Irreversible Phosphorus Sorption in

Septic System Plumes?

by W.D. Robertson

Abstract
The mobility of phosphorus (P) in septic system plumes remains a topic of debate because of the considerable
reactivity of this constituent. In this study, a septic system plume in Ontario was monitored over a 16-year period
with detail that clearly shows the advancing frontal portion of the P plume. This monitoring record provides
insight into the extent of secondary P attenuation in the ground water zone beyond that available from previous
studies. A P plume 16 m in length developed over the monitoring period with PO4-P concentrations (3 to 6 mg/L)
that approached the concentrations present under the tile bed. Simulations using an analytical model showed that
when first-order solute decay was considered to account for the possibility of secondary P attenuation in the
ground water zone, field values could only be matched when decay was absent or occurred at an exceedingly slow
rate (half-life greater than 30 years). Thus, hypothesized secondary P attenuation mechanisms such as slow re-
cystallization of sorbed P into insoluble metal phosphate minerals, diffusion into microsites, or kinetically slow
direct precipitation of P minerals such as hydroxyapatite were inactive in the ground water zone at this site or
occurred at rates that were too slow to be observed in the context of the current 16-year study. Desorption tests on
sediment samples from below the tile bed indicated a PO4 distribution coefficient (Kd) of 4.8, which implies a P
retardation factor of 25, similar to the field apparent value of 37 determined from model calibrations. This exam-
ple of inactive secondary P attenuation in the ground water zone shows that phosphorus in some ground water
plumes can remain mobile and conservative for decades. This has important implications for septic systems
located in lakeshore environments when long-term usage scenarios are considered.

Introduction through mineral coatings onto internal sorption sites (van

Laboratory column and batch studies investigating
phosphorus (P) mobility in sediments invariably note
retardation of P due to surface sorption processes, but in
most cases also note additional loss of P mass and a ten-
dency for increasing irreversibility of P sorption with
time. This behavior has been attributed to the presence of
multiple pools of sediment P, including a labile pool,
which is easily desorbed, plus other secondary pools that
are more stable (Lookman et al. 1997). Several processes
have been suggested to account for the stable secondary
P, including molecular diffusion of P into micropores or
Department of Earth Sciences, University of Waterloo,
Waterloo, Ontario, Canada N2L 3G1; (519) 885-1211; fax (519)
746-7484; wroberts@sciborg.uwaterloo.ca
Received March 2007, accepted September 2007.
Copyright ª 2007 The Author(s)
Journal compilation ª 2007 National Ground Water Association.
doi: 10.1111/j.1745-6584.2007.00389.x
Vol. 46, No. 1—GROUND WATER—January–February 2008 (pages 51–60)
Riemsdijk et al. 1984; Willett et al. 1988), slow crystalli-
zation of sorbed P into insoluble metal phosphate miner-
als (Mansell et al. 1977; Lijklema 1980; Nagpal 1986;
Bleam et al. 1991; Lookman et al. 1997), or direct pre-
cipitation of metal phosphate minerals (Chen et al. 1973;
van Riemsdijk et al. 1975; Enfield et al. 1981; Martin
et al. 1988; Isenbeck-Schroter et al. 1993; Li and
Stanforth 2000; Parkhurst et al. 2003).
The occurrence of secondary attenuation has impor-
tant implications for the fate of P in ground water
plumes, particularly those generated by septic systems. In
a review of PO4 behavior in 10 well-characterized septic
system plumes in Ontario (Robertson et al. 1998), six
sites on sands had distinct PO4 plumes more than 10 m in
length, with PO4-P concentrations (1 to 6 mg/L) that were
orders of magnitude higher than threshold levels that
could stimulate algal growth and eutrophication in lakes
(Dillon and Rigler 1974). Although P migration in septic
system plumes is always highly retarded because of
51
sorption processes, typically by a factor of 10 to 100 in
sand aquifers (Robertson et al. 1998; Parkhurst et al.
2003), the degree of reversibility of P sorption reactions
in field settings has not been well established. The dis-
tinction between normal fast reversible sorption and other
secondary processes is important because reversible sorp-
tion does not permanently remove P from solution and
consequently this P mass remains ultimately mobile and
conservative in the flow system. On the other hand, if sec-
ondary processes are active, even at a slow rate, they
could be of considerable importance in immobilizing P
mass when considering the decade-scale time frames nor-
mally associated with septic system use and the slow
migration rate of PO4.
In our previous review of P mobility in septic system
plumes (Robertson et al. 1998), field evidence showed
that substantial P mass removal (25% to 99%) occurred
within the first 1 to 2 m of subsurface flow, largely coin-
cident with the unsaturated zone. This was the result of
rapid mineral precipitation reactions prompted by redox
changes occurring within this zone (Zanini et al. 1998).
Additional declines in P concentrations were also noted
farther along the plume flowpaths. However, the previous
studies were not sufficiently detailed to determine if this
was the result of secondary attenuation or other processes
such as hydrodynamic dispersion which could produce
a similar effect.
Compared to laboratory studies, long-term monitor-
ing of field-scale plumes should provide a better opportu-
nity to ultimately establish the importance of secondary P
attenuation processes. However, few such studies have
been undertaken. One site where long-term monitoring of
a ground water P plume has been undertaken is at the
Massachusetts Military Reserve on Cape Cod, Massachu-
setts. Here, a P plume approximately 600 m in length em-
anates from a set of large sewage infiltration beds that
have been in operation since 1936 (Walter et al. 1996).
Detailed monitoring of the P plume was initiated in 1993,
at which time the plume was already well established and
was discharging to a small kettle lake (Ashumet Pond).
Variations in PO4 concentrations along the length of the
plume, in part, reflect irregular sewage loading during ear-
lier decades (Parkhurst et al. 2003). Additionally, changes
in redox conditions along the plume cause changes in the
concentrations of Fe and Mn, which in turn affect satura-
tion conditions with respect to P minerals such as strengite
and vivianite (Parkhurst et al. 2003). This further compli-
cates the interpretation of P trends at the site. Both one-
dimensional (1D) streamtube modeling (Stollenwerk
1996) and multidimensional reactive transport modeling
incorporating field P mass distribution and plume geo-
chemistry (Parkhurst et al. 2003) have been undertaken in
an effort to predict future P loading to the lake.
The current study was initiated specifically to
address the possibility of secondary P attenuation in sep-
tic system plumes. One of the sites described previously
in our review paper (Long Point, tile bed 2) was selected
for this purpose because of several considerations. A
detailed monitoring network was installed at this site
prior to commissioning of the tile bed in 1990; thus, the
complete history of sewage loading and P plume
52 W.D. Robertson GROUND WATER 46, no. 1: 51–60
development over a 16-year period is available. In 2003/
2004, the existing monitoring network was augmented
with eight additional multilevel piezometer bundles tar-
geting specifically the advancing frontal portion of the P
plume. The septic system is a large flux system servicing
a seasonal use campground and generates a plume that is
easily distinguished from the relatively pristine back-
ground ground water at the site. The plume occurs within
a moderately homogeneous calcareous sand aquifer that
has a limited dispersive capacity; thus, the plume retains
a large core zone that is unaffected by dilution. In the
plume area where PO4 is present, NO3-N concentrations
are consistently greater than 10 mg/L and Fe remains less
than 0.1 mg/L. Thus, the P plume remains within a single
redox environment and the complexities of P fate at redox
boundaries, particularly when Fe concentrations change,
are avoided. Additionally, the seasonally intermittent
nature of the sewage loading at this site provided a mecha-
nism to directly measure the bulk horizontal ground water
velocity, which in many studies is difficult to measure.
The 16-year monitoring history at this site and the
sampling detail that shows the advancing frontal portion of
the P plume are not available in previous studies. This evi-
dence is crucial in establishing the importance of secondary
P attenuation, particularly if these reactions occur slowly.
Site Description
The Long Point septic system services a campground
with approximately 200 overnight campsites and is situ-
ated on the north shore of Lake Erie (Figure 1). A single
comfort station is present, serviced by two separate, simi-
lar-sized tile beds (~290 m2 each). The plume emanating
from the older tile bed (tile bed 1) has been described pre-
viously (Robertson and Cherry 1992). Tile bed 2, which is
the focus of the current study, has been used exclusively
since 1990 except for a 2-year period (1995 to1996) when
flow was diverted back to tile bed 1. The plumes from
both tile beds flow southward within an approximately
5-m-thick calcareous sand aquifer and discharge into Lake
Erie. Nitrate concentrations are high in the shallow zones
of both plumes, but at depth, nitrate is attenuated by deni-
trification. The behavior of NO3 (Robertson and Cherry
1995; Aravena and Robertson 1998) and PO4 (Robertson
et al. 1998; Robertson and Harman 1999) in the tile bed 2
plume has been discussed in previous studies.
Sewage is dosed intermittently to the tile bed
by pumping. Tile bed loading during heavy usage in July
was estimated previously at approximately 10 cm/d
(Robertson and Cherry 1995). A bromide tracer test car-
ried out on tile bed 2 during a lower usage period in
September and October 1990 indicated a loading rate of
approximately 3 cm/d at that time. The park is not open
during the winter season and no sewage loading occurs
from early November until early May each year.
Methods
Site characterization methods including sediment cor-
ing, piezometer installation, ground water sampling, and
Figure 1. Section A to A9 through Long Point tile bed 2 showing stratigraphy, monitoring point locations, septic system plume
location based on Cl concentrations, ground water residence time determined from Br tracer test undertaken in year 1 (1990),
and location of the PO4 zone in year 16 (2006).
tracer testing techniques have been described previously
(e.g., Robertson and Cherry 1992; Aravena and Rob-
ertson 1998). Piezometer bundles each consist of 3 to 13
variable-depth sampling tubes. Section A to A9 through
the center of tile bed 2 contains approximately 170 moni-
toring points (Figure 1), including 47 that were installed
in 2003/2004 (bundles 123 to 138, section B to B9; Fig-
ure 2) targeting specifically the PO4 plume.
Ground water samples were collected using a peristal-
tic pump and were filtered (0.45 lm) in-line prior to
atmospheric exposure. Samples were acidified to pH less
than 2 with HCl immediately after collection for cation,
metals, PO4, and NH4 analyses, or were left untreated
(other ions). Measurement of pH and Eh was completed
in-line, prior to atmospheric exposure, using field porta-
ble meters. Eh readings were checked against Zobell’s
solution (Nordstrom 1977) and pH readings were checked
against buffers of pH 4 and 7.
PO4 was analyzed at the University of Guelph, Ontario,
Soil and Nutrient Laboratory using a colorimetric technique
(cadmium reduction) with a Cobas FaraTM infrared spec-
trometer, which provided a detection limit of 0.01 to 0.05
mg/L P. NO3 and Cl were analyzed by ion chromatogra-
phy in the Earth Sciences Department, University of
Waterloo, which provided a detection limit of less than
0.05 mg/L for NO3-N and less than 0.5 mg/L for Cl.
Analyses of other ions were completed at Maxxam Ana-
lytics Inc., Mississauga, Ontario, using a variety of ana-
lytical techniques.
Sediment acid-extractable Ca, Fe, Al, and Mn con-
tents were analyzed at Activation Laboratories, Ancaster,
Ontario, using an Aqua Regis technique, which involved
extraction with concentrated HCl and HNO3 under heat.
This method liberates cations associated with carbonate,
hydroxide, and sulfide minerals but not silicate minerals.
Sediment readily desorbable or ‘‘plant-available’’ P
content was measured at the University of Guelph, Soil
and Nutrient Laboratory, by leaching with a 0.5 M NaH-
CO3 solution (McBride 1994).
Mineral Saturation Conditions
An interactive version of the chemical speciation
model MINTEQA2 (Allison et al. 1990) using a standard
data base (visual MINTEQ version 2.51, Gustafsson
2006) was used to establish equilibrium solute speciation
and mineral saturation indices (SI).
Simulation of P Migration
Ground water velocities within the local P plume
zone were likely to be relatively uniform. Under these
conditions, an analytical model can be used to simulate
solute transport and has the advantage of being computa-
tionally simple and free of numerical dispersion. The ana-
lytical model CMM (Sudicky et al. 1991) was used to
simulate 1D solute transport along the plume core flow-
path (streamtube model). The model solves the advection-
dispersion equation and considers solute retardation and
first-order decay. The modeling exercise involved assign-
ing the field indicated velocity value (28 m/year, see sub-
sequently) and then calibrating for PO4 retardation (R)
and hydrodynamic dispersion using the field data. Addi-
tional simulations then considered the effect of slow sol-
ute degradation. Degradation in this case is not process
specific but could be representative of any attenuation
W.D. Robertson GROUND WATER 46, no. 1: 51–60 53
Figure 2. Section B to B9, Long Point tile bed 2: (a) year 13 PO4 distribution (September 2003 to May 2004), (b) year 16 PO4
distribution (October 2006), and (c) year 16 NO3 distribution (October 2006).

mechanism that is concentration dependent, such as those
suggested previously to account for secondary P attenua-
tion (e.g., molecular diffusion, slow recystallization of
sorbed P, or slow direct precipitation). Selection of the
source P concentration was considered carefully for this
modeling exercise. The septic tank effluent was sampled
on 15 occasions during 1989 to 2004 and PO4-P concen-
trations remained consistent (6.3 6 1.4 mg/L). At most
septic system sites, however, some P attenuation occurs
abruptly in the first 1 to 2 m of subsurface flow in
response to changing redox conditions and ensuing rapid
mineral precipitation reactions (Zanini et al. 1998;
Robertson et al. 1998). Thus, it would be inappropriate
for a modeling exercise focused on slower secondary pro-
cesses to choose a starting point that included the initial P
54 W.D. Robertson GROUND WATER 46, no. 1: 51–60
‘‘rapid transformation zone.’’ Consequently, the 2.6-m-
deep piezometer at nest 123 was selected as the source
location for the modeling exercise. This piezometer is
located at the downgradient edge of the tile bed (Fig-
ures 1 and 2), 1.9 m below the infiltration pipes (~0.6 m
below the water table), and samples plume water that is
fully nitrified. Thus, sewage oxidation reactions are
largely completed at this depth. Additionally, elevated
ground water PO4-P concentrations, greater than 1 mg/L,
were measured at this location within 4 months after ini-
tiation of sewage loading in 1990 (Figure 3). For the
modeling exercise, the mean PO4-P concentration mea-
sured in this piezometer during the last three year-16 sam-
pling events (6.4 6 0.7 mg/L, 08-11/06) was used as the
source concentration. This value was similar to the mean

Figure 3. Distribution of phosphorus in the source area
underlying the tile bed: (a) depth profile of sediment ad-
sorbed P content, core 121 year 16 (08/06) and (b) depth pro-
files of ground water PO4 content (years 0.4 to 7, nest 106;
years 13 to 16.5, nest 123).
long-term effluent value of 6.3 6 1.4 mg/L, although
more recent effluent values have indicated a slightly high-
er value of 7.2 6 1.2 mg/L (n ¼ 3, 1995 to 2004). The
reason for the modest amount of P attenuation in the
vadose zone at this site compared to other sites (25% to
99%, Robertson et al. 1998) is uncertain but could be
related to the high sewage loading rate (10 cm/d during
peak use). In any case, the high PO4 concentrations pres-
ent should only increase the propensity for secondary re-
actions to occur along the plume flowpath.
Results
Aquifer Characteristics
Coring undertaken previously determined that the
surficial aquifer is 5 to 6 m thick and consists of silt-free
fine to coarse sand, underlain by a flat lying aquitard of
lacustrine clayey silt. The mean hydraulic conductivity of
the aquifer in the P plume zone is 0.02 cm/s (Table 1).
The aquifer sands are calcareous (6.3 6 0.8 wt % Ca)
and contain substantial amounts of acid-extractable
Al (0.13 6 0.02 wt %), Fe (2.5 6 1.2 wt %), and Mn
(0.21 6 0.03; Table 1). This presumably reflects the pres-
ence of hydroxide minerals, which should contribute to
PO4 sorption. The aquifer also contains trace quantities of
organic carbon (0.15 wt %) and reduced S (0.02 wt %,
probably as biogenic pyrite), both of which are suffi-
ciently labile to promote denitrification in deeper parts of
the aquifer (Robertson and Cherry 1992; Aravena and
Robertson 1998).
Ground Water Velocity
The seasonal nature of sewage loading at the site
provided an opportunity to directly measure the horizon-
tal ground water velocity. During the 6-month period
when no sewage loading occurs (November to April),
precipitation continues to recharge the aquifer, causing
the upper boundary of the plume to migrate progressively
deeper as the season progresses. With the initiation of
sewage loading in May, a new slug of waste water begins
to occupy the shallow water table zone near the tile bed
and then migrates laterally away from tile bed at a rate
reflecting the horizontal ground water velocity. Conse-
quently, the piezometer nests on section B to B9 each
have one to three shallow monitoring points that oscillate
seasonally between being in or out of the plume. The
background ground water (electrical conduction [EC] 300
to 600 lS) is easily distinguished from the plume water
(EC 800 to 1400 lS) and the boundary frequently occurs
over a vertical distance less than the piezometer spacing
(less than 0.5 m). In 2004, detailed EC monitoring was
undertaken on seven occasions, which revealed a rate of
advance of the plume of 40 m/year over the 140-d period
from May 1 to September 17. However, the ground water
velocity was likely higher than average during this period
due to the added sewage loading. If an arbitrary assump-
tion is made that the flow velocity for the remainder of
the year is 50% of the May 1 to September 17 rate (i.e.,
20 m/year), an average annual flow velocity of 28 m/year
is obtained. This estimate is considered more reliable
than previous estimates based on the Darcy equation
(e.g., 30 m/year, Aravena and Robertson 1998), although
values are similar. Additionally, the Br tracer test carried
out in 1990 revealed the vertical migration rate of the
sewage plume below the tile bed (4 cm/d) during a period
of moderate sewage loading in the fall (Figure 1).
Plume Development
Figure 1 shows the current position of the 16-m-long
PO4 zone, within the larger septic system plume. Figure 3
shows depth profiles of PO4 distribution below the tile
bed over the 16-year period beginning with commission-
ing of the tile bed in 1990. During the last decade, PO4-P
concentrations in the source area have remained consis-
tently in the 4 to 7 mg/L range. Figure 2 shows detailed
PO4 distribution within the P plume zone (section B to
B9) in years 13 (2003) and 16 (2006). By year 13, the P
plume had advanced to approximately 15 m from the
edge of the tile bed and by year 16, further advancement
of the P plume was evident. Although denitrification is
active at depth (Aravena and Robertson 1998), Figure 2c
shows that the P plume remains entirely within the
nitrate-rich zone (NO3-N greater than 10 mg/L).
The absence of NH4-N in the plume water (less than
0.1 mg/L) and low dissolved organic carbon (DOC) val-
ues (3 mg/L; Table 2) indicate that the effluent is well
oxidized. Acidity generated from the oxidation reactions
has been buffered by carbonate mineral dissolution as
indicated by increases in Ca and Mg concentrations in the
plume water compared to the effluent (Table 2). Other
plume constituents such as K (17 to 21 mg/L) and Na (34
to 46 mg/L) remain consistent along the plume and are
similar to the effluent values. Although these cations are
also subject to sorption reactions, their retardation factors
in sands are normally small (e.g., R less than 5; Dance
and Reardon 1983; Ceazan et al. 1989) compared to the
large number of pore volumes (~30) that have migrated
through the plume zone of interest here.
W.D. Robertson GROUND WATER 46, no. 1: 51–60 55
 Table 1
Sediment Properties: Hydraulic Conductivity Determined from Grain-Size Analyses of Core 137, 0.8 to 2.5 m
Depth (n ¼ 6); Acid-Extractable Cations (Al, Ca, Fe, and Mn) from Core 128, 1.0 to 1.8 m Depth (n ¼ 5), OC
Content (Aravena and Robertson [1998]), and CEC (Robertson and Cherry [1992])

Hydraulic Conductivity (cm/s) Al (wt %) Ca (wt %) Fe (wt %) Mn (wt %) OC (wt %) CEC (meq/100 g)

Aquifer 0.02 6 0.01 0.13 6 0.02 6.3 6 0.8 2.5 6 1.2 0.21 6 0.03 0.15 5.2

Note: OC ¼ organic carbon; CEC ¼ cation exchange capacity.

Simulation of PO4 Migration
Figure 4 shows model calibrations for the P retarda-
tion factor (R, ground water velocity/solute migration
velocity) and aquifer longitudinal dispersivity (aL) along
the plume core flowpath. Note that for the 2-year period
(1995/1996) when no sewage loading occurred to tile bed
2, a reduced ground water velocity of 20 m/year was
assumed (see previously). This value is 71% of the nor-
mal velocity; thus, this 2-year period is represented as
a shorter interval of only 1.4 years in the modeling exer-
cise. A good fit with the field data is obtained using a P
retardation factor of 37 and an aL value 0.5 m. This

Table 2
Geochemistry of the Septic System Plume with
Increasing Distance from the Tile Bed and SI for
Several Minerals That Could Influence
P Mobility

Piezometer Tank 123–2.0 m 138–1.9 m 136–2.3 m

Distance (m) 0.0 1.3 9.0 13.3

Sample date 10/04 10/06 10/06 10/06
Na (mg/L) 46 34 42 38
K 21 20 18 17
Ca 45 200 190 220
Mg 15 21 21 30
Alkaline 380 320 230 350
(as CaCO3)
SO4 49 54 30 28
Cl 71 55 58 55
SiO2 4.6 12.2 13.0 13.0
PO4-P 5.8 6.97 4.98 2.96
NO3-N 0.08 59 71 68
NH4-N 79 0.1 0.1 0.1
DOC 19.3 3 3 3
Al 0.14 <0.005 <0.005 <0.005
Fe 0.13 0.094 <0.05 <0.05
Mn 0.04 0.039 0.027 0.20
pH 7.3 6.7 6.8 7.0
Eh (mV) 2150 1400 1290 1360
Temperature (
C) 17 15 15 15
SI Values
Calcite 20.09 20.13 20.19 0.23
Siderite 20.68 22.3 <21.9 <22.3
Gibbsite 3.0 <1.6 <1.8 <1.8
Geothite 20.71 5.9 <5.0 <6.0
Ferrihydrite 23.6 3.0 <2.0 <3.0
Rhodocrosite 20.79 1.5 21.6 20.43 Figure 4. Calibration along the plume core flowpath for
Hydoxyapatite 7.4 7.5 7.5 8.1 (a) P retardation factor (R), (b) longitudinal dispersivity
b(Ca)3(PO4)2 0.78 1.1 1.0 1.2 (aL), and (c) P decay rate using 1D analytical model (CMM;
Vivianite 1.3 23.2 <21.9 <24.2 Sudicky et al. 1991). Time is year 16 (11/06). Plume values
Strengite 5.0 2.7 <1.5 <1.9 are means of last three year-16 sampling events (8/06, 10/06,
MnHPO4 2.3 2.0 1.8 2.5 11/06). Sewage value is mean of last three sampling events (10/
95, 10/03, 10/04). Unless otherwise noted, simulations use the
Note: SI values (log [ion activity product/solubility product]) determined base case; ground water velocity 28 m/year, aL ¼ 0.5 m, R ¼
using visual MINTEQA2 (Gustafsson 2006). 37. Monitoring point locations are indicated on Figure 2.

56 W.D. Robertson GROUND WATER 46, no. 1: 51–60

retardation value is typical for P in septic system plumes
in calcareous sands (Robertson et al. 1998) and the aL
value is a typical value for local-scale transport in a mod-
erately heterogeneous sand aquifer (e.g., Borden aqui-
fer, Sudicky 1986). Figure 4c shows the effect of solute
degradation considering the base case simulation (ground
water velocity of 28 m/year, R ¼ 37, aL ¼ 0.5 m, and
constant source concentration of 6.4 mg/L PO4-P). Using
a degradation coefficient with a half-life of less than
about 30 years produces simulated values that are less than
the field values. Figure 5 shows PO4 breakthrough in four
piezometers located at increasing distances from source. In
each case, observed values can be matched only when deg-
radation is absent or occurs very slowly. Thus, moderate-
fast attenuation in the ground water zone is unlikely; how-
ever, very slow attenuation (half-life greater than 30 years)
cannot be ruled out from these data.
Figure 5. Long-term breakthrough of PO4 with increasing
distance from the tile bed: (a) 7.8 m from source (piezometer
130–1.9), (b) 10.7 m (128–2.4), (c) 13.3 m (136–2.3), and (d)
16.4 m (135–1.9). Simulations are for base case considering
several rates of solute decay.
The P plume at this site does not currently discharge
to a surface water course; however, it is likely that nonre-
tarded plume solutes such as Cl are now discharging to
the Lake Erie shoreline, which is located 170 m downgra-
dient. It is of interest to consider longer term simulations
that examine the future arrival of the P plume at the Lake
Erie shoreline. Figure 6a shows that for the base case
with a constant source and no solute decay, the P plume
arrives at the lake after about 200 years and reaches a dis-
charge concentration the same as the source shortly there-
after. However, Figure 6a also shows that even very slow
solute degradation can have an increasingly important
effect on plume development as travel distances increase.
The very slow degradation rate (half-life ¼ 30 years) re-
sults in a much diminished peak concentration of only
0.04 mg/L PO4-P reaching the lake after approximately
250 years.
The normal lifespan of a tile bed is about 20 years;
thus, century-scale simulations with constant loading
may be unrealistic. Figure 6b considers the case where
the source is decommissioned at year 20. Without degra-
dation, longitudinal dispersion alone is not enough to sub-
stantially reduce peak concentrations reaching the lake,
and a period still occurs (~year 200) when P flux to the
lake is about the same as for the constant input case.
Figure 6. Simulation of long-term P migration considering
(a) a constant source and (b) source decommissioned at year
20. Simulations are for base case considering several rates of
solute decay.
W.D. Robertson GROUND WATER 46, no. 1: 51–60 57
However, slow degradation again dramatically reduces P
loading to the lake (Figure 6b).
It should be noted that although increasingly large
dispersion parameters may be encountered as travel dis-
tances increase (Sudicky 1986), and transverse dispersion
will have an increasing opportunity to dilute plume core
concentrations, dispersion is a conservative process that
will not affect overall mass loading to the lake. Thus 1D
simulations such as these still provide useful insight, even
at larger scales.
Mineral Saturation
One of the processes that could contribute to second-
ary attenuation in the ground water zone is the possibility
of kinetically slow P mineral precipitation. Table 2 shows
major ion chemistry and mineral SI for a number of miner-
als that could influence P persistence. The plume water is
near saturation with respect to calcite. Several Fe(III) min-
erals (geothite, ferrihydrite) exhibit supersaturation, includ-
ing the Fe(III)-P mineral strengite; thus, these minerals
could be precipitating in the plume. However, the amount
of Fe in solution is low (less than 0.1 mg/L; Table 2); thus,
the amount of P that could be removed by direct Fe-P min-
eral precipitation appears limited. The plume is highly
supersaturated with respect to hydroxyapatite (SI 7.4 to
8.1), and there is no indication of declining SI values along
the length of the plume (Table 2). Thus, hydroxyapatite is
an ineffective control on P concentrations in the current P
plume zone. Other more soluble Ca-P minerals such as
b(Ca)3(PO4)2 have also been suggested as possible controls
on P concentrations in calcareous water (Moore and Reddy
1994; Baker et al. 1998), but a consistent degree of super-
saturation is also maintained along the plume for this
mineral (SI 1.0 to 1.2; Table 2). Additionally, the Mn-P
mineral MnHPO4 exhibits supersaturation in the plume (SI
1.8 to 2.5) and Mn-P minerals have been implicated as
controls on P concentrations in several studies (Boyle and
Lindsay 1986; Martin et al. 1988). Thus, a number of P
minerals exhibit supersaturation in the plume, although
direct evidence of their occurrence was not established in
this study.
Discussion
The observation of P mass removal at many septic
system sites (e.g., 25% to 99%; Robertson et al. 1998)
demonstrates that secondary processes leading the devel-
opment of stable, less-mobile P pools undoubtedly play
an important role in the fate of P from septic systems.
However, field evidence from this and other sites suggests
that this effect may be limited to a relatively local area
coinciding with the vadose zone where changing redox
conditions promote the rapid precipitation of insoluble
metal P minerals (Zanini et al. 1998). Septic system P that
is not retained in this P rapid transformation zone, but
rather is transported into the ground water zone, will con-
tinue to be retarded by sorption processes but, as this
study has shown, can remain mobile and consequently
has the potential to ultimately impact downgradient sur-
face water bodies. In some jurisdictions (e.g., province of
Ontario), the normal minimum setback distance for septic
58 W.D. Robertson GROUND WATER 46, no. 1: 51–60
systems from lake shorelines is 20 m. This study shows
that in calcareous sands with typical ground water veloci-
ties, secondary P attenuation in the ground water zone
cannot be relied upon to attenuate P over this distance.
However, Ontario has also adopted a more stringent set-
back requirement for some lakes that are considered to be
‘‘at capacity’’ with respect to P loading. At these sites,
a much larger septic system setback distance of 300 m is
required in some cases. Simulations carried out in this
study show that with larger setback distances, possible
very slow secondary attenuation mechanisms have a much
greater opportunity to beneficially effect long-term lake-
shore P loading. For example, if over time, hydroxyapa-
tite equilibrium is eventually reached and other plume
conditions remain the same, MINTEQA2 modeling in-
dicates that a PO4-P concentration of only approximately
0.01 mg/L would result. However, septic system plumes
in calcareous terrain frequently exhibit a high degree of
supersaturation with respect to hydroxyapatite (Robertson
et al. 1998; Ptacek 1998); thus, the ability of this mineral
to control P in septic system plumes appears weak from
the existing evidence.
The absence of secondary P attenuation in the ground
water plume at this site provides insight into the mecha-
nisms that could account for the phenomenon of secondary
attenuation commonly observed in laboratory studies. Pro-
cesses such as molecular diffusion of P through mineral
coatings or into soil microsites should have been active at
the high aqueous phase P concentrations present in the
plume. Although such processes could be occurring at
rates that are slow enough to impart an effect of irrevers-
ibility in laboratory experiments which are relatively short
term (days to months), this study shows that there is no
effect on longer term (multiyear) plume behavior. Thus,
such P would not be permanently immobilized and conse-
quently this process has limited relevance to the fate of P
in field-scale plumes. Likewise, the hypothesized process
of slow recrystallization of sorbed P into insoluble metal
phosphate minerals should have been active at the high
sorbed P concentrations present in the plume. It was not.
Overall, the plume behavior at Long Point supports the
likelihood that any long-term P mass removal that has
occurred is the result of direct mineral precipitation reac-
tions that occurred rapidly and reached equilibrium before
the pore water arrived at the water table.
Prediction of P Migration from Sediment Desorption Tests
The Long Point field data show no evidence of sec-
ondary P attenuation; thus, P migration in the ground
water zone at this site is controlled primarily by revers-
ible sorption reactions. Under these conditions, the well
known ‘‘retardation’’ equation, which assumes reversible
sorption and a linear relationship between the adsorbed
and aqueous phase concentrations (Langmuir isotherm),
should remain useful for predicting P mobility at this site
(Freeze and Cherry 1979):
R ¼ 1 1 ðpb=hÞ Kd ð1Þ
where R ¼ retardation factor (ground water velocity/
solute velocity), pb ¼ sediment bulk density, h ¼ effective
porosity, Kd ¼ distribution coefficient (mass adsorbed/
mass in solution).
To test this possibility, a single sediment core was
retrieved from the source area in year 16 (nest 121; Fig-
ure 1) and selected sediment samples from the P plume
zone were subjected to desorption tests. The desorption
procedure (see previously) liberated only loosely bound
plant-available P, which is the P mass available from
readily reversible sorption sites. Sediment from the P
plume zone (1.9 to 3.2 m depth; Figure 3) had a mean de-
sorbable P content of 31 6 8 mg/kg (n ¼ 5). Considering
an aqueous phase P concentration of 6.4 mg/L (source
value used in the modeling exercise), a Kd value of 4.8
(31/6.4) is obtained. Assuming sediment particle density
of 2.65 g/cc and effective porosity of 0.35, a P retardation
factor of 25 is predicted using Equation 1. This is in good
agreement with the field apparent value of 37 determined
from the modeling study.
Conclusions
The phenomenon of secondary slow or irreversible
sorption of P widely reported in laboratory studies was
not evident during long-term monitoring of P migration
in the Long Point septic system plume. None of the pro-
cesses suggested to account for secondary attenuation,
such as P diffusion into soil particle microsites, slow re-
cystalization of sorbed P to insoluble metal phosphate
minerals, or kinetically slow direct precipitation of P
minerals, was active enough to be observable in the
ground water zone at this site.
A phosphorous Kd value of 4.8 was obtained from
desorption tests on core samples from the source area.
Use of this value with the well-known retardation equa-
tion, which assumes that fast reversible sorption is the
only attenuation mechanism operating, provided an accu-
rate prediction of the P migration rate at the site. This
success opens up the possibility of using simple field tests
to provide estimates of P migration rates at septic system
sites.
The absence of an ‘‘irreversible’’ P attenuation
mechanism at this site has important implications for the
long-term fate of P in septic system plumes. Without a
secondary process, P in the ground water zone remains
ultimately mobile, and thus a downgradient surface water
course, in this case, Lake Erie, will eventually be im-
pacted. Although hydrodynamic dispersion may dampen
peak concentrations arriving downgradient, dispersion by
itself cannot lower mass loading rates. The modeling
exercise did demonstrate, however, that even very slow
secondary attenuation, beyond that which might be evi-
dent in the context of the current 16-year study, could still
provide important P mass reductions over longer travel
distances. Further research should continue to investigate
this possibility.
Similar comprehensive observations of P plume
advancement have not been presented in previous studies,
with the possible exception of the Cape Cod site. None-
theless, the Long Point plume is typical of septic system
plumes generated by domestic waste water, and results
should thus be applicable to many other sites where prop-
erly functioning septic systems occur in sandy calcareous
terrain.
Acknowledgments
The manuscript benefited from comments provide by
S. Panno and one other anonymous reviewer. Access to
the study site and enthusiastic support for the project was
provided by J. Marchington.
References
Allison, J.D., D.S. Brown, and K.J. Nova-Gradac. 1990.
MINTEQA2/PRODEFA2, A geochemical assessment
model for environmental systems: Version 3.0 user’s man-
ual. Athens, Georgia: U.S. EPA.
Aravena, R., and W.D. Robertson. 1998. Use of multiple isotope
tracers to evaluate denitrification in ground water: Study of
nitrate from a large-flux septic system. Ground Water 36,
no. 6: 975–982.
Baker, M.J., D.W. Blowes, and C.J. Ptacek. 1998. Laboratory
development of permeable reactive mixtures for removal of
phosphorus from onsite wastewater disposal systems. Envi-
ronmental Science and Technology 32, no. 15: 2308–2316.
Bleam, W.F., P.E. Pfeffer, S. Goldberg, R.W. Taylor, and R.
Dudley. 1991. A 31P solid state nuclear magnetic resonance
study of phosphate adsorption at the boehmite/aqueous
solution interface. Langmuir 7, no. 8: 1702–1712.
Boyle, F.W., and W.L. Lindsay. 1986. Manganese phosphate
equilibrium relationships in soils. Soil Science Society of
America Journal 50, no. 3: 588–593.
Ceazan, M.L., E.M. Thurman, and R.L. Smith. 1989. Retarda-
tion of ammonium and potassium transport through a con-
taminated sand and gravel aquifer: The role of cation
exchange. Environmental Science and Technology 23, no.
11: 1402–1408.
Chen, Y.S.R., J.N. Butler, and W. Stumm. 1973. Kinetic study of
phosphate reaction with aluminum oxide and kaolinite.
Environmental Science and Technology 7, no. 4: 327–332.
Dance, J.T., and E.J. Reardon. 1983. Migration of contaminants
in groundwater at a landfill: A case study, 5. Cation migra-
tion in the dispersion test. Journal of Hydrology 63, no. 1–2:
109–130.
Dillon, P.J., and F.H. Rigler. 1974. The phosphorus–chlorophyll
relationship in lakes. Limnology and Oceanography 19, no.
5: 767–773.
Enfield, C.G., T. Phan, D.M. Walters, and R. Ellis Jr. 1981.
Kinetic model for phosphate transport and transformation
in calcareous soils: I. Kinetics of transformation. Soil Sci-
ence Society of America Journal 45, no. 6: 1059–1064.
Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Englewood
Cliffs, New Jersey: Prentice-Hall Inc.
Gustafsson, J.P. 2006. Visual MINTEQ version 2.51.
www.lwr.kth.se/English/OurSoftware/vminteq/.
Isenbeck-Schroter, M., U. Doring, A. Moller, J. Schroter, and G.
Matthess. 1993. Experimental approach and simulation of
the retention processes limiting orthophosphate transport in
groundwater. Journal of Contaminant Hydrology 14, no. 2:
143–161.
Li, L., and R. Stanforth. 2000. Distinguishing adsorption and
surface precipitation of phosphate on goethite (a -FeOOH).
Journal of Colloid and Interface Science 230, no. 1:
12–21.
Lijklema, L. 1980. Interaction of orthophosphate with iron (III)
and aluminum hydroxides. Environmental Science and
Technology 14, no. 5: 537–540.
Lookman, R., P. Grobet, R. Merckx, and W.H. Van Riemsdijk.
1997. Application of 31P and 27Al MAS NMR for phos-
W.D. Robertson GROUND WATER 46, no. 1: 51–60 59
 phate speciation studies in soil and aluminum hydroxides:
Promises and constraints. Geoderma 80, no. 3–4: 369–
388.
Mansell, R.S., H.M. Selim, P. Kanchanasut, J.M. Davidson, and
J.G.A. Fiskell. 1977. Experimental and simulated transport
of phosphorus through sandy soils. Water Resources
Research 13, no. 1: 189–194.
Martin, R.R., R.S.C. Smart, and K. Tazaki. 1988. Direct obser-
vation of phosphate precipitation in the goethite/phosphate
system. Soil Science Society of America Journal 52, no. 5:
1492–1500.
McBride, M.B. 1994. Environmental Chemistry of Soils. New
York: Oxford University Press.
Moore, P.A., and K.R. Reddy. 1994. Role of Eh and pH on phos-
phorus geochemistry in sediments of L. Okeechobee, Florida.
Journal of Environmental Quality 23, no. 5: 955–964.
Nagpal, N.K. 1986. Effect of soil and effluent characteristics on
phosphorus sorption in dosed columns. Journal of Environ-
mental Quality 15, no. 1:73–78.
Nordstrom, D.K. 1977. Thermodynamic redox equilibria of
Zobell’s solution. Geochimica et Cosmochimica Acta 41,
no. 12: 1835–1841.
Parkhurst, D.L., K.G. Stollenwerk, and J.A. Colman. 2003.
Reactive transport simulation of phosphorus in the sewage
plume at the Massachusetts Military Reservation, Cape
Cod, Massachusetts. USGS Water-Resources Investigation
Report 03-4017. Northborough, Massachusetts: USGS.
Ptacek, C.J. 1998. Geochemistry of a septic-system plume in
a coastal barrier bar. Point Pelee, Ontario, Canada. Journal
of Contaminant Hydrology 33, no. 3–4: 293–312.
Robertson, W.D., and J.A. Cherry. 1995. In-situ denitrification
of septic-system nitrate using reactive porous media bar-
riers: Field trials. Ground Water 33, no. 1: 99–111.
Robertson, W.D., and J.A. Cherry. 1992. Hydrogeology of an
unconfined sand aquifer and its effect of the behavior of
nitrogen from a large flux septic system. Applied Hydro-
geology 1, no. 1: 32–44.
Robertson, W.D., and J. Harman. 1999. Phosphate plume persis-
tence at two decommissioned septic system sites. Ground
Water 37, no. 2: 228–236.
60 W.D. Robertson GROUND WATER 46, no. 1: 51–60
Robertson, W.D., S.L. Schiff, and C.J. Ptacek. 1998. Review of
phosphate mobility and persistence in 10 septic system
plumes. Ground Water 36, no. 6: 1000–1010.
Stollenwerk, K.G. 1996. Simulation of phosphate transport in
sewage-contaminated groundwater, Cape Cod, Massachu-
setts. Applied Geochemistry 11, no. 1–2: 317–324.
Sudicky, E.A. 1986. A natural-gradient experiment on solute
transport in a sand aquifer: Spatial variability of hydraulic
conductivity and its role in the dispersion process. Water
Resources Research 22, no. 13: 2069–2082.
Sudicky, E.A., J.B. Kool, and P.S. Huyakorn. 1991. CMM—A
semi-analytical computer model for simulating ground
water fate and transport of contaminants subject to chained-
decay reactions. Technical report prepared for U.S. EPA
Office of Solid Waste, Washington, DC.
van Riemsdijk, W.H., L.J.M. Boumans, and F.A.M. de Haan.
1984. Phosphate sorption by soils: I. A model for phos-
phate reaction with metal-oxides in soil. Soil Science Soci-
ety of America Journal 48, no. 3: 537–541.
van Riemsdijk, W.H., F.A. Westrate, and G.H. Bolt. 1975. Evi-
dence for a new aluminum phosphate phase from reaction
rate of phosphate with aluminum hydroxide. Nature 257,
no. 5526: 473–474.
Walter, D.A., B.A. Rea, K.G. Stollenwerk, and J. Savoie. 1996.
Geochemical and hydrological controls on phosphorus
transport in a sewage-contaminated sand and gravel aquifer
near Ashumut Pond, Cape Cod, Massachusetts. USGS
Water-Supply paper 2463. Washington, DC: USGS.
Willett, I.R., C.J. Chartres, and T.T. Nguyen. 1988. Migration of
phosphate into aggregated particles of ferrihydrite. Journal
of Soil Science 39, no. 2: 275–282.
Zanini, L., W.D. Robertson, C.J. Ptacek, S.L. Schiff, and
T. Mayer. 1998. Phosphorus characterization in sediments
impacted by septic effluent at four sites in central Canada.
Journal of Contaminant Hydrology 33, no. 3–4: 405–429.
Editor’s Note: The use of brand names in peer-reviewed
papers is for identification purposes only and does not
constitute endorsement by the authors, their employers, or
the National Ground Water Association.