Diagnostic Analysis of Annual Phosphorus Loading and Pelagic Primary Production in
Flathead Lake, Montana.
Rathead Lake Clean Lakes Project, Phase One
FLBS Open File Report 132-94
by
Jack A. Stanford, Bonnie K. Ellis, Drury G. Carr, Geoffrey C. Poole,
James A. Craft and Dale W. Chess
Funding Provided by
Water Quality Bureau
Montana Department of Health and Environmental Sciences
Room A-206 Cogswell Building
Helena, Montana 59620
Loren L. Bahls, Project Officer
in cooperation with
Clean Lakes Program
U. S. Environmental Protection Agency
EPA Region 8
999 18th St., Suite 500
Denver, Colorado 80202
David Rathke, Project Officer
April 1, 1994
Reference:
Stanford, J. A., B. K. Ellis, D. G. Carr, G. C. Poole, J. A. Craft and D. W. Chess. 1994.
Diagnostic analysis of annual phosphorus loading and pelagic primary production in
Flathead Lake, Montana. Flathead Lake Clean Lakes Project, Phase One. Open File
Report 132-94. Flathead Lake Biological Station, The University of Montana, Poison,
Montana.
N

-------
f%5~
INTRODUCTION
Flathead Lake, Montana is thought to be one of the most pristine large lakes in the
temperate latitudes of the world. The water column is very transparent due to the paucity of plant
growth nutrients incoming annually to the lake. Secchi disk (15 cm diameter) readings in the fall
and winter usually exceed 15 meters. Owing to rapid increase in human habitation (> 2% per year
since 1970) of the lake's shoreline and catchment basin, sustaining the high quality of water in the
lake is an important public concern (Stanford and Ellis 1988).
Pelagic primary production at a single midlake site was quantified in relation to mass flux
of nitrogen and phosphorus through the lake during the period 1977 - 1993. (Stanford et al. 1992)
showed that primary production increased significantly, which strongly suggested chronic decline
in water quality. Earlier analyses of the data bases (Stanford and Ellis 1988, Stanford et al. 1983)
suggested that pollution from human sources was a primary factor in the observed decline. In an
effort to curtail the phosphorus (P) mass reaching the lake, the State water quality regulatory
agency instituted a phosphorus reduction strategy for the catchment that included P removal from
sewage effluents and a ban on sale of phosphorus containing detergents (Water Quality Bureau
1985). The rationale was that lake wide primary production was determined by availability of
phosphorus and that significant P reduction was achievable at the sewage treatment plants. The
present study was undertaken to further examine inter-annual variation in P loading and pelagic
primary production in a process - response context and to determine if the pelagic site was truly
representative of lake wide conditions.
We report herein time series measures of limnological variables that describe conditions at
six sites compared to the long-term data base at the midlake monitoring station. Inferences about
the utility of the midlake site in describing long term trends in water quality in Flathead Lake are
provided from the intersite comparisons. We also provide preliminary calculations of phosphorus
mass flux through Flathead Lake, based on all data collected during 1977 - 1993. The major
sources (i.e., precipitation on the lake surface, river tributaries, urban sewage discharges) of P are
identified and annual loads from each source are quantified. Special consideration was given to the
2

-------
importance of Ashley Creek as a source of P, because the largest sewage treatment plant in the
catchment discharges into that creek. Storage and release capacity of the sediments in Ashley
Creek were estimated from in vitro studies. Finally, preliminary analyses of annual P load as a
determinant of annual pelagic primary production are provided.
STUDY SITE
Flathead Lake and its catchment basin comprise 18,379 km2 (Figure 1). Seven major
tributaries drain the catchment, but the 3 forks of the Flathead River produce most of the water that
enters Flathead Lake annually (Table 1). About 65% of the annual inflow occurs from mountain
snowmelt during the spring freshet, which generally peaks between 15 May and 10 June in the
mainstem Flathead River at the gauging station near Columbia Falls. Minimum flows generally
occur during the mid-winter as a result of ice formation in the tributaries. The pattern of water flux
is controlled in part by Hungry Horse and Kerr Dams; consequences of hydromanipulation, while
likely very significant, are not discussed herein (but see Stanford and Hauer 1992).
Flathead Lake is deepest on the east shore and relatively shallow on the west side (Figure
2). Owing to the large volume of water stored (Table 2), the lake has a large heat budget and rarely
freezes over in winter. Seasonal cooling and heating (2° - 20°C annual amplitude at the surface)
interacts with very strong coriolis circulation to produce dramatic longshore flow in a
counterclockwise direction. Longshore flow is moderated by the inflow and outflow currents and
complex wind-generated surges and currents. Hence, movement of river water into the lake and
subsequent mixing is very complex, related to temperature and density patterns interacting with
convective (internal) and advective (external) forces .
About 60,000 people live within the catchment basin including the shoreline of the lake and
upstream areas. Ninety five percent of the population lives within the area of the box in Figure 1.
Therefore, a large share of the nutrient pollution that reaches Flathead Lake likely is derived within
that relatively small portion of the drainage. Kalipsell, Whitefish (including the Big Mountain
3

-------
BRITISH
COLUMBIA
MONTANA
to COLUMBIA RIVER
forkJ^
&
&
CLARK FORK
Figure 1. Flathead River Basin, Montana and British Columbia. Major urban centers are shown
by the cross hatching; urban sewage treatment plants that discharge into Flathead Lake exist at K
(Kalispell), C (Columbia Falls), W (Whitefish) and B (Bigfork). Open circles are locations of
field sampling sites. The Flathead Lake Biological Station is located at S. Rectangle includes 95%
of the catchment's population.
4

-------
Table 1. Basin area and discharge characteristics of major tributaries contributing flow through
Flathead Lake (compiled from U.S. Geological Survey records and maps).

Basin
Area
Total
Volumea
Maximum
Flow
Minimum
Flow
Period of
Record*5
Tributary
(km2)
(m3 X 106)
(m3/sec)
(m3/sec)
(yrs)
South Fork
4,307
3,190
1,310
0.2ld
53
North Fork
4,009
2,670
1,960
5.61
50
Middle Fork
2,921
2,630
3,960
4.90
42
Swan
1,881
1,040
252
5.47
29
Stillwater
875
301
123
1.13
29
Whitefish
440
172
45
1.08
30
Ashley Creek^
520
29
~
—
5
Flathead River
at Lake Outlet
18,372
10,500
2,340
0.14^
74
a Average annual discharge
b For calculation of mean total volume
c Data collected by the Flathead Lake Biological Station
d Due to dam closure
winter sports complex), Columbia Falls and Bigfork are sewered communities that discharge
effluent into the lake (Bigfork) or the larger tributaries. Other, much smaller sewage treatment
plants exist in the basin but effluents are irrigated on crop land and assumed not to reach the lake.
The Biological Station is served by a very small STP that discharges tertiary effluent directly into
the lake. Rural areas include homes served by septic systems, and fertilizers are used on some
crop lands. Water quality effects from these diffuse sources are problematic in the lower reaches
of the Stillwater and Whitefish Rivers, Ashley Creek, Stoner Creek (and other small tributaries of
5

-------
STONER CREEK
FUTh£A0 R1VE8
A
SWAN RIVER
QIGFQRK SAY
WOODS BAY
U. MT
aoi-OftCAL STATiOw(D)
lj/yEL\£& BA££K
L
YELLOW SAY
SK1D00 SAY
POLSOM SAY
ftATKEAO RIVER
V
Figure 2. Bathymetry of Flathead Lake, Montana, and sampling sites. The long term sampling
sites are A - E, whereas intersite comparisons were made at G - L.
6

-------
Table 2. Morphometric and hydrologic features of Flathead Lake, Montana, based on
measurements at lake elevation of 879 m above mean sea level.
Maximum length
43.9 km
Maximum width
24.9 km
Shoreline length
301.9 km
Maximum depth
113.0 m
Mean depth
50.2 m
Area
495.9 km2
Volume
23.2 km3
the lake) and the alluvial aquifer of the Flathead River in the Evergreen area (Noble and Stanford
1986, Stanford et al. unpubl.). The lake shoreline is intensively developed with homes served by
septic systems; shoreline sources of sewage breakout from septic systems have been documented
(Hauer 1988). Timber harvest and associated road building have affected large areas of the
catchment and may influence water, sediment and nutrient yield (Hauer 1991, Hauer and Blum
1991, Spencer 1991). Episodic wildfires also generate measurable nutrient loading in small
tributaries and for short time periods (Spencer and Hauer 1991).
Approximately 60 percent of the land mass in the basin is included in Glacier National Park
and various National Forest Wilderness Areas and roadless areas; this land mass also includes the
highest elevations and highest precipitation in the basin. Hence, most of the water that reaches
Flathead Lake is derived from a very pristine land mass and is therefore of very high quality. This
is extremely important in terms of the mass flux of materials in the lake, since the volumetric
renewal time (period required for tributary inflow and lake outflow to equal the volume of the lake
basin) is less than 3 years (Stanford et al. 1983).
7

-------
METHODS
Intersite Comparisons
Six sites (G-L, Figure 2) on Flathead Lake were selected for collection of limnological
data for comparison with the midlake site (E, Figure 2). Locations of sites (Figure 2) were
determined by shipboard global positioning system (GPS):
•	G, Lakeside, 20m depth;
•	H, Midlake North, 90m;
•	I, Painted Rocks, 50m;
•	J, Ross Deep, 30m;
•	K, Skidoo Bay, 60m;
•	L, Yellow Bay, 20m; and,
•	E, Midlake Deep, 110m, the long term monitoring site.
These sites were sampled monthly from August, 1991 - September, 1992 and April -
August, 1993. All work was done from the Biological Station's research vessel, Jessie B. Sites
were sampled within a one or two day period each month.
The sampling protocol at each site was as follows:
•	discrete samples for chemistries (Table 3) at 5 m (or 1 m) and near bottom;
•	a single, integrated (0 - 30 m or near bottom) sample, subsampled for:
1.)	analytical chemistries (Table 3),
2.)	chlorophyll a\
•	duplicate chlorophyll a samples from the depth of maximum fluorescence (as determined in
situ using a shipboard fluorometer);
•	depth profiles using electronic instrumentation to measure relative fluorescence,
photosynthetically-active radiation, specific conductance, pH, temperature, dissolved
oxygen and water clarity (% transmission); and,
•	secchi depth using a standard black and white secchi disk.
8

-------
Precision of the analytical analyses of water samples was determined by ±1 sd of replicated
analyses on individual samples, whereas accuracy was determined by 110% > x > 90% recovery
of a known addition of standard solution to selected samples. These quality control criteria were
tested on approximately 1 out of every 15 samples run in the Freshwater Research Laboratory at
the Biological Station. Analytical performance of the lab personnel was also evaluated about every
6 months by analyses of quality control (unknown concentrations) samples provided by the U.S.
Environmental Protection Agency. These performance evaluations are on file at the Biological
Station. The lab was able to achieve the correct analytical answer over 90% of the time in spite of
the fact that the unknown samples provided by EPA were often 1 or more orders of magnitude
more concentrated than samples routinely received by the lab from collections in the solute-poor
waters of the Flathead Basin; thus, errors almost always resulted from inaccurate dilution of the
unknowns into the lab's normal working ranges, which approach the analytical detection limits of
the various methods. All sample data, laboratory standard curves and quality control information
were archived by G. Poole (FLBS Data Manager) in the Biological Station's data storage and
retrieval system.
Ashley Creek Sediment Assays
The objective was to determine the extent of sorption or uptake, and desorption or release
of soluble forms of phosphorus and nitrogen to and from the bottom sediments of Ashley Creek.
Two experiments were done. In the first experiment, we utilized the commonly employed method
of creating a synthetic equilibration solution or reagent (see Klotz 1988) which was used in
combination with intact cores of sediment, rather than suspending sediments in a flask. In our
second experiment we attempted to better mimic natural conditions by adding filtered creek water
(collected above the STP outfall) to intact cores of sediments from Ashley Creek (below STP
outfall) for same-day incubation.
9

-------
Initial Experiment -13 November, 1992
All materials used in the field and laboratory were thoroughly cleaned using 10%
hydrochloric acid, followed by repeated rinsing with deionized water.
A 24 cm length sediment core was extracted from the bottom of Ashley Creek, downstream
of the Kalispell STP, approximately midstream. The sediment core was collected in a clear 6.4 cm
diameter PVC tube. The sediment column consisted of a fine silty layer of detritus, approximately
2 cm in depth, overlying a 16 cm layer of coarse sandy loam and a 6 cm layer of sand at the
bottom. The bottom of the tube was sealed (watertight) using a 6.4 cm diameter rubber stopper
with a metal screw-on compressor wrapped in parafilm. The tube was filled with stream water
from the same location and capped using a rubber stopper wrapped in parafilm. The sediment core
remained undisturbed and intact throughout collection.
Temperature and oxygen content of the stream water were measured on site. The sediment
core was transported to the laboratory within 45 minutes and immediately placed in an
environmental chamber at 5° C (the temperature of Ashley Geek at the time of core sampling).
The water column above the sediment core was carefully extracted using a peristaltic pump, leaving
50 ml of water in order to minimize disturbance at the sediment/water interface. The creek water
was immediately replaced with 250 ml of prepared reagent by slowly pumping the solution into the
top of the core incubation apparatus. An L-shaped glass tube was used to distribute the water to
the side of the incubation apparatus to reduce disturbance of the upper sediment layer. In order to
mimic nutrient and specific ion concentrations (Ca) in Ashley Creek a reagent was prepared with
deionized, bacteria-free water and was amended with NaHPC>4 (final concentration of 19.2 |ig/l
SRP), buffered with sodium acetate/acetic acid to a pH of 8.07 and CaCl added for a final
concentration of 39 mg/1 Ca. For the period of record 1977-1992, the mean concentration of Ca
and pH in Ashley Creek at the site of core collection was 39 mg/1 and 8.04, respectively; the mean
concentration of SRP in Ashley Creek above the STP was 19.2 |ig/l. We chose a site on Ashley
Creek above the STP which might represent a minimum level of SRP likely to occur in Ashley
Creek (i.e., should the STP remove all SRP from wastewater inflow).
10

-------
Initial incubation of the reagent and the sediment core proceeded for 2 hours, then 200 ml
was removed by peristaltic pump and filtered for subsequent analysis of SRP, NO2/3-N and NH3-
N. Immediately following removal of the water, another 250 ml of the reagent was added to the
core incubation apparatus for continued incubation. The same removal, filtration and addition
procedures were performed 4, 6, 8,10, 17, 22, 32, 52 and 72 hours after the initial addition of the
reagent to the core incubation apparatus.
Second Experiment - 25 March, 1993
Four benthic cores were collected from Ashley Geek in the same vicinity as the initial core
(i.e., below Kalispell STP). The cores were obtained equi-distant from one another in a cross
section of the stream. Each core was approximately 20 cm in length and included a clay layer
below the silt and sandy loam layers. All field preparations were identical to the initial run. A grab
sample from the water column of Ashley Creek was obtained for chemical analysis at the same site
and at the same time of core sampling.
The cores were handled in much the same way as the initial run. However, instead of
using a synthetic solution in the incubations, water collected from Ashley Creek below Smith
Lake, a site well above the Kalispell STP, was filtered for use in the experiment. The incubation
was also extended 34 hours to assess potential equilibration; samples of incubating water were
collected 3, 6, 8, 12, 24, 45, 65, 85, and 106 hours after initiation of the experiment. In an
attempt to diminish any effects from other soiption/desorption variables such as dissolved oxygen
content, pH and ionic strength, additional water was collected from Ashley Creek (same site below
Smith Lake) approximately every 48 hours for use in the incubation (a total of three new batches).
Dissolved oxygen and pH of each of the filtered water samples did not change more than 5%
during the period of use. All cores were incubated at a constant temperature of 4° C (the
temperature of Ashley Creek at the time of core sampling).
The removal and addition of water to the core incubation apparatus and filtration and
analysis of samples were the same as that described for the initial run. Filtered Ashley Creek water
11

-------
samples were utilized according to the following time table: 1) first sample was used for the 3, 6,
8, 12, 24, and 45 hour incubations, 2) second sample was used for the 65 and 85 hour
incubations, 3) third sample was used for the 106 hour incubation.
Long Term Phosphorus Loading and Pelagic Primary Productivity
Monitoring sites (Figs. 1 and 2) where long term phosphorus data were obtained:
•	Ashley Creek below the Kalispell sewage treatment plant outfall (Figure 1);
•	Stillwater River in Evergreen below the confluence of the Whitefish River (Figure 1);
•	Flathead River near Holt (Sportsmen Bridge), the primary upstream tributary (A, Figure 2);
•	Swan River at Bigfork, upstream from the outfall of the sewage treatment plant (B, Figure 2);
•	Flathead Lake at the outlet sill near the Highway 93 bridge in Poison (C, Figure 2);
•	the bulk precipitation collector located on the dock at the Flathead Lake Biological Station (D,
Figure 2);
•	midlake deep (110 m depth) ca. 1 mile west of Yellow Bay Point in a pelagic area of Flathead
Lake (E, Figure 2);
•	and, Stoner Creek near Lakeside (F, Figure 2).
These sites were selected so that all of the major inputs of water and materials into Flathead
Lake could be quantified (e.g., the Kalispell sewage effluent via Ashley Creek) and related to
biophysical dynamics within the lake measured in time series at the midlake deep site and at the
lake outlet. Flow data were obtained from the US Geological Survey (Table 1) except for Ashley
and Stoner Creeks, where we operated continuous recording devices to estimate flow.
Long term biophysical data were collected 6-15 times per year from October, 1977 -
October, 1993. Sampling frequency was increased to twice monthly during spring in years where
complete data sets were obtained. Some years funding limitations prevented collection on the
regular schedule. Hence, we report herein annual estimates based only on those years in which a
full set of data were obtained (i.e., at least 12 monthly samples per year).
12

-------
Biophysical variables were as in Table 3 with some minor modifications over the years.
For example, the lab switched from long light, path spectrophotometry to an auto analyzer for
colorimetric analyses in 1988. However, quality assurance criteria, described above for the
intersite comparisons, remained constant throughout the period of record. Primary productivity
and chlorophyll a were measured at the midlake site only, not on the tributaries. Estimates of
input of phosphorus from the atmosphere on the surface of the lake were obtained from collections
of bulk precipitation at the Biological Station. Duplicated samplers collected wet and dry
deposition and the volume of precipitation was determined by averaging Poison and FLBS
standard NOAA rain gauge volumes as well as the water mass obtained in the collectors.
Mass flux of phosphorus was estimated by summing the daily load values for the water
year. Daily loads at the various sites were determined by multiplying water volume by TP
concentrations. Concentrations for intervals between sampling dates were estimated by averaging
nearest measured concentrations weighted by flow. Periods of high flow in the Flathead River at
Holt were characterized by high total suspended solids (TSS) loads and correspondingly high total
phosphorus concentrations (see Figure 3 for TP to TSS relationship). Ellis and Stanford (1986)
previously showed that only 10% of the phosphorus in samples with high suspended sediment
concentrations was biologically available. Therefore, corrections were applied to all Flathead River
data where total suspended solids exceeded 10 mg/1, which is the point where TP concentration in
Flathead River samples becomes a positive linear function. The same correction was used at other
tributary sites when TSS values exceeded 10 mg/1. Ellis and Stanford (1986) also showed that
>95% of the total phosphorus in bulk precipitation was bioavailable at all times of the year, so no
corrections were applied to those data.
13

-------
300 -T™
250 --
200
3
l-
O
JG
a.
C/J
o
150 --
a.
"3 100
50 --
0
0
50 100 150 200 250 300 350
total suspended solids (mg/1)
400
450
500
Figure 3. Total phosphorus as a linear function of total suspended solids for all data collected in the Flathead
River at Holt near the confluence with Flathead Lake.

-------
Table 3. Biophysical variables and methods used in monitoring water quality in Flathead Lake.
Detection
Variable (units)	Method (references)	limit
Analyses of water samples


phosphorus Gig/l-P)


total
persulfate digestion; modified
0.4

automated ascorbic acid (1)

soluble total
filtration; persulfate dig.;
0.4

mod. auto, ascorbic acid (1)

soluble reactive
filt.; mod. auto, ascorbic acid (1)
0.4
nitrogen (jig/l-N)


total persulfate
persulfate digestion (2);
20.0

auto, cadmium reduction (1)

nitrite + nitrate
auto, cadmium reduction (1)
2.0
ammonia
auto, phenate (1)
0.5
sulfate (mg/l-S04)
ion chromatography (1)
0.05
dissolved silica (mg/l-Si02)
auto, molybdate-reactive silica (1)
0.2
carbon (mg/l-C)


non-dissolved organic
persulfate dig.; infrared C02
0.10

detection (3)

dissolved inorganic
acid liberation; infrared C02
0.10

detection

carbonate alkalinity (mg/l-CaCO^)
titration (1)
0.5
turbidity (NTU)
nephelometry (1)
0.10
soluble total metal
filtration;

Ca; Mg; Na; K (mg/l-AW)
flame atomic absorption (4)
0.10
Biological analvses


chlorophyll a (mg/m^)
acetone extraction (1,5)
1.00
relative fluorescence (units)
continuous flow in situ
0.05

fluorometry (6)

photosynthetically active
submarine/deck quantum meter (7)
0.01
radiation (^Ein stein s/m2/sec)
14C uptake in light and dark

phytoplankton primary

productivity
bottles; acid-bubbling


technique (8)

15

-------
Table 3 (continued).
Physical profiles


temperature (°C)
thermistor (9)
0.15
dissolved oxygen (ppm)
electrode (9)
0.20
pH (units)
electrode (9)
0.1
conductivity (nmhos/cm)
electrode (9)
1.5
secchi depth (m)
secchi disk
NA
UPHA, 1985
2D'Elia et al., 1977
^Menzel and Vaccaro, 1964
^Perkin-Elmer, 1976
^Marker et al., 1980
^Turner Designs, 1981
^Licor 188 integrating quantum meter
^Theodorssen and Bjarnason, 1975; Wetzel and Likens, 1991
^ measured in situ using Hydrolab Surveyor HI
RESULTS AND DISCUSSION
Intersite Comparisons
We observed very little intersite variation in terms of thermal patterns and stratification (e.g.,
Figure 4, additional profiles are given in Appendix 1). No significant differences between sites or depth
categories were detected in depth profiles of pH, specific conductance, percent transmittance and
photosynthetically active radiation, except during spring runoff when turbid river water was present as an
overflow plume. Turbidity was always highest near the river mouth and on the west shore of the lake as
reported in more detail by Stanford et al. (1983). Turbidity also persisted at the Lakeside site owing to
wind resuspension of sediments at that shallow site. A multi-way MANOVA with repeated measures
showed significant differences (p<.05) in turbidity at Lakeside in comparison to the other shallow sites
(i.e., Yellow Bay, Ross Deep and Midlake Deep samples from < 30m; see Table 4). Significant
differences in TSS were also observed between Painted Rocks and Midlake Deep (Table 5). Only those
analysis results where significant differences were found are presented in Tables 4 and 5.
16

-------
Table 4. Results of a multi-way with repeated measures and post hoc Tukey HSD test for all chemical data
collected at 4 sites on Flathead Lake, August, 1991 - September, 1992 and April - August, 1993. The
shallow sites of Lakeside (L), Yellow Bay (Y), Ross Deep (R) and <30m samples from Midlake Deep (M)
were included in the analyses. Significant differences in concentrations between sites, depths (t = top, i =
integrated, b = bottom) or a significance due to an interaction of site and depth are noted. "Bottom"
samples for the Midlake Deep site were from 30m for comparison with the other shallow sites. A
description of the significance is presented by bars which overlay sites or depths with no significant
difference.
Variable	Site	Depth	Site x Depth	Description
Chi a
X


YMLR
NDOC
X


ML YR


X

b i t
NO2/3

X

t i b



X
For R & M: t i b



X
Forb: MRLY
Si02
X


LYMR


X

t i b



X
For R: t i b



X
Forb: YLMR
S04


X
For b: L Y R M



X
Fori: L YMR



X
Fort: LRYM
SP

X

t b i
Turb
X


Y R M L


X

t i b



X
Fort: YRML



X
Forb: YRML
17

-------
Table 5. Results of a multi-way MANOVA with repeated measures and post hoc Tukey HSD test for all
chemical data collected at 4 sites on Flathead Lake, August, 1991 - September, 1992 and April - August,
1993. The deep sites of Midlake North (N), Painted Rocks (P), Skidoo Bay (S) and Midlake Deep (M)
were included in the analyses. Significant differences in concentrations between sites, depths (t = top, i =
integrated, b = bottom) or a significance due to an interaction of site and depth are noted. A description of
the significance is presented by bars which overlay sites or depths with no significant difference.
Variable
Site
Depth
Site x Depth
Description
Alk

X

t i b
NDOC

X

b i t
NO2/3

X

t i b



X
Forb: PSNM
Si02
X


NPSM


X

t i b
S04

X

t i b
TPN

X

t i b
TSS
X


MSNP



X
ForN: i b t



X
For P: i t b



X
For t: M S P N



X
Fori: MSNP
However, during the summer period of thermal stratification, a significant dissolved oxygen sag
developed in the hypolimnion at the Ross Deep site (Figure 5). The deficit was observed all three
summers of the study. This phenomenon occurred no where else in the lake and, to our knowledge,
dissolved oxygen values significantly less than saturation were never before reported for Flathead Lake.
18

-------
0
10
20
30
40
50
60
70
80
90
100
temperature (deg C)
5.0 10.0 15.0 20.0 25.0
Midlake
Deep
Lakeside
Midlake
North
Painted
Rocks
temperature (deg C)
0.0 5.0 10.0 15.0 20.0 25.0
Midlake
Deep
Ross Deep
Yellow Bay
0
10
20
30
40
50
60
70
80
90
100
4. Depth profiles of temperature in August, 1992 at various sites on Flathead Lake.

-------
6
0
10
20
30
40
50
60
70
80
90
100
dissolved oxygen (mg/1)	dissolved oxygen (mg/I)
7.00 8.00 9.00 10.00 11.00	6.00 7.00 8.00 9.00 10.00 11.00
5. Depth profiles of dissolved oxygen in August, 1992 at various sites on Flathead Lake.

-------
For easy comparison of the P and N variables over the 18 sampling dates, data for depth integrated
samples from the photic zone for each site were plotted in time series with the midlake data (Figures 6 - 8).
The entire data set is provided in Appendix 2. A pronounced nitrate + nitrite spike was observed at
Lakeside and Painted Rocks in June, 1993 (Figure 8), but was likely due to the presence of the river
plume. Summer depletion of nitrate + nitrite occurred uniformly at all sites (Figure 8), except Ross Deep
where the depletion began in April. This event suggested that microbial uptake of nitrate was progressing
at higher rates and for longer periods of time at that site. Site alone was not a significant predictor of the
variation in NO2/3, but for both the shallow and deep sites, depth and site x depth were significant
predictors (Tables 4 and 5). Bottom NO2/3 concentrations were significantly different from top and
integrated values and the interaction of site and depth indicated that for bottom samples, Lakeside, Yellow
Bay and Painted Rocks NO2/3 values were significantly different from Midlake values. No other
statistically significant differences in N or P concentrations were observed between sites. Both ammonium
and soluble reactive phosphorus remained below or very near the detection limits at all sites.
A few other instances of significant deviation from the midlake data were observed at the other
sites. Concentrations of NDOC at Ross Deep were significantly different from Midlake Deep data (Table
4). Dissolved silica concentrations at Lakeside, Yellow Bay, Midlake North and Painted Rocks were also
significantly different from Midlake Deep concentrations; for bottom samples, Ross Deep SiC>2 was
significantly different from Midlake Deep SiC>2 (Tables 4 and 5). The interaction of site and depth
indicated that Lakeside SO4 concentrations were significantly different from Yellow Bay, Ross Deep and
Midlake Deep concentrations at each depth (Table 4).
Higher bioproduction at the Ross Deep site in Big Arm Bay was corroborated by presence of
significantly higher chlorophyll values on nearly all dates (Figure 9; see also Table 4). We think that
plankton fallout from the epilimnion and subsequent microbial decomposition in the hypolimnion may
explain the oxygen deficit noted above. We assume this phenomenon is likely driven by shoreline nutrient
loading in the Big Arm area, since other measurements were consistent with other sites. Entrainment of
the overflow plume of the nutrient-rich spring runoff event may also be a factor in the observed decline in
21

-------
0-
H
14
12
10
8 4
a 6-
4 -
2 --
0
Midlake Deep
Lakeside
Midlake North
Painted Rocks
Aug-91 Dec-91 Apr-92
Aug-92
date
Dec-92 Apr-93 Aug-93
14
12
10
4 -
2 -
0
Midlake Deep
Ross Deep
Skidoo Bay
Yellow Bay
Aug-91 Dec-91 Apr-92 Aug-92 Dec-92 Apr-93 , Aug-93
date
Figure 6. Intersite comparisons of total phosphorus ((J.g/1-P) within the photic zone of the water
column (i.e., integrated samples from 0-30m, or bottom) for the period of record in the clean lakes
diagnostic study.
22

-------
250 -
c 200

-------
70 -r
60 --
50 -
40 -
Z
r3
o
± 30 +
"bi)
20 -1-
10 -
0
Midlake Deep
Lakeside
Midlake North ^
Painted Rocks 11
Aug-91 Dec-91 Apr-92
Aug-92
date
Dec-92 Apr-93 Aug-93
Midlake Deep
Ross Deep
Skidoo Bay
Yellow Bay
Aug-91 Dec-91 Apr-92
Aug-92
date
Dec-92 Apr-93 Aug-93
Figure 8. Intersite comparisons of nitrate+nitrite (jxg/l-N) within the photic zone of the water
column (i.e., integrated samples from 0-30m, or bottom) for the period of record in the clean lakes
diagnostic study.
24

-------
Midlake Deep
date
		Midlake Deep
date
Figure 9. Intersite comparisons of chlorophyll a (jag/1-Chl a) within the photic zone of the water
column (i.e., integrated samples from 0-30m, or bottom) for the period of record in the clean lakes
diagnostic study.
25

-------
water quality. Otherwise, except for a very few instances, the chlorophyll data at other sites were
remarkably similar to midlake.
Ashlev Creek Sediment Assays
Phosphorus and nitrogen loading to Ashley Creek from the Kalispell sewage treatment
plant was very high prior to 1989; thereafter concentrations decreased (Figure 10) in response to
the ban on phosphorus containing detergents and better technology at the treatment plant.
Although no measurements have been made of the P and N content of the Ashley Creek sediments,
they are also believed to be very high. The concern in this portion of the study was that a reduction
in the ambient nutrient concentrations in the water column due to improved removal of phosphorus
and possibly nitrogen by the Kalispell STP may accelerate the desorption of nutrients from the
streambed and offset the gains made by improving the phosphorus removal technology.
Initial Experiment -13 November 1992
The greatest concentrations of SRP, NO2/3-N and NH3-N were observed at the end of the
first 2 hour incubation (Figs. 11 and 12) indicating an initial dramatic release of both nitrogen and
phosphorus from the sediments. SRP concentrations remained above the synthetic water reagent
concentration of 19.2 |ig/L until the 10 hour interval. The reduction of SRP below the level of
added SRP indicated that either adsorption or microbial uptake of SRP occurred 8-10 hours after
addition of the reagent to the sediment core. Release of SRP into the water column increased after
10 hours and continued to steadily increase throughout the experiment. The concentration of SRP
was 15 (J.g/1 higher than reagent concentration at the final 72-hour interval. This discovery led to
an increase in the duration of the second experiment to assess the extent of desorption.
Release of NH3 from the sediments into the water column was extremely high throughout
the incubation (Figure 11). Unlike the SRP incubation, no NH3 was added to the reagent water;
analysis confirmed levels of NH3-N below the detection limits (<0.5 (ig/1) in the reagent water.
26

-------
1400 -r
1200 --
1000 --
800 --
Jan-86

Jan-88
Jan-90
Jan-92
Jan-94
Figure 10. Total phosphorus (mg/l-P) in Ashley Creek downstream from the Kalispell sewage treatment
plant

-------
After only 2 hours of incubation, 3538 |ig/l NH3-N was detected in the reagent water and levels
never dropped below 1500 (ig/1 during the experiment. The release of NH3 decreased after the
initial pulse and did not begin to increase until after the 8-hour interval, similar to the pattern
observed for SRP. NH3-N concentrations were still increasing in the water column at the end of
the incubation, with a final value of 2646 |ig/l measured after 72 hours.
Nitrate+nitrite nitrogen was not added to the reagent solution and analysis confirmed a
concentration below detection limits (<2.0 |ig/l). Approximately 239 |ig/l NO2/3-N was released
from the Ashley Creek sediments during the first 2 hours of the incubation (Figure 12). A rapid
reduction in the release of NO2/3 was observed after 2 hours and concentrations in the water
column were down to 13 |ig/l by 8 hours. From the 17 to 72 hour intervals, NO2/3-N was
concentrations were steady at approximately 14 jo.g/1. Although the intervals of incubation varied
greatly in length (i.e., 2-20 hours), a fairly constant level of NO2/3 was released into solution.
Second Experiment - 25 March 1993
The experimental design for the second core incubation was modified to include an
appropriate equilibration solution by using Ashley Creek water collected above the wastewater
treatment facility. By using filtered creek water collected during the period of core sampling,
ambient concentrations of SRP, NO2/3 and NH3 as well as important sorption variables such as
ionic strength, specific ionic concentrations, dissolved oxygen, and pH were represented. Such a
change in conditions served to dramatically affect the uptake and release of nutrients observed in
the second experiment.
The ambient concentration of SRP in Ashley Creek at the site below the Kalispell STP
during the time of sediment core sampling was 119 jag/1. The ambient concentration of SRP in the
reagent water prepared for incubation (i.e., filtered water from Ashley Creek above the Kalispell
STP) ranged from 6.3 to 7.6 fig/1 (Figure 13). Release of SRP from the sediment cores was
somewhat variable within the first 24 hours . Release of SRP from core A into the reagent creek
water peaked at 147 (ig/1 at 6 hours while core D water peaked at 47 jag/1 at 24 hours. Levels in the
28

-------
hours
hours
Figure 11.. Concentration of SRP (upper graph) and NH3-N (lower graph) in water samples
collected at varying intervals from the water column above a core of Ashley Creek sediments. The
water reagent was amended with 19.2 |ig/l SRP.
29

-------
hours
Figure 12. Concentration of NO2/3-N in water samples collected at varying intervals from the
water column above a core of Ashley Creek sediments.
other two cores peaked at only twice the level present in the reagent creek water. Concentrations
of SRP in water from all cores decreased to concentrations below or near ambient conditions (i.e.,
reagent creek water concentration) as early as 45 hours after initial incubation. SRP concentrations
in cores C and D remained below ambient levels throughout the remainder of the incubation,
indicating possible adsorption or uptake of SRP by the sediment column. Core A concentrations
remained above ambient SRP, while core B concentrations fluctuated below and then above
ambient levels.
The ambient concentration of NH3-N in Ashley Creek below the STP at the time of
sediment core sampling was 167 (ig/1. Reagent creek water collected three times during the period
of incubation indicated quite variable NH3-N levels at the site above the STP, ranging from 103
|ig/l for the first 45 hours, to 83.7 during the next 48 hours, to 52.7 by the 106 hour interval
(Figure 13). During the first 45 hours of the incubation, when the reagent creek water had a high
30

-------
hours
hours
Figure 13. Concentration of SRP (upper graph) and NH3-N (lower graph) in water samples
collected at varying intervals from the water column above 4 cores of Ashley Creek sediments.
The concentration of filtered Ashley Creek water (reagent water) used in the experiment is also
given.
31

-------
ambient level of NH3-N (103 |ig/l), levels of NH3 increased above ambient concentrations in some
core incubations and decreased below ambient in others. After 45 hours, the reagent creek water
used in the incubations contained less NH3 and the response appeared to be less variable, as three
of the four cores exhibited a release of NH3 from the sediments (i.e., 7 to 90 [ig/1 above ambient,
excluding core C at 85 hours). The peak in NH3 release in core C at the 85 hour interval was
likely due to accidental disturbance of the sediment/water interface during extraction of the sample
from the core incubation apparatus.
The ambient concentration of NO2/3 -N in Ashley Creek below the STP at the time of
sediment core sampling was 468 jj.g/1, while the ambient concentration in the reagent water
collected above the STP was significantly lower but did increase throughout the period, ranging
from 39.9 to 57.1 (ig/1 (Figure 14). Concomitant with the increase in NO2/3 concentrations
observed at the upper site was a steady decrease in NH3 (Figure 13). The greatest release of
NO2/3 fr°m sediment cores was observed during the first 3 hours, when NO2/3 -N in the
reagent creek water ranged from 25 to 46 p.g/1 above ambient concentrations. A significant release
of NO2/3 from the sediments continued in all core incubations throughout the 45 hour interval.
Levels of NO2/3 in the core incubations were at or near ambient reagent water concentrations after
45 hours. However, during this period higher NO2/3-N was present in the reagent water. The
slight increase in NO2/3 level in core C at 85 hours may have been due to disturbance of the
sediment/water interface during extraction of the water sample.
Implications of the Sediment Assays
Bottom sediments are a potential source for regulation of soluble nitrogen and phosphorus
concentrations in freshwater streams. Several mechanisms regulate the exchange of nutrients
across the sediment/water interface, which may include the ambient nutrient concentration in both
sediment and water, oxygen-dependent interactions, ion-exchange processes and activities of
bacteria, fungi, invertebrates and algae in both the sediments and the water column. The nitrogen
dynamics of the sediments are not well understood. Several studies have characterized the
32

-------
	a	core A
hours
Figure 14. Concentration of NO2/3-N in water samples collected at varying intervals from the
water column above 4 cores of Ashley Creek sediments. The concentration of filtered Ashley
Creek water (reagent water) used in the experiment is also given.
phosphorus sorption-desorption processes in specific stream sediments (Green et al. 1978, Klotz
1988). Most studies have quantified the phosphate sorption index or the equilibrium phosphorus
concentration by suspending sediments in various solutions over a given time period (usually 1
hour) and measuring the concentration of phosphorus at the end of the incubation. Few studies
have attempted to characterize sorption and release of both nitrogen and phosphorus using intact
streambed cores.
The initial core incubation using a synthetic water reagent indicated that the bottom
sediments from Ashley Creek were capable of releasing significant quantities of soluble
phosphorus and nitrogen. The greatest release of all the nutrients was within the first 2 hours of
the incubation. Although the ambient concentration of SRP in the synthetic reagent was higher
than that in the reagent creek water used in the second incubation (i.e., 19 versus 7 (ig/1), the
sediments of the preliminary experiment continued to release SRP (15 jj.g/1 higher than ambient at
33

-------
72 hours), whereas release and adsorption or uptake of SRP was near equilibrium in the second
incubation within 45 hours. The initial release of SRP was 30 [J.g/1 higher than ambient using the
synthetic reagent, while initial release was as high as 147 (ig/1 using natural stream water. The
dissolved oxygen concentration in the preliminary incubation using the synthetic reagent ranged
from 69% at 0 hours to 34.7% saturation at 72 hours, whereas the dissolved oxygen concentration
in the second incubation was 49.3% (± 5%) saturation. Perhaps the greater reduction in dissolved
oxygen in the preliminary experiment resulted in greater release of SRP from the sediment core.
The effectiveness of the oxidized microzone of the surface layer of bottom sediments in
preventing significant release of phosphorus from the porewaters of the sediments to the overlying
water column has been well documented (Mortimer 1971). The oxidized layer serves as a trap for
manganese, iron and phosphate, and thus greatly reduces transport of materials into the water. At
the sediment surface, a difference of a few millimeters in oxygen penetration is critical in regulating
the exchange of materials between the sediment and water. Oxygen penetration into the sediments
is governed by the oxygen content of the overlying water, turbulent mixing of superficial
sediments and the oxygen demand of the sediments. Adsorbed phosphate, manganese and iron are
quickly mobilized when the redox potential decreases, as the oxidized microzone barrier weakens.
If the core incubations were repeated under anaerobic conditions, it is very likely that enormous
loading of phosphorus would occur. Continuous recording of dissolved oxygen concentrations in
Ashley Creek in combination with additional core incubation experiments under anaerobic
conditions would provide valuable information concerning possible loading of SRP from Ashley
Creek sediments. However, under aerobic conditions the sediments may serve to regulate
concentrations of SRP in the water column. Such regulation has been suggested in other studies
(e.g., Klotz 1988).
The initial core incubations utilizing the synthetic reagent may provide an indication of the
maximum amount of NH3 and NO2/3 that would be released from Ashley Creek sediments. The
synthetic reagent did not contain any NH3 or NO2/3 and it appeared that under these conditions a
massive amount of NH3 was released (3538 (ig/1 initially and 2646 |ig/l at 72 hours). When a high
34

-------
ambient concentration of NH3 was present in the water column, as in the second incubation, the
response was variable; but when the ambient level of NH3 was reduced during that incubation,
more NH3 was released from the sediments (i.e., up to 90 |ig/l).
Although the magnitude was not the same as that observed for NH3, the response of NO2/3
to ambient levels of nutrients in the water column was similar. In the incubation with the synthetic
reagent (no NO2/3), 239 p.g/1 was released in the first 2 hours, then a steady release of about 14
fj.g/1 was observed from 17 to 72 hours. However, in the incubation utilizing natural stream water
(40-57 |ig/l ambient NO2/3) the initial release was only 25-46 (J.g/1 in the first 3 hours. Continual
release of NO2/3 was observed while ambient concentrations in the water column were at their
lowest, but when the levels increased, much less NO2/3 was released. In fact, when ambient
concentrations were at their highest, three of the cores showed loss of NO2/3 fr°m solution.
Much of the nitrogen present in sediments of fresh waters is immobilized and sorbed to
inorganic particles. The porewaters of the sediments usually contain a much higher concentration
of NH3-N and organic N than that of the overlying water. The oxidized microzone of the
sediments is crucial to the solubility and sorption properties of the sediments for ammonia and can
alter the rates of microbial transformations (Wetzel 1983). NH3 can accumulate when appreciable
amounts of organic matter are deposited. Under anaerobic conditions, bacterial nitrification of
NH3 to NO2/3 ceases and with the loss of the oxidized microzone a marked release of NH3 from
the sediments can occur. Nitrate may diffuse into the water column following bacterial nitrification
in well-oxygenated surficial sediments. The massive release of ammonia observed in the
incubations is perplexing if the microzone was indeed oxygenated. Water samples from the core
indicated an oxygenated water column, but lack of circulation as would occur in the creek may
have resulted in a reduction of oxygen at the interface. However, a greater release of SRP would
be expected under such conditions.
The regulation of NH3 and NO2/3 by sediments is poorly understood (Wetzel 1983).
Conditions in a flowing stream are obviously quite different than the conditions presented in this
experiment. The degree of turbulence at the sediment/water interface caused by stream flow would
35

-------
no doubt affect the extent of oxygenation of the surface sediments and thus the many chemical,
physical and biological mechanisms controlling the release, sorption or uptake of nutrients.
Long Term Phosphorus Loading and Pelagic Primary Productivity in Flathead Lake
Over the period of record, pelagic primary production measured annually at the midlake site
increased (Figure 15). Until 1989, the rate of increase was very steep corresponding to at least
two major limnological phenomena. First, note that inputs of P and N from the sewage treatment
plants in the basin were very high during 1977 - 1988 and runoff was below normal most years.
Second, establishment of Mysis relicta in the lake in 1981 and subsequent, rapid expansion of the
population to peak numbers in 1986 and 1987 reduced the crustacean zooplankton to very low
levels (Spencer et al. 1991). Crustacean zooplankton numbers remained low in comparison to pre-
Mysis years and significant shifts in the abundance of zooplankton species occurred, with the
disappearance of two species in 1988 and the appearance of a new species in April 1991 (Chess
and Stanford, unpubl.). Hence, substantially reduced herbivory related to the trophic cascade
caused by the establishment of Mysis may explain the high level of primary production in 1988.
Chlorophyll values reached a maximum in 1988 as well (Figure 16). After 1987, the Mysis
populations declined dramatically and numbers stabilized, possibly in relation to size selective
predation by fishes or a density-dependent effect from low zooplankton numbers. Primary
productivity also declined and stabilized (Figure 15). Indeed, inclusion of the 1993 primary
production data reduced the statistical significance of the linear relation in the long term data to P <
0.1. Prior to 1993 the increasing trend in primary production over the period of record was highly
significant (P < 0.01).
Were the observed dynamics in pelagic primary productivity due to food web effects or
nutrient loading or both? We examined the annual phosphorus flux through the lake for clues to
this question.
36

-------
water year
Figure 15. Mean annual pelagic primary productivity over the period of record at the midlake site (E, Figure 2). Bars
represent minimum and maximum yearly estimates.

-------
3 -r
2.5 -
"eft
(J-)
oo
1.5 -
0.5 -
Apr-85 Apr-86 Apr-87 Apr-88 Apr-89 Apr-90 Apr-91 Apr-92 Apr-93 Apr-94
date
Figure 16. Concentrations of chlorophyll a in the photic zone (0-30m, integrated samples) over the
period of record of Flathead Lake at the midlake site (E, Figure 2).

-------
Daily input of bioavailable phosphorus from the Flathead River was distinctly correlated
with flood flows (Figure 17), as expected. And, a general tendency, although not statistically
significant, for higher TP values in the lake to occur in years of high inflow was observed,
especially in the surface waters (Figure 18).
In all years the Flathead River was the largest contributor of bioavailable phosphorus to
Flathead Lake (Figure 19). However, bulk precipitation was a major source as well (10 - 38% of
the total mass input); and, while total load was clearly related to the Flathead River flow (i.e., total
load was highest in the years of highest runoff), inputs from bulk precipitation were independent
of riverine water yield. In fact, bulk precipitation load was highest in low water years.
Over half of the incoming load of bioavailable phosphorus each year was retained in the
lake (Figure 20). Phosphorus annually retained in the lake is stored in the bottom sediments at the
mud-water interface and is held there by the oxidation equilibrium (Ellis and Stanford 1988), as
noted above for Ashley Creek. Should the water column above the sediments become anoxic,
thereby reversing the redox gradient, a substantial amount of the stored phosphorus would be
released back into the lake and water quality could decline precipitously (Wetzel 1983). This
further underscores the concern for the hypolimnetic oxygen deficit observed in Ross Deep,
although values observed to date do not approach anoxia.
Stoner Creek, the largest shoreline tributary, was not a significant source of bioavailable
phosphorus on a lake wide basis. We made no attempt to add other lake shore sources into the P
mass flux calculations because the data were insufficient for accurate load calculations and the very
small additional loads clearly would not change the observed pattern in annual loading. We also
assumed that ground water input was minimal. However, inputs of labile phosphorus and
nitrogen from shoreline sources, while small in comparison to lake wide mass flux, do produce
localized bursts of bioproduction, particularly in the form of benthic algae mats. Indeed, on
several instances sewage breakouts have been discovered by presence of these mats (Stanford et al.
unpubl.). Shoreline nutrient pollution is a problem in Flathead Lake in some areas, such as Big
Arm Bay (Hauer 1988), and should not be trivialized by the fact that shoreline sources of
39

-------
Figure 17. Mass of biologically active phosphorus discharged daily over the period of record into Flathead Lake
by the Flathead River, based on measures made in the river at the Holt sampling site.

-------
40 T
O
¦ Surface Waters (<30m)
(Mean, S.D. =5.75±2.56)
35 --
o Deep Waters (>30m)
(Mean, S.D. = 6.33±3.82)
30 --
25 --	o
o
Jan-77 Jan-79 Jan-81 Jan-83 Jan-85 Jan-87 Jan-89 Jan-91 Jan-93
Figure 18. Concentrations of total phosphorus observed in the photic and aphotic zones of Flathead Lake
over the period of record. Box shows the long-term means and standard deviations.

-------
250 -r
-r 1400U
200 -
150 --
Of)
G
¦3
C3
o ^
2	«J
3	
® fc.
¦5 ^
Q> a.
 w
•4—1
(/i
3
"O
C3
50 --
0 --
-- 12000


p





sw


KttA5»
A •1"

H


1


1
92
93
10000
8000
- 6000
4000
L- 0
^ s
3 g
S B
5-S.
3 2}
« o
2, ^
<"> N*
c 2
a* »
K* <¦*
o cr
3 ft
M
R «¦
25 w
•© rt>
ft> "1
"• to
<-«* <-~
ft M
65 sc
—i o
- 2000
water year
Figure 19. Mass of biologically available phosphorus by source (histograms) reaching Flathead Lake annually in relation to
annual inflow from the Flathead River (closed squares). S = Swan River, S/W = Stillwater River (below confluence
with the Whitefish River), F = Flathead River (at Holt), A = Ashley Creek (from 1978 - 1988 Ashley Creek values were
included in the Flathead River values because flow data were unavailable for the Creek), P = bulk precipitation, St =
Stoner Creek.

-------
250 -r-
200 --
150

b.
CU
H
50 --I?
0
79
~ Retained TP
H Discharged TP
80
81
83 84 85 86 87
water year
88
89
90
91
92
93
Figure 20. Phosphorus (corrected for bioavailability) mass balance for Flathead Lake (i.e., solid histograms represent annual phosphorus
export whereas the open histograms represent retention in the lake). Asterisk denotes a mass balance calculation which lacked precipitation
data.

-------
bioavailable P are very small in terms of whole lake flux. Moreover, shoreline periphyton growth
in the littoral zone of Flathead Lake is limited by the quantity of phosphorus in the water column
(Bauman 1982, Marks and Lowe 1993) and lake shore residents often express concern over the
very visible "ring around the lake" caused by periphyton growths. Long term data are not available
for shoreline periphyton productivity but should be expected to respond to annual P flux in similar
manner to pelagic primary productivity. An often-expressed complaint by long time shoreline
residents is that the ring around the lake is significantly more visible in the last decade (comments
made at Flathead Lakers annual meetings).
The dramatic reduction in point source loading of bioavailable phosphorus from the urban
sewage treatment plants as a consequence of the P-reduction strategy (Figure 21) seems
particularly important. Indeed, the percent of total load declined from well over 20% in 1978 to
less than 5% since 1989 (Figure 21, top panel). Phosphorus in point source sewage discharges is
95+% bioavailable (Ellis and Stanford 1986). Reducing these sources of P loading may be a
major reason why primary productivity stabilized in recent years. However, note the increase in
total phosphorus loading from the Whitefish STP from 1989 to 1993. Although advances in the
removal of P at the Whitefish STP have greatly reduced TP concentration in the effluent,
population expansion in the Whitefish area has resulted in an increase in the total volume of sewage
and thus the Whitefish STP accounts for about half the TP load from Flathead Valley sewage
treatment plants.
We were unable to demonstrate a statistical relationship between pelagic primary production
and bioavailable phosphorus load on the basis of annual load calculations (Figure 22). However,
if elevated levels of productivity in 1988 were largely a result of the Mysis reducing crustacean
zooplankton that eat algae, a relationship does exist. On the other hand, none of the many
bioassays done in vitro suggest top down (consumer mediated) controls on primary productivity
(Spencer 1992). As noted above with respect to the declining P loads from the sewage treatment
plants, the pattern and timing of water delivery to the lake may also be important. In the 1990's the
operation of Hungry Horse Dam was such that large volumes of water were delivered to the lake
44

-------
during the summer stratification period (Stanford and Hauer 1992). Since the water was
discharged from the bottom of the reservoir, the water mass arriving at Flathead Lake was
significantly colder (7 - 10 C) and more dense than the surface layers of the lake (10 - 12 C).
Hence, the river water had to flow under the epilimnion into the hypolimnion. Since the lake was
full during the summer, considerable water from Kerr Dam had to be released to prevent lake shore
flooding as summer releases from Hungry Horse arrived in the lake. That water had to come from
the epilimnion of the lake. The result could have been enrichment of the hypolimnion which would
produce effects after breakdown of thermal stratification or perhaps even the following spring.
Clearly, the loading relationship is more complicated than can be discerned from annual loading
calculations.
Moreover, primary production in Flathead Lake is influenced by processes other than water
and nutrient flux. Depth and duration of the mixed layer in relation to patterns of thermal
stratification, food web dynamics related to the establishment of opossum shrimp, Mysis relicta,
(Spencer et al. 1991) and seasonal or shorter term dynamics in bioavailability of P, N and
inorganic carbon in relation to photosynthetically active radiation (Dodds et al. 1989, Dodds et al.
1991, Dodds 1989, Ellis and Stanford 1986, Spencer and Ellis 1990, Stanford et al. unpubl. data)
are some of the more important processes that have been examined quantitatively to date. Final
determination of process and response relationships requires synthesis of these interactive
determinants in relation to P loading in both short (seasonal) and long (annual) time scales.
Nonetheless, based on all previous study and inferences of the data presented herein, our
preliminary conclusion is that P flux likely is a primary determinant of pelagic primary production
(and hence water quality), though clearly other biophysical factors are important.
The ultimate goal of the Clean Lakes Program at Flathead Lake is to develop and
implement a plan for management of daily nutrient loads (N and P) from point (e.g., sewage
treatment plants) and non-points (diffuse) sources for the purpose of preventing cultural
eutrophication of Flathead Lake. The phosphorus reduction strategy implemented in 1984 by the
Water Quality Bureau clearly is working (cf. Figure 21). However, volume of sewage reaching
45

-------
25% T
E 2
o c
£ «
^ o*
3
a c
c v
*" E
On -w
t-H «
0
«<— i_
© t-
-*j v
C WD
a> a
Z £
QJ O
D. w
20%
15%
10%
5% -
~	Bigfork
~	Columbia Falls
H Whitefish
¦ Kali spell
0%
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93
water year
~	Bigfork
~	Columbia Falls
H Whitefish
¦ Kalispell
78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93
water year
Figure 21. Mass of bioavailable phosphorus from the urban sewage treatment plants
expressed as percent of the total annual mass input from all sources (top graph). Mass of
bioavailable phosphorus from the urban sewage treatment plants discharged annually
into Flathead Lake or its tributaries (bottom graph).
46

-------
Figure 22. Annual primary productivity (solid squares) plotted in relation to annual input of biologically
active phosphorus from all sources (histograms).

-------
the waste treatment plants and non-point discharges likely is increasing in relation to increasing
human population and may be offsetting the gains made by reducing N and P concentrations in
treatment plant effluents. Data bases at the Flathead Lake Biological Station are currently being
used to more precisely allocate the loads to upstream point and diffuse sources (i.e., Flathead Lake
Total Maximum Daily Load Study).
CONCLUSIONS
•	Limnological parameters vary in similar ways at the midlake site in comparison to other sites,
except Ross Deep, and to a lesser extent, at the shallow Lakeside site. Long term data collected
at the midlake site are representative of the lake as a whole. Continuation of the long term
collection at the midlake monitoring site is critically important to assessment of water quality in
the lake and its tributaries.
•	Parts of the lake, such as Ross Deep, appear to be changing and routine monitoring is
warranted. Other sites are important too because, like Ross Deep, they may show signs of
deterioration before Midlake Deep.
•	Pollution of Ashley Creek by the Kalispell Sewage Treatment Plant has significantly abated as
a result of installation of new treatment technology and the ban on phosphorus containing
detergents, but the sediments of Ashley Creek contain a great deal of potentially exchangeable
phosphorus and nitrogen. Hence, sediment release may somewhat offset improvements in
water quality attributable to better sewage treatment, at least for several years. Maintaining oxic
conditions in the creek (e.g., by eliminating pollution sources above the effluent from the
Kalispell Treatment Plant) is very important because P storage in the sediments of the creek is
accelerated by oxidation equilibrium at the mud-water interface.
•	Bioavailable phosphorus mass from the urban sewage plants declined more than 15% after
1988 as a result of the implementation of the phosphorus reduction plan of the Montana Water
48

-------
Quality Bureau. Reduced discharge of P from the waste treatment facilities in the basin may be
responsible, at least in part, for stabilization of annual primary productivity in recent years.
Shorter term (seasonal) analyses of loading are required to verify a relationship.
Bioavailable phosphorus mass reaching Flathead Lake is largely determined by river flow (i.e.,
annual catchment runoff) and no significant changes have occurred over the period of record
with respect to the allocation of load by tributary.
•	Bulk precipitation is a significant source of bioavailable phosphorus (10 - 38% of the total
mass input) that varies independently of river inflow. Contrary to river input which occurs
predominantly in the spring runoff period, sporadic events producing deposition of bulk
precipitation occur year around (though peak P inputs are common in both spring and fall).
Bioavailable P input via airshed deposition occurs lake wide, while river inflow occurs at the
north end of the lake.
•	Annual input of bioavailable phosphorus alone is not enough to predict primary productivity
with any accuracy. However, algae growth rates do increase in response to added labile
phosphorus in bioassays of Flathead Lake water. Interactions with food web (Mysis
establishment) and other biophysical dynamics (depth and duration of the mixed layer;
photosynthetically active radiation, DOM and DON concentrations in the photic zone,
epilimnetic withdrawal of water by Kerr Dam) may influence primary production measures in
ways that mask lake wide responses to phosphorus loading. Careful analysis of intra-annual
variation is warranted.
49

-------
RECOMMENDATIONS
Based on the results of this Phase I diagnostic study, the following activities are
recommended for protection of the high values of water quality in Flathead Lake and its catchment
basin.
1.	The long term water quality monitoring program for Flathead Lake should be continued
in order to document any new changes in the lake's status and to assist the ongoing TMDL effort.
New monitoring is needed to more accurately demonstrate ground water pollution from septic
systems on the lake shore and the chemistry of bulk precipitation on a lake wide basis (i.e., the one
bulk precipitation site at the Biological Station may be insufficient to accurately determine airshed
sources of nutrient pollution.
2.	Research should be funded to determine the cause of the hypolimnetic oxygen deficit in
Big Arm Bay. Shoreline sources of pollution and the possibility of entrainment of river water in
the bay are possible explanations that should be investigated thoroughly.
3.	The P reduction strategy of the Montana Water Quality Board should continue to guide
management actions in the Flathead Basin to reduce nutrient pollution. Greater attention should be
paid to documentation and reduction of non-point sources, especially with respect to increasing
fertility of the Stillwater and Whitefish Rivers and the Evergreen alluvial aquifer as they flow
through the Flathead Valley and discharge into the Flathead River. Additional water quality
monitoring sites are needed to determine how land use activities are associated with non-point
source inputs into these water bodies. Predictive knowledge of sewage leachates from septic
systems and soil capacities for new systems is badly needed in view of very rapid growth of
homes and businesses outside of the sewer districts in the Flathead Valley.
4.	All federal, state and local statutes that control water quality should be strictly enforced.
5.	Introduction or planting of non-native biota into the waters of the Flathead Basin should
be strictly forbidden.
50

-------
6.	Reduce the full pool level in Flathead Lake one foot to curtail shoreline erosion as
recommended by (Lorang 1993).
7.	Efforts by the Flathead Basin Commission to develop a water quality management
program for the Flathead Basin should be responsive to the conclusions and recommendations of
this study. Control of non-point sources of nitrogen and phosphorus should be emphasized in all
planning processes. The Basin Commission should encourage best management practices for all
land use activities in the basin. The main goal in all cases should be to conduct activities in such
manner to minimize sediment and nutrient inputs to surface and ground waters that reach Flathead
Lake. Use of development setbacks, buffer strips, limitations on fertilizer uses and animal waste
and road runoff management systems (e.g., settling ponds and wetlands to retain sediments and
nutrients) should be emphasized.
51

-------
LITERATURE CITED
American Public Health Association. 1985. Standard methods for the examination of water and
wastewater. (16th ed.). American Public Health Association, Washington, District of
Columbia. 1193 pp.
Bauman, C. H. 1982. Effects of nutrient enrichment and lake level fluctuation on the shoreline
periphyton of Flathead Lake, Montana. Master of Science Thesis, The University of Montana,
Missoula, Montana.
/
D'Elia, C. F., P. A. Steudler and N. Corwin. 1977. Determination of total nitrogen in aqueous
samples using persulfate digestion. Limnology and Oceanography 22:760-764.
Dodds, W. K., K. R. Johnson and J. C. Priscu. 1989. Simultaneous nitrogen and phosphorus
deficiency in natural phytoplankton assemblages: Theory, empirical evidence and implications
for lake management. Lake and Reservoir Management 5(l):21-26.
Dodds, W. K., J. C. Priscu and B. K. Ellis. 1991. Seasonal uptake and regeneration of
inorganic nitrogen and phosphorus in a large oligotrophic lake: size-fractionation and antibiotic
treatment. Journal of Plankton Research 13(6): 1339-1358.
Dodds, W. K. and J. C. Priscu. 1989. Ammonium, nitrate, phosphate, and inorganic carbon
uptake in an oligotrophic lake: seasonal variations among light response variables. Journal of
Phycology 25:699-705.
Ellis, B. K. and J. A. Stanford. 1986. Bioavailability of phosphorus fractions in Flathead Lake
and its tributary waters. Project Completion Report. Open File Report 091-86. U.S.
Environmental Protection Agency, Duluth, Minnesota and Flathead Lake Biological Station,
The University of Montana, Poison, Montana.
Ellis, B. K. and J. A. Stanford. 1988. Nutrient subsidy in montane lakes: fluvial sediments
versus volcanic ash. Verhandlungen der Internationalen Vereinigung fur Theoretische und
Angewandte Limnologie 23:327-340.
Green, D. B., T. J. Logan and N. E. Smeck. 1978. Phosphate adsorption-desorption
characteristics of suspended sediments in the Maumee River Basin of Ohio. Journal of
Environmental Quality 7(2):208-212.
Hauer, F. R. 1988. Study of shoreline sewage leachates in Flathead Lake, Montana. Open File
Report 099-88. Flathead Lake Biological Station, The University of Montana, Poison,
Montana. 23 pp.
Hauer, F. R. 1991. An analysis of the effect of timber harvest on streamflow quantity and
regime: An examination of historical records. Flathead Basin Commission, Kalispell,
Montana. 42 pp. .
Hauer, F. R. and C. O. Blum. 1991. The effect of timber management on stream water quality.
Open File Report 121-91. Flathead Lake Biological Station, The University of Montana,
Poison, Montana.
Lorang, M. S., P. D. Komar and J. A. Stanford. 1993. Lake level regulation and shoreline
erosion in Flathead Lake, Montana: a response to the redistribution of annual wave energy.
Journal of Coastal Research 9(2):494-508.
52

-------
Klotz, R. L. 1988. Sediment control of soluble reactive phosphorus in Hoxie Gorge Creek, New
York. Canadian Journal of Fisheries and Aquatic Science 45:2026-2034.
Marker, A. F. H., E. A. Nusch, H. Rai and B. Riemann. 1980. The measurement of
photosynthetic pigments in freshwater and standardization of methods: Conclusions and
recommendations. Ergebnisse der Limnologie. 14:91-106.
Marks, J. C. and R. L. Lowe. 1993. Interactive effects of nutrient availability and light levels on
the periphyton composition of a large oligotrophic lake. Canadian Journal of Fisheries and
Aquatic Sciences 50(6):1270-1278.
Menzel, D. W. and R. F. Vaccaro. 1964. The measurement of dissolved organic and particulate
carbon in seawater. Limnology and Oceanography 9:138-142.
Mortimer, C. H. 1971. Chemical exchanges between sediments and water in the Great Lakes -
speculations on probable regulatory mechanisms. Limnology and Oceanography 16:387-404.
Noble, R. A. and J. A. Stanford. 1986. Groundwater resources and water quality of the
unconfined aquifers in the Kalispell Valley, Montana. Open File Report No. 177. Montana
Bureau of Mines and Geology. Montana College of Mineral Science and Technology, Butte
Montana. Open File Report 093-86. Flathead Lake Biological Station, The University of
Montana, Poison, Montana.
Perkin-Elmer. 1976. Analytical methods for atomic absorption photometry. Norwalk,
Connecticut.
Spencer, C. N. 1991.. Evaluation of historical sediment deposition related to land use through
analysis of lake sediments. Open File Report 123-91. Flathead Lake Biological Station, The
University of Montana, Poison, Montana. 37 pp.
Spencer, C. N. 1992. Role of Mysis, zooplankton and nutrients in regulation of phytoplankton
populations in Flathead Lake, Montana. Open File Report 127-92. Flathead Lake Biological
Station, The University of Montana, Poison, Montana. 41 pp.
Spencer, C. N. and B. K. Ellis. 1990. Co-limitation by phosphorus and nitrogen, and effects of
zooplankton mortality, on phytoplankton in Flathead Lake, Montana, U.S.A. Verhandlungen
der Internationalen Vereinigung fur Theoretische und Angewandte Limnologie 24:206-209.
Spencer, C. N. and F. R. Hauer. 1991. Phosphorus and nitrogen dynamics in streams during a
wildfire. Journal of the North American Benthological Society 10(l):24-30.
Spencer, C. N., B. R. McClelland and J. A. Stanford. 1991. Shrimp stocking, salmon collapse,
and eagle displacement: cascading interactions in the food web of a large aquatic ecosystem.
Bioscience 41(1):14-21.
Stanford, J. A. and B. K. Ellis. 1988. Water quality: status and trends, pp. 11-32. IN: Our
Clean Water—Flathead's Resource of the Future. Proceedings of a Water Quality Conference,
April 25-26, 1988, Kalispell, Montana. Flathead Basin Commission, Governor's Office,
Helena, Montana. Open File Report 098-88. Flathead Lake Biological Station, Poison,
Montana.
53

-------
Stanford, J. A., B. K. Ellis, D. W. Chess, J. A. Craft and G. C. Poole. 1992. Monitoring water
quality in Flathead Lake, Montana. 1992 Progress Report. Open File Report 128-92.
Flathead Lake Biological Station, The University of Montana, Poison, Montana. 31 pp.
Stanford, J. A. and F. R. Hauer. 1992. Mitigating the impacts of stream and lake regulation in
the Flathead River Catchment, Montana, USA: An ecosystem perspective. Aquatic
Conservation 2:35-63.
Stanford, J. A., T. J. Stuart and B. K. Ellis. 1983. Limnology of Flathead Lake. Flathead
River Basin Environmental Impact Study. U.S. Environmental Protection Agency, Helena,
Montana. Open File Report 076-83. Flathead Lake Biological Station, The University of
Montana, Poison, Montana.
Theodorsson, P. and J. O. Bjarnason. 1975. The acid-bubbling method for primary productivity
measurements modified and tested. Limnology and Oceanography 20:1018-1019.
Turner Designs. 1981. Field Fluorometry: Operating and Service Manual. Mountain View,
California.
Water Quality Bureau. 1984. Strategy for limiting phosphorus in Flathead Lake. Water Quality
Bureau, Montana Department of Health and Environmental Sciences, Helena, Montana.
Wetzel, R. G. 1983. Limnology (2nd Edition). Saunders College Publishing, Orlando, Florida.
Wetzel, R. G. and G. E. Likens. 1991. Limnological Analyses. Springer-Verlag, New York,
NY.
54

-------
APPENDIX 1. Depth profiles of dissolved oxygen and temperature.
Al-1

-------
dissolved oxygen (mg/l)
8.00	9.00	10.00	11.
Sep-91
dissolved oxygen (mg/l)
8.00	9.00	10.00	11.

-------
dissolved oxygen (mg/l)
8.00	9.00	10.00	11.00
Nov-91
dissolved oxygen (mg/l)
8.00	9.00	10.00	11.00

-------
dissolved oxygen (mg/l)
6.50 7.50 8.50 9.50 10.50 11.
Jul-92
dissolved oxygen (mg/l)
6.50 7.50 8.50 9.50 10.50 11.

-------
dissolved oxygen (mg/l)
6.50 7.50 8.50 9.50 10.
Sep-92
dissolved oxygen (mg/l)
6.50 7.50 8.50 9.50 10.50

-------
dissolved oxygen (mg/l)
8.5 9.5 10.5 11.5 12.5
Jun-93
dissolved oxygen (mg/l)
8.5	9.5 10.5 11.5 12.5

-------
dissolved oxygen (mg/l)
8	9	10	11	12
Aug-93
dissolved oxygen (mg/l)
8	9	10	11	12

-------
temperature (deg C)
0.0 5.0 10.0 15.0 20.0
Sep-91
temperature (deg C)
0.0	5.0 10.0 15.0 20.0

-------
temperature (deg C)
3.0 5.0 7.0 9.0 11.0 13.0 15.
Nov-91
temperature (deg C)
3.0 5.0 7.0 9.0 11.0 13.0 15.0

-------
temperature (deg C)
0.0	5.0 10.0 15.0 20.0
Jul-92
temperature (deg C)
0.0 5.0 10.0 15.0 20.0 25.0

-------
temperature (deg C)
0.00 5.00 10.00 15.00 20.00
Sep-92
temperature (deg C)
0.00 5.00 10.00 15.00 20.00

-------
temperature (deg C)
0.0 5.0 10.0 15.0 20.0
Jun-93
temperature (deg C)
0.0	5.0 10.0 15.0 20.0

-------
temperature (deg C)
0.0	5.0 10.0 15.0 20.0
Aug-93
temperature (deg C)
0.0	5.0 10.0 15.0 20.0

-------
APPENDIX 2. Summary of chemical characteristics.
A2-1

-------
Table 1. Chemical characteristics of grab samples from a depth of lm (September 1991 to January 1992) or 5m (February 1992 to August 1993) at Flathead Lake Clean Lakes water
quality monitoring sites. Means, standard deviations (±1.0 S.D.) and ranges (minimum - maximum) are given for n samples.
Site (n)

ADc
Turb
TSS
Si02
S04
SRP
SP
TP
NH3
N02/3
TPN
NDOC

mg/l-CaCC>3
NTU
mg/1
mg/l-Si02
mg/l-S04
Hg/l-P
M-P
Hg/l-P
Hg/l-N
Hg/l-N
Hg/l-N
mg/l-C
Midlake Deep (27)
Mean
84.7
1.2
0.7
4.8
3.05
1.0
2.9
5.1
5.1
26.9
88.1
0.25
S.D.
2.8
3.0
0.4
0.4
0.27
0.7
1.3
1.8
3.2
21.2
35.1
0.07

Min.
80.3
0.2
<0.5
4.0
2.71
<0.4
1.0
3.1
1.2
<2.0
39.4
0.14

Max.
90.2
16.0
1.1
5.5
3.37
3.4
7.4
11.7
14.1
62.2
215.0
0.39
Midlake North (18)
Mean
85.5
1.6
1.1
4.8
3.04
0.9
2.9
5.2
3.6
30.8
101.0
0.24
S.D.
2.9
3.9
0.5
0.3
0.23
0.5
1.1
1.4
1.9
21.1
44.3
0.06

Min.
80.3
0.3
0.5
4.0
2.60
<0.4
1.6
3.2
0.5
<2.5
31.4
0.13

Max.
90.2
17.0
1.6
5.3
3.42
2.6
5.7
7.6
7.3
57.4
236.0
0.38
Lakeside (17)
Mean
84.9
4.2
2.3
4.8
2.99
0.8
3.0
6.3
3.1
31.4
104.7
0.26
S.D.
4.5
13.6
1.7
0.4
0.28
0.5
1.1
2.6
1.4
21.3
41.6
0.06

Min.
73.7
0.4
0.8
4.1
2.60
<0.4
1.1
2.3
<1.2
<2.6
35.5
0.17

Max.
92.4
57.0
5.3
5.5
3.42
2.8
5.8
12.6
6.3
63.8
224.0
0.43
Painted Rocks (18)
Mean
84.7
2.5
1.0
4.8
3.04
0.8
3.4
6.4
2.9
29.6
97.8
0.26
S.D.
3.7
7.6
0.7
0.4
0.20
0.4
2.3
3.7
1.4
20.8
41.1
0.07

Min.
77.0
0.2
<0.5
3.9
2.58
<0.4
1.3
3.2
0.5
<2.6
35.7
0.17

Max.
90.2
33.0
2.2
5.3
3.27
1.6
10.4
19.0
5.4
55.2
213.0
0.44
Ross Deep (18)
Mean
85.3
1.2
0.8
4.8
3.02
0.7
2.7
5.7
2.8
20.3
92.0
0.30
S.D.
3.7
2.5
0.2
0.4
0.16
0.2
1.1
1.4
1.6
16.7
32.5
0.07

Min.
77.0
0.2
<0.5
3.9
2.67
<0.4
1.3
3.3
<0.5
<2.5
26.1
0.15

Max.
90.2
11.0
0.9
5.7
3.30
1.4
5.2
8.0
7.4
53.7
195.0
0.40
Yellow Bay (18)
Mean
85.5
0.6
0.6
4.8
3.04
0.7
2.4
5.4
2.7
28.0
93.3
0.25
S.D.
3.1
0.3
0.1
0.3
0.15
0.2
0.9
1.8
1.3
20.9
31.7
0.11

Min.
80.3
0.3
<0.5
4.0
2.79
<0.4
1.0
2.9
<1.2
<2.5
20.8
0.14

Max.
89.7
1.2
0.8
5.2
3.30
1.0
4.7
9.5
5.8
59.5
140.0
0.59
Skidoo Bay (17)
Mean
85.6
1.3
0.8
4.8
3.03
0.7
2.8
5.5
3.0
29.0
96.3
0.25
S.D.
3.2
3.0
0.2
0.4
0.20
0.2
1.2
1.9.
1.6
20.4
34.2
0.11

Min.
80.3
0.3
0.6
3.9
2.56
<0.4
1.6
3.2
<1.4
<2.5
33.3
0.07

Max.
90.2
13.0
1.0
5.6
3.27
1.3
6.3
10.6
7.0
59.2
154.0
0.57
Osprey Nest (1)

79.2
0.5

4.8
2.93
0.5
4.2
5.1
1.5
<5.0
38.9
0.30
Somers Bay (1)

67.1
0.6
-
4.7
3.21
0.6
3.3
11.0
3.6
<5.0
41.5
0.27
Narrows (1)

80.3
0.5
-
4.8
2.93
0.8
3.8
4.4
1.4
<5.0
37.3
0.30
East Bay (1)

78.1
1.4
-
5.3
2.80
0.9
5.1
8.2
3.0
<5.0
92.5
0.37

-------
Table 1 Chemical characteristics of grab samples from a depth of about 1 m from the bottom at each Flathead Lake Clean.Lakes water quality monitoring sites. Data are also given for ait
additional mid-water site at Midlake Deep (30m). Means, standard deviations (±1.0 S.D.) and ranges (minimum - maximum) are given for n samples.
Site (n)
30m
Midlake Deep(27)
90m
Lakeside (17)
Ross Deep (18)
Yellow Bay (18)
Skidoo Bay (17)
Osprey Nest (1)
Somers Bay (1)
Narrows (1)

Alk
Tuib
TSS
Si02
S04
SRP
SP
TP
NH3
N02/3
TPN

mg/l-CaC03
NTU
mg/1
mg/l-Si02
mg/l-S04
Hg/I-P
Hg/l-p
Hg/I-P
hr/i-n
Hg/I-N
W5/I-N
Mean
86.4
0.8
0.5
5.0
3.12
0.68
19
5.8
3.26
55.6
126.8
S.D.
2.3
0.5
0.0
0.2
0.10
0.17
0.8
1.5
132
8.4
55.2
Min
81.4
0.3
<0.5
4.7
3.03
<0.40
2.2
4.5
0.58
45.3
80.6
Max
89.1
1.9
0.5
5.4
3.22
1.01
4.6
8.7
9.30
71.2
247.0
Mean
86.4
0.7
0.5
5.5
335
0.89
3.1
5.4
4.79
64.8
116.3
S.D.
1.8
0.3
0.0
0.5
0.08
0.42
1.4
1.6
3.63
15.5
43.9
Min
82.5
0.3
<0.5
4.6
3.31
0.40
1.3
3.3
<0.5
36.9
66.9
Max
91.3
1.4
0.5
7.1
3.47
2.17
7.4
9.8
14.97
95.5
254.0
Mean
87.3
1.0
0.7
5.2
3.10
0.87
3.3
5.6
4.34
63.4
116.2
S.D.
1.5
1.0
0.3
0.4
0.17
0.3
1.2
1.4
3.3
14.6
36.9
Min
85.5
0.3
<0.5
4.6
2.88
0.5
<1.6
3.7
<1.2
39.8
79.6
Max
91.3
4.6
1.1
5.8
3.52
1.8
5.8
9.3
13.2
87.3
238.0
Mean
86.1
3.7
1.9
5.0
2.97
0.73
3.3
6.3
4.51
39.1
104.1
S.D.
2.8
9.9
1.2
0.3
0.30
0.4
1.2
16
4.4
14.4
35.0
Min
81.4
0.4
0.8
4.4
131
<0.4
1.6
3.7
<1.2
<2.6
416
Max
90.2
42.0
3.4
5.5
3.39
2.1
5.7
14.4
20.2
55.1
216.0
Mean
87.9
1.9
0.9
5.2
3.08
0.70
3.0
5.7
3.95
59.3
114.8
S.D.
3.8
3.6
0.4
0.3
0.21
0.2
1.8
1.9
3.0
15.4
27.6
Min
83.6
0.5
<0.5
4.5
2.67
0.4
1.2
3.7
<1.2
33.8
73.1
Max
101.2
16.0
1.3
6.0
3.50
1.3
8.4
110
11.5
82.2
196.0
Mean
86.9
0.7
0.8
5.4
3.06
0.72
3.1
6.3
3.64
50.7
116.6
S.D.
2.0
0.3
0.4
0.5
0.19
0.2
1.1
11
2.5
18.7
27.9
Min
82.5
0.3
0.5
4.6
176
<0.4
1.7
3.5
1.0
21.0
715
Max
90.2
1.4
1.4
6.4
3.34
1.1
5.9
116
11.3
77.9
165.0
Mean
86.2
0.6
0.7
4.9
3.10
0.69
2.6
5.2
4.11
37.2
95.7
S.D.
2.8
0.3
0.4
0.3
0.10
0.3
0.8
1.2
4.1
17.4
23.4
Min
80.3
0.3
<0.5
3.9
2.91
<0.40
1.4
3.4
0.7
7.9
36.3
Max
90.2
1.3
1.2
5.3
3.22
1.6
4.9
7.6
17.7
60.6
148.0
Mean
87.1
0.6
0.6
5J
3.14
0.72
3.4
6.0
4.17
59.8
116.1
S.D.
11
0.2
0.3
0.5
0.16
0.2
1.9
19
3.9
14.6
27.0
Min
83.6
0.4
<0.5
4.9
2.79
<0.4
1.5
4.0
0.9
35.3
68.5
Max
91.3
1.2
1.0
6.5
3.43
1.3
9.2
14.3
14.9
86.1
177.0

84.7
0.9

5.4
3.25
1.6
8.2
8.6
33
62.2
93.9

84.7
2.3

5.2
3.31
0.6
5.1
8.1
2.7
46.7
89.6

81.4
0.7
-
4.8
193
0.8
4.5
7.4
2.8
<5.0
45.0
NDOC
mg/l-C
0.17
0.02
0.14
0.22
0.13
0.04
0.10
0.29
0.14
0.04
0.08
0.24
0.24
0.06
0.17
0.42
0.17
0.03
0.12
0.24
0.25
0.08
0.16
0.49
0.23
0.10
0.12
0.50
0.14
0.03
0.11
0.21
0.15
0.34
0.37

-------
Table 3. Chemical and biological characteristics of integrated photic zone samples collected from 0 to 30m (except Lakeside and Yellow Bay, (V20m; Somen, 0-15m; Narrows 0-8m;
East Bay 0-2m) at Flathead Lake Gean Lakes water quality monitoring sites. Means, standard deviations (±1.0 S.D.) and ranges (minimum - maximum) are given for n samples.
m	Turb	T55	5T755	5(54 510s 5P TP FTJE RC373 TPR RD5C Chil Phi^
mg/l-CaC03 NTU mg/1 mg/l-Si02 mg/l-S04 Hg/I-P Hg/l-P Hg/l-P fig/l-N Hg/l-N Hg/l-N mg/l-C Hg/1 M^g/I
85.1
1.3
0.5
4.9
2.70
0.7
2.7
5.1
4.3
31.9
96.3
0.23
0.91
044
>
s>
I
Lakeside (17)
Ross Deep (18)
Yellow Bay (17)
Skidoo Bay (17)
Osprey Nest (1)
Somers Bay (1)
Narrows (1)
S.D.
15
3.3
0.0
0.3
0.91
0.4
0.9
1.6
3.0
16.6
34.5
0.06
0.32
0.26
Min.
80.3
0.3
0.5
4.2
0.00
<0.4
<0.8
2.8
<0.5
4.8
64.0
0.13
0.56
0.09
Max.
89.7
17.5
0.6
5.6
3.17
1.8
4.6
8.7
12.3
59.9
241.0
0.33
2.02
0.97
Mean
86.1
1.7
0.8
4.8
2.87
0.7
3.1
5.4
3.7
35.7
103.1
0.22
0.92
0.52
S.D.
2.7
3.7
0.2
0.3
0.81
0.2
1.5
1.6
1.5
15.8
36.7
0.05
0.36
0.28
Min.
81.4
0.3
0.5
4.1
0.00
<0.4
1.5
3.8
<1.2
6.8
49.2
0.13
0.32
0.11
Max.
90.2
16.0
1.0
5.1
3.27
1.1
8.5
10.9
6.1
58.6
229.0
0.34
1.82
1.11
Mean
85.1
3.8
1.8
4.8
3.02
0.8
3.4
5.9
3.1
316
1016
0.25
1.00
0.51
S.D.
4.1
11.4
1.5
0.4
0.23
0.3
1.6
1.5
1.1
19.9
36.4
0.06
0.39
0.40
Min.
74.8
0.4
0.5
4.1
2.58
0.4
1.1
3.1
<1.2
<16
35.4
0.15
0.48
0.07
Max.
91.3
48.0
4.4
5.3
3.32
1.6
8.6
8.1
5.0
610
205.0
0.39
1.85
1.41
Mean
85.6
2.5
1.2
4.9
2.80
0.7
2.6
5.3
3.3
35.1
106.3
0.24
0.98
0.40
S.D.
3.5
6.8
0.8
0.3
0.80
0.3
0.7
1.4
1.3
16.4
34.0
0.0«
0.49
0.30
Min.
79.2
0.3
<0.5
4.1
0.00
<0.4
1.6
3.7
1.3
7.6
46.8
0.15
0.48
0.05
Max.
90.8
29.0
2.3
5.3
3.24
1.6
4.0
10.2
5.4
55.7
198.0
0.36
2.30
1.37
Mean
86.3
1.4
0.6
5.0
2.89
0.7
2.7
5.4
3.2
29.3
101.3
0.27
1.28
046
S.D.
2.6
2.6
0.1
0.3
0.81
0.2
0.7
1.5
10
111
35.4
0.06
0.53
0.23
Min.
81.4
0.3
<0.5
4.1
0.00
0.3
1.8
3.9
<0.5
9.3
41.4
0.16
0.63
0.16
Max.
90.2
11.5
0.8
5.5
3.24
1.0
3.9
9.7
7.7
56.6
218.0
0.40
2.38
0.94
Mean
86.0
0.7
0.8
4.8
2.89
1.0
3.4
5.7
3.3
30.4
101.9
0.25
0.95
0.33
S.D.
2.9
0.4
0.4
0.4
0.81
1.0
2.3
2.1
1.6
20.2
25.7
0.11
0.32
0.36
Min.
81.4
0.3
<0.5
4.0
0.00
<0.4
1.2
2.9
<1.2
<15
310
0.16
0.40
0.01
Max.
90.2
1.7
1.4
5.4
333
4.8
10.1
10.5
7.3
57.3
135.0
0.57
1.66
1.66
Mean
S.D.
Min.
Max.
86.0
3.1
79.2
90.2
1J
3.4
0.3
14.0
0.7
0.2
<0.5
1.0
4.9
0.4
4.1
5.6
2.88
0.84
0.00
3.90
0.7
0.3
<0.4
1.4
2.9
1.8
1.6
8.9
5.1
13
2.5
13.2
3.3
1.8
<1.2
8.0
35.4
16.3
3.9
59.5
100.9
28.4
47.3
163.0
0.23
0.09
0.12
0.49
0.93
0.34
0.40
1.70
1.19
0.71
0.71
0.95
0.35
0.22
0.09
1.08
0.21
0.22
0.10
0 25

-------