PDF Archive

Easily share your PDF documents with your contacts, on the Web and Social Networks.

Share a file Manage my documents Convert Recover PDF Search Help Contact



J Environ Eng Sci 4, 2005, 369 383 .pdf


Original filename: J Environ Eng Sci 4, 2005, 369-383.pdf
Title: I:\
Author: Judy Riopelle (NRC Ottawa ON) 3031 2000 Jun 14 16:02:34

This PDF 1.5 document has been generated by DVIPSONE 2.2.3 http://www.YandY.com / Acrobat Distiller 6.0.1 (Windows), and has been sent on pdf-archive.com on 01/11/2015 at 18:05, from IP address 71.17.x.x. The current document download page has been viewed 625 times.
File size: 499 KB (15 pages).
Privacy: public file




Download original PDF file









Document preview


369

Polybrominated diphenyl ethers in an
advanced wastewater treatment plant.
Part 2: Potential effects on a unique aquatic
system
Sierra Rayne and Michael G. Ikonomou

Abstract: Concentrations of the mono- through deca-substituted polybrominated diphenyl ether (PBDE) flame retardants
were determined in the aqueous effluent from a tertiary-level wastewater treatment plant (WWTP) that uses post-filtration
ultraviolet light disinfection. The WWTP is located in a semi-arid region of British Columbia. Subsequent limnological
modeling of receiving waters examined the potential long-term effects of various PBDE-loading scenarios on this unique
aquatic system. Over the three decades from 2002 to 2031, total PBDE concentrations in the water column and in
suspended and surficial sediments are expected to increase to >120 pg·L−1 and ∼1 ng·g−1 wet weight, respectively.
Following implementation of a hypothetic halt on PBDE releases into the aquatic system, individual PBDE congener
concentrations in the water column and sediments declined by <35% over the ensuing 17-year modeling period after the
ban, illustrating the potential long-term problem arising from continued PBDE inputs into aquatic systems worldwide. The
results also suggest that PBDEs represent one of the single largest halogenated aromatic loadings to Canadian lakes, rivers,
and streams from wastewaters, and their worldwide use continues to increase exponentially.
Key words: polybrominated diphenyl ethers (PBDEs), flame retardants, municipal wastewater treatment effluent,
contaminant fluxes, limnologic modeling.
Résumé : Les concentrations de produits ignifugeants à base d’éthers diphényliques polybromés monosubstitués à
décasubstitués ont été déterminées dans l’effluent aqueux d’une station d’épuration des eaux usées de niveau tertiaire
munie d’un système de désinfection UV, située dans une région semi-aride de Colombie-Britannique, Canada. La
modélisation limnologique subséquente des eaux réceptrices a permis d’étudier les effets potentiels à long terme de divers
scénarios de charge d’éthers diphényliques polybromés sur ce système aquatique unique. Au cours de trois décennies, de
2002 à 2031, les concentrations totales d’éthers diphényliques polybromés dans la colonne d’eau et dans les sédiments
en suspension et de surface devraient augmenter à respectivement >120 pg·L−1 et ∼1 ng·g−1 , poids humide. Suivant
l’implantation d’une interdiction hypothétique sur la libération d’éthers diphényliques polybromés dans le système
aquatique, les concentrations de congénères individuels d’éthers diphényliques polybromés dans la colonne d’eau et les
sédiments diminuera par moins de 35 % au cours de la période de modélisation de 17 ans suivant l’interdiction, illustrant
le problème potentiel à long terme soulevé par les intrants continus d’éthers diphényliques polybromés dans les systèmes
aquatiques dans le monde. Les résultats suggèrent également que les éthers diphényliques polybromés représentent l’une
des plus importantes charges en hydrocarbures aromatiques halogénés dans les lacs, les rivières et les ruisseaux canadiens
provenant des eaux usées et leur utilisation mondiale continue à s’accroître de manière exponentielle.
Mots clés : éthers diphényliques polybromés, produits ignifugeants, effluent du système d’épuration des eaux usées
municipales, flux de contaminants, modélisation limnologique.
[Traduit par la Rédaction]

Introduction
In the previous paper of this series (Rayne and Ikonomou
2005), we reported on the concentrations and patterns of the
brominated flame retardants, polybrominated diphenyl ethers

(PBDEs), within a tertiary-level municipal wastewater treatment plant (WWTP) that uses post-filtration ultraviolet (UV)
disinfection. The continuity of PBDE mass balance within the
plant demonstrated that PBDEs do not appear to be substantially

Received 17 February 2004. Revision accepted 5 October 2004. Published on the NRC Research Press Web site at http://jees.nrc.ca/ on
2 September 2005.
Sierra Rayne. Department of Chemistry, University of Victoria, Victoria, BC V8W 3V6, Canada.
M.G. Ikonomou.1 Marine Environment and Habitat Science Division, Pacific Region, Institute of Ocean Sciences, Fisheries and Oceans
Canada, 9860 West Saanich Road, P.O. Box 6000, Sidney, BC V8L 4B2, Canada.
Written discussion of this article is welcomed and will be received by the Editor until 31 January 2006.
1

Corresponding author (e-mail: IkonomouM@pac.dfo-mpo.gc.ca).

J. Environ. Eng. Sci. 4: 369–383 (2005)

doi: 10.1139/S04-067

© 2005 NRC Canada

370

degraded by municipal WWTP processes, with the 93% removal of PBDEs from the WWTP influent due to sorption onto
wastewater sludges. However, despite significantly reduced
PBDE concentrations in the WWTP effluent (∼26 ng·L−1 ),
high rates of municipal WWTP effluent discharges result in
potentially large PBDE fluxes into receiving waters. Although
efforts have been made to understand the long-range transport
potential and toxicological effects of PBDEs, little ecological
modeling has been done using available box models for aquatic
systems to examine the potential effects of various PBDE use
and release scenarios. The present study aims to illustrate the
possible effects of PBDE discharges via domestic wastewaters
on the unique receiving water environment of Okanagan Lake
in the semi-arid interior of British Columbia.

Decription of the study area
Lake hydrology
In the Okanagan Basin where the WWTP study took place
(see Fig. 1 in Rayne and Ikonomou (2005) for a map of the
study area), water quality is an important issue owing to the
region’s arid climate and the long hydraulic residence times
(HRTs) of lakes that provide both drinking water and fisheries
habitat. The unique hydrologic character of the Okanagan Basin
has been the subject of extensive study over the past century
because the observed range of lake and stream morphologies
within a single watershed is quite unusual. Of particular interest is the linkage of the main valley lakes. Starting at the
headwaters, the Wood–Kalamalka lakes sub-basin discharges
into a northern arm of Okanagan Lake, the largest of the mainstem lakes. Okanagan Lake subsequently discharges through
Okanagan River into Skaha, Vaseaux, and Osoyoos lakes at the
southern end of the Okanagan Basin, with a total drop in elevation of ∼113 m, from 391.4 m above sea level (ASL) at
Wood Lake to 278.3 m ASL at Osoyoos Lake (Anon. 1974b).
As with most regions of the Columbia River system, the four
major lakes in the Okanagan Basin system (Okanagan, Kalamalka, Skaha, and Vaseaux) have dams to control water flows.
From both a hydrologic and contaminant science perspective,
Okanagan Lake (49◦ 30 –50◦ 22 N, 119◦ 20 –119◦ 45 W, 342 m
ASL) is considered the “master lake” of the Okanagan Basin
system, and its large volume (2.62 × 1010 m3 ) and long HRT
(52–60 years) cause it to attenuate and “store” the effects of
upstream lakes on the downstream portion of the system.
Of particular importance with regard to contaminant loadings, the HRT of the Lake is very long compared with other
lakes around the world (Anon. 1974b). Hydraulic residence
times >5 years are very unusual for lakes; even a very large
lake such as Lake Erie has an HRT of only 2.6 years. Other
lakes of similar size (e.g., Kootenay Lake, 1.6 years; Kamloops
Lake, 60 d; Shuswap Lake, 2.1 years) (Anon. 2003) located in
southern British Columbia have HRTs more than an order of
magnitude less than that of Okanagan Lake, making the lake
particularly interesting in terms of contaminant loadings and
ecological modeling. Furthermore, the long HRT also means

J. Environ. Eng. Sci. Vol. 4, 2005

that changes in the water quality of Okanagan Lake will influence that of Skaha, Vaseaux, and Osoyoos lakes (and the
more heavily populated downstream portion of the Columbia
River) for long periods of time. Although the large volume and
long exchange time of Okanagan Lake is a short-term “boon”
for attenuating anthropogenic impacts, these properties could
also be a long-term “bane” should water quality be allowed to
deteriorate (Anon. 1974b).
Okanagan Lake is a deep (dmax = 242 m, dmean = 76 m),
long (∼113 km), and narrow (wmax = 5.2 km) oligotrophic water body with a perimeter length of ∼270 km (Anon. 1974b; City
of Kelowna 2003c) comprising three major basins with approximately equal drainage areas: (i) the north basin from the “ends”
of Vernon and Armstrong arms south to Kelowna (drainage area
[AD1 ] = 1370 km2 ), (ii) the central basin from Kelowna south
to Squally Point (AD2 = 2341 km2 ), and (iii) the south basin
from Squally Point south to Penticton (AD3 = 1373 km2 ) (Oldham and Kennedy 1974). As recently as 1909, 50 years after
European settlement of the Okanagan Valley, Okanagan Lake
outflow was still controlled by a natural bar (341.3 m ASL) at
Penticton. Over the next 44 years, however, control dams were
constructed at the outlet with progressively lower sill heights
until the present structure was completed in 1953 at 339.75 m
ASL to allow lowering for flood control and summer storage
for irrigation (Andrusak et al. 2000).
Similar to the majority of other interior lakes in the Pacific
Northwest, most water input to Okanagan Lake occurs from
April to June (Truscott and Kelso 1979) during the freshet
season, with maximum surface-water temperatures reaching
>25 ◦ C in late August. There is wide inter-annual variation
in lake inflow, from <100 × 106 to >1000 × 106 m3 ·year−1
during dry and wet years, respectively, with a record inflow of
1400 × 106 m3 ·year−1 in 1997. In addition, most tributaries (by
number, not volume) of Okanagan Lake are ephemeral (no sustained year-round flow) (Andrusak et al. 2000), thereby making
detailed long-term modeling on this system particularly difficult. Although the lake is frequently classified as dimictic (two
periods of isothermal circulation of the complete water column
each year) (Truscott and Kelso 1979), it is more accurately considered monomictic with thermal stratification occurring from
May through October and isothermal circulation taking place
from November through April (City of Kelowna 2003c), except
in unusually cold years when the lake carries a complete ice surface (a less than 1-in-40-year event) (Anon. 1974b). Because
of this summer thermal stratification, the hypolimnetic (<15–
20 m (Andrusak et al. 2000)) temperatures remain at ∼4–6 ◦ C
year-round.
Lake trophic, sediment, and soil characteristics
Based on nutrient concentrations ([PO4 3− ] ≈ 2 µg·L−1 ,
[NO3 − ] ≈ 20 µg·L−1 (Anon. 1974b)) and dissolved oxygen
(DO) data (typically at saturation values at all depths and most
locations throughout the year), Okanagan Lake is classified as
oligo-mesotrophic (Anon. 1974b) with a trophic gradient from
mesotrophic conditions at the north end to an oligotrophic state
© 2005 NRC Canada

Rayne and Ikonomou

in the south (Andrusak et al. 2000). Total phosphorus and nitrogen concentrations have risen steadily over the past three to
four decades, and particularly over the past decade, because of
rapidly increasing population levels. Deep-water surface sediments in Okanagan Lake are made up of approximately equal
proportions of silt (4–63 µm) and clay (<4 µm) and are accumulating at ∼1 mm·year−1 (Anon. 1974b) with organic carbon
contents (presently at 2%–4%) increasing rapidly over the past
century, which — in concert with pollen and nutrient studies
— further testify to the influence of anthropogenic activities in
the region (Anon. 1974b). In addition, soils in the study area
appear to have a low capacity for sorption of hydrophobic contaminants, thus increasing the potential for leachate from septic
fields (of which a substantial portion of the rural region is still
served) to reach local water bodies. The soils tend to be highly
porous, coarse, and composed mainly of gravels, in contrast
to the preferred silt-loams, gravelly sandy loams, and sandy
loams for subsurface wastewater disposal (Anon. 1974c). Of
these soil types, nutrient, and presumably hydrophobic organic
contaminant, retention capacity decreases with increasing sand
and gravel content (Oldham and Kennedy 1974). At the present
time, domestic wastewater discharges to the ground are unknown; the last estimate appears to be 2.95 × 109 L·year−1 in
1974 (Anon. 1974c). Work at that time indicated that the three
most important factors determining the amount of applied nutrients to soil via wastewaters that finally reach surface water
were the depth of unsaturated soil above the water table, the
soil characteristics, and the proximity of septic fields to surface
waters (Oldham and Kennedy 1974). All three factors appear
to favor subsurface contaminant transport to groundwater and
surface-water resources, since most of the population lives on
coarse, sandy, well-drained soils with groundwater tables near
the land surface.
Historical water quality concerns
Declining water quality due to nutrient discharges from point
and non-point sources in the Okanagan Valley main lakes system during the 1960s led to the first joint federal–provincial
government watershed study of its kind in Canada. This work
culminated in a series of limnological studies during the early
to mid-1970s that resulted in major upgrades to regional wastewater treatment systems in the larger communities from the
late 1970s through the early 1980s. Furthermore, extensive
dichlorodiphenyltrichloroethane (DDT) use during the 1960s
and 1970s also demonstrated the high propensity for this system to reflect anthropogenic influences and its susceptibility to
organohalogen contamination. In the early 1970s, it was found
that ∼14% of all fish sampled from the Okanagan Valley lakes
had DDT concentrations >5.0 mg·kg−1 (the allowable limit
at the time for human consumption), with concentrations of
15 to >50 mg·kg−1 DDT found in rainbow trout from Kalamalka Lake (Northcote et al. 1972). Although DDT use has
been eliminated over the ensuing three decades, the effects of
more contemporary contaminants on this aquatic system have
not been investigated extensively.

371

Current threats to water quality and lake fauna
Following these ecosystem-level problems, in 1985 the Province of British Columbia declared the Okanagan Basin an “environmentally sensitive area”, thereby opening up special funding for further control of wastewater discharges. Subsequently,
several smaller communities (e.g., Westbank, Peachland, Summerland) upgraded their WWTPs to tertiary-level treatment and
installed deep-water outfalls in Okanagan Lake (Andrusak et
al. 2000; Northcote 1996). Despite these advances in regional
wastewater treatment and a halt to DDT use, there are still several threats to the water quality of Okanagan Lake. These include continuing use of pesticides in this agricultural region,
pathogens from agricultural and storm-water runoff and WWTP
inputs (a Cryptosporidium parvum outbreak during the summer of 1996 caused 10 000 to 15 000 illnesses in Kelowna),
and the relatively unknown threat from emerging wastewater
contaminants. Moreover, the fisheries resource, a population of
land-locked “kokanee” sockeye salmon (Oncorhynchus nerka),
has declined by ∼70% over the past decade (Andrusak et al.
2000). While changes in trophic structure of Okanagan Lake
and introduction of other species (e.g., Mysis relicta, a freshwater shrimp) that compete with the kokanee salmon for food
have been implicated (Andrusak et al. 2000; Ashley et al. 1999),
little attention has been paid to potential contaminant influences
in the fisheries decline.
Domestic and agricultural water needs also place great demands on the quantity of regional water supplies. The total population of the Okanagan Valley is presently >350 000, a 12.5fold increase since 1935, and some estimates see this figure rising to >1 000 000 by 2020 based on current population growth
rates (Andrusak et al. 2000; Northcote 1996) in this popular
tourist destination (>1 000 000 visitors per year). Three major
population centers on the Canadian side of the Okanagan Basin
are Kelowna (population of 96 288 in 2001), Vernon (population of 33 494 in 2001), and Penticton (population of 30 985 in
2001). Major industries in the Okanagan Valley are associated
directly or indirectly with agriculture and forestry, hence there
is a high economic value on water quantity and quality (City
of Kelowna 2003c). In total, water licenses on Okanagan Lake
grant a diversion volume of 110 × 106 m3 ·year−1 , or ∼23% of
the total lake outflow, of which the City of Kelowna withdraws
∼47 × 106 m3 ·year−1 , mainly for municipal water-supply purposes (Andrusak et al. 2000). Adding to the stress is the high
per capita water consumption in this region, with residents in
the City of Kelowna using >570 L·d−1 , which is well above the
Canadian average of 326 L·d−1 (City of Kelowna 2003b); all
of this water comes from Okanagan Lake. Although the source
drinking water quality is relatively high in terms of conventional
physical, chemical, and biological parameters, no work to date
has looked at the potential effects of emerging contaminants.
Despite large quantities of pesticide use in this agricultural region, none of the 67 pesticides analyzed for in the source water
has been found above detection limits (although the methods
used are not particularly sensitive, with method detection limits
from 50 to 250 µg·L−1 (City of Kelowna 2003a)).An additional
© 2005 NRC Canada

372

threat to the water quality of Okanagan Lake is the large amount
of direct storm-water runoff from Kelowna, and although the
city’s storm water is of typical urban runoff quality, several parameters do exceed aquatic life guidelines (City of Kelowna
2003c).

Methods
The limnological model for Okanagan Lake and the resulting estimated PBDE concentrations dissolved in the water column and adsorbed on suspended sediments were obtained using AQUATOX Release 1.1 (US EPA 2001), with the default
lake model forming the base system. Hydrological and physicochemical properties used in the model are described briefly in
the manuscript and are available as Supplementary material.2
Details on sample collection, analysis, and PBDE concentrations and patterns in the WWTP under study are available in
the companion manuscript to this work (Rayne and Ikonomou
2005).

Results and discussion
Rationale for modeling efforts
Despite known discharges of relatively high PBDE concentrations into marine, fluvial, and lacustrine systems, little modeling work has attempted to examine the effects of continuing PBDE discharges and of changes on discharge regimes
(e.g., use and disposal bans or increasing product demand)
into aquatic systems over the coming decades. For the current
model, WWTP effluent discharges into Okanagan Lake from
the Kelowna WWTP gave a daily flux of 0.76 g·d−1 PBDE
(Rayne and Ikonomou 2005). To examine the potential longterm effects on PBDE concentrations freely dissolved in the water column and adsorbed to suspended sediments over the next
three decades, a simple limnological model of Okanagan Lake
was constructed using available data from the literature, local
professionals, and theAQUATOX and EPI Suite environmentalmodeling software programs. Although not considered in detail here, the large PBDE fluxes into aquatic environments via
wastewater would also present a threat to groundwater and
surface-water supplies through septic tank disposal. Previously,
we have implicated septic fields as important sources of PBDEs
to aquatic systems in rural regions (Rayne et al. 2003a), and although the current proportion of the Okanagan Valley still on
septic fields remains unknown, the number of residents using
subsurface wastewater disposal is likely still in the tens of thousands. Abiotic and biotic degradation of trace organic contaminants in groundwater is typically much slower than that observed in WWTPs (i.e., expected to be negligible for PBDEs,
2

Supplementary data for this article are available on the Web
site or may be purchased from the Depository of Unpublished
Data, Document Delivery, CISTI, National Research Council Canada, Ottawa, ON K1A 0R6, Canada. DUD 4017. For more
information on obtaining material refer to http://cisti-icist.nrccnrc.gc.ca/irm/unpub_e.shtml.

J. Environ. Eng. Sci. Vol. 4, 2005

given the lack of degradation within the WWTP) (Barber et al.
1988; Bouwer and McCarty 1980), and thus their disposal into
low-carbon, permeable aquifers — which occurs in much of the
Okanagan Valley — may present a threat for decades to come to
regional subsurface-water supplies that are either used directly
as drinking water or that discharge into local water bodies.
Population modeling
The first step in model construction was to consider the effects of regional population growth on potential PBDE loadings.
In 2001, the last census year, the population of Kelowna was
96 288, of which 69 734 (65.2%) were served by the WWTP
(Table 1 and references therein). In the model, we considered only the influence of populations living directly along the
lakeshore or whose effluent could be discharged into Okanagan
Lake by traveling <1–2 km through groundwater or a tributary
stream. Because the City of Vernon treats its wastewater using
a secondary trickling filter process and discharges it via a spray
irrigation system located east of the city on lands draining into
a tributary of Kalamalka Lake, the sorption onto the soils of this
agricultural region and the outlet dam on Kalamalka Lake were
expected to prevent significant PBDE transport into the Vernon
Arm of Okanagan Lake from Vernon’s wastewater disposal.
Similarly, the City of Armstrong (population of 4256 in 2001),
located adjacent to Deep Creek at the north end of Okanagan
Lake on Armstrong Arm, ceased discharging wastewater (often
at a low flow dilution ratio of 1:1 with stream flow) into Deep
Creek in 1993 and instead now has a spray irrigation facility
for wastewater disposal (Swain 1993). Although not likely to
contribute to Okanagan Lake PBDE loadings, spray irrigation
of PBDE-containing reclaimed wastewater onto crops and soils
may pose a human and environmental health threat (Janssens
et al. 1997). Furthermore, PBDE loadings from the populations living along Kalamalka, Wood, and Ellison lakes (mostly
within the District Municipality of Lake Country (population of
9267 in 2001), which discharges its wastewater into Wood Lake
after treatment by a secondary biological nutrient removal process) were not considered because the extent of trans-lacustrine
PBDE transport could not be estimated reliably owing to the
control structure at the outlet of Kalamalka Lake that largely
prevents discharge of significant quantities of suspended sediments (to which the majority of PBDEs in the water column are
associated). If the assumed zero PBDE loadings from Vernon,
Armstrong, and Lake Country are incorrect, then PBDE loadings to Okanagan Lake have been underestimated by up to 31%.
This estimate is based on a similar per capita PBDE loading to
wastewater and unimpeded (i.e., 100%) transmission of current
loadings via surficial leaching and drainage and (or) direct input
into Coldstream Creek, Deep Creek, and Wood Lake and subsequent transport into Kalamalka Lake, Vernon Creek, and eventually Okanagan Lake. Given the unlikelihood of such efficient
PBDE transmission, any underestimate of PBDE loadings is
likely much <31%. Penticton, at the southern end of Okanagan
Lake, discharges its tertiary-treated municipal wastewater into
the Okanagan River approximately midway between Okanagan
© 2005 NRC Canada

Rayne and Ikonomou

373

Table 1. Population parameters used in the limnological model for estimating regional contributions to Okanagan Lake
wastewater inputs.
Region

2001 Data

City of Kelowna: population
Total
Served by Kelowna WWTP
Regional District of Central Okanagan: population
City of Summerland: population
Okanagan – Similkameen Regional District: population
Total potential population discharging wastewater directly into Okanagan Lake

96 288a (62.2%)b
69 734c (65.2%)d
154 854e
10 713f (7.5%)g
80 327e
107 001h

Annual wastewater flow in the City of Kelowna
Total
Per capita
Estimated annual wastewater flow into Okanagan Lake

9819 × 106 L·year−1c (26.9 × 106 L·d−1 )
140 806 L·year−1i (385.8 L·d−1 )
15066 × 106 L·year−1

a Statistics Canada (2001a).
b Percentage of the total population residing in the Regional District of Central Okanagan.
c City of Kelowna (2003c).
d Percentage of regional population potentially discharging wastewater directly into Okanagan Lake.
e Province of British Columbia (2003).
f Statistics Canada (2001b).
g Percentage of the total population residing in the Okanagan – Similkameen Regional District.
h Calculated as the sum of the total population in the Regional District of Central Okanagan and in the City of Summerland.
i Calculated as the total annual wastewater flow in the City of Kelowna divided by the population of the City of Kelowna.

and Skaha lakes and thus is not relevant for the model. Stormwater discharges of PBDEs into Okanagan Lake from the communities of Kelowna, Westbank, Peachland, and Summerland,
as well as from regional highways, may also be significant.
To our knowledge, no studies have looked at PBDEs in urban runoff, and although polychlorinated biphenyls (PCBs) are
known to be associated with storm-water runoff, they were not
detected previously in runoff from Kelowna streets in the early
1980s, possibly because detection limits were fairly high at
400 ng·L−1 . Hence, PBDE loadings into aquatic systems from
storm-water runoff cannot yet be estimated reliably. Although
the current percentage of Kelowna served by storm sewers is
unknown, it was ∼15%–20% in 1982, with approximately one
half of this area discharging into Okanagan Lake (Swain 1982).
Consequently, PBDE loadings may be underestimated (by a factor difficult to determine) because potential storm-water inputs
have not been considered.
Hence, the population potentially discharging PBDE contaminated wastewater into Okanagan Lake is made up of those
within the Regional District of Central Okanagan (RDCO),
which includes all residents of the OkanaganValley along Okanagan Lake from near the relatively unpopulated northern arm
south to Peachland, and the City of Summerland located in the
Okanagan–Similkameen Regional District (OSRD). This results in an estimated discharging population of 103 844 in 2002
(Tables 1 and 2). The projected discharging population to the
year 2031 was then calculated using the projected growth rates
for the RDCO and OSRD, assuming that the relative discharging
populations towards the total populations of these two regional
districts remained constant over this period. Urbanization in the
Okanagan Valley is taking place, and municipalities are steadily

acquiring a greater share of the total regional population versus
rural areas. Thus, PBDE loadings may be underestimated because of the inability to account fully for differential population
growth in urban versus rural regions of the study area.
Population growth in the study area will see the estimated
discharging population increase to ∼169 000 in 2031, a projected increase of 62.4%. Polybrominated diphenyl ether loadings were consequently modeled as increasing in step with population growth, assuming that per capita wastewater production
remains constant at ∼386 L·d−1 over this period, such that
PBDE concentrations in wastewater effluent will remain constant over this period and that the projected increases in PBDE
loadings arise from increasing wastewater volumes, not increasing PBDE concentrations in wastewater. Some variability in per
capita PBDE loadings to wastewater throughout the discharging population is expected, given the variability among WWTP
configurations within the study area. However, all WWTPs in
the region use secondary treatment processes and are expected
to provide an effluent quality with a sufficient degree of similarity to that of Kelowna’s (i.e., approximately the same per capita
wastewater production and PBDE loadings) to validate the modeling approach. Because of this population growth, PBDE
loadings into Okanagan Lake are estimated to increase from
1.06 g·d−1 in 2002 to 1.74 g·d−1 in 2031, with loadings of seven
major individual congeners (BDEs 15, 28, 47, 99, 100, 153,
209) increasing proportionally, assuming the congener profile
in wastewater remains constant over this period.
Lake modeling
Next, a rudimentary limnological model of Okanagan Lake
was constructed using published data and the AQUATOX mod© 2005 NRC Canada

157 239
159 790
162 749
165 770
168 949
172 232
175 757
179 344
182 951
186 535
190 251
193 994
197 777
201 551
205 279
208 995
212 659
216 314
219 967
223 574
227 127
230 684
234 223
237 730
241 199
244 641
248 055
251 437
254 783
258 092

2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031


1.6
3.5
5.4
7.4
9.5
11.8
14.1
16.4
18.6
21.0
23.4
25.8
28.2
30.6
32.9
35.2
37.6
39.9
42.2
44.4
46.7
49.0
51.2
53.4
55.6
57.8
59.9
62.0
64.1

Increaseb
(%)

80 550
80 818
81 268
81 826
82 435
83 412
84 458
85 530
86 613
87 685
88 783
89 903
91 036
92 148
93 246
94 397
95 525
96 645
97 760
98 848
99 893
100 931
101 956
102 964
103 952
104 927
105 884
106 833
107 767
108 681

Populationa

OSRD


0.3
0.9
1.6
2.3
3.6
4.9
6.2
7.5
8.9
10.2
11.6
13.0
14.4
15.8
17.2
18.6
20.0
21.4
22.7
24.0
25.3
26.6
27.8
29.1
30.3
31.5
32.6
33.8
34.9

Increaseb
(%)
103
105
107
109
111
113
115
117
120
122
124
127
129
132
134
137
139
141
144
146
148
151
153
155
157
160
162
164
166
168

844
451
325
246
269
384
655
967
291
601
995
407
845
276
677
075
438
796
151
477
765
055
333
590
822
036
232
406
558
684

Estimated
discharging
population
4.01E+07
4.07E+07
4.14E+07
4.21E+07
4.29E+07
4.37E+07
4.46E+07
4.55E+07
4.64E+07
4.73E+07
4.82E+07
4.92E+07
5.01E+07
5.10E+07
5.20E+07
5.29E+07
5.38E+07
5.47E+07
5.56E+07
5.65E+07
5.74E+07
5.83E+07
5.92E+07
6.00E+07
6.09E+07
6.17E+07
6.26E+07
6.34E+07
6.43E+07
6.51E+07

Average
discharge
volume
(L·d−1 )
1.06
1.08
1.10
1.12
1.14
1.16
1.18
1.21
1.23
1.26
1.28
1.31
1.33
1.36
1.38
1.41
1.43
1.46
1.48
1.50
1.53
1.55
1.58
1.60
1.62
1.65
1.67
1.69
1.71
1.74

Projected
PBDE
loadingc
(g·d−1 )
0.0165
0.0168
0.0171
0.0174
0.0178
0.0181
0.0185
0.0189
0.0193
0.0196
0.0200
0.0204
0.0208
0.0212
0.0216
0.0220
0.0224
0.0228
0.0232
0.0235
0.0239
0.0243
0.0247
0.0250
0.0254
0.0257
0.0261
0.0265
0.0268
0.0272

Projected
BDE15
loading
(g·d−1 )
0.0696
0.0708
0.0721
0.0734
0.0748
0.0763
0.0778
0.0794
0.0810
0.0826
0.0843
0.0859
0.0876
0.0893
0.0909
0.0926
0.0942
0.0958
0.0974
0.0990
0.1006
0.1022
0.1037
0.1053
0.1068
0.1083
0.1099
0.1114
0.1128
0.1143

Projected
BDE28
loading
(g·d−1 )

Note: RDCO, Regional District of Central Okanagan; OSRD, Okanagan – Similkameen Regional District.
a Province of British Columbia (2003).
b Relative to projected 2002 population.
c PBDE inputs to Okanagan Lake calculated assuming PBDE loading per capita to municipal wastewater remains constant until 2031.

Populationa

Year

RDCO

0.247
0.251
0.256
0.261
0.266
0.271
0.276
0.282
0.288
0.293
0.299
0.305
0.311
0.317
0.323
0.329
0.334
0.340
0.346
0.351
0.357
0.363
0.368
0.374
0.379
0.385
0.390
0.395
0.400
0.406

Projected
BDE47
loading
(g·d−1 )
0.128
0.130
0.132
0.135
0.137
0.140
0.143
0.146
0.149
0.151
0.154
0.157
0.161
0.164
0.167
0.170
0.173
0.176
0.179
0.182
0.184
0.187
0.190
0.193
0.196
0.199
0.201
0.204
0.207
0.210

Projected
BDE99
loading
(g·d−1 )
0.0346
0.0352
0.0358
0.0365
0.0372
0.0379
0.0387
0.0395
0.0402
0.0410
0.0419
0.0427
0.0435
0.0443
0.0452
0.0460
0.0468
0.0476
0.0484
0.0492
0.0500
0.0507
0.0515
0.0523
0.0531
0.0538
0.0546
0.0553
0.0560
0.0568

Projected
BDE100
loading
(g·d−1 )
0.0107
0.0108
0.0110
0.0112
0.0115
0.0117
0.0119
0.0122
0.0124
0.0127
0.0129
0.0132
0.0134
0.0137
0.0139
0.0142
0.0144
0.0147
0.0149
0.0152
0.0154
0.0157
0.0159
0.0161
0.0164
0.0166
0.0168
0.0171
0.0173
0.0175

Projected
BDE153
loading
(g·d−1 )

0.240
0.244
0.248
0.253
0.258
0.263
0.268
0.273
0.279
0.284
0.290
0.296
0.301
0.307
0.313
0.319
0.324
0.330
0.335
0.341
0.346
0.352
0.357
0.362
0.368
0.373
0.378
0.383
0.388
0.393

Projected
BDE209
loading
(g·d−1 )

Table 2. Projected regional populations in the Okanagan Basin potentially discharging wastewater directly into Okanagan Lake with the associated PBDE loadings until the year
2031.

374
J. Environ. Eng. Sci. Vol. 4, 2005

© 2005 NRC Canada

Rayne and Ikonomou

375

Table 3. Parameters used for Okanagan Lake in the AQUATOX limnological model.
Parameter
First day of study
Last day of study
Data storage step
Relative error
Latitude
Volume
Surface area
Mean depth
Maximum depth
Maximum length (or reach)
Net runoff on lake basin (without
lake)
Precipitation on lake
Evaporation from lake
Abstraction from lake
Return flow from abstraction
Groundwater inputs to lake
Outflow at Penticton
Flushing rate
Mean water residence time
Average epilimnetic temperature
Epilimnetic temperature range
Average hypolimnetic temperature
Hypolimnetic temperature range
pH
Total suspended solids
Dissolved oxygen
Nitrogen, nitrate+nitrite
Nitrogen, ammonia
Phosphorus, dissolved
Carbon, inorganic
Carbon, organic
Percent particulate OC in suspended
and dissolved detritus
Percent refractory OC in suspended
and dissolved detritus

Symbol

Value

Reference

L
V
Ao
h
hmax
R
QR

1 January 2003
1 January 2031
100 d
0.0010
49.5◦ N
2.62 × 1010 m3
3.48 × 108 m2
76 m
242 m
112.8 km
780 × 106 m3 ·year−1

Anon. 1974a
Anon. 1974a
Anon. 1974a
Anon. 1974a
Anon. 1974a
Andrusak et al. 2000

QP
QE
QA
QRF
QG = QR + QP + QRF −
QE − QA − QO
QO
kw = QO /V
τw = kw−1
Tepi,avg
Tepi,range
Thypo,avg
Thypo,range
pH
TSS
DO
NO3 − -N
NH3 -N
PO4 3− -P
CO2
OC
OCpart

100 × 106 m3 ·year−1
330 × 106 m3 ·year−1
96 × 106 m3 ·year−1
62 × 106 m3 ·year−1
46 × 106 m3 ·year−1

Andrusak et al. 2000
Andrusak et al. 2000
Andrusak et al. 2000
Andrusak et al. 2000

470 × 106 m3 ·year−1
1.79 × 10−2 year−1
55.7 years
14.2 ◦ C
22.4 ◦ C
5.0 ◦ C
1.0 ◦ C
8.00
1 mg·L−1
11 mg·L−1
0.04 mg·L−1
0.01 mg·L−1
0.006 mg·L−1
27.4 mg·L−1
3 mg·L−1
10%

Andrusak et al. 2000

OCrefr

80%

eling software. Modeling took place from 1 January 2003 to
1 January 2031, with a data storage step of 100 d. Values for
all parameters input into the AQUATOX model are presented in
Tables 2–4 and will be discussed only briefly here. Okanagan
Lake lies at ∼49.5◦ N latitude with a volume of 2.62 × 1010 m3 ,
surface area of 3.48 × 108 m2 , mean depth of 76 m, maximum depth of 242 m, and maximum length of 112.8 km (Anon.
1974b). The hydrologic balance for the lake and measured outflow at Penticton of 470×106 m3 ·year−1 (Andrusak et al. 2000)
give a calculated mean water residence time of ∼55.7 years,
within the 52 to 60 year range commonly quoted in the literature
(Anon. 1974b; Ashley et al. 1999; Bryan 1990, 2003). Summer
thermal stratification occurs annually in Okanagan Lake, with
epilimnion temperatures reaching into the mid- to upper-20 ◦ C
range and the hypolimnion remaining at ∼5 ◦ C. During the fall
through spring period, the lake approaches isothermal condi-

Andrusak et al. 2000
Andrusak et al. 2000
Andrusak et al. 2000
Andrusak et al. 2000
Bryan 1990
Bryan 1990
Bryan 1990
Bryan 1990
Bryan 1990
Bryan 1990
Bryan 1990

tions at ∼4–6 ◦ C with complete mixing (Andrusak et al. 2000).
Control runs of the limnological model provided stratification
periods that approximated observed patterns.
Contaminant physicochemical property modeling
Finally, the physicochemical properties for the analytes of
interest were input to the model (Table 4). With no acidic aromatic hydrogens and because the ether linkage cannot be protonated at ambient pH (i.e., pKa <1), the dissociation constant
(Ka ) is not relevant for PBDEs. Water solubility for the seven
congeners from mono- through deca-brominated ranges widely
from 7.90 × 10−1 to 1.30 × 10−8 mg·L−1 , as do values for vapor pressure (7.50 × 10−5 to 4.07 × 10−13 mm Hg (1 mm
Hg = 133.322 4 Pa)), the Henry’s Law constant (4.10 × 10−5
to 4.29 × 10−11 atm·m3 ·mol−1 (1 atm = 101.325 kPa)), and
log Kow (5.48–11.15), attesting to the complexities involved in
© 2005 NRC Canada

5.80
4.14×104
7.17×103
0
0
0
0
0
0
0
0
0f

5.48
2.26×104
7.17×103
0
0
0
0
0
0
0
0
0f

Log Kow
KSD
EA
kan
kaer
kN
kA
kB
kP
kox
Cw,t=0
Cair,t=0

0
3.40×10−12

5.03×10−12

0
0

0

0

0

0

0

7.17×103

2.54×105

6.76

5.21×10−6

2.72×10−7

564.7
NR
3.89×10−2

BDE99

0

0
0

0

0

0

0

0

7.17×103

1.26×105

6.39

1.08×10−5

1.61×10−6

485.8
NR
9.47×10−2

BDE47

0.49×10−12

0

0
0

0

0

0

0

0

7.17×103

1.64×105

6.53

3.79×10−6

2.76×10−7

564.7
NR
5.41×10−2

BDE100

0.19×10−12

0

0
0

0

0

0

0

0

7.17×103

4.64×105

7.08

3.38×10−6

6.65×10−8

643.6
NR
1.67×10−2

BDE153

<0.10×10−12

0

0
0

0

0

0

0

0

7.17×103

1.01×109

11.15

4.29×10−11

4.07×10−13

959.2
NR
1.30×10−8

BDE209

(Strandberg et al. 2001)

(Palm et al. 2002)

(Wania and Dugani 2003)

(Palm et al. 2002; Wania
and Dugani 2003)

(Palm et al. 2002; Wania
and Dugani 2003)
(Palm et al. 2002; Wania
and Dugani 2003)
H = PL /Sw

References

major commercial penta-BDE mixtures (see Rayne and Ikonomou 2002) and thus likely to be produced only via environmental debromination processes from higher brominated congeners (see
Rayne et al. 2003b).
f Initial gas-phase concentrations are assumed to be zero (although they would be more volatile than the other five congeners examined above).

a Not relevant as ether linkage cannot be protonated at ambient pH values.
b Calculated by the AQUATOX program.
cAssumed to be negligible.
dAssumed to be negligle as no reported water concentrations to date with values above detection limits (see Palm et al. 2002).
eValues for BDEs 47, 99, 100, 153, and 209 taken as average of rural sites (Strandberg et al. 2001). No reported concentrations of BDEs 15 and 28 to date; they are minor constituents of the two

1.89×10−5

4.10×10−5

H

Henry’s Law constant
(atm·mol3 ·mol−1 )
Log octanol-water partition
coefficient
Sediment/detritus water partition
coefficient (kg−1 )b
Activation energy for
temperature (cal·mol−1 )
Rate of anaerobic microbial
degradation (d−1 )c
Rate of aerobic microbial
degradation (d−1 )c
Uncatalyzed hydrolysis constant
(mol−1 ·d−1 )c
Acid catalyzed hydrolysis
constant (mol−1 ·d−1 )c
Base catalyzed hydrolysis
constant (mol−1 ·d−1 )c
Photolysis rate (d −1 )c
Oxidation rate constant
(L·mol−1 ·d−1 )c
Initial water concentration
(mg·L−1 )d
Initial gas phase concentration
(g·m−3 )e

1.17×10−5

7.50×10−5

PL

Vapor pressure (mmHg)

BDE28
406.9
NR
3.34×10−1

BDE15
328.0
NRa
7.90×10−1

MW
Ka
Sw

Symbol

Molecular weight (g·mol )
Dissociation constant
Water solubility (mg·L−1 )

−1

Table 4. Physicochemical parameters and initial conditions used for the seven congeners under study in the AQUATOX model.

376
J. Environ. Eng. Sci. Vol. 4, 2005

© 2005 NRC Canada

Rayne and Ikonomou

modeling this contaminant class. Rates of anaerobic and microbial degradation were assumed to be negligible, given the lack of
studies and the high concentrations apparently now building up
in natural systems. Our previous report that involved anaerobic
microbial degradation of two mono- and di-bromo congeners
(BDEs 3 and 15) took place under high biomass and severe
anaerobic conditions more representative of an attached growth
wastewater treatment process than natural systems (Rayne et al.
2003b). Because aerobic microbes do not commonly degrade
halogenated aromatics (Zitomer and Speece 1993), and because
aerobic conditions (DO ≥ 1 mg·L−1 ) prevail year-round in
Okanagan Lake at all depths, biodegradation in the water column can likely be neglected. Some recent computer modeling
(using linear and nonlinear BIOWIN approaches) suggests that
PBDE half-lives in soils and sediments may range from 300 to
3000 h for di-BDEs up to >30 000 h (>3.4 years) for hexaBDEs (Gouin and Harner 2003). However, we emphasize that
these are untested modeling results, and field studies are needed
to calibrate the models on a few select congeners before such
results can be accepted. Anaerobic degradation may take place
in benthic sediments at nontrivial rates. However, our model
sought to examine potential concentrations in suspended and
(aerobic) surficial sediments that may be more important with
regard to contaminant exposure for humans (through ingestion
of contaminated fish and drinking water) and wildlife. Hydrolysis of the aryl–ether or aryl–bromine linkages was also assumed
to be negligible, as would be expected at ambient pH (∼8.0 for
Okanagan Lake). Photodegradation of PBDEs may be an important fate pathway in natural systems in light of the previous work
that shows these compounds are quite reactive after exposure
to UV light (Hua et al. 2003; Raloff 2003; Rayne et al. 2003b);
however, field and laboratory studies on PBDE photolysis are
needed in natural aqueous solvents to test this hypothesis and
to provide preliminary rate constants. Polybrominated diphenyl
ether concentrations in Okanagan Lake at the start of the modeling period were assumed to be negligible because there are, to
our knowledge, no peer-reviewed PBDE concentrations above
detection limits in water samples, although some early results
from Lake Ontario and Lake Michigan suggest PBDE concentrations in the range of 6–158 pg·L−1 (reviewed in Hale et
al. 2003). Initial gas-phase concentrations were taken to be the
average of the values given for rural sites in the Great Lakes
region (Strandberg et al. 2001), with the exception of BDEs
15 and 28, for which ambient gas-phase concentrations have
not been reported, and thus their initial concentrations in the
model were assumed to be zero. These two congeners comprise 1.6% and 6.6% by mass of PBDEs leaving the WWTP,
and they are expected to be more volatile than the other congeners in the model, such that future work should include them
in atmospheric assessments.
Forecast for constant per capita polybrominated
diphenyl ether loadings
Over the three decades from 2002 to 2031, PBDE concentrations (sum of the seven modeled congeners) in the water

377

column are expected to increase to >120 pg·L−1 , with concentrations of the individual congeners ranging from ∼0.2 to
∼70 pg·L−1 (BDEs 15 and 209, respectively) (Fig. 1). Similarly, PBDE concentrations in suspended (and surficial) sediments are expected to increase to ∼1 ng·g−1 wet weight (ww),
with concentrations of the individual congeners ranging from
∼0.005 to ∼0.35 ng·g−1 ww (BDEs 15 and 209, respectively).
Although this modeling approach remains to be validated by
field work in systems where both PBDE loadings and concentrations are known on a decadal time scale, perhaps the more
interesting insight is the relative temporal patterns in water and
suspended sediment concentrations over time, given a constant
input profile. In both the freely dissolved form and suspended
sediments, concentrations reach a pseudo-steady state much
more rapidly for the lower brominated congeners (BDEs 15, 28,
47, 99, 100, and 153) than for the deca-brominated BDE209.
For a particular congener, freely dissolved and suspended sediment concentrations reach steady state at approximately the
same rates, with rates decreasing with increasing bromination:
BDEs15 (∼3–4 years), 28 (∼5–6 years), 47 ( ∼7–8 years), 99
(∼8–12 years), 100 (∼8–14 years), 153 ( ∼10–16 years), and
209 (no apparent steady state). When the individual congener
concentrations are summed, no steady state is reached (after an
early apparent steady state from 8 to 12 years) because BDE209
dominates the congener profile in both the freely dissolved form
and suspended sediments after ∼2010 (see Fig. 2). This predomination and subsequent persistence of the highly brominated PBDEs (e.g., BDE209) in Okanagan Lake over the ensuing years arises because of their more hydrophobic characteristics relative to the lighter members of this contaminant class.
The higher brominated PBDEs reside almost exclusively on suspended and benthic sediments that are not flushed from the lake
system. Because of their negligible water solubility, even compared with the lower brominated PBDEs (e.g., BDE15), which
are also relatively insoluble, the higher brominated congeners
cannot be significantly removed from the lake by water exports.
Thus, there is essentially “no limit” to the concentrations of
these more brominated PBDEs that may build up in aquatic systems (at least not within the 30-year modeling period), whereas
the lower brominated congeners are sufficiently sparingly soluble to facilitate reaching a steady state between flushing via
the water column and inputs from wastewater sources.
Thus, if an approximately constant PBDE source profile over
time is assumed and modeling software is used, the distance
from steady state of an environmental system can be estimated
through an examination that compares the congener profiles observed in various compartments with the predicted profiles. If
a similar PBDE source profile over time is assumed, the higher
brominated congeners gradually become more dominant, such
.
that in the limit as t = ∞, the contribution of BDE209 approaches 100%. Indeed, over the course of the modeling period, the contribution of BDE209 is expected to increase from
15% to 57% in the freely dissolved form and from 15% to
63% in suspended sediments. Corresponding decreases in the
contributions of the mono- through tetra-brominated congeners
© 2005 NRC Canada


Related documents


j environ eng sci 4 2005 369 383
j environ eng sci 4 2005 353 367
j environ sci health 40 2005 535 551
environ sci technol 37 2003 2847 2854
preliminary brett results poster 4sep06
october 2006 brett project update poster 4nov06


Related keywords