 |
Alum Bench Test Results
|
Executive Summary
In the summer of 2002, Honeoye Lake experienced a
mid-summer blue-green algae bloom of a significant proportion.
The bloom resulted in clarity, odor and aesthetic impacts, which
impaired the use and enjoyment of the lake.
Blooms of the nature and magnitude experienced in Honeoye Lake are
typically associated with highly productive, eutrophic waterbodies.
Although eutrophication is a natural process, its negative affects
on water quality and lake aesthetics, especially intense algae blooms, are
cause for concern. Human
activities, in particular watershed development, can greatly accelerate
the process leading to significant water quality and lake user
impairments. As most of the
problems arising from eutrophication are the result of excessive nutrient
loading, especially the influx of biologically available phosphorus,
emphasis is placed on the control of phosphorus and the implementation of
lake management measures that control the rate and amount of its loading
to the affected lake.
The
magnitude of the 2002 algae bloom stimulated increased interest in
assessing and identifying those factors contributing to the eutrophication
of Honeoye Lake. Fortunately,
a considerable amount of water quality data have been collected over the
past decade as part of studies conducted by the Finger Lake Community
College (FLCC) and the New York State Department of Environmental
Conservation (NYSDEC), as well as by local volunteer and stakeholder
groups, such as the Honeoye Valley Association (HVA).
A detailed examination of the Honeoye Lake historic water quality
database revealed that the lake at times experiences deep-water anoxia.
That is, during the summer, for limited periods of time, the bottom
waters of the lake (> 7 meters in depth) become devoid of dissolved
oxygen. Although this only
affects a small percentage of the overall lake, during these episodes of
anoxia significant amounts of phosphorus are released from the lake’s
sediments. Because the lake
does not strongly stratify, this phosphorus-rich water is subsequently
re-circulated into the lake’s upper layers where it then becomes
available for uptake and assimilation by algae.
This appears to be the most likely cause of the lake’s mid- to
late-summer blue-green algae blooms.
Although a comprehensive watershed and pollutant loading analysis
must be conducted for the lake, the evidence and data thus far developed
indicate that control for the lake’s internally regenerated phosphorus
load must be dealt with immediately.
This is best accomplished through the application of alum. This
report provides details of an alum treatment bench test conducted for
Honeoye Lake to determine the appropriate alum dose needed to control the
internally regenerated phosphorus load.
It also discusses the theory of alum lake treatments and the
success of such treatments in the control of the eutrophication process in
other lakes nationwide. The
use of alum in Honeoye Lake is proposed for the purpose of maintaining
sediment-bound phosphorus sequestered in the lake’s sediments, even
during periods of deep water anoxia.
By making this source of phosphorus unavailable for algal
assimilation, a reduction in mid-summer, intense, nuisance algal blooms is
predicted. The proposed
treatment of the lake is not intended to strip the entire water column, as
is often a technique used to clarify lakes and bind whole water column
concentrations of total phosphorus. In
Honeoye Lake this is neither necessary nor ecologically desirable.
Rather, alum use in the lake would focus totally on the management
of the lake’s internal load. This
will require the application of alum in such a manner that the alum floc
becomes incorporated into the lake’s deep-water sediments.
This in effect creates what is commonly referred to as an “alum
blanket”. Based
on the results of the titration tests, the theoretical appropriate alum dose for the lake is 1.17
mg alum / L (0.105 mg of Al+3 / L) as based on the results of
the alum/phosphate titration bench test. This equates to a surface
application of 74 gallons/acre. Assuming
only 50% efficiency, this yields a projected alum application rate of 150±
gallons/acre. Within the
report three additional techniques are used to project an alum rate for
the lake. The results of
these three alternative computational techniques, including a treatment
formula developed by a leading alum treatment applicator, are all in
relative agreement. Thus the
projected application rate of 150 gallons alum/acre appears accurate and
reasonable. There
will be the need to obtain permits from the NYSDEC in advance of the alum
treatment. Even with the
introduction of alum and the control of the lake’s internal phosphorus
load, the need for watershed management, macrophyte control and other
related lake management and restoration activities, including the
preparation of a comprehensive watershed management plan will not be
eliminated. However, the
proposed alum treatment will address the most likely cause of the recently
experienced nuisance blue-green algae blooms and greatly aid in the
overall proper management of the lake’s water quality. 1.0
Introduction - The Use of Alum in Lake Restoration Alum (aluminum sulfate) has been
used in the water treatment industry for over five decades to clarify raw
drinking water. These applications typically involve the large-scale
introduction of aluminum sulfate within a drinking water treatment plant
in order to create a flocculent (floc), which as it settles, strips the
water column of particulate matter. The
ionic binding of fine sediments to the alum molecule causes this stripping
effect. However, alum is capable of achieving
more than the simple clarification of water and the binding of particulate
matter. In natural lake
environments, alum has been used to reduce the amount of phosphorus in
the water column. Phosphorus
is recognized as the nutrient present in limiting amounts in most
freshwater lake ecosystems. This
means, the amount of phosphorus available for uptake and assimilation by
primary producers, in particularly phytoplankton (algae), will dictate the
overall productivity of the lake. Simply
stated, as phosphorus levels increase, lake productivity can be expected
to increase. Because lake
productivity is largely a function of algal photosynthesis, the more
phosphorus present in a lake, the greater the amount of algal biomass.
Although algae are essential to the maintenance of a healthy,
balanced lake ecosystem, in excessive amounts, algae can create water
quality impacts, especially reductions in lake clarity and aesthetics that
impair recreational use of the lake.
Since
the early 1980’s, alum has been used in the management of recreational
lakes and drinking water reservoirs to “inactivate” biologically
available inorganic phosphorus. As
noted above, reducing
phosphorus concentrations in lake water decreases the availability of this
vital nutrient for uptake and utilization by phytoplankton.
A reduction in phytoplankton production leads to an increase in the
clarity of the lake. Because phosphorus may enter a lake ecosystem from
external (stormwater runoff, septic tank effluent, atmospheric deposition,
waterfowl) and internal (decaying aquatic vegetation, phosphorus released
from bottom sediments) simply treating the water column alone may not
result in sufficient phosphorus controls to achieve adequate control of
lake productivity and the desired water quality and lake aesthetic
benefits. In addition,
because treatment of lake involves the introduction of large quantities of
alum in a natural environment, it is imperative to investigate and assess
whether the alum treatment will elicit negative impacts on the biota and
chemistry of the treated lake. As
noted above, in every lake ecosystem some of the phosphorus utilized by
algae and phytoplankton is internally generated.
Of particular concern, because of the potential magnitude
associated with this source, is the phosphorus released from the lake’s sediments.
When there is ample dissolved oxygen in the water column (even
concentrations as little as 1.0 mg/L) phosphorus is relatively stable in
the sediments, remaining bound to iron.
However, under anoxic conditions (when dissolved oxygen levels are
at or close to zero) the phosphorus-iron is disrupted resulting in the
liberation of sedimentary phosphorus back into the water column.
Anoxic conditions may occur on a regular basis in lakes that even
weakly undergo thermal stratification.
Thermal stratification is a natural process caused by the seasonal
warming of the water column. This phenomenon is common for lakes with maximum depths
exceeding 10 feet. When a
lake stratifies, the density of the water column is no longer uniform. The warmer, upper layers of the lake are “lighter” (less
dense) than the cooler “heavier” (more dense) deeper layers. When the density difference between layers becomes
significant, these layers can no longer freely mix.
In the summer, this results in the deeper portions of the lake
becoming segregated from the upper layers.
This inability of the water column to freely mix impairs the
reoxygenation of the lakes deeper layers.
Bacteria quickly utilize the limited available oxygen in the
lake’s deeper layers, resulting in the onset of anoxia.
Phosphorus is then released from the sediments at higher than
normal rates. The
destratification of the water column results in its free mixing and the
subsequent circulation or “up welling” of the phosphorous rich water
from the lake’s dark depths, to the sunlit surface.
Once in the sunlit portions of the lake, where algal photosynthesis
occurs, this phosphorus becomes available for biological assimilation.
If this internally regenerated phosphorus load is significant, an
algae bloom will likely be experienced.
Control of this internal phosphorus recycling is therefore
critical. Initially
alum was used in the restoration of lakes and ponds as simply a stripping
or flocing agent, similar in effect to the manner it is used in water
treatment plants. These
applications of alum were discovered to readily clarify the treated
waterbody, decreasing the amount of suspended material present in the
water column. This included inorganic sediment and fine debris as well as
algae and phytoplankton. In
the late 1970s and early 1980s, alum became used increasingly in natural lake systems for the
control of deep-water phosphorus released from under anoxic conditions.
These treatments essentially capitalized on alum’s ability to
bind phosphorus. In these
applications, the applied alum was utilized for more than simply water
column stripping; it was used for nutrient management.
Alum utilized in this manner decreased the amount of phosphorus
available in the lake for algal and phytoplankton assimilation.
Thus, although these treated lakes remained anoxic and stratified,
the released phosphorus was not available for uptake and utilization by
algae and phytoplankton. The
use of alum for nutrient management purposes increased in frequency
following a series of carefully executed deep-water alum applications
conducted by Kent State under the direction of Dr. Dennis Cooke.
Those projects demonstrated that when properly administered, alum
could sequester the released phosphorus, resulting in lower overall lake
productivity, which translated to lakes less prone to algae blooms.
This practice of deep-water alum treatments, for the purpose of
controlling phosphorus released from anoxic hypolimnetic sediments, was
colloquially termed the creation of an “alum blanket”.
Essentially the “blanket” is the resulting alum floc and
sludge, which has settled from the water column and has become integrated
into the lake’s sediments. It does not create a discernable cap or “plastic”
barrier. Rather, it results
in the integration of active alum into the lake’s sediment, with this
alum being responsible for the binding of the majority of the phosphorus
released from the sediments under periods of anoxia.
Cooke’s research demonstrated that this alum could remain in an
active form, effectively fixing the released phosphorus for ten years or
more in certain situations (Welsh and Cooke, 1995).
More
recent work conducted by Drs. Stephen Souza and Fred Lubnow of Princeton
Hydro has shown that alum “blankets” can be successfully used to
control the internal phosphorus load of relatively shallow, polymictic
lakes such as Honeoye Lake. Similar
successful applications of alum for relatively shallow lake ecosystems
have been reported by Pilgrim (2004).
Given the data collected over the past decade by Finger Lake
Community College, under the direction of Dr. Bruce Gilman, and data
compiled by NYSDEC, it appears the lake’s internal phosphorus load is
significant. Interest has
thus been raised regarding the use of an alum blanket to control the
lake’s internal phosphorus load. 2.0
How Alum Works Whether alum is used to strip and clarify
a lake’s water column or to control internally released and recycled
sedimentary phosphorus, the process of creating the alum floc is
responsible for the subsequent results.
On contact with water, alum forms a “fluffy” amorphous aluminum
hydroxide precipitate. This
precipitate is the result of the dissociation
of the alum and liberation of aluminum ions which are immediately
hydrated, and through a progressive series of hydrolyzes, form aluminum
hydroxide.
Aluminum hydroxide is the principle ingredient in many common antacids. The
resulting colloidal amorphous floc also has high coagulation and
phosphorus adsorption and binding properties. Because the floc is heavier
than water, it settles out of the water column.
As the floc slowly settles out of the water
column, phosphorus, especially inorganic phosphorus, binds to the alum
floc and becomes, in effect, inactivated or unavailable for biological
uptake by algae and phytoplankton. The
resulting aluminum-phosphorus compounds are insoluble in water. Unlike the phosphorus-iron bond, the aluminum-phosphorus
bond is extremely strong. Of
particular importance, is phosphorus will not be released from the
aluminum even under anoxic conditions.
As noted above, once the alum floc settles on the bottom of the lake it
becomes integrated into the sediments and subsequently reacts with
phosphorus released from the sediments. This
binding of sediment-released phosphorus further decreases the overall
availability of phosphorus for uptake and utilization by algae and
phytoplankton. The overall
effect of the alum treatment is a reduction in lake productivity.
With less algae present in the water column, the lake is clearer
and more transparent. 3.0
Eliciting Care When Applying Alum in Natural Lake Ecosystems Concerns
are often raised regarding the impact of alum treatments on lake biota.
These concerns tend to arise due to: 1.
The potential “smothering” of benthic fauna, especially aquatic
insect larvae, attributable to the settled floc, and 2.
The potential acute toxicity to fish caused by elevated
concentrations of dissolved aluminum in the water column. With
respect to impacts to benthic fauna, one of the most comprehensive studies
on this matter, Narf, 1990, reported that alum treated lakes showed
no long term negative effects on the diversity or assemblage of benthic
infauna. Although not
published, data collected by Princeton Hydro at Lake Mohawk show the same
results. With
respect to dissolved aluminum toxicity, the proper use of alum dictates
that the amount applied to a lake be sufficient to control phosphorus
availability, but not so great to cause the pH to shift below 6.0.
The control of pH shifts is critical because the solubility of
aluminum in water is directly related to pH.
The relationship is such that increased dissolution of aluminum
occurs at pH levels below 5.5 and above 9.0.
It is the lower pH though that is typically of concern, because
alum (aluminum sulfate) is acidic. The
introduction of large quantities of alum, over a short period of time, as
occurs during a lake treatment, has the potential to cause a major
depression of the lake’s pH.
As the lake’s pH declines below 5.5, an increasing proportion of
the aluminum contained in the applied alum becomes present in a dissolved,
rather than in a particulate state. As
noted above, this is problematic as dissolved aluminum, at high enough
concentrations, is toxic to aquatic biota, in particular fish. On the
basis of continuous flow bioassays conducted to evaluate the toxicity of
dissolved aluminum on trout and other fish, Freeman
and Everhart (1971) concluded that concentrations of dissolved aluminum
below 52 μg Al/L have no negative impacts on fish.
This
is an extremely low concentration and is evidence of the potential
toxicological impacts of improper alum use on aquatic life.
However, because the solubility of aluminum in water is a function
of not only dose, but also of baseline pH and alkalinity, a simple upper
threshold concentration is not in itself a suitable means of predicting or
avoiding impact to aquatic life. As such, guidance developed for the use
of alum (largely based on the studies of Cooke) conclude that measures
must be taken to ensure the pH of the treated lake system remains in a
range of 6.0 to 9.0 during and immediately following application of the
alum. Doing so avoids the
situation where aluminum is present largely in a dissolved state, thus
circumventing dissolved aluminum impacts to the lake’s fish community
(Cooke, et al, 1978). The
most appropriate way to determine the degree and extent of the effect of
alum on a lake’s pH is to conduct controlled, laboratory scaled
assessments, commonly referred to as “bench tests”.
Such a bench test was conducted for Honeoye Lake for the specific
purpose of assessing the affects of the introduction of alum on the pH of
the lake and the feasibility of an alum treatment for the control, through
inactivation, of available phosphorus. 4.0 Honeoye
Lake Alum Treatment Bench Scale Test To
ascertain both the appropriate dose (the amount needed to
effectively inactivate available phosphorus) and the safe dose (the
amount below which no impact will result to the lake’s biota), an alum
bench test was conducted for Honeoye Lake.
The remainder of this report discusses the results of that test and
addresses the feasibility of using alum to inactive a large portion of the
lake’s internal phosphorus load. Specifically,
this report discusses: 1.
Whether alum [the salt aluminum sulfate (Al2(SO4)3
. 14H2O)] can be safely used to bind biologically
active phosphorus, 2. Determine the appropriate safe dose of alum needed to treat and inactivate the lake’s deep water internal load, 3.
Develop initial estimates for the cost of such an alum treatment. Sampling
of Honeoye Lake was conducted by Princeton Hydro staff on 20 July 2004.
A variety of physical, chemical and biological water quality data
was collected as part of the sampling effort.
All sampling was conducted at a single sampling station (Figure 1)
located along the lake’s east central shoreline in an area recognized to
be the deepest portion of the lake. In-situ sampling of
temperature, dissolved oxygen (DO), pH and conductivity was conducted at
1.0 meter intervals from the surface to the bottom of the lake.
Discrete water samples were collected at 0.5 meters, mid-depth
(approximately 4 meters) and near the lake’s bottom at a depth of
approximately 8.5 meters. These
samples were analyzed for total phosphorus, soluble reactive phosphorus,
dissolved aluminum, hardness and alkalinity (as CaCO3).
Finally, approximately 10 gallons of water were collected from the
mid depth and bottom depth stations and utilized to conduct the alum bench
test. On
20 July, the in-situ data documented that Honeoye Lake was well
mixed (only weakly thermally stratified).
It was fairly well oxygenated from the surface to approximately 8
meters, but anoxic near the bottom. The
lake, at the time of sampling had a pH of 7.6.
Alkalinity from surface to bottom was fairly uniform ranging from
65.0 mg/L at the surface to 69.5 mg/L at the bottom. Likewise, the lake displayed relatively uniform hardness
values from surface to bottom ranging from 79.6 mg/L at the surface to
80.6 mg/L at the bottom. Dissolved
aluminum was non-detectable at all sampled depths. The
lake’s total phosphorus (TP) concentration was 0.02 mg/L at the surface,
0.03 mg/L at mid-depth and 0.05 mg/L at the bottom.
This suggests active release of phosphorus from the lake’s
deep-water sediments. The
lake’s soluble reactive phosphorus (SRP) concentrations were 0.004 mg/L
at the surface, 0.005 mg/L at mid-depth, and 0.029 mg/L near the bottom.
This again indicates active phosphorus release from the lake’s
deep-water sediments. In
addition, these data indicate that a large fraction (approximately 60-70%)
of the total phosphorus measured near the lake bottom was in a dissolved,
soluble form. Such forms of
phosphorus are readily available for algal and phytoplankton uptake and
assimilation. |
|
Table 1 - Discrete Water Quality Data for Honeoye Lake
|
|||
|
Water
Quality Parameter |
Surface
(0.5 m) |
Mid
(~ 4.0 m) |
Bottom
(>8.0 m) |
|
Alkalinity
(mg CaCO3 / L) |
65.0 |
63.5 |
69.5 |
|
Total
Phosphorus TP
(mg / L) |
0.02 |
0.03 |
0.05 |
|
Soluble
Reactive Phosphorus SRP
(mg / L) |
0.004 |
0.005 |
0.029 |
|
Hardness |
79.6 |
69.4 |
80.6 |
|
|
|
5.0 Feasibility
Assessment of Alum Treatments Introduction For the past decade, Honeoye Lake has been the
subject of studies conducted by the NYSDEC, Finger Lake Community College
(FLCC), the Honeoye Valley Association and others engaged in the
restoration and management of the lake.
This extensive dataset has focused mostly on the quantification and
characterization of in-lake conditions, encompassing the lake’s biology,
water chemistry and physical attributes.
Princeton Hydro had an opportunity to review these data and discuss
our findings with the HVA and Dr. Bruce Gilman of FLCC. The results of Princeton Hydro’s 2003
introductory assessment of Honeoye Lake suggested that a significant
internal phosphorus load exists in the lake, and that this load, because
of the polymictic nature of the lake (i.e., weakly stratified), can be a
primary cause for major mid- to late-summer algae blooms.
Because of the magnitude and effect of this internal load, an
appropriate management measure would be the control of this internally
regenerated and recycled phosphorus. This project, the Alum Bench Test, was conducted to further evaluate the environmental and economic feasibility of applying alum to deep-water anoxic sediments. As proposed herein, this form of alum application would involve the introduction of alum at a rate of at least 150 gallons per acre to that portion of the lake at least eighteen feet (18 feet) deep. As discussed below, there is no need to apply alum in the shallower sections of the lake. First, these shallow areas do not experience anoxia, and therefore are not major sources of internally recycled phosphorus. Second, it is within these shallow areas of the lake (which includes the littoral zone) where the majority of the lake’s biota live. Avoiding the application of alum in these areas reduces the likelihood of environmental impact. Finally, not applying alum in these shallower areas will negate the chance that the alum blanket will be disturbed or circulated to the lake’s surface. The primary consideration in determining
the ecological feasibility of such an alum application for Honeoye Lake is
to ensure no toxicological or other deleterious environmental side effects
will occur. This was the
primary reason for conducting the alum bench test.
Specifically, the potential for aluminum toxicity, caused by a
rapid, sustained decline in lake pH must be evaluated in advance of any
proposed alum treatment. This
was accomplished by conducting a laboratory bench test to determine the
maximum amount of alum that could be safely applied to Honeoye Lake.
The
bench test was also conducted to determine the appropriate dose of alum
needed to inactivate the lake’s soluble reactive phosphorus (SRP) load
and inactivate other available forms of phosphorus.
If done correctly, this could result in at least four to five
years, and perhaps even greater, control of the lake’s internal
phosphorus load. The key to
successful selection of the dose rate is to introduce alum in the amount
needed to inactivate the measured and/or projected internal load, yet not
introduce more alum than is actually necessary.
In doing so, the application can be conducted in a cost-effective,
ecologically responsible manner. Alum Treatments for the Control of Internal Phosphorus Loading
Sediments usually release phosphorus into the overlying
water; however, this process substantially increases when the overlying
waters are anoxic. Under oxic
conditions the net flux of phosphorus is into the sediments, primarily in
the form of HPO4-2.
Under anoxic conditions, the net flux of phosphorus, as phosphate
(HPO4-2) is from the sediments into the overlying
water, with this flux being largely the result of the dissolution of iron
bound phosphorus. The rate is
typically one to two orders of magnitude (10 – 100 times) greater under
anoxic versus oxic conditions, reaching rates as great as 200 mgP/M2/d. As such, a significant amount of phosphorus can be liberated
from the sediments into the overlying water column even during periods of
short-term anoxia. Because a
large proportion of the sediment stored phosphorus is contained in the
upper 10 cm of the sediment (Want, et. al., 2004) it is not necessary for
anoxic conditions to persist for an extended period in order for such
conditions to have dire consequences.
In addition, the release occurs over a relatively short time frame.
As reported by Armstrong (1979), approximately 50% of the
phosphorus released from anoxic sediments occurs within a few hours of the
onset of anoxia. Thus, even
short-term, sporadic periods of anoxia can
result in a major phosphorus load to the lake.
It is widely recognized that phosphorus readily
binds with a variety of cations, including Calcium (Ca+2), Iron
(Fe+3) and aluminum (Al+3). While a
variety of phosphorus-inactivating compounds were initially considered for
Honeoye Lake, the aluminum-based products were considered over others such
as iron and calcium salts. Funk and Gibbons (1979) provides a
thorough overview of the history and applicability of nutrient
inactivation techniques. Although some of the first in-lake nutrient
control measures were conducted using ferric chloride and other iron-based
products, the most effective and large-scale nutrient inactivation
projects have focused on the use of alum, aluminum hydroxide. Unlike
the other compounds, aluminum-based products remain bound to phosphorus
over a relatively wide range of pH (6.0 to 9.0) and under both oxic (with
oxygen) and anoxic (no or very little dissolved oxygen) conditions.
For Honeoye Lake, the only technically feasible nutrient
inactivation project is one involving the introduction of alum. It
was concluded, based on the initial data collected and analyzed by
Princeton Hydro in 2003 that an appropriate alum dose for Honeoye Lake
should be in the range of 100 to 175 gallons/acre, with the alum applied
to approximately 800 acres of the lake’s total surface area.
This essentially equates to that portion of the lake greater than
18 feet in depth. Treatment of sections of the lake less than 8 feet in
depth is greatly discouraged, and in fact is not required. As noted above, the application of alum in shallow, to even
moderately deep, sections of the lake is strongly discouraged.
The most serious likely problem to occur following shallow water
alum applications is the resurfacing of the alum floc.
This happens because of wind and wave related turbulence, as well
as turbulence attributable to boating, all of which act to circulate the
settled alum floc to the surface of the lake.
The resurfaced floc in turn causes significant aesthetic impacts.
However, more importantly, there is little need to treat these
areas because they contribute little to the overall internal phosphorus
recycling dynamics of the lake. Thus,
there is no need to treat the shallow, oxic sediments of the lake. Alum Bench Test and Computation of Alum Treatment Dose
A
bench test was conducted of Honeoye Lake water to determine how much alum
could be safely added to the lake without producing aluminum toxicity.
Some of the basic data utilized in assessing the lake’s safe and
effective alum doses are provided below.
As noted, the introduction of alum in Honeoye Lake is limited to
the anoxic hypolimnetic reaches of the lake.
In our analysis, a cut-off depth of 18 feet (5.5 meters) was
utilized to define the treatment zone.
It was also assumed, for the purpose of dose computations, that the
hypolimnion of the lake remains anoxic for 90 days.
This is an obvious over assumption given the lake’s polymictic
nature. |
|
|||||||||||||||||||||||||||||||
