Persistent Groundwater Contaminant Plumes: Processes, - PowerPoint PPT Presentation

Persistent Groundwater Contaminant Plumes: Processes, Characterization, and Case Studies UA-SRP & USEPA Seminar Series- Webinar February 24, 2014 Mark L. Brusseau School of Earth & Environmental Sciences University of Arizona 1

Constrained Mass Removal & Plume Persistence “ significant limitations with currently available remedial technologies persist that make achievement of MCLs throughout the aquifer unlikely at most complex groundwater sites in a time frame of 50-100 years. ” * “Complex” groundwater sites are defined as those that have DNAPL present (e.g., chlorinated solvents) and that have substantial subsurface heterogeneity, including the presence of extensive lower-permeability units or fractured media. • Why does this situation exist? • What options are available? * National Research Council (NRC). 2013. Alternatives for Managing the Nation's Complex Contaminated Groundwater 2 Sites. Wash., DC

Outline • Chlorinated-solvent sites- prevalence and issues • Constrained mass removal and plume persistence: Impact of DNAPL source zones • Constrained mass removal and plume persistence: Impact of mass storage in lower-K zones & hydraulic factors • Constrained mass removal and plume persistence: Impact of sorbed mass • Summary 3

~1600 SUPERFUND Sites ~80% have Chlorinated-Solvent Contaminants 4

Arizona Sites Chlorinated- Solvents Presence: State: 31/35 Federal: 13/15 5

Groundwater Contamination Sites in Tucson Chlorinated-Solvent Contaminants are Primary Concern 6 at all 9 Sites

Groundwater Remediation Standard Method = Pump and Treat Very effective for plume containment 7

Impact of P&T on Water Resources • Analysis for Tucson [Brusseau & Narter, 2013]- year 2010 • Compare aggregate volume of groundwater extracted for all P&T systems to total metropolitan groundwater withdrawal • Total groundwater withdrawal for all P&T systems = 16.6 M m 3 • This is ~20% of the total groundwater withdrawal in Tucson • Treated water used primarily for potable water or re-injection • Represents ~6% of total potable water supply 8

Three Chlorinated-Solvent Sites in Arizona • TCE is Primary COC • Very Low Retardation (R<2) • No Measurable Transformation Processes • V. Low Biogeochemical Attenuation Capacity 4.5 KM Large Plumes (several km long) 9 KM 11 KM 9

Pump & Treat CMD Data Composite Measure: CMD = Q * C - OU1 Q = pumping rate C = concentration ~90% Reduction Currently ~ 1 kg/d Asymptotic conditions ~2 equivalent pore volumes displaced 10

Constrained Mass Removal & Plume Persistence Potential Factors: • Uncontrolled DNAPL Sources • Plume-scale Lower-K Zones and Mass Storage (diffusive mass transfer- “back diffusion”) • Plume-scale Sorbed-phase Mass Storage (sorption/desorption processes) • Hydraulic Factors (P&T well-field, etc) • Low Attenuation Capacity • Other (Institutional, Analytical, etc) 11

Constrained Mass Removal & Plume Persistence Long Known: 1989 Need to Determine Relative Significance of Each Factor, and Site-dependent Functionality 12

Tucson International Airport Area Superfund Site • TCE/DCE Contamination Identified in 1981 • Site Placed on Superfund NPL in 1983 • Pump and Treat started in 1987 (south plume) • Source-zone Remediation efforts [SVE, ISCO] • UA Collaboration since 1993 13

Composite CMD: AFP44 High-resolution Temporal Data set Asymptotic conditions Start Pump & Treat 1987 14

Constrained Mass Removal & Plume Persistence Question: What is the relative significance of each of the various Persistence/Attenuation factors for this site? Conducted an integrated Laboratory, Field, and Modeling study 15

Plume-scale Modeling Effort Known Inputs Conduct series of scenario-testing sensitivity analyses ~50 km 2 16

17

Impact of Transport Processes K Variability & Diffusive Mass Transfer (back diffusion) 18

Impact of Transport Processes Sorption-desorption (nonlinear, rate limited) [Sims include physical heterogeneity] 19

Impact of Transport Processes DNAPL in Source zones Controlling Factor for Early Phase 20

21 Source-zone Architecture, DNAPL Dissolution, and Mass Removal Multi-scale Investigations of Systems Pore Core Intermediate ~2 m ~10 cm ~6 mm APS

DNAPL Source Behavior Column Experiments Pore-scale Imaging: 10 um resolution 1 1 2 3 4 NAPL Dissolution Control [1-4]: Non-uniform accessibility 2 Relative Concentration 3 4 Desorption Control No-NAPL Expt 2 3 4 2 3 4 22

23 DNAPL Source Behavior Laboratory Experiments DNAPL S n Imaging - Known DNAPL distributions - Permeability variability - Measure DNAPL in situ Flow-cell Experiments 100000 Control- Homogeneous Mixed Source 10000 Heterogeneous Concentration (mg/L) Heterogeneous-2 1000 100 10 1 0 20 40 60 80 100 120 140 160 180 Pore Volume

DNAPL Source Behavior Difficult to conduct comparative analysis Variables: • Domain size Field Data [20 vs 10,000 m 2 ] 10 • Gradient & Q AFP44- 3 3 Hangers [natural vs induced] Dover- Surf 1 • Initial DNAPL Mass CMD (Kg/d) 0.1 0.01 0.001 0.0001 0 50 100 150 200 24 Time (month)

Data Analysis & Interpretation - Employ contaminant mass discharge (CMD) metric - Determine relationship between reduction in mass discharge and reduction in mass 1 - Enhances comparative analysis Fractional Mass Discharge Reduction 0.9 0.8 1 Relative Concentration or Relative CMD 1:1 0.7 Minimal Reduction Maximal Reduction 1:1 (First order) 0.6 0.1 Minimal Reduction (efficient 0.5 mass removal) 0.4 Maximal Reduction (inefficient mass removal) 0.3 0.01 0.2 0.1 0.001 0 5 10 15 20 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 Relative Time Fractional Mass Reduction 25

DNAPL Source Behavior Contaminant Mass Distribution [Accessibility] {source architecture, site age (mass removed)} Flow-cell Experiments Field Data 1 1 Control-homogeneous Mixed Source Heterogeneous-1 Fractional Reduction in CMD 0.8 0.8 Heterogeneous-2 Fraction Reduction in CMD Increasing fraction 0.6 0.6 of poorly accessible mass) TIAA-1 0.4 0.4 TIAA-2 Visualiz. Dover-CSF 0.2 0.2 Dover-Surf Borden-1 Borden-2 0 0 0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1 Fractional Reduction in Mass Fraction Reduction in Mass 26

Post Source-zone Remediation Persistence Factors: • Residual DNAPL Sources (incomplete removal/containment) • Plume-scale Lower-K Zones and Mass Storage (diffusive mass transfer- “back diffusion”) • Plume-scale Sorbed-phase Mass Storage (sorption/desorption processes) • Hydraulic Factors (P&T well-field, etc) • Other (Institutional, Analytical, etc) 27

Composite CMD: AFP44 Impacts from Source Remediation efforts Pre SZR CMD = 2 kg/d ~90% Reduction Current CMD = 0.2 kg/d SVE End SVE ISCO Start Start ISCO End 1987 28

Plume Persistence after Source Remediation Predictions for AFP44 Site 1.0 Simulated- Non-source Factors: 0.9 Contaminant Mass Discharge (Kg/d) Remediation *Plume-scale* Simulated- No 1. Mass in Lower-K Zones 0.8 Remediation 2. Sorbed Mass Measured 0.7 3. Hydraulic Factors (well field) *Ideal case- all source 0.6 mass removed 0.5 0.4 0.3 0.2 0.1 0.0 0 2 4 6 8 10 12 14 Time (Y) 29

Lower-permeability Zones & Diffusion 100 Model Simulations 10 90 Variance of lnK Stochastic (random K fields) Mass Remaining (%) Mass Removed (%) 0 1 99 vs. 1 3.5 Discrete (homogeneous, orthogonal) layers Modflow: Clay-Sand-Clay 0.1 99.9 (MODFLOW) 99.99 0.01 0.001 0 5 10 15 20 25 Relative Time 1 0 Fraction Reduction in Mass Discharge 1 3.5 0.8 Modflow: Clay-Sand-Clay 0.6 0.4 0.2 0 0 0.2 0.4 0.6 0.8 1 30 Fraction Mass Reduction

Well-field Configuration Model Simulations 1 3 longitudinal wells 3 downgradient 3–Layer system (Clay-Sand-Clay) (transverse) wells 9 uniform-dist wells Relative Concentration [MODFLOW] 0.1 Natural gradient (equiv Q) 0.01 0.001 0 5 10 15 20 25 Relative Time 31

Sorption-Desorption Processes Column Experiments 1 Non-Linear, Rate-Limited Sorption 0.1 Linear, Rate-Limited Sorption Non-Linear, Equilibrium Sorption Relative Concentration 0.01 Measured 0.001 RLS >> NLS 0.0001 0.00001 0.000001 0.0000001 0 500 1000 1500 2000 2500 3000 3500 4000 Pore Volumes Causative Mechanisms? Extensive Elution Tailing • Observed for all media • Occurs with short contact times • Need continuous-distribution domain model 32

Sorption-Desorption Processes 1 Eustis 2 PV Interaction with Hard Carbon 4 PV 0.1 8 PV 20 PV 0.01 [sorbate permeation within, and Relative Concentration 100 PV Progressive increase 1000 PV sorbate-induced deformation of, 0.001 Aged 30 days in resistance with the HC matrix] Aged 420 days increasing contact 0.0001 Aged 4 years time 0.00001 0.000001 0.0000001 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 Pore Volumes 1 Replication Exp 1 Exp 2 0.1 Exp 3 98% quartz sand Exp 4 Relative Concentration 0.01 Simulation 2% clay (kaolinite- non-expanding) 0.001 0.38% organic carbon 0.0001 0.14% hard carbon (kerogen, bc) 0.00001 Non-linear sorption 0.000001 Competitive sorption 0.0000001 0 500 1000 1500 2000 2500 3000 3500 4000 33 Pore Volumes