Waste Management 34 (2014) 2312–2320
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Case study of landfill leachate recirculation using small-diameter vertical wells Pradeep Jain a,c,1, Jae Hac Ko b,c,2, Dinesh Kumar d,c,3, Jon Powell a,c,1, Hwidong Kim e,c,4, Lizmarie Maldonado a,1, Timothy Townsend c,⇑, Debra R. Reinhart f,5 a
Innovative Waste Consulting Services, LLC, 6628 NW 9th Boulevard, Suite 3, Gainesville, FL 32605, USA School of Environment and Energy, Peking University Shenzhen Graduate School, Shenzhen 518055, China Department of Environmental Engineering Sciences, University of Florida, P.O. BOX 116450, Gainesville, FL 32611-6450, USA d Engineering Department, Municipal Corporation of Delhi, 9th Floor, Dr. SPM Civic Centre, Jawahar Lal Nehru Marg, New Delhi 110 002, India e Environmental Science and Engineering, Gannon University, 109 University Square, Erie, PA 16541-0001, USA f Civil and Environmental Engineering Department, University of Central Florida, P.O. BOX 162450, Orlando, FL 32816, USA b c
a r t i c l e
i n f o
Article history: Received 2 January 2014 Accepted 8 July 2014 Available online 24 August 2014 Keywords: Bioreactor landfill Leachate recirculation Municipal waste Vertical wells Well performance Injection pressure
a b s t r a c t A case study of landfill liquids addition using small diameter (5 cm) vertical wells is reported. More than 25,000 m3 of leachate was added via 134 vertical wells installed 3 m, 12 m, and 18 m deep over five years in a landfill in Florida, US. Liquids addition performance (flow rate per unit screen length per unit liquid head) ranged from 5.6 108 to 3.6 106 m3 s1 per m screen length per m liquid head. The estimated radial hydraulic conductivity ranged from 3.5 106 to 4.2 104 m s1. The extent of lateral moisture movement ranged from 8 to 10 m based on the responses of moisture sensors installed around vertical well clusters, and surface seeps were found to limit the achievable liquids addition rates, despite the use of concrete collars under a pressurized liquids addition scenario. The average moisture content before (51 samples) and after (272 samples) the recirculation experiments were 23% (wet weight basis) and 45% (wet weight basis), respectively, and biochemical methane potential measurements of excavated waste indicated significant (p < 0.025) decomposition. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Bioreactor landfills are designed and operated to optimize the municipal solid waste (MSW) stabilization process rather than to simply contain the wastes as prescribed by most regulations (Reinhart and Townsend, 1997; Reinhart et al., 2002). Addition of liquids is the most common approach in bioreactor operations; leachate is the most common liquid supply, but other moisture sources can be used. Subsurface moisture addition techniques are the more feasible approach compared to surface application at large landfills. In subsurface systems, the liquids are added to the
⇑ Corresponding author. Tel.: +1 352 392 0846; fax: +1 352 392 3076. E-mail addresses: [email protected] (P. Jain), [email protected] (J.H. Ko), [email protected] (D. Kumar), [email protected] (J. Powell), kim008@gannon. edu (H. Kim), [email protected] (L. Maldonado), ttown@ufl.edu (T. Townsend), [email protected] (D.R. Reinhart). 1 Tel.: +1 352 331 4828; fax: +1 352 331 4842. 2 Tel.: +86 0755 26033289; fax: +86 0755 26033226. 3 Tel.: +91 (11) 23225918. 4 Tel.: +1 814 871 7025. 5 Tel.: +1 407 823 2315; fax: +1 407 882 2819. http://dx.doi.org/10.1016/j.wasman.2014.07.005 0956-053X/Ó 2014 Elsevier Ltd. All rights reserved.
landfilled waste through devices such as trenches, wells, and galleries (Benson et al., 2007; Bareither et al., 2010; Barlaz et al., 2010), with pressure commonly applied to promote moisture distribution (pressurized addition is supplied by a pump or a standing head of liquid). Horizontal leachate injection trenches are common but must be constructed at multiple elevations during waste placement to achieve even moisture distribution throughout the waste. Vertical wells allow the operator to retrofit landfill areas that are already filled with waste and minimize the interference of liquids addition device construction with routine landfill operations by waiting until waste placement is complete. While some field experience and experimental results from buried horizontal systems have been reported, similar data for vertical well systems have not been reported, although the practice of using vertical wells for leachate recirculation has been described (Benson et al., 2007; Bareither et al., 2010; Barlaz et al., 2010). This paper provides a case study of liquids addition experience using small-diameter vertical wells at an operating MSW landfill in Florida (New River Regional Landfill [NRRL]). Results from various experiments at the site have been reported previously (e.g., estimating in-situ air permeability and hydraulic conductivity of
P. Jain et al. / Waste Management 34 (2014) 2312–2320
landfill waste, Jain et al., 2005, 2006; assessment of in-situ moisture sensors, Kumar et al., 2009; experience with air addition, Powell et al., 2006; Ko et al., 2013). In this study, 5 years of performance data and lessons learned from the operation of a vertical well liquids addition system are summarized. The paper reports on vertical well performance with respect to well depth, various leachate seep control techniques (e.g., installation of seep control collar, and limiting liquids head in wells), system operating issues, and an assessment of the degree of waste stabilization. The information presented is valuable for design, operation and monitoring of future liquids addition systems at landfills. 2. Material and methods 2.1. Site description The NRRL receives mixed residential, commercial, and light industrial waste at a rate of approximately 700 metric tons (MT) per day. Clayey–sandy soil mined on site was used as daily cover. Approximately 4 hectares of Cell 1 and part of Cell 2, containing approximately 550,000 MT of MSW, were retrofitted to operate as a bioreactor landfill, as shown in Fig. 1. The leachate collection system for these cells, which are not hydraulically separated, was constructed with a saw-tooth pattern, with a total of eight leachate
2313
collection pipes each connected to a dedicated sump. Only 6 pipes are shown in Fig. 1 and the remaining two pipes were located in the eastern section of Cell 2 not shown in Fig. 1. Leachate from these sumps gravity drained to a wet well where it is then pumped to one of two 1900-m3 capacity leachate storage basins (Fig. 1) – note that leachate from all other lined cells at the site was also pumped from different sumps to the storage basins. The maximum waste depth of the tested area was 22 m at the beginning of bioreactor landfill operation. Systems for moisture addition, air injection (air injection experiments in this site were reported by Powell (2005) and Powell et al. (2006)), gas extraction, as well as a range of instruments, were installed as part of research to examine several technical aspects of full-scale bioreactor landfill operation. More information on the bioreactor landfill and related infrastructure can be found elsewhere (Jain, 2005; Jain et al., 2005, 2006; Powell et al., 2006; Kumar et al., 2009; Jonnalagadda et al., 2010). Due to concerns of potential accumulation of more than 30 cm of leachate on the liner, the state permit specified a maximum daily recirculation volume of 218 m3 and 132 m3 for the first 250 days and after 250 days, respectively. The permit also required cessation of liquids addition if the daily leachate collection volume from any of the eight leachate collection pipes exceeded 24 m3. These limits were not exceeded at any point during the study period. The leachate inflow rate into each sump was measured using a 22.5° V-notch weir box and an ultrasonic meter (Ametek
Fig. 1. Bioreactor well field layout.
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Landfill Surface
18 m
12 m
Drexelbrook, Inc., Series 505-1320) mounted above the box. Each weir box was also equipped with a battery-operated pump for leachate sampling. The ultrasonic meters transmitted water level in weir boxes (4–20 mA signal) to a Campbell Scientific data logger (CR10X) installed at an on-site research trailer; instantaneous leachate generation rates for each of sump were recorded once every ten minutes. Magnetic meters (Unipulse 4411, ISCO, Inc.) tracked the total volume of leachate and groundwater recirculated, as well as the volume of leachate collected from the bioreactor.
6m
2314
While designing the vertical well system, it was hypothesized that most of the liquid would flow through the bottom portion of a well screened across the entire landfill depth due to greater water column head and potential preferential flow to the leachate collection system (LCS). Therefore, a cluster of three wells screened at different depths was designed instead of a single fully-screened well. Furthermore, it was postulated that liquid could be injected under pressure through these wells rather than by just maintaining a standing water column. A total of 134 5-cm diameter wells were installed in 2001 in 44 clusters of three wells (plus one cluster with two wells) in a 100 m by 90 m grid using a rig (CME 85) with a 10 cm open-flight auger. Each cluster had Schedule 40 polyvinyl chloride (PVC) wells installed at different depths: 6 m (the bottom 3 m was slotted pipe), 12 m (bottom 6 m slotted), and 18 m (bottom 6 m slotted). The 6-m, 12-m, and 18-m deep wells are also referred to as shallow, middle, and deep, respectively (see Fig. 2). A threaded 5-cm-diameter PVC pipe section was lowered as soon as the borehole was completed and the drill stem was removed – no permeable media was placed in the area surrounding the slotted pipe. The top 0.3–0.6-m section of the annulus was backfilled with hydrated bentonite pellets. A 10-cm PVC (Schedule 80) distribution manifold was used to convey liquids from the leachate storage basins to the well field. The wells were connected to the distribution manifold with a 2.5-cm flexible hose. The 4-ha recirculation study area was covered with a 1-mm linear low density polyethylene (LLDPE) textured geomembrane underlain by an approximately 1-m thick soil layer. Note that the portion of Cell 2 that was not part of the recirculation area was covered with a vegetated 12-in. thick (minimum) clayey soil and was not operated during the recirculation study period. Once the exposed geomembrane cover (EGC) was installed, gas monitoring wells, leachate wells, and monitoring ports (for a total of 300 penetrations) were extended above the landfill surface by cutting a hole in the EGC around the pipes. The gap between the pipe and the EGC was sealed using a prefabricated boot designed to accommodate differential settlement. A series of resistivity-based and time domain reflectometry sensors were installed between rows of the wells and at depths of 4.5, 9, and 15 m to monitor the in-situ moisture content before the EGC cap installation. An evaluation of these sensors and moisture monitoring results were reported by Kumar et al. (2009) and Jonnalagadda et al. (2010). Construction photographs of sensor packs are presented in Figs. S9 and S10 in the Supplementary Material. Also, type-T thermocouple wire (Omega, Inc., CT, USA) was fitted to the bottom end of each well and included in the moisture sensor packs to allow for frequent temperature measurement. Groundwater was used as a supplemental source of moisture if sufficient leachate was not available. 2.3. Initial operation of the landfill bioreactor to evaluate hydraulic performance The liquids addition experiments described herein took place over two years at 26 (2003 through 2005) well clusters (total of 78 wells) located in zone 1 as shown in Fig. 1; the wells in zone
3m
2.2. Vertical well field construction
Bottom Liner Fig. 2. Schematic of a vertical well cluster.
2 were also designed to add air to the waste and only limited liquids addition was practiced in this zone. Recirculation was accomplished by pumping leachate from the leachate storage basin to the recirculation cells via a single distribution manifold, and flow rate at each well was controlled using an in-line valve. Initially, recirculation occurred in all wells in a cluster at once during the operating hours of the facility (generally 8:00 AM to 5:00 PM). Subsequently, recirculation proceeded in 22 clusters (66 wells) starting with the 18-m well, followed by the 12-m well and the 6-m well; liquids were added in only one well at a time in each cluster. In the remaining four clusters, leachate was first added to the 6-m wells followed by the 12-m and 18-m deep. Liquid heads at the well bottom was measured using a submersible pressure transducer (GE Druck, Inc.) with a 4–20 mA output – some wells had dedicated pressure transducers that were connected to a data logger (Campbell Scientific CR10X) for continuous measurements (once per minute), while the liquid head in the remaining wells was measured manually once per day using a pressure transducer (GE Druck, Inc.) and a handheld readout device (UPS II, GE Druck, Inc.). Instantaneous flow rate was measured continuously in some wells with an ABB single-jet impeller meter with a digital output connected to the data logger, while instantaneous and cumulative liquids addition was measured at the remaining wells once per day with ABB single-jet impeller meters permanently connected to the vertical wells. The objective of the tests conducted was to estimate the achievable flow rates and the associated pressure for the wells installed at different depths. The liquids addition system was continuously operated except during times when the system was stopped to change the test locations, clean the inline filters and liquid distribution manifold, and during severe weather conditions.
2.4. Operation of the landfill bioreactor to evaluate impact of well radius and seep collar on hydraulic performance An additional series of experiments were conducted in 2007 with 10 newly-constructed vertical wells to assess the performance of a concrete pad in controlling seeps around the wells and to assess the impact of well radius on hydraulic performance of a vertical well. Two trenches (9.4 m long by 1.8 m wide by 0.6 m deep) were excavated and five vertical wells were installed using the same open flight auger procedure described previously.
2315
P. Jain et al. / Waste Management 34 (2014) 2312–2320
2.5. Waste sampling Waste samples were collected prior to liquids addition experiments (51 samples in 2001 during vertical well construction) and following liquids addition experiments (272 samples in 2007) using a truck-mounted power flight auger equipped with a 10cm diameter solid stem open flight auger. Samples in 2007 were collected across the well field in 3-m composite fractions at depths of 0–3 m, 4.5–7.6 m, and 9.1–13.7 m and were analyzed for moisture content and the Biochemical Methane Potential (BMP) assay. For the BMP assay, the collected samples were sieved and sorted. The BMP assay was conducted only for the biodegradable fractions (mostly paper), non-identifiable fractions, and fine fractions. More details on BMP assay sample collection, analysis and results for the site can be found elsewhere (Kim and Townsend, 2012). 3. Results 3.1. Liquids addition and leachate collection Fig. 3 presents the cumulative volume of liquid (leachate and groundwater) added over the duration of the study. A total of approximately 22,000 m3 of liquids were added during the initial set of experiments in 26 well clusters, while approximately 3000 m3 of leachate were added during the experiments with the 10 new vertical wells installed with seep collars. Approximately 60% of the total liquid volume (16,000 m3) was added in 2004 (day 371–731) primarily in Zone 1 of the bioreactor, using multiple well clusters operating continuously. No considerable impact on the overall leachate collection rate was observed as shown in Fig. 3. None of the 6 collection pipes shown on Fig. 1 registered an increase in leachate collection rates – Fig. 4 presents data from one of the two pipes that showed an increase; both of these pipes were located in the eastern section of Cell 2 (not shown in Fig. 1). It can be seen that the leachate collection rate increased approximately a year (approximately day 600 on Fig. 3) after the commencement of leachate recirculation (from 0.27 to 0.83 m3 per day). The leachate collection from these pipes represented a small fraction (5–10%) of the overall leachate collection rate measured in the area of Cell 1 and Cell 2 corresponding to the recirculation system. The limited impact of recirculation on leachate production rates likely reflects absorption of moisture by the landfilled waste and the low vertical hydraulic conductivity of the waste reported by
Cumulative Liquids Volume (m3)
30000
25000 Added
20000
15000
10000 Collected
5000
0 0
500
1000
1500
2000
Time (Days) (Day 1= 1/1/2003) Fig. 3. Liquids addition and leachate collection cumulative volume.
Leachate Collection rate (m3/day)
The wells were installed in two clusters, which each consisted of a 5-cm diameter well installed at 3 (bottom 1.5 m slotted), 6 (bottom 4.5 m slotted), 9 (bottom 7.5 m slotted), and 12 (bottom 10.5 m slotted) deep and a 7.5-cm diameter well installed 9 (bottom 7.5 m slotted) deep. The top 0.3 m section of the annulus of each well was backfilled with hydrated bentonite pellets. Ready-mixed concrete was poured to build a 15-cm thick concrete slab (referred here into as seep collar) covering the entire trench area. The excavated cover soil was backfilled and compacted to cover the trenches. Each PVC pipe was extended approximately 0.6 m above the landfill surface and connected to a valve, flow meter, and liquids distribution manifold – construction photos are provided in the supplementary material. Leachate addition was performed intermittently during landfill operating hours. A submersible transducer was installed at the bottom of each well following the initiation of the experiments, and once the liquid level in the well was above the landfill surface (based on transducer measurements), liquids addition was performed in only selected wells and the area was closely monitored for surface seeps.
1.0 0.8 0.6 0.4 0.2 0.0 0
200
400
600
800
Time (Days) (Day 1=1/1/2003) Fig. 4. Temporal trend of leachate generation rate in a leachate collection pipe that experienced an increase in leachate generation rate during the experimental period.
Jain et al. (2006), who reported a range at this site from 5.4 106 cm/s to 6.1 105 cm/s. The total amount of liquid added (25,000 m3) represents 16% of the moisture retention capacity of the waste estimated based on the initial moisture content (described further later) of 23% and an assumed waste field capacity of 40%. 3.2. Leachate addition per well screen at different depths The daily average liquids addition rate and pressure head, measured at the bottom of 6 m, 12 m, and 18 m deep wells at a single cluster location, are plotted as a function of time in Fig. 5. The pressure at the bottom of the well first increased rapidly followed by a more gradual increase as liquids addition progressed, which was attributed to the high pressure gradient required to add liquids at a given specific flux to a media of very low relative permeability; the relative permeability of the medium depth wells is low because of low waste saturation. The subsequent gradual increase in pressure for a constant flux over time was ascribed to the increase in flow path length as the moisture front moved away from the well. Other factors such as landfill gas pressure, clogging, and changes in waste hydraulic conductivity associated with waste decomposition and consolidation also may cause a reduction in flow rate for a given injection pressure or increase in liquid head in the well for a given flow rate. During all tests, liquids addition was attempted at a constant flow rate for the duration required to achieve steady-state conditions, defined as attainment of a steady pressure head for a constant flow rate for a well. The flow rate in many cases was reduced to minimize the chance of surface seeps if the liquid level in the well approached or exceeded the surface of the landfill.
2316
P. Jain et al. / Waste Management 34 (2014) 2312–2320 0.38
15
Pressure (m of w.c.) Flow rate (x10-4m3/s)
-4
3
0.25 9
Flow rate (x10 m /s)
Pressure (m w.c.)
12
6 0.13 3
0
0.00 10
20
30
40
50
60
70
Time (Days)
(a) 15
0.50
Pressure (m of w.c.) Flow rate (x10-4m3/s)
3
9
-4
Pressure (m w.c.)
0.38
Flow rate (x10 m /s)
12
0.25 6 0.13 3
0
0.00 10
20
30
40
50
60
70
Time (Days)
(b) 0.50
15
3
Pressure (m of w.c.) Flow rate (x10-4m3/s)
9
-4
Pressure (m w.c.)
0.38
Flow rate (x10 m /s)
12
0.25 6 0.13 3
0
0.00 10
20
30
40
50
60
70
80
Time (Days)
(c) Fig. 5. Typical temporal variation of head measured at the bottom of a well in response to moisture addition for (a) shallow well (Day 1 = 08/17/2004), (b) middle well (Day 1 = 06/02/2004), and (c) deep well of a well cluster (Day 1 = 03/08/2004). The system was turned off for maintenance or severe weather from day 17 to day 27, from day 9 to day 19, and from day 33 to day 36 for the shallow, middle, and deep wells, respectively.
The temporal variation of flow and pressure for other wells is presented elsewhere (Jain, 2005). Based on the dimensionless variables and design charts developed by Jain et al. (2010) for the design of a vertical well system, the liquids volume added to a well was analyzed as the number of well volumes and the flow rate was normalized with the liquid head in the well bottom and the screen length to generalize the results. The number of well volumes is the ratio of the actual volume added to the screened well section that is distributing flow to the surrounding waste mass. The normalized flow rate (herein referred to as fluid conductance) closely resembles specific capacity,
which is used to assess the performance of groundwater pumping or injection from or into the aquifer using vertical wells; specific capacity is defined as water withdrawal or injection rate per unit head drawdown or buildup, respectively (Vecchioli and Ku, 1972). Fig. 6 shows the variation of fluid conductance as a function of liquids volume added and number of well volumes. The top, middle, and bottom lines of the boxes represent 75th percentile, median and 25th percentile of the data, respectively. The top and the bottom whiskers present the 90th and 10th percentile, respectively, while the dots represents data outliers beyond the 90th and 10th percentile. The number inside each box corresponds to
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Number of Well Volumes
-4
10
16,181
24,272
32,362
40,453
48,544
56,634
-5
10
-
-
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
8,091
30 27 29
-6
10
5 24
7
11 20
4
10
13
2
x¼ 1
-7
10
1
(a) Deep Wells
-8
10
100
200
300
400
500
600
700
Cumulative Volume (m³)
Number of Well Volumes
-4
10
8,091
12,136
16,181
20,227
24,272
28,317
-5
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
4,045
-
10
26
-
25
-6
6
23
10
6
13
3 1
-7
10
-8
10
(b) Middle Wells -9
10
0
50
100
150
200
250
300
350
Cumulative Volume (m³)
Number of Well Volumes
-3
10
8,065
16,129
24,194
32,258
40,323
48,387
56,452
-4
10
-
-
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
volumes into wells with greater fluid conductance. The number of wells decreased with liquid volume added in Fig. 6 as the larger volumes were only achieved in a small number of good performing wells. Based on the dimensionless variables developed by Jain et al. (2010), the fluid conductance is also equivalent to xKr, where x is defined as follows:
26
0
-5
23
10
18 14
10 7 1 6
1
-6
10
-7
10
(c) Shallow Wells -8
10
0
50
2317
100
150
200
250
300
350
Cumulative Volume (m³) Fig. 6. Fluid conductance as a function of liquids volume added for (a) deep wells, (b) middle wells, and (c) shallow wells. The top, middle, and bottom of each box represents the 75th, median, and 25th percentile of the data, the upper and bottom bars represent the 90th and 10th percentile, and the dots represent data outliers beyond the 90th and 10th percentile. The number inside of each box corresponds to the number of wells for that range.
the number of wells. The median fluid conductance ranged from 5.6 108 to 3.6 106 m3 s1 per m screen length per m liquid head at the well bottom for all wells and showed only slight variation with liquids volume. An increasing trend in fluid conductance with liquids volume is reflective of an ability to add larger liquid
pq Q p w2 K r ; with q ¼ 2v pId ¼ I ; g ¼ 2 pId g prw K z w rw K z
where Qv is the liquids addition rate to a well (L3T1), rw is the well radius (L), w is the screen length (L), pI is the injection pressure head at the bottom of the well (L), and Kz and Kr are the vertical and radial hydraulic conductivity, respectively (LT1). The variable g is expected to be more than 104 for the system evaluated due to a large w/rw ratio and waste anisotropy. Based on Design Chart 1 presented by Jain et al. (2010), at steady-state conditions x ranges from 0.86 (for g = 107) to 1.39 (for g = 104) for shallow and middle wells and from 0.92 (for g = 106) to 1.48 (for g = 104) for deep wells for the operating conditions (i.e., ratio of the liquid head at the well bottom to screen length ranges from 2 to 3) used in this study. The radial hydraulic conductivity was estimated for median fluid conductance values for liquid addition amount greater than 16,000 well volumes, which, for a waste porosity of 0.5, is equivalent to 32,000 pore volumes as defined by Jain et al. (2010); a pore volume of 32,000 is greater than the volume needed to achieve steady state for g = 104. The range of median radial hydraulic conductivity of the waste layers corresponding to the screened sections for 6-m, 12-m, and 18-m deep wells were 1.3 104 to 4.2 104, 2 105 to 9.9 105, and 3.5 106 to 1.2 104 cm/s, respectively. The estimated radial hydraulic conductivity in the shallow layer was greater than the middle layer and the deep layer, which is consistent with the trend reported by Jain et al. (2005) and Jain et al. (2006). Based on the number of well volumes added and the estimate of the number of pore volumes needed to achieve steady state (from Jain et al., 2010), the radial hydraulic conductivity is overestimated as many of the wells did not reach steady state. The range is in good agreement with the radial hydraulic conductivity estimated based on air injection tests conducted at the site previously (Jain et al., 2005) but is one to two orders of magnitude less than that estimated using the borehole permeameter tests in Jain et al. (2006). Although the radial hydraulic conductivity influences the flow rate achievable through a vertical well under a given liquid head, the hydraulic conductivity in the vertical direction would be expected to dictate the flow rate at steady state. The values reported by Jain et al. (2006), which were based on use of the borehole permeameter equation for isotropic media, therefore, probably correspond to hydraulic conductivity in vertical direction. The hydraulic conductivity estimates presented in this paper are based on numerical modeling data for anisotropic media. The anisotropy of the in-place waste at the site was estimated to range from 10 to 100, based on the Kr range presented above and Kz range reported by Jain et al. (2006). The estimated anisotropy range was consistent with that reported by Singh et al. (2014). The experiments showed that operating a well with a pressure head greater than the depth of the well resulted in a surface seep around the well cluster. The wells must be operated with the leachate level in the wells near the surface (but not above) to maximize liquids addition while avoiding surface seeps. The anticipated advantage of wells installed at different depths was not realized since shallow wells could not operate under higher pressures. Our results show that the fluid conductance achievable in shallow wells was comparable to the middle and deep wells, suggesting that a single well screened across the depth of the landfill
2318
P. Jain et al. / Waste Management 34 (2014) 2312–2320
would provide equivalent performance to a cluster of multiple wells installed with screened sections at varying depths.
More than 95% of the moisture sensors located within 8–10 m from a recirculation well indicated an interception of the wetting front. More than 50% of the 9- and 15-m deep moisture sensors and 20% of the 4.5-m deep sensors, all located approximately 25–30 m away from well clusters, showed an interception of the wetting front. Based on these results, the extent of radial moisture movement using vertical wells in this study was at least 8 m. Several issues need to be considered when interpreting data on the lateral movement of moisture based on the sensors. First, most of the sensors located 8–10 m away from the wells intercepted the moisture front within 1–2 weeks of initiating liquids addition; however, not enough moisture was added in this period to fully saturate the waste within a radial distance of 8– 10 m from the well. Assuming a porosity of 0.5 (v/v), approximately 300 m3 and 600 m3 liquids would be needed to saturate an 8-m cylindrical zone around slotted section of a shallow well (3 m slotted section) and middle (6 m slotted section) (and deep) well, respectively. Similarly, not enough moisture was added, even considering the entire study period, to saturate the waste within 25–30 m of the radial distance from the wells, suggesting that moisture traveled to these sensors primarily along preferential paths rather than as a uniform wetting front. Secondly, the waste in the vicinity of the wells generated gas in the process of decomposition; by potentially trapping gas, the waste may have exhibited an effective saturation less than 100%. The lateral extent of moisture movement based on actual saturation is expected be greater than the previous calculation, which was based on an assumption of 100% saturation. Moreover, as the generated gas moved away from the wells it would be expected to carry moisture to areas even farther from the wells. Finally, the spacing of the wells should not be solely decided based on the extent of lateral movement of moisture from the well. Because of low flow rates achievable through the small-diameter well, conditions may be such that more wells than calculated on the basis of lateral moisture movement are required to add liquids within a reasonable timeframe.
Fluid Conductance (m3s-1m-1m-1)
3.3. Extent of radial moisture movement
10-4 5-cm dia. wells Trench 1 7.5-cm dia. wells Trench 1 5-cm dia. wells Trench 2 7.5-dia. wells Trench 2
10-5
10-6
10-7 40
60
80
100
120
140
160
Cumulative Volume (m3) Fig. 7. Comparison of fluid conductance as a function of liquids volume for 5-cm and 7.5-cm diameter wells.
2.7 m3/d. In the experiments held starting in 2007 (day 1500), leachate seepage occurred when the liquid head was 1–2 m above the surface of the landfill, despite the presence of the concrete seep collar and the liquids addition rates had to be reduced to maintain a liquid level beneath the landfill surface.
3.5. Operational issues
3.5.2. Clogging of meters and piping During limited, preliminary field tests in early 2003, suspended solids (e.g., small pieces of plastic) were found to clog the meters used to measure flow rate and resulted in reduced flow rates. To remedy this situation, a high-capacity, self-cleaning 0.3-mm screen water filter was installed at the inlet of the liquids distribution manifold in mid-2003. The filter required routine flushing and disassembly and more rigorous cleaning during periods of recirculation system downtime. Routine maintenance thereafter (approximately monthly) involved spot-checking of measured flow rates by disconnecting the flow meter and emptying the recirculated liquid into a 500-mL graduated cylinder. If the resultant flow rate into the graduated cylinder differed by approximately 10% from that measured with the flow meter, the meter was taken off-line, cleaned, and returned to service. Generally, the frequency that meters had to be taken off-line for cleaning was limited (2–3 times during the study period). The flexible hose that was used to connect individual injection wells to the distribution manifold was transparent and was prone to biological growth during 2003 and 2004, particularly when long-term, low-flow recirculation was ongoing. Although biological growth did not impact the achievable flow rate, it occasionally caused clogging of the water meters and required monthly cleaning. The use of opaque flexible hoses should be contemplated in future designs to reduce the occurrence of biological growth during leachate recirculation. Because of low flow rates through the 10cm distribution headers, suspended solids passing through the main inlet filter settled in the distribution manifold and eventually reduced the flow carrying capacity over the study period; the distribution headers were occasionally backflushed with groundwater to remedy this.
3.5.1. Seeps During the initial recirculation testing period (through day 850), surface seeps around the well clusters occurred (as indicated by soft spots below the EGC and near the well cluster) when the level of leachate rose above the surface of the landfill. Attempts to reduce this phenomenon by decreasing flow rates (particularly in shallow wells) were still unsuccessful even at flow rates less than
3.5.3. Differential settlement Settlement of the waste beneath and surrounding the recirculation wells occurred during the study, largely in proportion to the amount of waste beneath the well (e.g., the 6-m deep wells settled more than the 12-m deep wells) and caused the wells to be further raised above the landfill surface (see Fig. S20 in Supplemental Material). When a well extended 1 or 2 additional meters above
3.4. Evaluation of well diameters Two diameters (5 cm and 7.5 cm) of 9-m-deep wells were tested for their performance in 2007 (day 1500 in Fig. 3). Fig. 7 shows the fluid conductance of 5-cm-diameter and 7.5-cm-diameter as a function of liquids volume added. The fluid conductance of 7.5-cm-diameter wells were similar to those of the 5-cm-diameter wells in Trench 1. In the case of the wells in Trench 2, the performance of the large diameter well was greater than that of the small diameter well as expected based on design charts presented by Jain et al. (2010). The difference in fluid conductance between the same diameter wells in two trenches suggests that the spatial variability in the waste hydraulic conductivity has a much greater impact on fluid conductance than the well diameter on the achievable flow rates for a given liquids addition pressure.
2319
P. Jain et al. / Waste Management 34 (2014) 2312–2320
3.5.4. Leachate collection rate measurement using ultrasonic meters Several operational and maintenance issues were encountered with the calibration and functioning of the ultrasonic meters. The leachate collection rate through each of these laterals was measured based on the time taken to fill a known volume of a corresponding (emptied) weir box. The ultrasonic meters were adjusted to read the measured flow rate. The background leachate collection rates of the laterals were less than the sensitivity and accuracy of the meters. Calibration of the ultrasonic meters was verified by adding water at a known flow rate to the leachate collection pipes through clean-outs and comparing the flow rate recorded by the meter to the actual flow rate. The ultrasonic meters did not have a built-in surge protection system and the circuits of many ultrasonic meters suffered damage due to lightning during the Florida rainy season and needed frequent repair. The V-notch of the weir boxes frequently became encrusted with biofilm or suspended solids from the leachate and caused the ultrasonic meter to measure erroneously high leachate collection rates, and also required frequent cleaning with a stiff bristle brush. 3.6. Waste moisture content and degree of decomposition To investigate the impacts of bioreactor operation on waste decomposition, landfilled waste was augured before and after the bioreactor operation and waste samples were analyzed for moisture content and methane production using the BMP assay. Moisture distribution in different depths was compared with collected waste samples from 2001 and 2007, as shown in Fig. 8. Leachate recirculation activities increased moisture levels of waste considerably in all depths excluding waste samples collected from 0 to 3 m depths. The average moisture content in different depths ranged from 22% to 26% (wet weight basis) in 2001 but increased to a range of 28–48% (wet weight basis) in 2007. The average initial and final spatial gravimetric moisture contents measured from the 51 samples collected in 2001 and 272 samples collected in 2007 were 23% and 45% (wet weight basis), respectively. Fig. 9 presents the distribution of the BMPs for samples collected before (in 2001) and after (in 2007) liquids addition system operation. The mean methane yields of the 2007 samples were significantly less than the samples collected in 2001 (p < 0.025). The moisture content and methane yield data suggests that the vertical well system was effective in wetting the landfilled waste and enhancing the waste decomposition process. It is noted that no ‘‘control’’ samples were collected, so a comparison of degradation in the leachate recirculation area to an area with no recirculation was not possible.
4. Summary and conclusions The performance of small diameter vertical wells was evaluated for liquids addition at a full-scale bioreactor landfill in Florida. Over 5 years of operation, approximately 25,000 m3 of liquids (leachate and groundwater) were added to a 4-ha landfill cell through injection wells installed at different depths. The cell-wide leachate collection rate did not increase significantly over the course of the study. Based on the data from moisture sensors installed around
2001 2007
0-3 m
3-6 m
6-9 m
9-12 m
12-15 m
0
20
40
60
80
Moisture Content (% wet weight basis) Fig. 8. Distribution of waste moisture content across landfill depth in 2001 and 2007.
0.16
0.14
Biochemical Methane Potential (L/g)
the surface, the well was cut and the 1–2 m section was removed and the PVC piping was joined together with a coupling. Although prefabricated geomembrane boot seals were used around each well to accommodate settlement, the boot sleeve could not slide over the PVC coupling, which required labor-intensive cutting of the boot and re-welding after slipping the boot below the sleeve.
0.12
0.10
0.08
0.06
0.04
0.02
0.00
Before Liquids Addition (2001)
After Liquids Addition (2007)
Fig. 9. BMP assay results of waste excavated in 2001 and 2007.
wells, the extent of lateral moisture movement was estimated to be at least 8 m. Results suggested that a single well screened across the entire depth can provide equivalent performance to multiple wells screened at discrete depths, which would translate to lower construction costs and fewer surface penetrations that will ultimately reduce the likelihood of surface seepage. The fluid conductance of wells was found to be inversely related to waste depth, and the estimated radial hydraulic conductivity ranged from 3.5 106 to 4.2 104 cm/s. Measured moisture content in waste samples collected before and after recirculation experiments demonstrated effective wetting of the waste and statistically significant decomposition of the waste. The experiments revealed that operational issues such as leachate surface seeps occur when the liquid level within a well was above the surface of the landfill, and mitigative measures such as the installation of a concrete seep collar were not effective in reducing the occurrence of seeps. Thus, to achieve higher flow rates than that used in this study, vertical well system design modifications (such as burying the vertical well inside of the landfill rather than having a surface stick-up) may be required.
2320
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Acknowledgements The Florida Bioreactor Demonstration Project was funded by the Florida Department of Environmental Protection and the Florida Center for Solid and Hazardous Waste Management (now the Hinkley Center for Solid and Hazardous Waste Management). The authors acknowledge the support of the project sponsors, the New River Solid Waste Association, and the site engineers Jones, Edmunds & Associates, Inc. William Matthew Farfour, Sreeram Jonnalagadda, and Ravi Kadambala provided valuable assistance in the collection of data presented in this paper. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.wasman.2014. 07.005. References Bareither, C.A., Benson, C.H., Barlaz, M.A., Edil, T.B., Tolaymat, T.M., 2010. Performance of North American bioreactor landfills. I: leachate hydrology and waste settlement. J. Environ. Eng. – ASCE 136 (8), 824–838. Barlaz, M.A., Bareither, C.A., Hossain, A., Saquing, J., Mezzari, I., Benson, C.H., Tolaymat, T.M., Yazdani, R., 2010. Performance of North American bioreactor landfills. II: chemical and biological characteristics. J. Environ. Eng. – ASCE 136 (8), 839–853. Benson, C.H., Barlaz, M.A., Lane, D.T., Rawe, J.M., 2007. Practice review of five bioreactor/recirculation landfills. Waste Manage. 27 (1), 13–29. Jain, P., 2005. Moisture addition at bioreactor landfills using vertical wells: mathematical modeling and field application (Ph.D. Dissertation). University of Florida, Gainesville, FL.
Jain, P., Powell, J.P., Townsend, T.G., Reinhart, D.R., 2005. Air permeability of waste in a municipal solid waste landfill. J. Environ. Eng., ASCE 131 (11), 1565–1573. Jain, P., Powell, J.P., Townsend, T.G., Reinhart, D.R., 2006. Estimating the hydraulic conductivity of landfilled municipal solid waste using the borehole permeameter test. J. Environ. Eng., ASCE 132 (6), 645–652. Jain, P., Tolaymat, T., Townsend, T., 2010. Steady state design of vertical wells for liquids addition at bioreactor landfills. Waste Manage. 30 (11), 2022–2029. Jonnalagadda, J., Kumar, D., Jain, P., Townsend, T., Reinhart, D., 2010. Comparison of resistivity and time domain reflectometry sensors for assessing moisture content in bioreactor landfills. Geotech. Test. J. 33 (3). Kim, H., Townsend, T., 2012. Wet landfill decomposition rate determination using methane yield results for excavated waste samples. Waste Manage. 32 (7), 1427–1433. http://dx.doi.org/10.1016/j.wasman.2012.03.017. Ko, J., Powell, J., Kim, H., Jain, P., Townsend, T., Reinhart, D., 2013. Case study of controlled air addition into landfilled municipal solid waste: design, operation, and control. ASCE J. Toxic Hazard. Radioact. Waste 17 (4), 351–359. http:// dx.doi.org/10.1061/(ASCE)HZ.2153-5515.0000183. Kumar, D., Jonnalagadda, S., Jain, P., Gawande, N.A., Townsend, T.G., Reinhart, D.R., 2009. Field evaluation of resistivity sensors for in situ moisture measurement in a bioreactor landfill. Waste Manage. 29 (5), 1547–1557. Powell, J., 2005. Trace gas quality, temperature control and extent of influence from air addition at a bioreactor landfill (M.E. Thesis). University of Florida, Gainesville, Florida, 8/2005. Powell, J., Jain, P., Kim, H., Townsend, T., Reinhart, D., 2006. Changes in landfill gas quality as a result of controlled air injection. Environ. Sci. Technol. 40 (3), 1029– 1034. Reinhart, D., Townsend, T., 1997. Landfill Bioreactor Design & Operation. CRC Press, Boca Raton, FL. Reinhart, D., McCreanor, P., Townsend, T., 2002. The bioreactor landfill: its status and future. Waste Manage. Res. 20, 172–186. Singh, K., Kadambala, R., Jain, P., Townsend, T., 2014. Anisotropy estimation of compacted municipal solid waste using pressurized vertical well liquids injection. Waste Manage. Res.. http://dx.doi.org/10.1177/0734242X14532003. Vecchioli, J., Ku, H.F.H., 1972. Preliminary results of injecting highly treated sewageplant effluent into a deep sand aquifer at Bay Park, New York. Geological Survey Professional Paper 751-A.
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Case study of landfill leachate recirculation using small-diameter vertical wells Pradeep Jain a,c,1, Jae Hac Ko b,c,2, Dinesh Kumar d,c,3, Jon Powell a,c,1, Hwidong Kim e,c,4, Lizmarie Maldonado a,1, Timothy Townsend c,⇑, Debra R. Reinhart f,5 a
Innovative Waste Consulting Services, LLC, 6628 NW 9th Boulevard, Suite 3, Gainesville, FL 32605, USA School of Environment and Energy, Peking University Shenzhen Graduate School, Shenzhen 518055, China Department of Environmental Engineering Sciences, University of Florida, P.O. BOX 116450, Gainesville, FL 32611-6450, USA d Engineering Department, Municipal Corporation of Delhi, 9th Floor, Dr. SPM Civic Centre, Jawahar Lal Nehru Marg, New Delhi 110 002, India e Environmental Science and Engineering, Gannon University, 109 University Square, Erie, PA 16541-0001, USA f Civil and Environmental Engineering Department, University of Central Florida, P.O. BOX 162450, Orlando, FL 32816, USA b c
a r t i c l e
i n f o
Article history: Received 2 January 2014 Accepted 8 July 2014 Available online 24 August 2014 Keywords: Bioreactor landfill Leachate recirculation Municipal waste Vertical wells Well performance Injection pressure
a b s t r a c t A case study of landfill liquids addition using small diameter (5 cm) vertical wells is reported. More than 25,000 m3 of leachate was added via 134 vertical wells installed 3 m, 12 m, and 18 m deep over five years in a landfill in Florida, US. Liquids addition performance (flow rate per unit screen length per unit liquid head) ranged from 5.6 108 to 3.6 106 m3 s1 per m screen length per m liquid head. The estimated radial hydraulic conductivity ranged from 3.5 106 to 4.2 104 m s1. The extent of lateral moisture movement ranged from 8 to 10 m based on the responses of moisture sensors installed around vertical well clusters, and surface seeps were found to limit the achievable liquids addition rates, despite the use of concrete collars under a pressurized liquids addition scenario. The average moisture content before (51 samples) and after (272 samples) the recirculation experiments were 23% (wet weight basis) and 45% (wet weight basis), respectively, and biochemical methane potential measurements of excavated waste indicated significant (p < 0.025) decomposition. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Bioreactor landfills are designed and operated to optimize the municipal solid waste (MSW) stabilization process rather than to simply contain the wastes as prescribed by most regulations (Reinhart and Townsend, 1997; Reinhart et al., 2002). Addition of liquids is the most common approach in bioreactor operations; leachate is the most common liquid supply, but other moisture sources can be used. Subsurface moisture addition techniques are the more feasible approach compared to surface application at large landfills. In subsurface systems, the liquids are added to the
⇑ Corresponding author. Tel.: +1 352 392 0846; fax: +1 352 392 3076. E-mail addresses: [email protected] (P. Jain), [email protected] (J.H. Ko), [email protected] (D. Kumar), [email protected] (J. Powell), kim008@gannon. edu (H. Kim), [email protected] (L. Maldonado), ttown@ufl.edu (T. Townsend), [email protected] (D.R. Reinhart). 1 Tel.: +1 352 331 4828; fax: +1 352 331 4842. 2 Tel.: +86 0755 26033289; fax: +86 0755 26033226. 3 Tel.: +91 (11) 23225918. 4 Tel.: +1 814 871 7025. 5 Tel.: +1 407 823 2315; fax: +1 407 882 2819. http://dx.doi.org/10.1016/j.wasman.2014.07.005 0956-053X/Ó 2014 Elsevier Ltd. All rights reserved.
landfilled waste through devices such as trenches, wells, and galleries (Benson et al., 2007; Bareither et al., 2010; Barlaz et al., 2010), with pressure commonly applied to promote moisture distribution (pressurized addition is supplied by a pump or a standing head of liquid). Horizontal leachate injection trenches are common but must be constructed at multiple elevations during waste placement to achieve even moisture distribution throughout the waste. Vertical wells allow the operator to retrofit landfill areas that are already filled with waste and minimize the interference of liquids addition device construction with routine landfill operations by waiting until waste placement is complete. While some field experience and experimental results from buried horizontal systems have been reported, similar data for vertical well systems have not been reported, although the practice of using vertical wells for leachate recirculation has been described (Benson et al., 2007; Bareither et al., 2010; Barlaz et al., 2010). This paper provides a case study of liquids addition experience using small-diameter vertical wells at an operating MSW landfill in Florida (New River Regional Landfill [NRRL]). Results from various experiments at the site have been reported previously (e.g., estimating in-situ air permeability and hydraulic conductivity of
P. Jain et al. / Waste Management 34 (2014) 2312–2320
landfill waste, Jain et al., 2005, 2006; assessment of in-situ moisture sensors, Kumar et al., 2009; experience with air addition, Powell et al., 2006; Ko et al., 2013). In this study, 5 years of performance data and lessons learned from the operation of a vertical well liquids addition system are summarized. The paper reports on vertical well performance with respect to well depth, various leachate seep control techniques (e.g., installation of seep control collar, and limiting liquids head in wells), system operating issues, and an assessment of the degree of waste stabilization. The information presented is valuable for design, operation and monitoring of future liquids addition systems at landfills. 2. Material and methods 2.1. Site description The NRRL receives mixed residential, commercial, and light industrial waste at a rate of approximately 700 metric tons (MT) per day. Clayey–sandy soil mined on site was used as daily cover. Approximately 4 hectares of Cell 1 and part of Cell 2, containing approximately 550,000 MT of MSW, were retrofitted to operate as a bioreactor landfill, as shown in Fig. 1. The leachate collection system for these cells, which are not hydraulically separated, was constructed with a saw-tooth pattern, with a total of eight leachate
2313
collection pipes each connected to a dedicated sump. Only 6 pipes are shown in Fig. 1 and the remaining two pipes were located in the eastern section of Cell 2 not shown in Fig. 1. Leachate from these sumps gravity drained to a wet well where it is then pumped to one of two 1900-m3 capacity leachate storage basins (Fig. 1) – note that leachate from all other lined cells at the site was also pumped from different sumps to the storage basins. The maximum waste depth of the tested area was 22 m at the beginning of bioreactor landfill operation. Systems for moisture addition, air injection (air injection experiments in this site were reported by Powell (2005) and Powell et al. (2006)), gas extraction, as well as a range of instruments, were installed as part of research to examine several technical aspects of full-scale bioreactor landfill operation. More information on the bioreactor landfill and related infrastructure can be found elsewhere (Jain, 2005; Jain et al., 2005, 2006; Powell et al., 2006; Kumar et al., 2009; Jonnalagadda et al., 2010). Due to concerns of potential accumulation of more than 30 cm of leachate on the liner, the state permit specified a maximum daily recirculation volume of 218 m3 and 132 m3 for the first 250 days and after 250 days, respectively. The permit also required cessation of liquids addition if the daily leachate collection volume from any of the eight leachate collection pipes exceeded 24 m3. These limits were not exceeded at any point during the study period. The leachate inflow rate into each sump was measured using a 22.5° V-notch weir box and an ultrasonic meter (Ametek
Fig. 1. Bioreactor well field layout.
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Landfill Surface
18 m
12 m
Drexelbrook, Inc., Series 505-1320) mounted above the box. Each weir box was also equipped with a battery-operated pump for leachate sampling. The ultrasonic meters transmitted water level in weir boxes (4–20 mA signal) to a Campbell Scientific data logger (CR10X) installed at an on-site research trailer; instantaneous leachate generation rates for each of sump were recorded once every ten minutes. Magnetic meters (Unipulse 4411, ISCO, Inc.) tracked the total volume of leachate and groundwater recirculated, as well as the volume of leachate collected from the bioreactor.
6m
2314
While designing the vertical well system, it was hypothesized that most of the liquid would flow through the bottom portion of a well screened across the entire landfill depth due to greater water column head and potential preferential flow to the leachate collection system (LCS). Therefore, a cluster of three wells screened at different depths was designed instead of a single fully-screened well. Furthermore, it was postulated that liquid could be injected under pressure through these wells rather than by just maintaining a standing water column. A total of 134 5-cm diameter wells were installed in 2001 in 44 clusters of three wells (plus one cluster with two wells) in a 100 m by 90 m grid using a rig (CME 85) with a 10 cm open-flight auger. Each cluster had Schedule 40 polyvinyl chloride (PVC) wells installed at different depths: 6 m (the bottom 3 m was slotted pipe), 12 m (bottom 6 m slotted), and 18 m (bottom 6 m slotted). The 6-m, 12-m, and 18-m deep wells are also referred to as shallow, middle, and deep, respectively (see Fig. 2). A threaded 5-cm-diameter PVC pipe section was lowered as soon as the borehole was completed and the drill stem was removed – no permeable media was placed in the area surrounding the slotted pipe. The top 0.3–0.6-m section of the annulus was backfilled with hydrated bentonite pellets. A 10-cm PVC (Schedule 80) distribution manifold was used to convey liquids from the leachate storage basins to the well field. The wells were connected to the distribution manifold with a 2.5-cm flexible hose. The 4-ha recirculation study area was covered with a 1-mm linear low density polyethylene (LLDPE) textured geomembrane underlain by an approximately 1-m thick soil layer. Note that the portion of Cell 2 that was not part of the recirculation area was covered with a vegetated 12-in. thick (minimum) clayey soil and was not operated during the recirculation study period. Once the exposed geomembrane cover (EGC) was installed, gas monitoring wells, leachate wells, and monitoring ports (for a total of 300 penetrations) were extended above the landfill surface by cutting a hole in the EGC around the pipes. The gap between the pipe and the EGC was sealed using a prefabricated boot designed to accommodate differential settlement. A series of resistivity-based and time domain reflectometry sensors were installed between rows of the wells and at depths of 4.5, 9, and 15 m to monitor the in-situ moisture content before the EGC cap installation. An evaluation of these sensors and moisture monitoring results were reported by Kumar et al. (2009) and Jonnalagadda et al. (2010). Construction photographs of sensor packs are presented in Figs. S9 and S10 in the Supplementary Material. Also, type-T thermocouple wire (Omega, Inc., CT, USA) was fitted to the bottom end of each well and included in the moisture sensor packs to allow for frequent temperature measurement. Groundwater was used as a supplemental source of moisture if sufficient leachate was not available. 2.3. Initial operation of the landfill bioreactor to evaluate hydraulic performance The liquids addition experiments described herein took place over two years at 26 (2003 through 2005) well clusters (total of 78 wells) located in zone 1 as shown in Fig. 1; the wells in zone
3m
2.2. Vertical well field construction
Bottom Liner Fig. 2. Schematic of a vertical well cluster.
2 were also designed to add air to the waste and only limited liquids addition was practiced in this zone. Recirculation was accomplished by pumping leachate from the leachate storage basin to the recirculation cells via a single distribution manifold, and flow rate at each well was controlled using an in-line valve. Initially, recirculation occurred in all wells in a cluster at once during the operating hours of the facility (generally 8:00 AM to 5:00 PM). Subsequently, recirculation proceeded in 22 clusters (66 wells) starting with the 18-m well, followed by the 12-m well and the 6-m well; liquids were added in only one well at a time in each cluster. In the remaining four clusters, leachate was first added to the 6-m wells followed by the 12-m and 18-m deep. Liquid heads at the well bottom was measured using a submersible pressure transducer (GE Druck, Inc.) with a 4–20 mA output – some wells had dedicated pressure transducers that were connected to a data logger (Campbell Scientific CR10X) for continuous measurements (once per minute), while the liquid head in the remaining wells was measured manually once per day using a pressure transducer (GE Druck, Inc.) and a handheld readout device (UPS II, GE Druck, Inc.). Instantaneous flow rate was measured continuously in some wells with an ABB single-jet impeller meter with a digital output connected to the data logger, while instantaneous and cumulative liquids addition was measured at the remaining wells once per day with ABB single-jet impeller meters permanently connected to the vertical wells. The objective of the tests conducted was to estimate the achievable flow rates and the associated pressure for the wells installed at different depths. The liquids addition system was continuously operated except during times when the system was stopped to change the test locations, clean the inline filters and liquid distribution manifold, and during severe weather conditions.
2.4. Operation of the landfill bioreactor to evaluate impact of well radius and seep collar on hydraulic performance An additional series of experiments were conducted in 2007 with 10 newly-constructed vertical wells to assess the performance of a concrete pad in controlling seeps around the wells and to assess the impact of well radius on hydraulic performance of a vertical well. Two trenches (9.4 m long by 1.8 m wide by 0.6 m deep) were excavated and five vertical wells were installed using the same open flight auger procedure described previously.
2315
P. Jain et al. / Waste Management 34 (2014) 2312–2320
2.5. Waste sampling Waste samples were collected prior to liquids addition experiments (51 samples in 2001 during vertical well construction) and following liquids addition experiments (272 samples in 2007) using a truck-mounted power flight auger equipped with a 10cm diameter solid stem open flight auger. Samples in 2007 were collected across the well field in 3-m composite fractions at depths of 0–3 m, 4.5–7.6 m, and 9.1–13.7 m and were analyzed for moisture content and the Biochemical Methane Potential (BMP) assay. For the BMP assay, the collected samples were sieved and sorted. The BMP assay was conducted only for the biodegradable fractions (mostly paper), non-identifiable fractions, and fine fractions. More details on BMP assay sample collection, analysis and results for the site can be found elsewhere (Kim and Townsend, 2012). 3. Results 3.1. Liquids addition and leachate collection Fig. 3 presents the cumulative volume of liquid (leachate and groundwater) added over the duration of the study. A total of approximately 22,000 m3 of liquids were added during the initial set of experiments in 26 well clusters, while approximately 3000 m3 of leachate were added during the experiments with the 10 new vertical wells installed with seep collars. Approximately 60% of the total liquid volume (16,000 m3) was added in 2004 (day 371–731) primarily in Zone 1 of the bioreactor, using multiple well clusters operating continuously. No considerable impact on the overall leachate collection rate was observed as shown in Fig. 3. None of the 6 collection pipes shown on Fig. 1 registered an increase in leachate collection rates – Fig. 4 presents data from one of the two pipes that showed an increase; both of these pipes were located in the eastern section of Cell 2 (not shown in Fig. 1). It can be seen that the leachate collection rate increased approximately a year (approximately day 600 on Fig. 3) after the commencement of leachate recirculation (from 0.27 to 0.83 m3 per day). The leachate collection from these pipes represented a small fraction (5–10%) of the overall leachate collection rate measured in the area of Cell 1 and Cell 2 corresponding to the recirculation system. The limited impact of recirculation on leachate production rates likely reflects absorption of moisture by the landfilled waste and the low vertical hydraulic conductivity of the waste reported by
Cumulative Liquids Volume (m3)
30000
25000 Added
20000
15000
10000 Collected
5000
0 0
500
1000
1500
2000
Time (Days) (Day 1= 1/1/2003) Fig. 3. Liquids addition and leachate collection cumulative volume.
Leachate Collection rate (m3/day)
The wells were installed in two clusters, which each consisted of a 5-cm diameter well installed at 3 (bottom 1.5 m slotted), 6 (bottom 4.5 m slotted), 9 (bottom 7.5 m slotted), and 12 (bottom 10.5 m slotted) deep and a 7.5-cm diameter well installed 9 (bottom 7.5 m slotted) deep. The top 0.3 m section of the annulus of each well was backfilled with hydrated bentonite pellets. Ready-mixed concrete was poured to build a 15-cm thick concrete slab (referred here into as seep collar) covering the entire trench area. The excavated cover soil was backfilled and compacted to cover the trenches. Each PVC pipe was extended approximately 0.6 m above the landfill surface and connected to a valve, flow meter, and liquids distribution manifold – construction photos are provided in the supplementary material. Leachate addition was performed intermittently during landfill operating hours. A submersible transducer was installed at the bottom of each well following the initiation of the experiments, and once the liquid level in the well was above the landfill surface (based on transducer measurements), liquids addition was performed in only selected wells and the area was closely monitored for surface seeps.
1.0 0.8 0.6 0.4 0.2 0.0 0
200
400
600
800
Time (Days) (Day 1=1/1/2003) Fig. 4. Temporal trend of leachate generation rate in a leachate collection pipe that experienced an increase in leachate generation rate during the experimental period.
Jain et al. (2006), who reported a range at this site from 5.4 106 cm/s to 6.1 105 cm/s. The total amount of liquid added (25,000 m3) represents 16% of the moisture retention capacity of the waste estimated based on the initial moisture content (described further later) of 23% and an assumed waste field capacity of 40%. 3.2. Leachate addition per well screen at different depths The daily average liquids addition rate and pressure head, measured at the bottom of 6 m, 12 m, and 18 m deep wells at a single cluster location, are plotted as a function of time in Fig. 5. The pressure at the bottom of the well first increased rapidly followed by a more gradual increase as liquids addition progressed, which was attributed to the high pressure gradient required to add liquids at a given specific flux to a media of very low relative permeability; the relative permeability of the medium depth wells is low because of low waste saturation. The subsequent gradual increase in pressure for a constant flux over time was ascribed to the increase in flow path length as the moisture front moved away from the well. Other factors such as landfill gas pressure, clogging, and changes in waste hydraulic conductivity associated with waste decomposition and consolidation also may cause a reduction in flow rate for a given injection pressure or increase in liquid head in the well for a given flow rate. During all tests, liquids addition was attempted at a constant flow rate for the duration required to achieve steady-state conditions, defined as attainment of a steady pressure head for a constant flow rate for a well. The flow rate in many cases was reduced to minimize the chance of surface seeps if the liquid level in the well approached or exceeded the surface of the landfill.
2316
P. Jain et al. / Waste Management 34 (2014) 2312–2320 0.38
15
Pressure (m of w.c.) Flow rate (x10-4m3/s)
-4
3
0.25 9
Flow rate (x10 m /s)
Pressure (m w.c.)
12
6 0.13 3
0
0.00 10
20
30
40
50
60
70
Time (Days)
(a) 15
0.50
Pressure (m of w.c.) Flow rate (x10-4m3/s)
3
9
-4
Pressure (m w.c.)
0.38
Flow rate (x10 m /s)
12
0.25 6 0.13 3
0
0.00 10
20
30
40
50
60
70
Time (Days)
(b) 0.50
15
3
Pressure (m of w.c.) Flow rate (x10-4m3/s)
9
-4
Pressure (m w.c.)
0.38
Flow rate (x10 m /s)
12
0.25 6 0.13 3
0
0.00 10
20
30
40
50
60
70
80
Time (Days)
(c) Fig. 5. Typical temporal variation of head measured at the bottom of a well in response to moisture addition for (a) shallow well (Day 1 = 08/17/2004), (b) middle well (Day 1 = 06/02/2004), and (c) deep well of a well cluster (Day 1 = 03/08/2004). The system was turned off for maintenance or severe weather from day 17 to day 27, from day 9 to day 19, and from day 33 to day 36 for the shallow, middle, and deep wells, respectively.
The temporal variation of flow and pressure for other wells is presented elsewhere (Jain, 2005). Based on the dimensionless variables and design charts developed by Jain et al. (2010) for the design of a vertical well system, the liquids volume added to a well was analyzed as the number of well volumes and the flow rate was normalized with the liquid head in the well bottom and the screen length to generalize the results. The number of well volumes is the ratio of the actual volume added to the screened well section that is distributing flow to the surrounding waste mass. The normalized flow rate (herein referred to as fluid conductance) closely resembles specific capacity,
which is used to assess the performance of groundwater pumping or injection from or into the aquifer using vertical wells; specific capacity is defined as water withdrawal or injection rate per unit head drawdown or buildup, respectively (Vecchioli and Ku, 1972). Fig. 6 shows the variation of fluid conductance as a function of liquids volume added and number of well volumes. The top, middle, and bottom lines of the boxes represent 75th percentile, median and 25th percentile of the data, respectively. The top and the bottom whiskers present the 90th and 10th percentile, respectively, while the dots represents data outliers beyond the 90th and 10th percentile. The number inside each box corresponds to
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Number of Well Volumes
-4
10
16,181
24,272
32,362
40,453
48,544
56,634
-5
10
-
-
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
8,091
30 27 29
-6
10
5 24
7
11 20
4
10
13
2
x¼ 1
-7
10
1
(a) Deep Wells
-8
10
100
200
300
400
500
600
700
Cumulative Volume (m³)
Number of Well Volumes
-4
10
8,091
12,136
16,181
20,227
24,272
28,317
-5
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
4,045
-
10
26
-
25
-6
6
23
10
6
13
3 1
-7
10
-8
10
(b) Middle Wells -9
10
0
50
100
150
200
250
300
350
Cumulative Volume (m³)
Number of Well Volumes
-3
10
8,065
16,129
24,194
32,258
40,323
48,387
56,452
-4
10
-
-
-
Fluid Conductance (m³ s ¹ m ¹ m ¹)
volumes into wells with greater fluid conductance. The number of wells decreased with liquid volume added in Fig. 6 as the larger volumes were only achieved in a small number of good performing wells. Based on the dimensionless variables developed by Jain et al. (2010), the fluid conductance is also equivalent to xKr, where x is defined as follows:
26
0
-5
23
10
18 14
10 7 1 6
1
-6
10
-7
10
(c) Shallow Wells -8
10
0
50
2317
100
150
200
250
300
350
Cumulative Volume (m³) Fig. 6. Fluid conductance as a function of liquids volume added for (a) deep wells, (b) middle wells, and (c) shallow wells. The top, middle, and bottom of each box represents the 75th, median, and 25th percentile of the data, the upper and bottom bars represent the 90th and 10th percentile, and the dots represent data outliers beyond the 90th and 10th percentile. The number inside of each box corresponds to the number of wells for that range.
the number of wells. The median fluid conductance ranged from 5.6 108 to 3.6 106 m3 s1 per m screen length per m liquid head at the well bottom for all wells and showed only slight variation with liquids volume. An increasing trend in fluid conductance with liquids volume is reflective of an ability to add larger liquid
pq Q p w2 K r ; with q ¼ 2v pId ¼ I ; g ¼ 2 pId g prw K z w rw K z
where Qv is the liquids addition rate to a well (L3T1), rw is the well radius (L), w is the screen length (L), pI is the injection pressure head at the bottom of the well (L), and Kz and Kr are the vertical and radial hydraulic conductivity, respectively (LT1). The variable g is expected to be more than 104 for the system evaluated due to a large w/rw ratio and waste anisotropy. Based on Design Chart 1 presented by Jain et al. (2010), at steady-state conditions x ranges from 0.86 (for g = 107) to 1.39 (for g = 104) for shallow and middle wells and from 0.92 (for g = 106) to 1.48 (for g = 104) for deep wells for the operating conditions (i.e., ratio of the liquid head at the well bottom to screen length ranges from 2 to 3) used in this study. The radial hydraulic conductivity was estimated for median fluid conductance values for liquid addition amount greater than 16,000 well volumes, which, for a waste porosity of 0.5, is equivalent to 32,000 pore volumes as defined by Jain et al. (2010); a pore volume of 32,000 is greater than the volume needed to achieve steady state for g = 104. The range of median radial hydraulic conductivity of the waste layers corresponding to the screened sections for 6-m, 12-m, and 18-m deep wells were 1.3 104 to 4.2 104, 2 105 to 9.9 105, and 3.5 106 to 1.2 104 cm/s, respectively. The estimated radial hydraulic conductivity in the shallow layer was greater than the middle layer and the deep layer, which is consistent with the trend reported by Jain et al. (2005) and Jain et al. (2006). Based on the number of well volumes added and the estimate of the number of pore volumes needed to achieve steady state (from Jain et al., 2010), the radial hydraulic conductivity is overestimated as many of the wells did not reach steady state. The range is in good agreement with the radial hydraulic conductivity estimated based on air injection tests conducted at the site previously (Jain et al., 2005) but is one to two orders of magnitude less than that estimated using the borehole permeameter tests in Jain et al. (2006). Although the radial hydraulic conductivity influences the flow rate achievable through a vertical well under a given liquid head, the hydraulic conductivity in the vertical direction would be expected to dictate the flow rate at steady state. The values reported by Jain et al. (2006), which were based on use of the borehole permeameter equation for isotropic media, therefore, probably correspond to hydraulic conductivity in vertical direction. The hydraulic conductivity estimates presented in this paper are based on numerical modeling data for anisotropic media. The anisotropy of the in-place waste at the site was estimated to range from 10 to 100, based on the Kr range presented above and Kz range reported by Jain et al. (2006). The estimated anisotropy range was consistent with that reported by Singh et al. (2014). The experiments showed that operating a well with a pressure head greater than the depth of the well resulted in a surface seep around the well cluster. The wells must be operated with the leachate level in the wells near the surface (but not above) to maximize liquids addition while avoiding surface seeps. The anticipated advantage of wells installed at different depths was not realized since shallow wells could not operate under higher pressures. Our results show that the fluid conductance achievable in shallow wells was comparable to the middle and deep wells, suggesting that a single well screened across the depth of the landfill
2318
P. Jain et al. / Waste Management 34 (2014) 2312–2320
would provide equivalent performance to a cluster of multiple wells installed with screened sections at varying depths.
More than 95% of the moisture sensors located within 8–10 m from a recirculation well indicated an interception of the wetting front. More than 50% of the 9- and 15-m deep moisture sensors and 20% of the 4.5-m deep sensors, all located approximately 25–30 m away from well clusters, showed an interception of the wetting front. Based on these results, the extent of radial moisture movement using vertical wells in this study was at least 8 m. Several issues need to be considered when interpreting data on the lateral movement of moisture based on the sensors. First, most of the sensors located 8–10 m away from the wells intercepted the moisture front within 1–2 weeks of initiating liquids addition; however, not enough moisture was added in this period to fully saturate the waste within a radial distance of 8– 10 m from the well. Assuming a porosity of 0.5 (v/v), approximately 300 m3 and 600 m3 liquids would be needed to saturate an 8-m cylindrical zone around slotted section of a shallow well (3 m slotted section) and middle (6 m slotted section) (and deep) well, respectively. Similarly, not enough moisture was added, even considering the entire study period, to saturate the waste within 25–30 m of the radial distance from the wells, suggesting that moisture traveled to these sensors primarily along preferential paths rather than as a uniform wetting front. Secondly, the waste in the vicinity of the wells generated gas in the process of decomposition; by potentially trapping gas, the waste may have exhibited an effective saturation less than 100%. The lateral extent of moisture movement based on actual saturation is expected be greater than the previous calculation, which was based on an assumption of 100% saturation. Moreover, as the generated gas moved away from the wells it would be expected to carry moisture to areas even farther from the wells. Finally, the spacing of the wells should not be solely decided based on the extent of lateral movement of moisture from the well. Because of low flow rates achievable through the small-diameter well, conditions may be such that more wells than calculated on the basis of lateral moisture movement are required to add liquids within a reasonable timeframe.
Fluid Conductance (m3s-1m-1m-1)
3.3. Extent of radial moisture movement
10-4 5-cm dia. wells Trench 1 7.5-cm dia. wells Trench 1 5-cm dia. wells Trench 2 7.5-dia. wells Trench 2
10-5
10-6
10-7 40
60
80
100
120
140
160
Cumulative Volume (m3) Fig. 7. Comparison of fluid conductance as a function of liquids volume for 5-cm and 7.5-cm diameter wells.
2.7 m3/d. In the experiments held starting in 2007 (day 1500), leachate seepage occurred when the liquid head was 1–2 m above the surface of the landfill, despite the presence of the concrete seep collar and the liquids addition rates had to be reduced to maintain a liquid level beneath the landfill surface.
3.5. Operational issues
3.5.2. Clogging of meters and piping During limited, preliminary field tests in early 2003, suspended solids (e.g., small pieces of plastic) were found to clog the meters used to measure flow rate and resulted in reduced flow rates. To remedy this situation, a high-capacity, self-cleaning 0.3-mm screen water filter was installed at the inlet of the liquids distribution manifold in mid-2003. The filter required routine flushing and disassembly and more rigorous cleaning during periods of recirculation system downtime. Routine maintenance thereafter (approximately monthly) involved spot-checking of measured flow rates by disconnecting the flow meter and emptying the recirculated liquid into a 500-mL graduated cylinder. If the resultant flow rate into the graduated cylinder differed by approximately 10% from that measured with the flow meter, the meter was taken off-line, cleaned, and returned to service. Generally, the frequency that meters had to be taken off-line for cleaning was limited (2–3 times during the study period). The flexible hose that was used to connect individual injection wells to the distribution manifold was transparent and was prone to biological growth during 2003 and 2004, particularly when long-term, low-flow recirculation was ongoing. Although biological growth did not impact the achievable flow rate, it occasionally caused clogging of the water meters and required monthly cleaning. The use of opaque flexible hoses should be contemplated in future designs to reduce the occurrence of biological growth during leachate recirculation. Because of low flow rates through the 10cm distribution headers, suspended solids passing through the main inlet filter settled in the distribution manifold and eventually reduced the flow carrying capacity over the study period; the distribution headers were occasionally backflushed with groundwater to remedy this.
3.5.1. Seeps During the initial recirculation testing period (through day 850), surface seeps around the well clusters occurred (as indicated by soft spots below the EGC and near the well cluster) when the level of leachate rose above the surface of the landfill. Attempts to reduce this phenomenon by decreasing flow rates (particularly in shallow wells) were still unsuccessful even at flow rates less than
3.5.3. Differential settlement Settlement of the waste beneath and surrounding the recirculation wells occurred during the study, largely in proportion to the amount of waste beneath the well (e.g., the 6-m deep wells settled more than the 12-m deep wells) and caused the wells to be further raised above the landfill surface (see Fig. S20 in Supplemental Material). When a well extended 1 or 2 additional meters above
3.4. Evaluation of well diameters Two diameters (5 cm and 7.5 cm) of 9-m-deep wells were tested for their performance in 2007 (day 1500 in Fig. 3). Fig. 7 shows the fluid conductance of 5-cm-diameter and 7.5-cm-diameter as a function of liquids volume added. The fluid conductance of 7.5-cm-diameter wells were similar to those of the 5-cm-diameter wells in Trench 1. In the case of the wells in Trench 2, the performance of the large diameter well was greater than that of the small diameter well as expected based on design charts presented by Jain et al. (2010). The difference in fluid conductance between the same diameter wells in two trenches suggests that the spatial variability in the waste hydraulic conductivity has a much greater impact on fluid conductance than the well diameter on the achievable flow rates for a given liquids addition pressure.
2319
P. Jain et al. / Waste Management 34 (2014) 2312–2320
3.5.4. Leachate collection rate measurement using ultrasonic meters Several operational and maintenance issues were encountered with the calibration and functioning of the ultrasonic meters. The leachate collection rate through each of these laterals was measured based on the time taken to fill a known volume of a corresponding (emptied) weir box. The ultrasonic meters were adjusted to read the measured flow rate. The background leachate collection rates of the laterals were less than the sensitivity and accuracy of the meters. Calibration of the ultrasonic meters was verified by adding water at a known flow rate to the leachate collection pipes through clean-outs and comparing the flow rate recorded by the meter to the actual flow rate. The ultrasonic meters did not have a built-in surge protection system and the circuits of many ultrasonic meters suffered damage due to lightning during the Florida rainy season and needed frequent repair. The V-notch of the weir boxes frequently became encrusted with biofilm or suspended solids from the leachate and caused the ultrasonic meter to measure erroneously high leachate collection rates, and also required frequent cleaning with a stiff bristle brush. 3.6. Waste moisture content and degree of decomposition To investigate the impacts of bioreactor operation on waste decomposition, landfilled waste was augured before and after the bioreactor operation and waste samples were analyzed for moisture content and methane production using the BMP assay. Moisture distribution in different depths was compared with collected waste samples from 2001 and 2007, as shown in Fig. 8. Leachate recirculation activities increased moisture levels of waste considerably in all depths excluding waste samples collected from 0 to 3 m depths. The average moisture content in different depths ranged from 22% to 26% (wet weight basis) in 2001 but increased to a range of 28–48% (wet weight basis) in 2007. The average initial and final spatial gravimetric moisture contents measured from the 51 samples collected in 2001 and 272 samples collected in 2007 were 23% and 45% (wet weight basis), respectively. Fig. 9 presents the distribution of the BMPs for samples collected before (in 2001) and after (in 2007) liquids addition system operation. The mean methane yields of the 2007 samples were significantly less than the samples collected in 2001 (p < 0.025). The moisture content and methane yield data suggests that the vertical well system was effective in wetting the landfilled waste and enhancing the waste decomposition process. It is noted that no ‘‘control’’ samples were collected, so a comparison of degradation in the leachate recirculation area to an area with no recirculation was not possible.
4. Summary and conclusions The performance of small diameter vertical wells was evaluated for liquids addition at a full-scale bioreactor landfill in Florida. Over 5 years of operation, approximately 25,000 m3 of liquids (leachate and groundwater) were added to a 4-ha landfill cell through injection wells installed at different depths. The cell-wide leachate collection rate did not increase significantly over the course of the study. Based on the data from moisture sensors installed around
2001 2007
0-3 m
3-6 m
6-9 m
9-12 m
12-15 m
0
20
40
60
80
Moisture Content (% wet weight basis) Fig. 8. Distribution of waste moisture content across landfill depth in 2001 and 2007.
0.16
0.14
Biochemical Methane Potential (L/g)
the surface, the well was cut and the 1–2 m section was removed and the PVC piping was joined together with a coupling. Although prefabricated geomembrane boot seals were used around each well to accommodate settlement, the boot sleeve could not slide over the PVC coupling, which required labor-intensive cutting of the boot and re-welding after slipping the boot below the sleeve.
0.12
0.10
0.08
0.06
0.04
0.02
0.00
Before Liquids Addition (2001)
After Liquids Addition (2007)
Fig. 9. BMP assay results of waste excavated in 2001 and 2007.
wells, the extent of lateral moisture movement was estimated to be at least 8 m. Results suggested that a single well screened across the entire depth can provide equivalent performance to multiple wells screened at discrete depths, which would translate to lower construction costs and fewer surface penetrations that will ultimately reduce the likelihood of surface seepage. The fluid conductance of wells was found to be inversely related to waste depth, and the estimated radial hydraulic conductivity ranged from 3.5 106 to 4.2 104 cm/s. Measured moisture content in waste samples collected before and after recirculation experiments demonstrated effective wetting of the waste and statistically significant decomposition of the waste. The experiments revealed that operational issues such as leachate surface seeps occur when the liquid level within a well was above the surface of the landfill, and mitigative measures such as the installation of a concrete seep collar were not effective in reducing the occurrence of seeps. Thus, to achieve higher flow rates than that used in this study, vertical well system design modifications (such as burying the vertical well inside of the landfill rather than having a surface stick-up) may be required.
2320
P. Jain et al. / Waste Management 34 (2014) 2312–2320
Acknowledgements The Florida Bioreactor Demonstration Project was funded by the Florida Department of Environmental Protection and the Florida Center for Solid and Hazardous Waste Management (now the Hinkley Center for Solid and Hazardous Waste Management). The authors acknowledge the support of the project sponsors, the New River Solid Waste Association, and the site engineers Jones, Edmunds & Associates, Inc. William Matthew Farfour, Sreeram Jonnalagadda, and Ravi Kadambala provided valuable assistance in the collection of data presented in this paper. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.wasman.2014. 07.005. References Bareither, C.A., Benson, C.H., Barlaz, M.A., Edil, T.B., Tolaymat, T.M., 2010. Performance of North American bioreactor landfills. I: leachate hydrology and waste settlement. J. Environ. Eng. – ASCE 136 (8), 824–838. Barlaz, M.A., Bareither, C.A., Hossain, A., Saquing, J., Mezzari, I., Benson, C.H., Tolaymat, T.M., Yazdani, R., 2010. Performance of North American bioreactor landfills. II: chemical and biological characteristics. J. Environ. Eng. – ASCE 136 (8), 839–853. Benson, C.H., Barlaz, M.A., Lane, D.T., Rawe, J.M., 2007. Practice review of five bioreactor/recirculation landfills. Waste Manage. 27 (1), 13–29. Jain, P., 2005. Moisture addition at bioreactor landfills using vertical wells: mathematical modeling and field application (Ph.D. Dissertation). University of Florida, Gainesville, FL.
Jain, P., Powell, J.P., Townsend, T.G., Reinhart, D.R., 2005. Air permeability of waste in a municipal solid waste landfill. J. Environ. Eng., ASCE 131 (11), 1565–1573. Jain, P., Powell, J.P., Townsend, T.G., Reinhart, D.R., 2006. Estimating the hydraulic conductivity of landfilled municipal solid waste using the borehole permeameter test. J. Environ. Eng., ASCE 132 (6), 645–652. Jain, P., Tolaymat, T., Townsend, T., 2010. Steady state design of vertical wells for liquids addition at bioreactor landfills. Waste Manage. 30 (11), 2022–2029. Jonnalagadda, J., Kumar, D., Jain, P., Townsend, T., Reinhart, D., 2010. Comparison of resistivity and time domain reflectometry sensors for assessing moisture content in bioreactor landfills. Geotech. Test. J. 33 (3). Kim, H., Townsend, T., 2012. Wet landfill decomposition rate determination using methane yield results for excavated waste samples. Waste Manage. 32 (7), 1427–1433. http://dx.doi.org/10.1016/j.wasman.2012.03.017. Ko, J., Powell, J., Kim, H., Jain, P., Townsend, T., Reinhart, D., 2013. Case study of controlled air addition into landfilled municipal solid waste: design, operation, and control. ASCE J. Toxic Hazard. Radioact. Waste 17 (4), 351–359. http:// dx.doi.org/10.1061/(ASCE)HZ.2153-5515.0000183. Kumar, D., Jonnalagadda, S., Jain, P., Gawande, N.A., Townsend, T.G., Reinhart, D.R., 2009. Field evaluation of resistivity sensors for in situ moisture measurement in a bioreactor landfill. Waste Manage. 29 (5), 1547–1557. Powell, J., 2005. Trace gas quality, temperature control and extent of influence from air addition at a bioreactor landfill (M.E. Thesis). University of Florida, Gainesville, Florida, 8/2005. Powell, J., Jain, P., Kim, H., Townsend, T., Reinhart, D., 2006. Changes in landfill gas quality as a result of controlled air injection. Environ. Sci. Technol. 40 (3), 1029– 1034. Reinhart, D., Townsend, T., 1997. Landfill Bioreactor Design & Operation. CRC Press, Boca Raton, FL. Reinhart, D., McCreanor, P., Townsend, T., 2002. The bioreactor landfill: its status and future. Waste Manage. Res. 20, 172–186. Singh, K., Kadambala, R., Jain, P., Townsend, T., 2014. Anisotropy estimation of compacted municipal solid waste using pressurized vertical well liquids injection. Waste Manage. Res.. http://dx.doi.org/10.1177/0734242X14532003. Vecchioli, J., Ku, H.F.H., 1972. Preliminary results of injecting highly treated sewageplant effluent into a deep sand aquifer at Bay Park, New York. Geological Survey Professional Paper 751-A.
