|
Background and Objectives:
There has been a dramatic increase in hog production in North America
in recent years and the fastest growth has occurred within Manitoba. In the last five
years, the industry has experienced 100% growth in the Province, and this trend of rapid
growth is expected to continue. In 1997, over three million hogs were raised in the
Province and sold for an estimated $1.25 billion (Source: Manitoba Pork). Typically,
earthen manure storages have been preferred by the industry for the storage of production
wastes because of the relative low capital cost. In 1998, the Livestock Manure and
Mortalities Management Regulation (Manitoba Regulation 42/98) was added to Manitoba’s
Environment Act in part to establish design criteria for animal manure storage structures.
The regulation recognizes the potential risk that the waste contained in these structures
poses to surface and ground water supplies, and aims to protect these supplies by setting
out construction and siting requirements. However, there is currently a lack of case
studies regarding the actual effects of earthen manure storages on underlying groundwater
supplies. Therefore, it is necessary to investigate current sites so that sound design
criteria, based on documented case studies and experience, can be established for the
future.
Research at the University of Manitoba has investigated six earthen
animal manure storage sites to date. This work is ongoing and is being conducted with the
financial assistance of the Agri-Food Research and Development Initiative, the Manitoba
Livestock and Manure Management Initiative, Inc. and the Natural Sciences and Engineering
Research Council. Manitoba Conservation, Water Resources Branch, has also provided
valuable assistance in this work. The investigation has included field and laboratory
tests, along with complementary mathematical modeling. The field component included the
removal of continuous sample cores from the base of the storages, and the characterization
of local geology and groundwater conditions. The laboratory component included the
geological classification of sample core materials, and the determination of the
concentration of waste contaminants at intervals along the continuous sample core.
Mathematical modeling was conducted so that contaminant seepage from the storages could be
predicted at times in the future. Chloride ion was used as the contaminant of interest for
this study. This component allows for the examination of potential groundwater quality
impacts at the sites, at various points in the future.
Procedure and Project Activities:
Location, Construction and Age of Sites
Six sites were examined in three distinct geological settings in
southern Manitoba. These settings included that of thick clay, till, or till with layered
sand and gravel overburden materials. Overburden material refers to any loose material
overlying bedrock. All of the earthen storages examined were constructed in a cut-and-fill
manner, meaning that the cell(s) of the storage were excavated in natural overburden
materials and filled with animal wastes without the intentional installation of a
recompacted or synthetic liner. At the time that the sample cores were removed from the
storage bases the storages ranged in age from 4 to 20 years. Figure 1 shows the
approximate location of the six sites examined in this study.
Figure 1: Approximate Site Locations
Site Geology and Hydrogeology
Subsurface conditions were determined for each site, since the type and
amount of material separating the storages from underlying aquifers has a direct bearing
on the extent of vertical contaminant seepage expected. This was accomplished by using
well driller’s reports, geological maps and examination of the removed sample cores.
At the Interlake 1 site, approximately 33 metres of till overburden
material separates the base of the storage from the underlying carbonate bedrock. This
carbonate bedrock forms an aquifer providing a source of both domestic and livestock water
at the site, and in the surrounding area. At the Sesco 1 site, approximately 23 metres of
clay, and till with interbedded sand and gravel overburden material separates the base of
the storage from the underlying carbonate bedrock. Similar to the Interlake 1 site, the
underlying carbonate bedrock forms the main aquifer in the Sesco 1 area, and is used as a
source of both domestic and livestock water at the site, and in the surrounding area.
Approximately 7 metres of clay and till overlies a sand and gravel aquifer at the Sesco 2
site. The sand and gravel aquifer at Sesco 2 was used to meet the operation’s
livestock requirements at the time of sample core removal. At the Southeast 1 site, 44
metres of layered till and sand with gravel separates the storage base from an underlying
sand and gravel aquifer. Sand and gravel aquifers satisfy both domestic and livestock
water requirements in the area surrounding Southeast 1, and at the site itself. Similar to
the Southeast 1 site, 15 metres of interbedded till with sand and gravel separates the
underlying sand and gravel aquifer from the storage base at the Southwest 1 site. The
storage base at this site was excavated directly into a thin sand and gravel unit. Sand
and gravel aquifers satisfy both domestic and livestock water requirements at Southwest 1,
and in the surrounding area. The TS earthen manure storage is underlain by a clay unit.
Below this clay is a layered till with sand and gravel sequence. Approximately 23 metres
of overburden material separates the underlying carbonate bedrock from the base of the
storage at the TS site. The carbonate bedrock forms the aquifer in the area and meets the
water requirements in the area surrounding the TS operation, and at the site.
Sampling Methods
Continuous sample cores were removed from the base of each of the six
earthen manure storages, either when sufficient ice had formed to support the weight of
the track mounted drill rig or when storage was emptied permitting access. A sample core
is simply a sample of geological materials obtained by drilling. A continuous sample core
refers to a sample core that is recovered continuously to a certain depth, meaning there
are no missing intervals. The continuous sample cores removed from each site were
sectioned into lengths varying from 0.05 m to 0.3 metres. These sections were sent to
local testing laboratories where analyses were conducted to determine the concentration
with depth of various ions known to have originated from the wastes contained in the
storage. These ions include chloride, ammonium, nitrate, nitrite, and phosphate. This work
examined only chloride concentrations observed in each of the sample cores as this is a
commonly used "conservative" tracer, meaning that it does not react or decay
within the soil column nor is it adsorbed onto mineral surfaces. Chloride concentrations
in the storage are often more than 10 times greater than chloride ion concentrations
occurring naturally in the pore water contained within the underlying soils, referred to
as background concentrations. The amount and extent of seepage is demonstrated by chloride
concentrations exceeding what is expected as a background concentration in the soil, at a
given depth. Based on the conservative nature of chloride, the seepage of this ion from
the storage provides the "worst-case" scenario of contaminant migration.
Results and Discussion:
The Interlake 1 storage was in operation for 4.7 years at the time of
sample core removal and laboratory testing. During these 4.7 years of storage operation,
there was approximately 1.4 metres of vertical chloride ion transport downward from the
storage base. Recall that there is an estimated 33 metres of till overburden material
separating the storage base from the underlying aquifer.
In the 15 years of storage operation at the Sesco 1 site, there has
been approximately 5 metres of chloride ion transport vertically downward from the
storage. Seepage of chloride at the time of sample core removal had remained confined to
the upper clay unit, and there remains over 18 metres of overburden material separating
the underlying carbonate bedrock aquifer from the observed extent of chloride seepage.
At the Sesco 2 site, there has been approximately 1.6 metres of
downward chloride ion transport into underlying overburden materials, below the storage
base, in 9 years of storage operation. This transport has remained confined to the
underlying clay unit below the storage, and there remains over 6 metres of clay and till
separating the observed extent of chloride seepage from the underlying sand and gravel
aquifer.
The Southeast 1 storage was in operation for approximately 3.7 years at
the time of sample core removal and laboratory testing. During this time interval,
approximately 2 metres of vertical chloride seepage occurred into underlying geological
materials. This seepage was confined to the upper till unit, and there is approximately 42
metres of layered till and sand with gravel separating this observed extent of chloride
seepage from the underlying sand and gravel aquifer.
At the Southwest 1 site, there has been about 3 metres of downward
chloride ion transport in 18 years of storage operation. Approximately 12 metres of till
separated the observed extent of seepage from the underlying sand and gravel aquifer at
the time of sample core removal and testing.
In the 20 years of storage operation at the TS site, approximately 6.5
metres of downward chloride ion transport had occurred at the time of sample core removal.
This transport was confined in the underlying clay unit, and over 15 metres of clay and
layered till and sand with gravel separated the top of the carbonate aquifer from the
observed extent of vertical chloride seepage.
Based on the results of the laboratory testing and field components, it
can be concluded that the extent of downward contamination transport is greater through
clay than through till in storages operating for a similar length of time. This is
contrary to what would be expected based on usual values of hydraulic conductivity for
Manitoba clay and till materials. Hydraulic conductivity can, in general, be thought of as
the ease with which water can pass through a layer of material. Clay generally has a lower
value of hydraulic conductivity than till, and as such it is expected that the water
containing the chloride ion will not be transported as quickly. These results suggest that
fractures, acting as conduits of fluid flow, exist in the clay units examined. This serves
to explain higher transport rates and fractures were observed in the clay units examined
in this study, and have been reported in other studies conducted in southern Manitoba.
Predicted Groundwater Quality Impacts
A mathematical model can be thought of as a simplified representation
of a real world system or process. The model used in this study was a one-dimensional
(downward transport only) mathematical model incorporating the processes of advection and
dispersion. Advection refers to the process of solute transport by the flow of
groundwater. Dispersion refers to the spreading of the contaminant plume due to mechanical
mixing and concentration differences. The solution to a particular mathematical model
depends on site-specific data used to solve the mathematical problem. A different solution
was obtained for each site, representing the theoretical transport occurring at that site,
by inputting site-specific data and conditions. In order to use a model to predict what
might happen in the future, there must be reasonable agreement between the chloride
concentrations observed at a particular depth at a known point in time, and those
predicted by the model at a particular depth at that same time.
Figure 2: Interlake 1 Observed and Model Predicted Chloride Ion Seepage
Profiles
Reasonable agreement was obtained for four of the six sites examined in
this study, namely the Interlake 1, Sesco 2, Southeast 1 and Southwest 1 sites. The most
promising results are obtained for the Interlake 1, Southeast 1 and Southwest 1 sites (see
Figures 2, 3 and 4, respectively), all of which are till or layered till with sand and
gravel sites. The diamonds represent the actual laboratory measured chloride
concentrations, while the solid line represents the model predicted chloride
concentrations along the sample core, at the time of sample core removal and laboratory
testing.
The model output provides an approximate contamination transport rate
for each of the sites. The bulk transport rate includes the effects of both advection and
dispersion. It includes the rate of transport with flow, as well as any decrease in this
rate due to dispersive effects. At the Interlake 1 site, this bulk transport rate is
predicted to be 0.19 metres per year by the model. At the Sesco 2 and Southeast 1 sites,
bulk transport is estimated to be 0.13 and 0.21 metres per year, respectively. Finally, at
the Southwest 1 site, bulk transport is estimated to be 0.13 metres per year.
Figure 3: Southeast 1 Observed and Model Predicted Chloride Ion Seepage
Profiles
Using bulk
transport rates, it is possible to estimate the time for elevated contaminant
concentrations to reach the underlying aquifers. At the Interlake 1 site, it is estimated
that elevated chloride concentrations can potentially reach the underlying aquifer in 170
years, while at the Sesco 2 sites, it is possible for elevated chloride concentrations to
reach the underlying aquifer in 60 years. At the Southeast 1 and Southwest 1 sites, it
could potentially take 210 years and 115 years, respectively, for elevated chloride
concentration to reach the underlying aquifers at these sites. It should be noted that
these are absolute worst-case scenario predictions due to the conservative nature of
chloride and the assumptions inherent in the model used.
Figure 4: Southwest 1 Observed and Model Predicted Chloride Ion Seepage
Profiles
In order to develop sound design criteria for earthen animal manure
storages, case study information regarding the effects of these structures and the
contained waste is required. Research is currently being conducted at the University of
Manitoba to obtain case study information and develop an understanding of the factors and
processes that influence the transport of contaminants from storages into underlying
groundwater supplies through the overburden materials typically encountered in southern
Manitoba.
Expected downward bulk transport rates have been established for clays
and tills and can be estimated as varying from 0.1 metres per year to 0.2 metres per year.
These transport rates are applicable to conservative contaminants only and provide a
worst-case transport rate. These transport rate estimations are subject to assumptions
inherent in the model used, as well as assumptions made regarding site conditions and soil
properties. These assumptions add a degree of uncertainty to the process. To reduce
uncertainty, a more extensive field and laboratory testing program is to be undertaken by
the University of Manitoba, which will include careful and routine monitoring of storage
and groundwater conditions, and periodic sample core removal at a number of sites. The
implementation of such a program requires the cooperation of producers, industry and
researchers alike, and the University is currently looking for additional sites where such
a program may be undertaken.
|
HomeContact
UsAbout
ManitobaDepartmentsLinksPrivacy 
