HYDRO GEO CHEM INC. by www.hgcinc.com Systematic Remedial Methodology for Chlorinated VOC Contamination of Soils and Groundwater Underlying Desert Landfills Harold Bentley, Stewart Smith Hydro Geo Chem, Inc. Tucson and Scottsdale, Arizona Presented at the Desert Remedial Action Technologies Workshop Phoenix, Arizona October 2-4, 2007
47
Embed
HYDRO GEO CHEM INC. by Systematic Remedial Methodology for Chlorinated VOC Contamination of Soils and Groundwater Underlying Desert Landfills.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
HYDROGEOCHEMINC.
by
www.hgcinc.com
Systematic Remedial Methodology for Chlorinated VOC Contamination of Soils and
Groundwater Underlying Desert Landfills
Harold Bentley, Stewart Smith
Hydro Geo Chem, Inc.
Tucson and Scottsdale, Arizona
Presented at theDesert Remedial Action Technologies Workshop
Phoenix, Arizona October 2-4, 2007
HYDROGEOCHEMINC.
First, you’ve got to understand the plumbing! ---King Hubbard, Father of American Hydrogeology
This presentation discusses a quantitative in-situ remediation methodology that relies on site-specific evaluation and numerical simulation to1. develop critical insights regarding the conceptual model of the contamination problem, and 2. to develop and optimize a remedial engineering design that meets corrective action goals at maximum efficiency and minimum expense.
HYDROGEOCHEMINC.
Landfill groundwater contamination by volatile chlorinated organic compounds (VCOCs) and Freons is pervasive throughout the desert southwest.
As an example of how serious this problem might be, the City of Tucson has concluded, based on a proactive series of extensive remedial investigations, that all of their 40-some unlined or partially-lined landfills have likely contaminated groundwater with VCOCs*.
*R. Murray, City of Tucson, personal communication
Statement of Problem
HYDROGEOCHEMINC.
Conceptual Model of Arid-Landfill Groundwater Contamination by
Volatile Chlorinated Organic Compounds (VCOCs)
HYDROGEOCHEMINC.
0
25
50
75
100
125
150
175
0 100 200 300 400PCE, ug/l
PCE Beneath Silverbell Landfill, Tucson
HYDROGEOCHEMINC.
The generally observed distribution of the VCOCs has some interesting characteristics:
1. VCOCs are found up as well as down the groundwater gradient, an observation often interpreted as evidence for an upgradient, non-landfill VCOC source
2. The groundwater plume tends to be depleted in the total organic carbon and semi-volatiles usually associated with landfill leachate. (deprived of electron donors and the resulting anaerobic biodegradation, the VCOC plume will be persistent).
HYDROGEOCHEMINC.
The generally observed distribution of the VCOCs has some interesting characteristics:
3. The vadose zone (soil gas) concentrations of VCOCS typically increase with depth, suggesting that the source is at depth rather than in the landfill itself
4. The groundwater concentrations of VCOCS are highest at the water table and decrease with depth, implying that the VCOC source is above the water table
HYDROGEOCHEMINC.
QUESTIONS:
1. Can The Observed Soil and Water VCOC (PCE) Distribution Beneath a Landfill Result From Vapor Phase Movement of PCE Introduced into the Landfill When Active?
2. If So, What is the Most Cost Effective Way to Remove this VCOC Source?
HYDROGEOCHEMINC.
Using Field Data and Numerical Simulation to Evaluate the
Conceptual Model of VCOC Groundwater Contamination
HYDROGEOCHEMINC.
The Numerical Model and its Assumptions
Model Code: TRAMP (Bentley and Travis, 1989)3-D variably saturated 2-Phase flow and transportuniform steady groundwater flowaerobic/anaerobic biodegradation
VCOC (PCE) initially within Landfill onlyModel parameters derived from site-specific permeabilities, porosities, water content; literature values of PCE anaerobic biodegradation ratesTransport Conditions:Advection resulting from landfill gas generationDiffusion Anaerobic biodegradation of PCE in landfillRun model for 20 years after Silverbell landfill
closure
HYDROGEOCHEMINC.
Roger Rd
Grant Rd
Prince Rd
Flo
win
g W
ells
Rd
Silverbell Rd
N
Plan View of 3-dimensional Model Grid
5000 feet
GROUNDWATER
FLOW DIRECTION
HYDROGEOCHEMINC.
Simulated Silverbell Landfill PCE Movement:
After 2 Years
Water tableDirection of Groundwater flow
HYDROGEOCHEMINC.
Simulated Silverbell Landfill PCE Movement: After 20 Yrs
Water tableDirection of Groundwater flow
HYDROGEOCHEMINC.
Measured and Simulated Silverbell Landfill PCE Concentrations 20 Years
after Closure
0 50 100 150 200 250 300
PC E in ug/l
-180
-160
-140
-120
-100
-80
-60
-40
-20
0
dept
h be
low
land
sur
face
(fe
et)
w ater tab le
sim ulated concentration
m easured soil gas concentration
m easured groundw ater concentration
approxim ate depth of base of landfill
HYDROGEOCHEMINC.
Results of Silverbell Landfill Gas-Phase PCE Transport Simulations
20 years after closure the original source area (the landfill) is relatively free of VCOC while deep vadose soils continue to have relatively high levels of PCE.Gas and liquid advection and diffusion and anaerobic biodegradation are all found to be important in reducing landfill PCE concentrations and increasing concentrations at depth.We conclude that vadose-zone VCOCs beneath the landfill are the source of past and continuing groundwater contamination.
HYDROGEOCHEMINC.
SVE is clearly the most cost-effective VCOC source removal option
Contamination is deep and covers a large areaVadose soils have a relatively high gas permeabilityContaminants are volatile and therefore amenable to removal by SVE
How best to implement SVE removal of the VCOC source is the rest of this presentation.
HYDROGEOCHEMINC.
Systematic Remediation of Deep Vadose-Zone PCE Contamination
(Harrison Road Landfill, Tucson, Arizona)
HYDROGEOCHEMINC.
Three-Dimensional Model Structure
Harrison Road Landfill, looking southwest
LFG WellsSVI-1: Multi-levelnested probes
SVE-1: Multi-level nested probes
VMW Multi-level nested probes
Base of Landfill
HYDROGEOCHEMINC.
Technical Issues Regarding Sub-Landfill SVE for Removing Deep
VCOCs
Must allow for LFG generation in overlying landfillMust operate in conjunction with LFG collection systemMust minimize air intrusion into landfill to prevent fires and maintain methanogenic efficiencyAn important goal is to minimize number of SVE wells. Cost of an SVE well, at this site, is greater than $50,000
HYDROGEOCHEMINC.
SVE Design Performance Criteria
Provide early removal of deep vadose VCOC source to groundwaterMinimize drawing VCOCs from shallower soils to deeper soilsMinimize air intrusion into the overlying landfill to maintain the landfill’s anaerobic characterBONUS: Removal of low-volatility organics from any leachate present beneath landfill by means of aerobic biodegradation (bioventing)
HYDROGEOCHEMINC.
Summary of Data Needed for Effective Design of Sub-Landfill SVE
Horizontal and Vertical Soil Air PermeabilityProvide Achievable Subsurface Air
Circulation Rate per SVE WellDetermine Minimum Number of SVE
Wells Needed to Achieve Desired Total Air Circulation Rate.
Ratio of Horizontal to Vertical Air Permeability Affects Surface Leakage and Lateral Effectiveness of Individual SVE Wells
HYDROGEOCHEMINC.
Summary of Data Needed for Effective Design of Sub-Landfill SVE
Subsurface VCOC Distribution Affects Location and Vertical Placement
of SVE wellsSoil Porosity, Moisture Content, Organic Carbon Content and VCOC Properties Affect Calculation of Total VCOC Mass
and Cleanup TimesDistributed Landfill Gas Generation Rates Affect induced SVE flow field
HYDROGEOCHEMINC.
Hydro Geo Chem’s Pneumatic Assessment
ToolboxThe numerical model TRAMP, a powerful 3-D integrated finite difference, distributed parameter model for assessing gas and liquid fate and transport. Includes aerobic and anaerobic biodegradation, thermodynamics, and liquid/gas phase changes. Capable of automatic parameter estimation and design optimization.ASAP, a proprietary pneumatic well test interpretation model. Includes automatic parameter estimation. The Baro-Pneumatic Method, an HGC-patented methodology for assessing landfill gas generation rates and permeabilities. Provides calibration data for numerical (TRAMP) model of landfill gas flow, useful for designing efficient gas collection and control systems.
HYDROGEOCHEMINC.
Obtaining Horizontal and Vertical Gas Permeabilities
Conduct Pneumatic Well TestsInstall wells as part of a trial SVE systemEmploy step-tests to determine well
efficienciesUtilize monitoring wells, if possible, as
observation wells Conduct Baro-pneumatic Tests Monitor barometric pressure and subsurface
pressure responses to changes in barometric pressure. Obtain vertical permeabilities and LFG generation rates
HYDROGEOCHEMINC.
SVE Wellhead
Construct Monitoring Well that can Later Double as SVE or SVI Well
APPR O VED D ATE R EFER EN C E FIG U R E
H Y D R OG E OC H E M , I N C .
MEASURED AND SIMULATED DRAW DOW NAT MP-12 DURING PUMPING OF SVE-A4
H :/796000/svetest/svea4/m p12.srf 3TS 2/26/04
Gas Porosity and Horizontal Permeability Estimates by Step-drawdown Test. (ASAP analysis) Gas
SVE Design Parameters Derived from SVE Performance
Optimization by Numerical Modeling
Need a total of 3 perimeter extraction wells and just one, central air injection well Injection well screened at deeper intervals than extraction wellsInjection wells and extraction wells operate at same rate of flow (250 scfm each.
HYDROGEOCHEMINC.
Simulation of VCOC Remediation Progress
(Month.Year)
SVI
SVE
SVE
SVE
HYDROGEOCHEMINC.
SVI-1
SVE-1
SVE-2 SVE-3
2000 2500 3000 3500 4000
easting in feet
2000
2500
3000
3500
4000
4500
nort
hing
in fe
et
15102030
SVI-1
SVE-1
SVE-2 SVE-3
2000 2500 3000 3500 4000
easting in feet
2000
2500
3000
3500
4000
4500
nort
hing
in fe
et
PC E in ug/l
S im u la ted V a d o se P C E C o n cen tra tio n s J u st A b o v e th e W a ter T a b le
in itia l conditions after 3 m onths sve operation
HYDROGEOCHEMINC.
SVI-1
SVE-1
SVE-2 SVE-3
2000 2500 3000 3500 4000
easting in feet
2000
2500
3000
3500
4000
4500
nort
hing
in fe
et
15102030
SVI-1
SVE-1
SVE-2 SVE-3
2000 2500 3000 3500 4000
easting in feet
2000
2500
3000
3500
4000
4500
nort
hing
in fe
et
PC E in ug/l
S im u la ted V a d o se P C E C o n cen tra tio n s J u st A b o v e th e W a ter T a b le
after 3 years sve operationafter 1 year sve operation
HYDROGEOCHEMINC.
Comparing Optimized Harrison Landfill SVE Design Layout to Conventional “Radius of Influence” (ROI) Layout
Based on SVE well tests, and interpretation by our well pneumatics software, ASAP, the achievable SVE pumping rate for each well is 250 standard ft3/min (scfm) and the ROI of each well is 200 feet (at a steady-state vacuum of 0.03 inches H2O).
Setting a well grid at 300 feet between wells provides 25% overlap of the circles defined by a 200-foot ROI. The resulting Harrison Landfill ROI well array is illustrated in the following slide.
● 200-FOOT ROI WELLS (TOTAL OF 30)
MODEL-OPTIMIZED SVI/SVE WELL LOCATIONS (TOTAL
OF 4)
●
●
●
●
VCOC-CONTAMINATED VADOSE ZONE
●
Site-specific and ROI SVE-Well Arrays
HYDROGEOCHEMINC.
Cost Comparison of Optimized Harrison SVE Design to Conventional “Radius of Influence”
(ROI) Design
The SVE well array resulting from the use of a 200-foot ROI is comprised of 30 SVE wells, each costing (in 1999) more than $50,000. Thus the total cost for this conventionally designed system’s well construction exceeds $1.5 million dollars, which compares poorly to the $200,000 well cost for the optimized 4-well system. This estimate does not account for the increased costs associated with valving, sumps, gas collection manifolds, and O&M of the more complex ROI-designed system.
HYDROGEOCHEMINC.
•Piping (PVC,ABS,Steel, HDPE)•SVE Blower (Lampson spark-proof)•Injection Blower (Roots positive displacement)•Electricity (NEMA4 Enclosures)•Sumps and Off-gas Treatment•Trenching and Cover
Source Removal Completed as of 8/2002; SVE System Shut Down
Deep SVE System Performance at Harrison Landfill
HYDROGEOCHEMINC.
ConclusionsMost unlined arid-zone landfills have and are contaminating groundwater with VCOCs.Landfill-origin chlorinated VCOCs reach groundwater via gas phase transport.Removal of the landfill origin vadose zone VCOCs sources can be accomplished by SVE.Using pneumatic data collection and SVE simulation to develop the engineering design results in Significant capital and operational cost
savingsHigh collection system efficiency and more
rapid remediation
HYDROGEOCHEMINC.
Suggested Further ReadingCan also be found in: http://www.hgcinc.com/papers.htm
Walter, Gary R. 2002. Fatal Flaws in Measuring Landfill Gas Generation Rates by Empirical Well Testing [PDF] J. Air & Waste Management, 2003 53, p 461Bentley, H.W., S. Smith, J. Tang, and G.R. Walter. 2003. A Method for Estimating the Rate of Landfill Gas Generation by Measurement and Analysis of Barometric Pressure Waves. Proceedings of the 18th International Conference on Solid Waste Technology and Management, Philadelphia, Pennsylvania, March 23-26, 2003 Walter, G.R., Geddis, A.M., Murray, R., Bentley, H.W. 2003. Vapor Phase Transport as a Groundwater Contamination Process at Arid Landfill Sites [PDF]. Proceedings of the 18th International Conference on Solid Waste Technology and Management, Philadelphia, Pennsylvania, March 23-26, 2003 Bentley, H.W., S.J. Smith, and T. Schrauf. 2005. Baro-pneumatic Estimation of Landfill Gas Generation Rates at Four Operating Landfills. Proceedings, SWANA’s 28th Annual Landfill Gas Symposium, March 7-10, 2005 Smith, S.J., H.W. Bentley, and K. Reaves. 2006. Systematic Design of Methane Migration Control Systems. Proceedings, 29th Annual SWANA Landfill Gas Symposium, St. Petersburg FL, March 27-30. 18 pp.
HYDROGEOCHEMINC.
Contact Information
Harold W. Bentley, Ph.D.Principal ScientistHydro Geo Chem, Inc.51 W. Wetmore Road