Experimental Set-up
CHAPTER
4
EXPERIMENTAL SET-UP
4.1Experimental Cell Design and Construction
4.2Instrumentation andMonitoring Program
4.3Leachate Recirculation Strategy
The engineering design, construction, supervision and instrumentation of the full-scale experimental cell constituted a significant part of this thesis, which laid down the required groundwork for the subsequent investigations. This chapter provides a detailed description of the experimental cell including its instrumentation and the related monitoring program. It also outlines and justifies the strategy of leachate recirculation employed in the test.
4.1 EXPERIMENTAL CELL DESIGN AND CONSTRUCTION
4.1.1General
The experimental cell is located in Browning-Ferris Industries Inc. (BFI) ’s Lyndhurst Sanitary Landfill site which is about 35km south-east of Melbourne city centre as shown in the location plan (Figure 4.1).
The existing pit used for landfilling, in common with many landfill sites in the same south-eastern sand belt region of Melbourne, was created by previous sand mining operations. The geology comprises a sequence of Tertiary age sands and clays of 15 to 35m depth underlain by granite rock.
The natural groundwater level is shallow at about 6m below original ground level, which is at approximately 12m AHD (Australian Height Datum). The regional hydraulic gradient is falling towards the west.
The site is at a latitude of 38.02o S. Historic climatic data from a meteorological station located 15km north of the landfill reveal that the mean annual Class-A pan evaporation (1227 mm) exceeds the mean annual rainfall (874 mm).
The cell commissioned for the full-scale experiment is Cell 3 at the north-western corner of the site as shown in the aerial photograph (Plate 4.1 – Appendix A). Table 4.1 summarises the progress of the experiment up to December 1997.
Table 4.1 - Progress of Experiment
4.1.2Size of Experimental Cell
The cell covers a footprint area of approximately 180m x 75m (about 1.4 hectares) as shown in the as-constructed survey plan (Figure 4.2). This cell size is regarded as typical compared with other operational cells in the same landfill.
The final capping level rises from +15m AHD at the north-western corner up to +20m AHD at the south-eastern corner as the landscape slopes up south-easterly (Figure 4.2). The surface of the base liner is at +4m AHD. The thickness of fill thus varies from 10m to 15m (excluding the 1m final capping).
Based on survey data, the as-constructed volume of the experimental cell is 180,400 m3. Total tonnage of MSW as recorded by weighbridge is 100,800 tonnes. The filling of the cell took about two years to complete (Table 4.1).
In plan, the cell is divided into two sections of roughly equal areas. The western half has been designated as the control section for dry landfilling and the eastern half as the test section for leachate recirculation (Figure 4.2).
Plate 4.2 (Appendix A) shows a general view of the completed experimental cell taken from the north-western corner.
4.1.3Waste Composition
As the composition of waste is one of the most vital factors that influence biodegradation and hence the enhancement strategy, it is important to be able to quantify and qualify the types of waste being investigated in the experiment. This information would also be essential to allow any findings to be cross-referenced with other similar studies.
Although the Lyndhurst Landfill is also licensed to accept prescribed wastes (as defined in SEPP 1991) and contaminated soil, for the purpose of this bioreactor experiment, only domestic garbage and non-hazardous/ non-toxic waste from the commercial/ industrial waste stream were used in the experimental cell.
A record of all the wastes placed in the experimental cell was kept according to their waste streams. It is shown graphically in Figure 4.3. The ratio of domestic to commercial/ industrial waste stream is 1:1.6. The filling thus constitutes a higher proportion of commercial/ industrial waste as this forms the bulk of the waste source of Lyndhurst Landfill.
Figure 4.3 – Record of Waste Disposed According to Waste Streams
In terms of composition, both domestic waste and industrial/ commercial waste streams from metropolitan Melbourne have been studied recently (Waste Management Council, 1995). Their compositions according to the survey are listed in Table 4.2. These figures should provide a close representation of the wastes disposed in the experimental cell. After being weighted according to the above 1:1.6 ratio, the overall waste composition was estimated. It is shown in Figure 4.4. Note that the percentages are expressed in terms of wet mass.
Table 4.2 - Compositions of Waste from Metropolitan Melbourne (WMC, 1995)
Components / Domestic Waste (%) / Commercial/ Industrial Waste (%)Residual / 20% / 10%
Plastics / 6% / 12%
Garden / 16% / 5%
Kitchen/Food / 25% / 17%
Paper / 19% / 22%
Metals / 3% / 5%
Textiles / 2% / 10%
Inert / 3% / 4%
Glass / 6% / -
Domestic / - / 2%
Timber/Wood / - / 11%
Others / - / 2%
Total / 100% / 100%
Figure 4.4Figure 4.5
Additional waste composition information was also determined by collecting continuous waste samples from seven augered holes immediately after final capping. These holes were drilled to install access tubes for subsequent moisture monitoring (refer Section 4.2.1 for locations and details). The samples were dried to determine their gravimetric moisture (refer Section 4.1.4) prior to sorting. The composition as sorted is presented in Figure 4.5. In contrast to Figure 4.4, it is expressed in terms of dry mass.
It is impossible to make a direct comparison between the compositions in Figure 4.4 and Figure 4.5, as the moisture contents of individual components are not available. Nevertheless, by assuming a reasonable moisture content for each component, the two compositions tend to agree reasonably well. The dry mass composition is always preferred as it is independent of moisture content.
As the control and test sections were filled up simultaneously and the ratio of domestic waste to commercial/ industrial waste was maintained fairly consistently (Figure 4.3), the composition of waste in the experimental cell (in macroscopic scale) can be considered to be reasonably uniform. For the same reason, the wastes in the control and test sections can be treated as identical, at least within the context of this experiment.
4.1.4Waste Moisture Content
The variations of moisture content (immediately after final capping) with depth for the seven sampling holes mentioned above are plotted in Figure 4.6. The moisture content range and distribution of all the samples are plotted in Figure 4.7. The volumetric values in the figures were calculated based on a bulk density of 0.83 tonne/ m3 (refer Section 4.1.6).
The results of the moisture content analysis are summarised in Table 4.3. Thus, for the purpose of the water balance analysis in Chapter 6, the mean value of 27% (volumetric) is taken to be the overall as-capped moisture content of the waste.
Figure 4.6 – Variation of Moisture Content with Depth in the Seven Sampling Holes:
(a) By Dry Mass; (b) By Wet Mass; (c) By Volume
Figure 4.7 – Moisture Content Range and Distribution of MSW:
(a) By Dry Mass; (b) By Wet Mass; (c) By Volume
Table 4.3 - Results of Waste Moisture Content Analysis
Moisture Content (%)By dry mass / By wet mass / By volume
Maximum / 173 / 63 / 53
Minimum / 15 / 13 / 11
Mean / 55 / 32 / 27
Std. Deviation / 38 / 13 / 11
4.1.5Daily/ Interim Covers
Due to licensing requirements as well as operational needs to control litter, birds and odour during filling, daily cover was used in a manner similar to other operational cells.
The licence requires a 150mm layer of earth material as daily cover during waste disposal. In addition, each completed vertical lift (of 2m) should be covered by an interim cover of 300mm earth material. This requirement was applied to the experimental cell.
In common with other sand-pit landfills in the region, semi-dry to dry slimes were used for daily and interim covers in the experimental cell. The slimes were generally a clayey sandy silt material left behind from previous sand washing (Yuen and Styles, 1995). They spread reasonably well at “spadable” dryness. While still moist, they exhibit low permeability to both gas and odour. As they dry out, they crack to form agglomerates. This would allow a reasonable permeability, which is desirable for moisture movement induced by later leachate recirculation.
To reduce the barrier effects of the daily and interim covers for later recirculation, permeability was improved by stripping and mixing the earth material with waste before placing the next lift.
Figure 4.8 provides a record of all cover material (excluding final capping) placed in the experimental cell. The cumulative waste volume is also plotted. From the figure, it can be deduced that the cover material occupies about 15% of the total waste volume. This figure compares well with the amount as stipulated by the licence requirement which is 17% as calculated based on the above configuration.
Figure 4.8 – Record of Cover Material against Total waste Volume
4.1.6Density and Porosity of Waste
The waste was compacted in vertical layers by a Caterpillar 826C landfill compactor with an operating weight of 32 tonnes. It is the same compaction method employed in other operational cells.
Table 4.4 lists all the weight and volume components and the procedure used to calculate the in-situ density of waste. The bulk density, accounting also for the daily/interim earth covers, is 0.83 tonne/m3.
Based on this density and the bulk moisture content (55% by dry mass) determined in Section 4.1.4, the dry density and porosity of the MSW was calculated, which are 0.54 tonne/m3 and 0.55 respectively. The literature suggests a porosity range between 0.5 to 0.6 (e.g. Korfiatis et al., 1984; Oweis et al., 1990; Zeiss and Major, 1992). In this case 0.55 reflects that the MSW in the experimental cell is reasonably well-compacted.
Table 4.4 - Calculation of in-situ waste density
Item / Description / Volume(m3) / Weight
(tonne) / Density
(tonne/m3)
a / Volume between top of liner and top of capping / 180,400
b / Volume of 1m thick final cap / 13,500
c / Volume of daily/interim covers / 24,750
d / Volume of leachate collection drainage layer / 4,160
e / Net volume of MSW ( = a - b - c - d ) / 137,950
f / Total MSW weight as recorded by weighbridge / 100,820
g / Total weight of daily interim covers as recorded / 34,650
h / As placed in-situ density of MSW ( = f / e)
(excluding daily/interim covers, drainage & cap) / 0.73
i / As placed in-situ density of combined MSW & cover material (= [f + g] / [a-b-d] ) / 0.83
4.1.7Cell Containment System
In common with all other operational cells, the experimental cell is lined with a 1m thick side and base clay liner. The compacted clay has a specified hydraulic conductivity of less than 10-9 m/s. The construction of the liner was subject to a quality assurance program as required by the licence.
Figure 4.9 – Details of Final Capping
Immediately upon completion of filling, a final capping was laid, the details of which are shown in Figure 4.9. The capping falls gently on a 1 on 7.5 gradient towards its north-western corner (refer Figure 4.2).
4.1.8Leachate Collection System
A more efficient leachate drainage system different from other operational cells is used. It aims to provide a better control of the hydraulic head on the base liner during leachate recirculation. The system comprises a 300mm thick drainage layer of 20/40mm basalt gravel immediately above the liner, with 90mm diameter slotted collector pipes installed at 15m centres. These then drain to a 150mm diameter header pipe falling at a 0.16% gradient into a leachate collection sump (Figure 4.10).
To enable both leachate quantity and quality from the control and test sections to be monitored separately, each of the two sections has its own collection system. The two systems are isolated by a compacted clay bund wall constructed on top of the base liner as shown in Figure 4.10.
Figure 4.10 – Schematic Plan showing Leachate Collection System
4.1.9Gas Extraction System
An active extraction system employing a small suction is used to collect gas. To enable both flow rate and composition in the control and test sections to be monitored independently, two isolated gas fields (Figure 4.11) are employed. Each field comprises nine collection wells spaced approximately 25m apart connecting to a manifold station (Plate 4.3 in Appendix A). As the amount of gas that can be extracted from each well may vary, each well has a separate control valve for vacuum adjustment at the manifold station.
The details of the gas wells in the control section are shown in Figure 4.12. For the test section, the wells are designed also for leachate injection (see also Section 4.1.10). Construction details are shown in Figure 4.13.
Further discussion on the gas collection system is provided in Section 8.2.
Figure 4.11 – Schematic Plan Showing Gas Extraction Wells and Combined Leachate injection/ Gas Extraction Wells
Figure 4.12 – Details of Gas Extraction Wells
Figure 4.13 – Details of Combined Gas Extraction/
Leachate Injection Well
4.1.10Leachate Recirculation System
A wide range of recirculation options was considered. They included surface irrigation, surface ponding, buried drip irrigation tubing, in-ground piping grid, sub-surface infiltration trench/field, and vertical recharge well.
With potential problems such as odour control and wind-blown misting, surface application of leachate by spray irrigation or ponding was ruled out at an early stage. The use of buried drip irrigation tubing or an in-ground piping grid was also excluded. This decision was based on previous trials which indicated that there are numerous difficulties in maintaining the connection of the system as the landfill subsides.
A combination of sub-surface horizontal infiltration trenches and deep vertical injection wells was finally selected for this experiment. Both devices were constructed after final capping to avoid interruption to waste filling. The integrated system is shown schematically in Figure 4.14. The design, including the sizing and spacing of wells and trenches, was based on a numerical model simulation as discussed later in Section 7.2.1.
The details of a sub-surface horizontal infiltration trench are shown in Figure 4.15. There are eight infiltration trenches (RT1 to RT8) strategically placed between each pair of wells (Figure 4.14).
As described previously in Section 4.9 (Figure 4.13), the nine leachate injection wells (RW1 to RW9 in Figure 4.14) also serve to extract gas in the test section.
Leachate is pumped from the collection sump (Plate 4.4 in Appendix A) into three storage/header tanks (Plate 4.5) with a total capacity of 27,000 litres. From there the leachate feeds the wells and trenches by gravity via a system of pipework and valves (Plate 4.6). The system has been designed to allow flexibility to inject either an individual well/trench or a group of selected wells/trenches.
Figure 4.14 – Integrated Leachate Recirculation System
Figure 4.15 – Details of Sub-Surface Horizontal Infiltration Trench
4.2INSTRUMENTATION AND MONITORING PROGRAM
The design of the right instrumentation is crucial to enable the collection of reliable data for later analysis. The following factors and constraints have been taken into consideration in the design: costs (both capital and running costs), compatibility with the landfill environment, reliability, and simplicity.
Table 4.5 lists all the items that are being monitored in the study with the corresponding method/ instrumentation employed. Figure 4.16 shows the location of instrumentation schematically in plan. It can be divided into four monitoring groups: Control A, Control B, Test A and Test B.
Table 4.5 - Instrumentation/ Method Employed To Collect Data
Items Required Monitoring / Instrumentation/ MethodWaste Moisture Distribution Profile / Using neutron probe to measure moisture changes via in-situ access tubes
Climatic Data / Automatic weather station
Surface Runoff / Collection by surface channels & measurement by flume with water level auto-logger
Landfill Settlement / Level survey on settlement plates
Waste Temperature / Stainless steel sheathed thermocouples
Leachate Level / Measure leachate levels in the sumps and open wells by a water level sensor
Leachate Volume / Combining the use of an ultrasonic flowmeter and tank measurement
Leachate Quality / Collect and test leachate samples from both leachate sumps and open wells
Landfill Gas Composition / Portable non-dispersive infra-red absorption landfill gas analyser.
Portable gas chromatograph (GC)
Landfill Gas Flow Rate / Pre-calibrated orifice plate
Groundwater Quality / Collect and analyse groundwater samples from adjacent monitoring bores
Figure 4.16 – Location Plan Showing Instrumentation of Experimental Cell
4.2.1In-situ Municipal Solid Waste Moisture Monitoring
Based on a parallel investigation as described later in Chapter 5, the combined use of a neutron probe and in-situ access tubes has been identified as a practicable tool for monitoring moisture of in-situ MSW. Seven in-situ access tubes each of 12m long have been installed as shown in Figure 4.17. They are used to monitor: seasonal moisture change in the control section (AC1 and AC2); moisture change adjacent to a recirculation well (AT1 to AT3); and moisture change adjacent to a trench in the test section (AT4 and AT5). The installation details of the access tubes are described in Section 5.5.1.
Figure 4.17 – In-situ Access Tubes for Monitoring Moisture Changes
4.2.2Climatic Data
An automatic weather station is installed on top of the experimental cell (Figure 4.16 and Plate 4.7). It is equipped with a data logger and the following sensors (with measurement height and instrument accuracy in bracket):
- Rain gauge (0.3m; 0.2mm tipping bucket)
- Air temperature sensor (1.4m; +0.2o )
- Relative humidity sensor (1.4m; +5%)
- Anemometer (2.0m; +0.1 km/hr)
- Solar radiation sensor (2.0m; +0.1 kJ/m2)
4.2.3Surface Runoff
All runoff from the experimental cell catchment is collected by two surface channels running along the north and north-western edges as show in Figure 4.16. Through two catch pits, all flow is then diverted to a single main channel. A flume equipped with two water level probes (one as backup) and a real-time logger is installed in-line with the main channel to measure flow rate (Plate 4.8).
The flume belongs to the RBC (Replogle, Bos and Clemmens) family of long throated flumes (Bos et al., 1984) which are designed for measurement of flow in open channels with minimum head loss and reasonable sediment and debris passage without affecting performance. They allow accurate measurement of head loss at the sill, ensure that critical flow occurs in the throat, and give an accuracy of about 2% over the range of measured discharge. They require only one measurement of flow depth at a point upstream of the sill, which simplifies the operation compared to other types of flume. Their full hydraulic design details are given by Bos et al. (1984).