41/9-4d 4/7/00 11:26 am Page 201

Improving secondary clarifier performance and capacity

Water Science and Technology Vol 41 No 9 pp 201–208 © IWA Publishing 2000

using a structured diagnostic approach
D.S. Parker, E.J. Wahlberg and H.Z. Gerges
Clarifier Optimization Services, Brown and Caldwell, P.O. Box 8045, Walnut Creek, CA, 94596-1220, USA

Abstract The causes of high effluent suspended solids (ESS) from secondary clarifiers are often
incorrectly identified resulting in improperly focused remedial actions that do not yield the desired results. In
this paper we demonstrate a structured diagnostic approach that allows relatively rapid problem isolation and
resolution. A sequence of tests, procedures or observations are described that allow problems to be
classified into one or more of the following categories: (1) denitrification or rising sludge, (2) high sludge
blankets, (3) flocculation or breakup problems, and (4) hydraulic problems. The following tools are used in
the approach: (1) direct observation, (2) blanket detection, (3) measurement of sludge thickening properties,
(4) solids flux or state point analysis, (5) dispersed suspended solids (DSS) test, (6) flocculated suspended
solids (FSS) test, (7) dye testing, and (8) hydrodynamic modeling. The structured diagnostic approach has
been applied and refined by the authors over nearly three decades and has proven extremely useful in rapidly
identifying the least-cost technology for reducing ESS. Case examples are presented where the structured
diagnostic approach is used to isolate the problems and identify the most appropriate remedy.
Keywords Activated sludge; blanket; effluent suspended solids; flocculation; denitrification; clarifier;
sedimentation

Introduction
The secondary clarifier often becomes the focus of attention for process improvement when
effluent requirements become more stringent or when activated sludge process operating
conditions change from those originally anticipated in design. This leads plant owners to
seek performance enhancements either in the form of reducing ESS or increasing tank
capacity or both. In our experience, reduction in ESS is usually accompanied with
“unlocking” additional tank capacity whether the capacity is specifically sought or not. The
causes of high effluent suspended solids (ESS) from secondary clarifiers are often incor-
rectly attributed and as a result improperly focused remedial actions do not yield the desired
results. Existing tanks vary widely in design features and operating problems, so prescrib-
ing “one size fits all” design or operations fixes will not necessarily correct perceived prob-
lems. Therefore, the authors have developed and refined a structured diagnostic approach
to assess problems in clarifier design and operations that lead directly to specific corrective
actions.

Structured diagnostic approach

We approach the identification and correction of high ESS problems much as a medical doc-
tor approaches diagnosis and specifies remedies. Measurements are taken to isolate the
probable illness and then a specific treatment program is recommended. The structured
diagnostic approach has been applied and refined by the authors over nearly three decades
and has been proven extremely useful in rapidly identifying the least-cost technology for
reducing ESS. The approach is outlined in Table 1. The problems that are most easily diag-
nosed are addressed first so as to eliminate the need for conducting a long list of diagnostic
tests for every situation whether needed or not. The first question concerns whether high
ESS is due to denitrification. This can usually be identified by examination of the surface of
the tanks. Under bright sunlight, small refractile gas bubbles attached to the floc will be 201
41/9-4d 4/7/00 11:26 am Page 202

Table 1 Outline of structured diagnostic approach

Problem Type No? Yes? If Yes, Necessary Actions

A. Is denitrification occurring? 1. Alter mode of operation

B. Is sludge blanket high? 1. Complete state point analysis (solids flux)
2. If operating change will reduce blanket level, alter
D.S. Parker et al.

mode of operation
3. If operating change will not reduce blanket level,
design changes to increase solids loading
capacity
C. Does DSS/FSS testing 1. If a physical problem, design clarifier
indicate a flocculation problem? flocculation zone
2. If a biological problem, alter mode of activated
sludge operation
D. Does DSS/FSS testing 1. Perform dye testing to establish flow through curve.
indicate a hydraulic problem? 2. Use High Accuracy Clarifier Model
3. Design clarifier improvements

observed carrying the floc to the surface. If this is found to be the case, the diagnosis can
stop at this point and the corrective action can be specified.
The next question concerns whether high blanket levels plague the operation. High blan-
ket levels can be detected by relatively simple test methods such as the Sludge Judge™ (a
thin, clear plastic tube with a foot valve, that when lowered into the tank samples the entire
water column) or more complex electronic methods such as blanket detection by light inter-
ference techniques. If high blankets are detected, the solids flux approach is used to analyze
the problem. We favor use of the state point analysis (Keinath, 1985; Wahlberg, 1996;
Ekama et al., 1997) as it is easily used to verify whether the secondary clarifier is critically
loaded or overloaded and where design or operational changes will result in reducing the
sludge blanket.
The application of solids flux theory to secondary clarifier analysis is best supported by
the use of on-site measurements of the sludge thickening parameters. For the most reliable
work, actual measurement of zone settling velocities as a function of TSS levels is recom-
mended using long column settling tests (1.5 m) in a temperature-controlled bath
(Wahlberg and Keinath, 1988). However, for preliminary diagnostic work, there are sever-
al published relationships that give zone settling velocities as a function of MLSS concen-
tration and one of the various sludge volume index (SVI) measures (see the review in
Ekama et al., 1997). The application of these equations should be with cognizance that dif-
ferent plant data sets have resulted in significantly different relationships, suggesting that
without site-specific confirmation they could be subject to considerable error.
The problem in application of solids flux theory (state point or otherwise) is that practi-
cal experience indicates that the predicted critical solids loading may not be practically
achieved, that is a blanket may form at lower solids loading rate (SLR) values than predict-
ed. This likely occurs due to a limitation not modeled by the flux theory, such as limitations
in sludge transport to the point of removal from the secondary clarifier. Unfortunately, there
have been relatively few determinations of this, but the available results are summarized in
Table 2.
The next questions distinguish between flocculation and hydraulic problems. The DSS
and FSS tests are described by Wahlberg et al. (1995); the same material is also presented in
202 Ekama et al. (1997). The DSS test is performed with a Kemmerer sampler. After the sample
41/9-4d 4/7/00 11:26 am Page 203

Table 2 Ratio of applied flux to predicted flux at which blanket forms

Ratio of
applied to
predicted
Type of tank Investigator(s) flux Comments

22 full-scale shallow circular tanks with Ekama and Marais 0.8 Sludge thickening parameters

D.S. Parker et al.

center feed and peripheral withdrawal; (1986) using from SVI related correlation
suction sludge removal. Dutch data rather than site data.
Full-scale circular tank with center feed Ekama et al. (1997) 0.67 Use of sloped rather than flat
with peripheral and radial weirs; conical using data of bottom likely posed an additional
bottom with “organ pipe” suction sludge Watts et al. (1996) thickening constraint.
removal.
Three full-scale shallow rectangular Ekama and Marais 1.15 Accuracy of prediction could
tanks, but most of critical data from a (1986) using easily have been affected by not
single tank. Dutch data using locally determined
thickening parameters; they
were based upon a SVI
correlation.
Full-scale rectangular tank with Göhle et al. (1996) 0.8
counter current scraper sludge
collection.

is collected with the Kemmerer unit, it is allowed to settle in the sampler for 30 minutes.
After settling, a sample of the supernatant is withdrawn for a suspended solids analysis.
The DSS is operationally defined as this supernatant suspended solids concentration and
represents the nonsettleable solids concentration. Because sampling and settling is per-
formed in one container with no breakup due to intermediate sample transfer, it can be
thought of as a “snapshot” of the state of flocculation of the suspension at the time the sam-
ple is collected. Sampling at several points along a process can be used to determine if floc-
culation or breakup is occurring. Results from the DSS test at the effluent weir have been
shown to closely approximate the ESS concentration from a well designed and operated
secondary clarifier; that is, a clarifier not subject to short-circuiting or denitrification
(Parker and Stenquist, 1986). If the ESS exceeds the DSS (measured at the clarifier weirs),
then there is a priori evidence that settleable flocs are being lost in the effluent.
The FSS test is performed by gently stirring a mixed liquor sample (50 revolution per
minute) in a square flocculation jar for 30 minutes followed by 30 minutes settling. After set-
tling, a sample of the supernatant is taken for suspended solids analysis. The FSS is opera-
tionally defined as this supernatant suspended solids concentration. Because the flocculation
potential of the sample is maximized and settling is performed in an ideal separation device
(i.e., one in which there are no secondary currents that impair settling), the FSS can be
thought of as the minimum possible secondary clarifier ESS concentration. Differences
between the FSS concentration of the mixed liquor entering a secondary clarifier and the ESS
quantify the flocculation and hydraulic inefficiencies occurring in the secondary clarifier.
DSS and/or FSS testing has proven useful in a number of related applications including
the following; (1) determining the effect of aeration tank turbulence level on effluent non-
settleable solids (Parker et al., 1970; Das et al., 1993), (2) assessing flocculation and
breakup phenomenon in conveyance channels (Parker et al., 1970; Matasci et al. 1986;
Parker and Stenquist, 1986; Das et al., 1993), (3) determining the benefits of flocculation in
minimizing ESS in activated sludge plants (Wahlberg et al., 1994), (4) assessing alternative
flocculator center well designs (Kinnear et al., 1998), and (5) stress testing to determine the
capacity of secondary clarifiers (Wahlberg et al., 1994). 203
41/9-4d 4/7/00 11:26 am Page 204

The two tests coupled with the ESS, can be used to make the following determinations:
Case 1: High ESS, High DSS, Low FSS. Flocculation not adequate, but it can be corrected.

Case 2: High ESS, Low DSS, Low FSS. Hydraulic problem in the secondary clarifier.

Case 3: High ESS, High DSS, High FSS. Flocculation not adequate due to a biology problem.
D.S. Parker et al.

Case 4: High ESS, Medium DSS, Low FSS. Combined hydraulic and flocculation problem.

Case 5: High ESS, Low DSS, High FSS. Impossible situation, repeat test.

When hydraulic problems within the secondary clarifier are identified, we use a High
Accuracy Clarifier Model (HACM, Gerges and McCorquodale 1997; Gerges 1997; Brewer
et al. 1998). The HACM represents the physical processes of real clarifiers. It solves the
differential equations of continuity, momentum, energy, and mass transport. This type of
model is often called a deterministic model since it reveals the role of natural laws in deter-
mining clarifier performance. The HACM has been verified and applied to secondary clari-
fiers at several sites in North America (Charlotte, NC; Denver, CO; Vacaville, CA; St. Paul,
MN; Hamilton, Ontario; and Windsor, Ontario).

Case examples
A denitrification problem: City of Tolleson, Arizona
Flocculator-clarifiers have demonstrated exceptionally stable performance and low ESS at
high surface overflow rates (SORs) and SLRs; this is in part due to their lack of significant
hydraulic problems (Parker, et al., 1995). However, two 34-m diameter flocculator clari-
fiers in Tolleson, Arizona were being stress tested at low SORs when denitrification prob-
lems confounded the attainment of low ESS (Parker and Stenquist, 1986). With two
clarifiers in service, floating sludge and rising floc were seen. Confirming this, the sludge
rose in the DSS test sampler after one hour of test set-up. Analysis confirmed high nitrate
levels in the influent to the secondary clarifiers. Corrective actions taken included increas-
ing the return rate and raising the dissolved oxygen (DO) of the mixed liquor entering the
secondary clarifier. While these actions to reduce denitrification were effective, they did
not completely eliminate the problem as the average ESS obtained was 11 mg/l at a SOR
averaging 0.61 m/h over a 10-day period. The DSS test measured at the effluent weirs aver-
aged 3.5 mg/l during this period, indicating that the SS difference was due to denitrification-
induced floating solids.
As further steps to reduce the denitrification problem, the solids residence time in the
secondary clarifier was reduced by taking a unit out of service and by increasing the speed
of the sludge removal mechanism. At an average SOR of 1.2 m/h, the average ESS was
9 mg/l for a one-month period. As evidence that the denitrification problem had been
corrected, the DSS measured at the effluent weirs averaged 7 mg/l.

A thickening problem: City of Atlanta, Georgia

High sludge blanket levels had plagued the secondary clarifiers at a Water Reclamation
Center (WRC) at a time when there was concern that effluent total phosphorus (TP) levels
of 0.75 mg/l would be exceeded. As the WRC did not yet have its planned effluent filters,
there was a concern that high blanket levels could elevate ESS values causing higher efflu-
ent TP concentrations. It was observed that even when SVI values were low (80 ml/g), blan-
kets persisted. Since SLRs were not high under these conditions, the presence of high
blankets was suspicious. A computer program based upon the state point approach was used
to model the process. To be conservative, the analysis was done under peak diurnal flow
204 conditions and only the effective thickening area was assumed to be the circular portion
41/9-4d 4/7/00 11:26 am Page 205

D.S. Parker et al.

Figure 1 State point analysis of Atlanta Secondary Clarifier (solid underflow operating line represents
actual operation condition, dashed line reflects reduced RAS rate)

swept by the suction pick up mechanism of the WRC’s square clarifiers. Since site-specific
thickening coefficients were not determined, the correlations for thickening coefficients of
Wahlberg and Keinath (1995) were used with a higher SVI value (150 ml/g) than actually
experienced.
Despite the conservative assumptions used in the state-point analysis, Figure 1 shows
that for the assumed operating conditions, the secondary clarifiers were underloaded and no
blanket buildup should have occurred. Based upon the analysis, it was determined that the
only likely variable not adequately accounted for was the return sludge rate. If the rate was
reduced to half the measured value, then the underflow operating line would fall tangent to
the flux curve, indicating the clarifier would form a blanket.
Potential causes of reduced RAS rates were investigated. Flow metering was checked by
a simple mass balance around the secondary clarifiers and found to be accurate.
Mechanical causes were explored next. The WRC’s secondary clarifiers are fitted with a
TowBro™ vacuum sludge removal system consisting of a rotating pipe with orifices.
There is a rotating “doughnut” at the center column that provides a connection to the under-
flow line; the doughnut is equipped with a 360º rotating seal. Recent inspection of one of the
clarifiers had shown that the seal was worn and therefore had been replaced. If the problem
existed in the other units, the concern was that clear supernatant could short-circuit into the
underflow line, reducing the effective RAS rate. Therefore, all of the units were inspected
and the majority were found to have leaking seals. Replacement of the seals led to a
significant reduction in blanket levels at the WRC.

A flocculation problem: Greeley Water Pollution Control Facility, Colorado

Evaluation of the Greeley facility determined that the least cost expansion from 45,000 to
60,000 m3/d would involve only minor additions to the aeration tanks and a single new sec-
ondary clarifier, as long as the existing secondary clarifiers could be upgraded to produce
lower ESS. To check problems with the three existing clarifiers, the diagnostic approach
was used.
DSS/FSS testing results for one of the unmodified clarifiers are shown in Table 3. The
testing showed that effluent DSS values were close to ESS values. Inasmuch as ideal set-
tling resulted in the same SS value as the ESS in the full-scale tanks, it could be conclud-
ed that flocs were not being swept out of the tanks due to inefficient tank hydraulics.
However, FSS values were well below effluent weir DSS values, indicating that floccula-
tion in the tanks was not optimal. High ESS was due to the poorly flocculated state of the
influent solids and inadequate flocculation occurring within the tank itself (a Case 1
problem). 205
41/9-4d 4/7/00 11:26 am Page 206

Table 3 DSS, FSS and ESS concentration before and after clarifier modifications*

Condition Inlet DSS, mg/l Effluent weir DSS, mg/l FSS, mg/l ESS, mg/l

Before modification 29.2 22.0 8.2 25.5

After modifications 16.7 5.3 7.0 6.3

*Data from Brischke et al., 1997

D.S. Parker et al.

Modifications to the three existing clarifiers involved converting the inlet structure to a floc-
culator center well of the type described by Parker et al. (1996). A floor was installed in the
existing center well by welding steel sheet to the existing spreader beams. Vertically orient-
ed slots were cut into the sidewalls of the inlet tub. Hinges were attached to the cutouts,
which were then mounted into the slots to serve as flow-directing gates.This conversion
allowed it to serve as an energy-dissipating/flow directing inlet tub. A flocculation zone was
created outside the inlet tub by constructing a skirt out of a steel frame and fitting the frame
with lightweight ribbed fiberglass sheets. This outer skirt contains the flow exiting the
inlet tub and provides optimal flocculation conditions.Total cost of the modifications was
50,000 USD per clarifier (construction year: 1996).
Table 3 reports the results after the modifications. It can be seen that the modifications
allowed the test clarifier to nearly equal the ideal performance predicted by the FSS test.
The reduction in DSS from inlet to effluent weir indicates that the poor flocculation
conditions occurring in the unmodified clarifier were corrected by providing a flocculation
zone. Long term effluent quality has also been substantially improved and moreover, its
variability has been reduced (Brischke et al., 1997).

A hydraulic problem: Central Marin Sanitation Agency (CMSA), California

The CMSA plant includes four secondary clarifiers, each 30.5-m in diameter with a side-
water depth of 3 m. Concerns that high ESS were being produced at peak flows lead to a
study by the plant staff using the DSS profiling technique. Measured DSS (mg/l) results are
compared to the ESS value below:
DSS, centerwell influent: 10.4
DSS, centerwell effluent: 11.0
DSS, effluent weir: 3.6
ESS: 8.5
It was apparent from the DSS testing that no flocculation was occurring in the existing cen-
terwell, owing to its small diameter. Yet some flocculation was occurring in the sedimenta-
tion tank as the DSS dropped from the influent DSS value to a low value at the effluent of the
tank. Also, the clarifiers had significant hydraulic problems as the DSS at the effluent weir
was significantly less than the ESS (indicating that flocs were being carried over the effluent
weirs). While the FSS was not measured during the test, it was concluded that it would have
had to have been of the same order of magnitude as the DSS measured at the effluent weirs.
Brown and Caldwell was retained to test the tanks, verify the HACM and use the HACM
to predict the effect of various modifications on ESS. This task involved establishing the
zone settling relationship with on-site testing as well as dye testing the existing tanks. The
model was verified by comparison of the measured dye flow through curve (FTC) with the
predicted response as shown in Figure 2.
An analysis of the FTCs shows the following:
• Time of arrival, which is considered a measure of short-circuiting, is well predicted by
the model (14 min versus 13 min actual). This indicates that the model accurately
206 predicts the highest velocity in the clarifier.
41/9-4d 4/7/00 11:26 am Page 207

D.S. Parker et al.

Figure 2 Measured and predicted flow through curves

Table 4 HACM performance predictions and estimated cost of various clarifier modifications*

Modification ESS, mg/l Cost, USD Cost, USD/mg/l

None 13.0 None None

Flocculation 9.5 113,000 32,000
Hydraulic 10.0 30,000 10,000
Flocculation and Hydraulic 8.5 141,000 31,000

* HACM predictions made at a SOR of 0.98 m/h; costs are for each individual clarifier

• Time to the peak concentration predicted by the model (33 min) is in good agreement
with the measured one (29 min). This indicates that the model accurately predicts the
stagnant zones in the clarifier.
• Normalized peak concentration, C/Co, predicted by the model (0.98) compares well
with the measured value (0.91). This indicates the model accurately predicts the
turbulent mixing levels in the clarifier.
The good agreement between the measured and predicted FTC also gives confidence to
the velocity field and suspended solids distribution predicted by the model. Modeling
showed that the existing units had too small a centerwell, forcing the entering fluid rapidly
downward toward the sludge blanket. The flow then turned and ran just above the sludge
blanket and impacted on the clarifier wall. It then turned upward, carrying solids towards
the peripherally placed weirs.
Predictions were made with the model for various potential modifications to the second-
ary clarifiers. Improvements were oriented to either improving the hydraulic problem (baf-
fling the effluent weirs with a Crosby baffle) or improving flocculation (retrofitting a
flocculator center well) or both. The model was used to optimize the dimensions of the
improvements. Table 4 below shows the results of the model predictions for the various
combinations of improvements. As can be seen, the most cost-effective improvement is
made with just the hydraulic modifications (effluent baffling), but none of the modifications
are exceptionally costly. In this case example, if both flocculation improvements and
hydraulic improvements were made, ESS could be reduced by one-third.

Conclusions
The structured diagnostic approach allows relatively quick identification of the problems
causing high ESS in secondary clarifiers and allows the next most appropriate step to be
readily identified. As shown by the case examples, the next step may involve relatively sim-
ple operating or maintenance changes. In other cases clarifier modifications can be directly 207
41/9-4d 4/7/00 11:26 am Page 208

specified immediately after DSS/FSS testing. In still other cases, hydrodynamic modeling
is necessary to recommend the specific clarifier design modifications. In all cases, the
approach is direct enough and the results conclusive enough so that unnecessarily complex
or lengthy investigations can be avoided, and owners of treatment facilities can proceed
directly to those actions that will yield performance improvements.
D.S. Parker et al.

References
Brewer, H.M., McCorquodale, J.A. and Crane, R. (1998). Modeling applied for optimization at the Dundas
WWTP. Proceedings of the Water Environment Federation 71st Annual Conference and Exposition,
Wastewater Treatment Research/Municipal Wastewater Treatment, 1, 653–664.
Brischke, K., Wahlberg, W., Dingeman, T. and Schump, D. (1997). Performance quantified: the impact of
final clarifier performance on effluent quality. Proceedings of the Water Environment Federation 70th
Annual Conference and Exposition. Management/Facility Operations/Current Issues, 5, 237–244.
Das, D., Keinath, T.M., Parker, D.S. and Wahlberg, E.J. (1993). Floc breakup in activated sludge plants.
Wat. Res., 65(2), 138–145.
Ekama, G.A, Barnard, J.L, Günthert, F.W., Krebs, P., McCorquodale, J.A., Parker, D.S. and Wahlberg, E.J.
(1997). Secondary Settling Tanks: Theory, Modelling, Design and Operation. IAWQ Scientific and
Technical Report No. 6., IAWQ, London, England.
Gerges, H.Z. (1997). A skew third order upwinding scheme (STOUS) for modeling settling tanks, Ph.D.
dissertation, University of Windsor, Windsor, Ontario, Canada.
Gerges, H.Z. and McCorquodale, J.A. (1997). Modeling of flow in rectangular sedimentation tanks by an
explicit third order upwinding technique. Inter J. Num. Methods in Fluids, 24, 537–561 .
Göhle, F., Finnson, A. and Hultman, B. (1996). Dynamic simulation of sludge blanket movements in a
full-scale rectangular sedimentation basin, Wat. Sci. Tech., 33(1), 89–99.
Kinnear, D.J., Williams, R., Olsen, C., Kennedy, K., Johnson, H., Pollack, E. and Vitasovic, C. (1988).
Theoretical and field comparison of feedwell sizing, proceedings of the Water Environment Federation
71st Annual Conference and Exposition. Wastewater Treatment Research/Municipal Wastewater
Treatment, 1, 845–866.
Keinath, T.M. (1985). Operational dynamics of secondary clarifiers, Journal Water Pollution Control
Federation, 57(7), 770–776.
Matasci, R.N., Benedict, A.H. and Parker, D.S. (1986). Trickling Filter/Solids Contact Process: Full Scale
Studies, USEPA Report, EPA/600/S2-86/046, May 1986.
Parker, D.S., Kaufman, W.J. and Jenkins, D. (1970). Characteristics of Biological Flocs in Turbulent
Regimes, SERL Report 70-5, University of California, Berkeley, CA.
Parker, D.S. and Stenquist, R.J. (1986). Flocculator-clarifier performance, Journal Water Pollution Control
Federation, 58(3), 214–219.
Parker, D., Butler, R., Finger, R., Fisher, R., Fox, W., Kido, W., Slapper, J. and Wahlberg, E. (1996). Design
and operations experience with Flocculator-Clarifiers in large plants, Wat. Sci. Tech., 33(12), 163–170.
Wahlberg, E.J. and Keinath, T.M. (1988). Development of settling curves using SVI. Journal Water
Pollution Control Federation, 60(12), 2095–2100.
Wahlberg, E.J., Keinath, T.M. and Parker, D.S. (1994). Influence of activated sludge flocculation time on
secondary clarification, Wat. Res., 66(6), 779–786.
Wahlberg, E.J., Augustus, M., Chapman, D.T., Chen, C-L., Esler, J.K., Keinath, T.M., Parker, D.S.,
Tekippe, R.J. and Wilson, T.E. (1994). Evaluating activated sludge secondary clarifier performance using
the CRTC protocol: four case studies, proceedings of the Water Environment Federation 67th Annual
Conference and Exposition. Biological Treatment Systems/Nutrient Removal, 1, 1–12.
Wahlberg, E.J., Merrill, D.T., and Parker, D.S. (1995). Troubleshooting activated sludge secondary
performance with simple diagnostic tests, proceedings of the Water Environment Federation 68th Annual
Conference and Exposition. Wastewater Treatment Research/Municipal Wastewater Treatment, 1,
435–444.
Wahlberg, E.J. (1996). Activated sludge solids inventory control using the state point concept, paper
presented at the Central States Water Environment Association meeting on Enhancing the Design and
Operation of Activated Sludge Plants, Madison, Wisconsin.
Watts, R.W., Svoronos, S.A. and Koopman, B. (1996). One-dimensional modeling of secondary clarifiers
using a concentration and feed velocity-dependent dispersion coefficient, Wat. Res, 30(9), 2112–2124.
208
 Reproduced with permission of copyright
owner. Further reproduction prohibited without
permission.