To be presented at Battelle's Sixth International Symposium on In Situ and On Site Bioremediation,
June 4-7, 2001, San Diego, California
A CIS TECHNIQUE FOR ESTIMATING NATURAL
ATTENUATION RATES AND MASS BALANCES
Neat D. Durant, P. Srinivasan, Charles R. Faust, Daniel K. Burnell,
and Katrina L. Klein (GeoTrans, Inc., Sterling, Virginia)
David S. Burden (U.S. EPA, Ada, Oklahoma)
ABSTRACT: Regulatory approval of monitored natural attenuation (MNA) as a
component for site remediation often requires a demonstration that contaminant
mass has decreased significantly over time. Successful approval of MNA also
typically requires an estimate of past and future natural attenuation rates.
Calculation of electron acceptor and/or donor mass budgets is also useful in
establishing the potential effectiveness of MNA. In this paper we present the
application of the GIS program (TINMASS) for quantifying and visualizing
contaminant and electron acceptor/donor mass-in-place. TINMASS uses a
triangulated irregular network (TIN) to interpolate dissolved contaminant mass
between monitoring points. The technique is best suited to sites where the plume
has reached a steady-state, and the existing network of monitoring wells includes
points located at or outside the perimeter of the plume, both horizontally and
vertically. Natural attenuation rates can be estimated from the slope of the line
formed by a log-linear plot of contaminant mass versus time. We present an
example application illustrating use of the method for mass-balance analysis of
sequential decay of trichloroethene. A second example is presented illustrating
use of the method for evaluating fuel hydrocarbon degradation by quantifying
masses of biodegradation end products.
INTRODUCTION
In April 1999, the U.S. Environmental Protection Agency (EPA) issued a
policy directive regarding the use of MNA at Superfund, RCRA corrective action,
and underground storage tank sites (OSWER Directive 9200.4-17P). The
directive described lines of evidence that are typically investigated for
determining the adequacy of MNA as a site remedy. One line of evidence is
historical groundwater and/or soil data that demonstrate a meaningful trend of
decreasing contaminant mass and/or concentration over time (EPA, 1999).
Another line of evidence is hydrogeologic and geochemical data that demonstrate
indirectly the types and rates of natural attenuation processes that are active at a
site. In most cases, biodegradation is recognized as the most important natural
attenuation process (for organic contaminants). As such, MNA evaluations are
typically designed to estimate natural biodegradation rates. Rate estimates are
used in analytical and/or numerical models to predict the time required to attain
cleanup standards under a MNA remediation scenario.
Establishing a decreasing trend in contaminant mass is sometimes difficult
because it can require multiple sampling events over several years, as well as a
monitoring network that is sufficient to define the limits of the plume vertically
and horizontally. Even at sites where the monitoring network is adequate,
synthesis and interpretation of historical groundwater monitoring data can be
-------
Durant et al. 2007 Battelle Bioremediation Conference
complicated by temporal and spatial variation. For example, declining trends may
be evident in some wells, but absent in others. Even if multiple wells at a site
show decreasing concentration trends, the local persistence of contamination can
sometimes jeopardize the approval of MNA as a site remedy.
A variety of methods are available for estimating site-specific intrinsic
biodegradation rates, most of which have been described in existing EPA
guidance manuals (Wiedemeier et al., 1998; EPA, 1998). The conservative tracer
normalization (dilution-correction) and Buscheck and Alcantar (1995) methods
are two of the more commonly employed techniques for estimating
biodegradation rates. Both of these techniques require the installation of a series
of monitoring points distributed along a linear transect that extends from the
contaminant source to downgradient, along the length of the plume. An
alternative approach to estimating attenuation/biodegradation rates is to quantify
the contaminant mass-in-place trend over time (Dupont et al., 1997).
In this paper we describe the development and application of TINMASS, a
simple GIS tool for visualizing natural attenuation trends and estimating natural
attenuation rates. TINMASS provides a means for estimating past and present
contaminant mass-in-place, as well an effective graphical framework for
visualizing mass reduction trends. Unlike the normalization and Buscheck and
Alcantar (1995) methods, TINMASS does not require the placement of wells
along a linear transect extending from source to downgradient. Not only can the
method be used to estimate mass loss trends for the contaminant, it can also be
used to estimate increases in the mass of inorganic end-products (e.g., Fe", HiS,
and CFLt), thereby facilitating a mass balance analysis. In addition, the method
allows for specification of spatial variation in porosity and aquifer thickness.
DESCRIPTION OF TINMASS
TINMASS is a Maplnfo® for Windows application that can be used to
calculate the mass of a dissolved constituent in groundwater based on water
quality data contained in a database (e.g., Microsoft Excel® or Access®). Input
data for TINMASS include the following parameters at each sample location: (1)
aquifer or interval thickness; (2) concentration of the constituent or chemical; and
(3) porosity. Coordinates of each sample are taken directly from the Maplnfo®
object representing the sample location. An example input parameter window for
TINMASS is shown in Figure 1.
Computational Technique. A triangulated irregular network (TIN) algorithm is
used to interpolate constituent mass between monitoring points. The technique is
a simple numerical integration approach commonly used to estimate volumes in
civil engineering applications. In the first step, an optimum network of triangles
connecting all sample locations is generated using the Dulauney Triangulation
procedure. Dulauney Triangulation connects points to form a network of triangles
that have as nearly equal angles at their vertices as possible. The procedure was
based on an algorithm described by Watson (1982).
In the second step, an estimate of the mass of dissolved contaminant
within each triangle is determined based on the assumption that aqueous
concentration (CM,), porosity (n), and aquifer/interval thickness (b) vary linearly
-------
Durant et al.
2001 Battelle Bioremediation Conference
between the sample locations. The dissolved mass within the aquifer bounded
horizontally by each triangle is determined by:
where «, Cw, and b vary linearly across the triangle. Details on the interpolation
and integration may be found in Istok (1989). The total mass of dissolved
constituent in the aquifer unit equals the sum of the mass from each triangle.
Running T1NMASS will create a TIN network as a Maplnfo® table and
calculate mass in units of pounds or kilograms. Data from the model run (input
data fields or values entered, number of nodes, TIN triangles generated, and total
mass) are stored in another Maplnfo® summary table indexed by a Run ID. The
Run ID may be used as an identifier of the contaminant and the sampling period.
Thus, results from multiple periods and different chemicals may be summarized
in one summary table. This approach supports the determination of temporal
mass-in-place trends.
• TinMass Piogiam
Input
Maplnfo da»a table
Node ID held
Porosity field/Value
|Well_id
03
Thickness feld/value [§
Concentration field
Btex
3 |ug/L
Mass Units
Units for calculated mass
Desired mass units for output
Conversion tactoi (multiplier]
ug/L " Cubic It
J2.832E-08
FIGURE 1. Example input parameter window for TINMASS.
Practical Use and Limitations. Mass-in-place estimates computed by
TINMASS can be used to estimate site-specific natural attenuation rates at sites
where sufficient historical data are available, and the existing monitoring network
has defined the vertical and horizontal limits of the plume. The natural
attenuation rate can be approximated as the slope of an exponential regression on
a log-linear plot of mass vs. time. When using TINMASS for this purpose it is
important to use a consistent set of monitoring locations across time. Even if the
number of available data points increased during the relevant period, the same set
of points should be used in order to limit the extent to which spatial/temporal
variability will bias the rate estimate.
-------
Durant et al. 2001 Battelle Bioremediation Conference
Use of TINMASS for estimating natural attenuation rates is best suited to
sites where the plume is receding or has reached a steady-state condition. The
method is not suitable for sites where a significant portion of the contamination
occurs outside the existing monitoring network. At such sites it would not be
possible to prove that temporal mass reductions were caused by attenuation (and
not due to plume migration beyond the TIN network).
TINMASS can estimate mass-in-place of a constituent in a variably thick
aquifer or within a discrete interval of an aquifer (depending on the layout of the
monitoring network). In order to estimate mass-in-place in three dimensions, it is
sometimes appropriate to estimate mass-in-place for separate layers (monitoring
depth zones) and sum the results.
Because concentrations typically decrease in an exponential manner away
from the source, the linear interpolation used in this procedure has a tendency to
overestimate mass-in-place. The potential for overestimation decreases as the
number of monitoring points in the TIN network increases. It should be noted
that TINMASS only examines mass trends for the dissolved phase. Where
present, non-aqueous phase liquids (NAPLs) will usually account for a dominant
portion of the total mass, but estimation of NAPL mass is subject to a very high
level of uncertainty. If undetected, NAPL presence may cause misinterpretation
of mass removed versus mass-in-place trends. Despite the uncertainty of the mass
estimate for a given sampling event, the uncertainty is relatively constant over
time when using TINMASS to assess mass-in-place trends for dissolved plumes
(assuming that the network is constant through time). Consequently, the natural
attenuation rate estimate for dissolved plume mass is not significantly affected by
errors associated with each individual mass estimate.
EXAMPLE APPLICATIONS
Example A. Natural attenuation is treating a portion of the dissolved chlorinated
solvent plume beneath the Sanitary Landfill (SLF) at the U.S. DOE Savannah
River Site (WSRC, 1997). Much of the trichloroethene (TCE) in the plume has
been reductively dechlorinated to c/s-l,2-dichloroethene (DCE) and vinyl
chloride (VC). An abundance of dissolved organic carbon is the likely reductant
supporting dechlorination. The landfill was capped in 1997, and a horizontal well
biosparging system is now being operated to treat the groundwater contamination.
A modeling analysis was recently completed to evaluate the performance of
natural attenuation for eliminating the residual contamination that escapes the
biosparging treatment zone. Estimation of site-specific natural biodegradation
rates was the first step in the natural attenuation modeling process. TINMASS
was used to specify a TIN network within the plume, and to quantify the
dissolved mass for TCE, DCE, and VC over time (see Figure 2). A uniform
porosity of 20% was specified as an input parameter. The unit thickness for each
point was specified as the difference between the water level and a datum
elevation. Although the monitoring network consisted of multi-level wells
screened at three depths, all the wells were pooled together for this analysis since
the aquifer formation and contaminant distribution are relatively homogeneous
-------
Durant et al.
2001 Battette Bioremediation Conference
(A)
10
5
i
Landfill
Penmeter
Oroundwater
Flow
M-90 S-91 J-93 J-94 O-95 M-97 J-98 D-99 A-01
Date
FIGURE 2. (A) Configuration of 1993 TCE plume Oig/L) and TIN network
used to estimate dissolved chloroethene mass at the Savannah River Site
SLF. Shading in TIN triangles is user-specified, and represents different
mass per volume quantities (values not shown). (B) Time series plots of
dissolved mass for TCE, DCE, and VC.
-------
Durant et at. 2001 Battelle Bioremediation Conference
over the interval (approximately 18 meters).
As illustrated in Figure 2, the TIN network covered reasonably well the
configuration of the TCE plume at the SLF. TINMASS was used to estimate
dissolved masses for TCE, DCE, and VC. The TINMASS estimates (kilograms)
were converted to a moles to allow a mass balance assessment of the sequential
decay of TCE over time (under anaerobic conditions, one mole of TCE decays to
one mole of DCE, and one mole of DCE decays to one mole of VC). The data
were plotted on a log-linear graph, and decay coefficients were approximated as
the slope of an exponential regression of the data (see Figure 2). From this
analysis, the estimated first-order biodegradation rate is 0.15 yr"1 TCE, and 0.55
yr"1 for DCE. Fewer data were available for the DCE rate estimate since
laboratory analysis for this constituent did not commence until 1995. Note that
these rates apply to the attenuation of the whole plume, without regard for spatial
variability in redox conditions.
Additional TINMASS analyses were performed to estimate
biodegradation rates for aerobic and anaerobic zones (data not shown). In
general, the anaerobic regions coincide with the footprint of the landfill perimeter,
and aerobic zones occur outside the landfill perimeter. Mass estimates from
TINMASS suggest that the DCE biodegradation rate is 0.44 yr"1 in the anaerobic
zone, and 0.66 yr"1 in the aerobic zone. The mass of VC appeared to show a net
increase directly underneath the landfill cap during the period analyzed, and the
rate of VC accumulation in the anaerobic zone was consistent with the rate of
DCE mass decay. In the aerobic zone downgradient of the landfill, the VC
biodegradation rate estimated from TINMASS is 0.29 yr"1.
The data in Figure 2B suggest that TCE and DCE are attenuating at the
SLF. Sufficient data are available to suggest that the TCE has degraded as a first
order trend. There is more scatter in the DCE trend, which likely reflects spatial
variability in the processes and conditions affecting DCE degradation.
Example B. During the 1980s, a jet fuel (JP-4) release introduced hydrocarbon
contamination into a shallow aquifer beneath a confidential site in the
Northeastern U.S. Sampling performed during the early 1990s detected benzene,
toluene, ethylbenzene, and xylene (BTEX), and naphthalene in the site
groundwater. More recent sampling events have confirmed that benzene and
naphthalene are the only two contaminants present at concentrations above clean-
up levels. Other straight-chain and branched hydrocarbons typical of JP-4 are
also present in the groundwater. These have not been differentiated, and are
characterized only as total petroleum hydrocarbon (TPH). The residual light non-
aqueous phase liquid (LNAPL) that was present in the 1990s likely still remains
in certain locations.
In addition to declining constituent concentrations, evidence of natural
attenuation includes the accumulation of biodegradation end-products (e.g., H2S,
CH4, and Fe ). TINMASS was used to estimate mass removed and attenuation
rates for the various contaminants. The mass of dissolved H2S, CH4, and Fe" was
also estimated with TINMASS, and these data were used with stoichiometric
relationships to estimate the potential amount of biodegradation associated with
-------
Durant et al.
2001 Battelle Bioremediation Conference
each biodegradation end product. Biodegradation rates were also estimated by
the method of Buscheck and Alcantar (1995) for the purpose of comparison with
the rates estimates from the mass-in-place calculations. The results of these
analyses are summarized in Table 1.
TABLE 1. Summary of dissolved mass-in-place and associated natural
attenuation rate estimates for groundwater constituents at JP-4 site.
BTEX
Benz
Naph
TPH
Fe"
H2S
CH4
Attenuation
Rate by
TINMASS
(y-1>
0.40
0.40
0.26
0.88
-
~
--
Degradation
Rate by
Buscheck and
Alcantar
(yr-1)
3.8
3.0
2.4
0.7
-
-
-
Mass Removed
1996-2000
(kg)
20
6
49
1050
--
-
--
Mass
Produced
by 2000
(kg)
-
—
—
-
184
120
98
Possible Incurred
Biodegradation
Based on
Stoichiometry1
(kg)
203
209
215
184
-
-
-
1. Stoichiometric relationships presented by Wiedemeier et al. (1998) were used to calculate the
amount of biodegradation possible - assuming no microbial growth - for the observed masses
of accumulated biodegradation end products. Octane was used as a surrogate for TPH.
As shown in Table 1, application of TINMASS allowed an estimate of the
dissolved mass of BTEX, naphthalene, and TPH degraded during the period of
1996 to 2000. This analysis indicates that TPH represents the greatest fraction
removed during this period. By comparison, the masses of BTEX and
naphthalene removed were significantly smaller. This finding is not surprising
given that BTEX and naphthalene together represent less than 5% by weight of
typical virgin JP-4.
The attenuation rates estimated from the TINMASS calculations for
BTEX, benzene, and naphthalene are within the range normally reported for
biodegradation of these compounds in aquifers. However, these estimates likely
underestimate the actual rate of intrinsic biodegradation since this did not account
for contributions from source dissolution. During the period of 1996 to 2000,
additional BTEX and naphthalene likely was introduced into the groundwater by
dissolution from the LNAPL source. If source dissolution •was had not occurred,
plume attenuation (and mass removal) would have been more rapid.
Contributions from source dissolution likely explain, in part, the differences in the
rate estimates between the TINMASS method and the method of Buscheck and
Alcantar. The Buscheck and Alcantar method focuses on biodegradation in the
plume, while the TINMASS method applies to both source and plume. As shown
in Table 1, rates estimated with the Buscheck and Alcantar method are nearly an
order of magnitude faster than attenuation rate estimates derived from the
TINMASS calculations (with the exception of TPH). Others have noted,
however, that the method by Buscheck and Alcantar is prone to overestimating
the rate of biodegradation (McNab and Dooher, 1998). As such, we conclude
-------
Durant et al. 2001 Battelle Bioremediation Conference
that the attenuation rate estimated by TINMASS method is a conservative
approximation to the biodegradation rate.
The last column in Table 1 provides an estimate of the mass of each
constituent that could have been degraded if all the H2S, CH4, and Fe" observed
was due to the biodegradation of that constituent. These data suggest that at least
a portion of the observed mass removed from 1996 to 2000 can be attributed to
biodegradation under ferrogenic, sulfidogenic, and methanogenic conditions.
Comparison of the mass removal estimates for BTEX, naphthalene, and TPH with
their calculated potential incurred biodegradation provides insight as to the
principal electron donors driving the accumulation of H2S, CFLi, and Fe1. This
comparison suggests that the biodegradation of benzene, BTEX, and naphthalene
alone during 1996 to 2000 does not account for all the H2S, CH4, and Fe that has
accumulated. It follows that biodegradation of the TPH has contributed to a
significant portion of the H2S, CFLi, and Fe ' that have accumulated.
SUMMARY
The software programs BIOSCREEN (Newell et al., 1996) and
BIOCHLOR (Aziz et al., 2000) provide a variety of tools for analyzing natural
attenuation, including an ability to estimate contaminant mass-in-place and mass
biodegraded. Although TINMASS is not an alternative to these powerful tools,
it offers a unique capability to estimate dissolved contaminant mass across a
variably contaminated domain. Unlike BIOSCREEN and BIOCHLOR,
TINMASS is not limited to symmetrical and homogeneous model domains, and
does not employ transport equations or flow parameters. Instead, TINMASS
focuses primarily on the trends in empirical data. TINMASS allows users to
estimate dissolved mass across formations of variable porosity, variable thickness,
and variable contaminant distribution. It also provides a simple framework for
integrating historical mass-in-place trends across multiple monitoring points.
In this paper, mass-in-place calculations were used to estimating site-
specific attenuation rates and assess mass balances between biodegradation parent
compounds and daughter products. These applications suggest that TINMASS is
a useful tool for natural attenuation analysis. Additional evaluation at other sites
is needed to further examine the use of this method for estimating site-specific
natural attenuation rates.
ACKNOWLEDGMENTS AND DISCLAIMER
The EPA Robert S. Kerr Environmental Research Laboratory and the
Westinghouse Savannah River Company (WSRC) funded portions of this work.
This paper has not been subjected to EPA peer and administrative review.
Therefore, no official endorsement may be inferred. Mention of commercial
products in this paper does not represent endorsement by EPA or WSRC.
REFERENCES
Aziz, C.E., C.J. Newell, J.R. Gonzales, P. Haas, T. P. Clement, and Y. Sun. 2000.
BIOCHLOR Natural Attenuation Decision Support System, User's Manual,
-------
Durant et al. 2001 Battelle Bioremediation Conference
Version 1.0, U.S. Environmental Protection Agency Report, EPA/600/R-00/008,
National Risk Management Research Laboratory, Ada, OK.
Buscheck, T.E., and C.M. Alcantar. 1995. "Regression Techniques and
Analytical Solutions to Demonstrate Intrinsic Bioremediation." In R.E. Hinchee,
J.T. Wilson, and D.C. Downey (Eds.), Intrinsic Bioremediation, pp. 109-116.
Battelle Press, Columbus, OH.
Dupont R.R., K. Gorder, D.L. Sorensen, M.W. Kemblowski, and P. Haas. 1997.
"Case Study: Eielson Air Force Base, Alaska." In Proceedings of the Symposium
on Natural Attenuation of Chlorinated Organics in Ground Water. U.S.
Environmental Protection Agency Report, EPA/540/R-97/504, Office of Research
and Development, Washington, D.C.
Istok, J. 1989. Groundwater Modeling by the Finite Element Method. Water
Resources Monograph No. 13, American Geophysical Union, Washington, D.C.
McNab, W.W., and B.P. Dooher. 1998. "A Critique of a Steady-State Analytical
Method for Estimating Contaminant Degradation Rates." Ground Water,
J6(6):983-987.
Newell, C.J., R.K. McLeod, and J.R. Gonzales. 1996. BIOSCREEN Natural
Attenuation Decision Support System, User's Manual, Version 1.3, U.S.
Environmental Protection Agency Report, EPA/600/R-96/087, National Risk
Management Research Laboratory, Ada, OK.
U.S. Environmental Protection Agency. 1998. Seminars on Monitored Natural
Attenuation for Ground Water. Report EPA/625/K-98/001.
U.S. Environmental Protection Agency. 1999. Use of Monitored Natural
Attenuation at Superfund, RCRA Corrective Action, and Underground Storage
Tank Sites. Office of Solid Waste and Emergency Response, Directive 9200.4-
17P, Washington, D.C.
Watson, D.F. 1982. "ACORN: Automated Contouring of Raw Data." Computers
and GeoSciences. S(l):97-101.
Wiedemeier, T.H., M.A. Swanson, D.E. Montoux, E.K. Gordon, J.T. Wilson,
B.H. Wilson, D.H. Kampbell, P.E. Haas, R.N. Miller, J.E. Hansen, and F.H.
Chapelle. 1998. Technical Protocol for Evaluating Natural Attenuation of
Chlorinated Solvents in Ground Water. U.S. Environmental Protection Agency
Report, EPA/600/R-98/128, National Risk Management Research Laboratory,
Ada, OK.
WSRC. 1997. Intrinsic Bioremediation of Landfills Interim Report.
Westinghouse Savannah River Company Report WSRC-RP-97-17, Rev. No.: 0.
-------
NRMRL-ADA-01308
1 . REPORT NO .
EPA/600/A-02/011
TECHNICAL REPORT DATA
2 .
4. TITLE AND SUBTITLE
A QIS TECHNIQUE FOR ESTIMATING NATURAL ATTENUATION RATES AND MASS
BALANCES
7. AUTHOR (S)
(1) Heal D. Durant, P. Srinivaaan, Charle* R. Fauct,
Daniel K. Burnell, and Katrina L. Klein
(2) David S. Burden
9 . PERFORMING ORGANIZATION NAME AN
(1) OeoTranB, Inc., 46050 Manek
Sterling, Virginia 20166
(2) USIPA, ORD, NRMRL, SPRD
P.O. Box 1198, Ada, Oklanom
D ADDRESS
in Plaza, Suite 100,
a 74821
12 . SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
National Riak Management Research Laboratory
Subsurface Protection & Remediation Division
P.O. Box 1198, Ada, Oklahoma 74821
15. SUPPLEMENTARY NOTES
Project Officer! Jerry N. Jone*
3. RECIPIENT'S ACCESSION NO.
5 . REPORT DATE
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
50106F
11. CONTRACT /GRANT NO.
In-Hou>e
13 . TYPE OF REPORT AND PERIOD COVERED
Symposium Paper
14. SPONSORING AGENCY CODE
EPA/600/15
(580)436-8593
16. ABSTRACT
ABSTRACT: Regulatory approval of monitored natural attenuation (MNA) as a component for site remediation often requires a
demonstration that contaminant mass has decreased significantly over time. Successful approval of MNA also typically requires an
estimate of past and future natural attenuation rates. Calculation of electron acceptor and/or donor mass budgets is also useful in
establishing the potential effectiveness of MNA. In this paper we present the application of the GIS program (TTNMASS) for quantifying
and visualizing contaminant and electron acceptor/donor mass-in-place. TTNMASS uses a triangulated irregular network (TIN) to
interpolate dissolved contaminant mass between monitoring points. The technique is best suited to sites where the plume has reached a
steady-state, and the existing network of monitoring wells includes points located at or outside the perimeter of the plume, both
horizontally and vertically. Natural attenuation rates can be estimated from the slope of the line formed by a log-linear plot of
contaminant mass versus time. We present an example application illustrating use of the method for mass-balance analysis of sequential
decay of trichloroethene. A second example is presented illustrating use of the method for evaluating fuel hydrocarbon degradation by
quantifying masses of biodegradation end products.
17.
A. DESCRIPTORS
18. DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
KEY WORDS AND DOCUMENT ANALYSIS
B. IDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLASS (THIS REPORT)
UNCLASSIFIED
20. SECURITY CLASS (THIS PAGE)
UNCLASSIFIED
C. COSATI FIELD, GROUP
21. NO. OF PAGES
9
22. PRICE
EPA FORM 2220-1 (REV.4-77)
PREVIOUS EDITION IS OBSOLETE
------- |