Proposed NJDEP Soil Standards
The New Jersey Department of Environmental Protection (NJDEP) is promulgating soil remediation standards as part of the N.J.A.C. 7:26D rules. This includes soil cleanup objectives protective of human health from the ingestion, inhalation, and groundwater pathways. The default cleanup objectives are designed to be extremely conservative. This is because regulatory default cleanup objectives represent the worst-case scenario, and as such they can safely be applied everywhere. Consultants can develop alternative soil remediation standards (ARS) that account for site-specific conditions. The NJDEP outlines four options for the development of ARS protective of groundwater quality.
|
Option I
|
Soil-water partition equation (SPE) using site-specific parameters
|
|
Option II
|
Synthetic precipitation leaching procedure (SPLP)
|
|
Option III
|
SESOIL vadose zone modeling
|
|
Option IV
|
SESOIL vadose zone and AT123D groundwater modeling
|
Option I, Soil-Water Partition Equation (SPE)
The SPE method consists of two parts. The first part uses a soil-water partitioning equation to establish a soil leachate concentration. A dilution attenuation factor (DAF) equation is then applied to produce a resulting groundwater concentration. The SPE method was used to establish the default NJDEP soil cleanup objectives for mobile organic contaminants. Essentially this method assumes that vadose zone soil contamination is located beneath the water table. In other words the soil contamination is instantaneously transported to groundwater.
Soil-Water Partition Equation
|
Parameter
|
Description
|
Units
|
Default
|
|
ARS
|
Alternative soil remediation standard protective of groundwater quality
|
mg/L
|
Chemical specific
|
|
GWQC
|
Groundwater quality criterion typically the maximum contaminant level (MCL)
|
mg/L
|
Chemical specific
|
|
Koc
|
Soil organic carbon-water partition coefficient
|
L/kg
|
Chemical specific
|
|
foc
|
Fraction organic carbon
|
Dimensionless
|
0.002
|
|
ϑw
|
Water filled soil porosity
|
Dimensionless
|
0.23
|
|
ϑa
|
Air filled soil porosity
|
Dimensionless
|
0.18
|
|
H′
|
Henry′s Law constant
|
Dimensionless
|
Chemical specific
|
|
ρb
|
Dry soil bulk density
|
kg/L
|
1.5
|
|
DAF
|
Dilution attenuation factor
|
Dimensionless
|
13
(See eq. below)
|
A DAF is then applied to produce a groundwater concentration. Default values for the DAF typically range from about 10 to 20. In reality using site-specific information it is difficult to justify a DAF much above 1.0.
Dilution Attenuation Factor Equation
|
Parameter
|
Description
|
Units
|
Default
|
|
DAF
|
Dilution attenuation factor
|
Dimensionless
mg/L
|
Site specific
|
|
Ki
|
Aquifer hydraulic conductivity (m/yr) multiplied by the hydraulic gradient
|
Dimensionless
|
30
|
|
d
|
Mixing zone depth
|
m
|
3.5
(See eq. below)
|
|
I
|
Infiltration rate
|
m/yr
|
0.28
|
|
L
|
Length of area of concern parallel to ground water flow direction
|
m
|
30.5
|
Aquifer mixing zone depth
|
Parameter
|
Description
|
Units
|
Default
|
|
d
|
Mixing zone depth
|
m
|
Site specific
|
|
L
|
Length of area of concern parallel to ground water flow direction
|
m
|
30.5
|
|
da
|
Aquifer thickness
|
m
|
3.5
|
|
I
|
Infiltration rate
|
m/yr
|
0.28
|
|
Ki
|
Aquifer hydraulic conductivity (m/yr) multiplied by the hydraulic gradient
|
Dimensionless
|
30
|
Complete SPE and DAF Equation
It is easy to assume that such a complex equation with so many input parameters must reliably account for a significant portion of the contaminant transport and fate processes. However it does not. In reality the results for benzene are typically within an order of magnitude of the following simplified equation.
Simplified SPE and DAF Equation
Option II, Synthetic Precipitation Leaching Procedure
Initially the NJDEP used the SPE to develop default values protective of groundwater quality for all contaminants of concern. However, it was determined that SPE results were overly conservative for immobile contaminants. So there are no default standards for immobile contaminants. Instead ARS for low mobility organic and inorganic contaminants are based on results of the synthetic precipitation leaching procedure (SPLP). The SPLP method directly measures the tendency of a contaminant to leach from soil. As with the SPE method these results are still rather conservative as it assumes that the leachate directly enters groundwater no matter the distance to the water table.
SPLP results produce a site-specific Kd. The Kd or “distribution coefficient” is similar to the simplified SPE equation presented above. This Kd can be either be used to produce an ARS, or as input to the SESOIL and AT123D models (see Option III and IV below).
Option III, SESOIL Vadose Zone Modeling
Option III provides for the development of site-specific ARS based on SESOIL vadose zone modeling. SESOIL is a one-dimensional vertical transport and fate model for the unsaturated (vadose) zone. Numerous regulatory agencies have selected the SESOIL model for the development of soil cleanup objectives. This was done for numerous reasons that remain consistent from state to state. These include the following.
- It simulates the principle processes controlling the transport and fate of contaminants in soil,
- It simulates seasonal influences on those transport and fate processes,
- The ability to vary soil properties with depth, and
- Model input parameters are collected as part of the standard site investigation.
SESOIL produces a leachate concentration not a groundwater concentration. As with the SPE and SPLP results a DAF can be applied to produce a resulting groundwater concentration. This is accomplished using the Summers model DAF. As with other simplistic DAF equations it does not adequately simulate the interaction between contaminant loading and groundwater flow. SESOIL must be linked to groundwater models such as AT123D and MODFLOW/MT3D to establish a groundwater concentration. AT123D is used as it utilizes the monthly groundwater loads produced by SESOIL. Although not as common MODFLOW/MT3D can also be used allowing SESOIL to be used at locations with complex flow conditions.
Option IV, SESOIL Vadose Zone and AT123D Groundwater Modeling
Option IV provides for the development of site-specific ARS based on both SESOIL and AT123D modeling. Linking SESOIL to AT123D produces more accurate groundwater concentrations. This is because groundwater concentrations produced by AT123D incorporate both the monthly SESOIL loads and groundwater flow. As with simple DAF methods such as the Summers model, AT123D should be used even if the point of compliance is directly below the contaminated soil.
Numerous regulatory agencies have selected the AT123D model for the development of groundwater cleanup objectives. As with SESOIL this was done for numerous reasons that remain consistent from state to state. There are three overriding factors for selecting AT123D.
- Simulation of the principle processes controlling the fate of contaminants in groundwater,
- Utilization of the monthly mass loads produced by SESOIL, and
- Model input parameters are collected as part of the standard site investigation.
AT123D can be linked to SESOIL or used on it’s own as a standalone model such as BIOSCREEN and MODFLOW/MT3D. It can even be used to verify the results of MODFLOW/MT3D modeling.
Summary
The NJDEP developed default soil cleanup objectives protective of groundwater quality. These values essentially represent the worst-case scenario thus ensuring protection of groundwater quality. Simplistic methods including the SPE, DAF, SPLP, and Summers model cannot adequately address the processes controlling the migration of contaminants. Such methods are useful as a first cut when dealing with low-level concentrations.
Given the complexities of contaminant transport it is not possible to develop a simple equation or laboratory test that replaces transport and fate modeling. Simplistic methods attempt this at the cost of accuracy. Instead transport and fate modeling represents the best option for the development of site-specific cleanup objectives. ARS based on SESOIL and AT123D modeling can be many orders of magnitude higher than the default concentrations and those produce via the SPE, DAF, SPLP, and Summers model methods. This is not because modeling is less conservative, but because it provides a more accurate depiction of what is occurring.
|
|
|
