.
LEACHATE
RECIRCULATION INEFFICIENCY
A 6-pump re-circulation
system was installed on site in 1996 to maintain low inter-site water table
levels of 7.5mAOD in order for the cut-off wall to horizontally contain
inter-site leachate from migrating off site. Figures 2 and 3 indicate that the
pumping system is not producing the desired results. The recent rise in
regional and on-site water table levels indicates that the leachate
re-circulation system may be strained, circulating leachate levels that are
above the capacity of the six-pump system.
Model simulations in
Scenario 1 and 2 indicated that the recharge value is an important parameter
increasing the water table and flux. Scenario 4 was not able to effectively
simulate all aspects of the leachate re-circulation system (evapotranspiration
and refuse degradation) however results showed that the pump's location, depth,
quantity and suction strength were important factors effecting water table
levels, flow direction and velocity. Modelling also showed that the geofin wall
played a highly conductive role in laterally channeling groundwater when pump
conditions were changed. Modelling emphasized the need to assess the
effectiveness of the leachate re-circulation system in Landfill Site A.
FIGURES
FIGURE 1: Indication of cut-off wall effectiveness: Northern ditch surface water samples shows leachate presence (ammonia) decreased after cut-off wall construction.
FIGURE 2 Ammonia concentrations as an indicator of leachate migration after cut-off wall construction at bore holes 19 -21, sampling taken from 05/1996 - 10/1997
FIGURE 4: Aerial view of Landfill Site A. Circular
area is model simulated problem area near Bore hole 20.
Figure 7: Pump depth is a sensitive parameter in determining flux
quantities. Flux through the ditch, geofin wall and refuse areas of the model
increased when the pump was higher or lower then 7.5 m AOD.
|
|
PRECIPITATION DATA |
|
Average precipitation |
659.9 mm/year |
Potential evapotranspiration
|
573.3 mm/year |
|
Actual evapotranspiration |
497.0 mm/year |
|
Effective precipitation |
163.0 mm/year |
|
Clay, sand & gravel hydraulic gradient |
0.0002 |
|
Lower Chalk Hydraulic Gradient |
0.004 |
TABLE 1:
(ENTEC, 1996)
|
|
CROSS SECTION DESCRIPTION |
|
1 |
·
Bh 20 Geologic lay out |
|
2 |
·
Bh 20 Geologic lay out , Cut-off wall |
|
3 |
·
Bh 20 Geologic lay out , Cut-off wall, Geofin wall |
|
4 |
·
Bh 20 Geologic lay out, Cut-off wall, Geofin wall ·
Leachate pump |
TABLE 2:
DESCRIPTION OF 4 MODEL SCENARIOS:
REFERENCES
Please note:
All references have been
listed however report titles have been altered to keep the name and location of
Landfill Site A anonymous. For further
information please contact The Centre for Waste and Pollution Research,
Geography Department, The University of Hull, Hull, UK, HU6 7RZ or telephone (44)1482465-371.
Booth C.J., Bethany C.P. (1989) Infiltration, soil moisture
and related measurements at a landfill with a fractured cover, Illinois.
Journal of Hydrology, 108, 175-188.
Cherry J.A., Gilham R.W., Anderson E.G, Johnson P.E. (1983a)
Migration of Contaminants in Groundwater at a Landfill: A Case Study, 1.
Groundwater flow and Plume Delineation. Department of Earth Sciences,
University of Waterloo, Canada. Elsevier Science Publishers B.V.
Haitjema H.M. (October, 1991) Groundwater hydraulics
considerations regarding landfills. Water Resources Bulletin, AWRA, 27(5).
Hanor J.S. (November 1993) Effective Hydraulic Conductivity
of Fractured Clay Beds at a Hazardous Waste Landfill, Louisiana Gulf Coast.
Water Resources Research, 29(11), 3691-3698.
Johansen O.J and Carlson D.A. (1976) Characterisation of
Sanitary Landfill Leachates. Water Research Journal, 10(11), 1129-1134.
Kjeldsen P. (June 1992) Groundwater pollution source
characterisation of an old landfill. Journal of Hydrology, 142, 349-371.
Lo Irene M-C. Liljestrand H. (1996) Laboratory conducted
sorption and hydraulic conductivity tests: evaluation of modified-clay
materials. Waste Management & Research 14, 297-310.
Pickens J.F., Lennox W.C. (April 1976) Numerical simulation
of waste movement in steady groundwater flow systems. Water Resources Research,
12(2).
Qasim S.R. and Burchinal J.C.(1970) Leaching from Simulated
Landfills. J. Water Pollution Control Federation, 43(3), 371-379.
Raveh A.. and Avnimelech.(1979). Leaching and Pollutants
from Sanitary Landfill Models. J. Water Pollution Control Federation, 51(11),
2705-2716.
Sulfida J.M., Gerba C.P, Ham R.K., Palisano A.C., Rathje
W.L., Robinson J.A.(1992) A Multidisciplinary Investigation: The World's
Largest Landfill. Environmental Science Technology, 26(8).
Sykes J.F., Soyupak S., Farquhar G.J.(February 1982a)
Modelling of leachate organic migration and attenuation in groundwater below
sanitary landfills. Water Resources Research, 18(1), 135-145.
Sykes J.F., Pahwa S.B., Lantz R.B., Ward D.S.(December
1982b) Numerical simulation of flow and contaminant migration at an extensively
monitored landfill. Water Resources Research, 18(6), 1687-1704.
Books
Domenico P.A., Schwartz F.W. Physical and Chemical
Hydro-geology. New York, John Wiley & Sons, 1990.
Department of Environment. Waste Mangement Paper 26B.
Landfill Design, Construction and Operational Practise. HSMO, London, 1995
Entec Consultants. (April 1996) Landfill Site A: Surface
Water Quality and Environmental Risk Assessment. UK.
CL Associates. (1995) Design Philosophy of Leachate
Management at Landfil Site A
CL Associates. (1995) Ground Investigation for Cells 1 and 2
of the proposed Extension to Landfill Site A
CL Associates. (1995) Specification
for Slurry Trench Cut Off Wall and Associated Site Engineering Works.
Smith C J. (1992) Landfill Site A- Extension Hydrological
Site Investigation Volumes 1,2 and 3.
Smith F W. (1993) Landfill Site A Extension Hydro-geological
Monitoring, March and April 1993 and Annual Summary.
Geography Department, The University of Hull
Dr. Andrew Ainsworth
Environmental Microbiologist
Dr. Richard Ankrett
Environmental Waste Mangement Monitoring Specialist
Richard Hanley
Landfill Hydrologist