2011 Alum Test Results-Grand Lake St Marys and comments from Ohio EPA’s Dina Pierce
TO HEAR COMMENTS WITH DINA PIERCE:
Experimental Treatments to Reduce Phosphorus and Algae in Nearshore Coves of Grand Lake St. MarysSpring 2011
Results Summary Technical Memo
Tetra Tech, Inc.
Four cove areas were isolated along the shore of Grand Lake St. Marys (GLSM) in western Ohio to test the effectiveness of in-lake algal controls during the first week in April 2011. The initial plan was to treat three coves, leaving the fourth untreated early in the spring, prior to the onset of the cyanobacteria bloom. The previous 2010 demonstration experiment did not start until September and showed that the presence of a large mass of algae inhibited the short-term water column effectiveness of alum, resulting in less than expected reductions in total phosphorus (TP) and chlorophyll (chl) in two coves and no reduction in a third cove. Although the 2011 experiment began at the beginning of April, an algal bloom composed in large part of cyanobacteria had already begun, as illustrated by the average TP concentration of 114±6 µg/L and chl concentration of 99±17 µg/L at the three open lake sites. Nevertheless, the experiment proceeded due to timing of funding and pressure to demonstrate that efforts to manage the water quality of the lake were underway. Two of these demonstration treatments in 2011 were dosed with nearly three times the amount of liquid alum as in 2010, and equivalent to the full alum dose needed to inactivate sediment phosphorus in the main lake at 86 mg Al/L.
Three phosphorus and algal control procedures were applied during the 2011 experiment:
1. Granular alum was applied at almost two times company recommendation, and higher than any previous reported dose at 300 lb/ac.ft. (Kozy Marina).
2. Buffered alum was applied at a dose of 86 mg Al/L (Windy Point marina).
3. Buffered alum was applied at 86 mg Al/L proceeded by peroxide at 200 lb/ac.ft. (State Park Bay). Pre – alum treatment with peroxide was used to oxidize and solubilize particulate P in the water column and sediments so it could be easily sorbed (inactivated) by alum. In addition, this peroxide pre-treatment was designed to reduce the algal interference to phosphorus precipitation from the water column due to high algal densities.
Unfortunately, a large storm brought 1.19 inches of precipitation the day before (4 April) the first treatment began which was preceded by heavy precipitation in the month of March, adversely affecting that treatment by increasing both TP loading at Kozy Marina and suspended solids loading to all sampling sites. Runoff from a large drainage area raised TP concentrations to high levels in Kozy Marina, while the other two sites (State Park Bay and Windy Point Marina) were initially unaffected. However, subsequent stormy weather (windy conditions and precipitation of 0.31 inches on 8 April) caused breaching of lake water through the barriers at all three sites effectively shortening the duration of observable impacts of the demonstration projects.
Kozy Marina (KM) was treated with granular alum on April 5th, 2011. Water samples were collected at four sites within the Marina daily for two days, weekly for two weeks and then after four weeks following treatment.
State Park Bay (SPB) was treated with peroxide on April 6th, 2011 and with liquid alum and sodium aluminate at 86 mg Al/L on April 7th, 2011. Water samples were collected from four sites in the Bay at the same frequency as at Kozy Marina.
Windy Point Marina (WPM) was treated with liquid alum and sodium aluminate at the same concentration as SPB on April 8th, 2011. Water sampling followed the same frequency after treatment as at KM and SPB. Windy Point Marina was not treated with peroxide.
Open Lake and Otterbein
Water quality was rather consistent in the open lake as well as at Otterbein-West, the control site. Transparency in the open lake was low, due to the large algal bloom, averaging 0.34 (±0.04 SD) m from late March through April. Otterbein was slightly clearer at 0.58±0.06 m. TP and chl averaged as follows:
Site March 28th – 31st April 13th – May 4th
TP (µg/L) Chl (µg/L) TP (µg/L) Chl (µg/L)
(n = 4*) 63 72 95±24 115±12
Open Lake (3 sites) (n = 9) 114±6 99±17 89±30 112±12
* Composites – 3 sites
These values show that bloom intensity was similar in the open lake and in the control cove.
Kozy Marina – Granular Alum
Runoff from the April 4th storm immediately affected this cove, lowering transparency to 0.08 m on April 5th. Transparency remained low throughout the experiment sampling period (April 20th) averaging 0.10±0.05 m, except at the bay dock control where it averaged 0.29±0.07 m for the whole period. Water color was dark brown, indicating suspended solids from soil erosion.
The high impact of suspended solids raised TP to an average of 470±187 µg/L in most of the cove throughout the experiment. Total Phosphorus at the sheltered control bay averaged 100±42 µg/L similar to concentrations in the open lake and Otterbein control. Chlorophyll was actually diluted out in the cove, decreasing from 57±3 µg/L on April 5th before treatment and after treatment to 14±2 µg/L by week 2 and 4. Chlorophyll was rather consistent at the sheltered control bay, averaging 107±32 µg/L, also similar to concentrations in the open lake and Otterbein. Therefore, the effectiveness of granular alum could not be evaluated due to interference from high stormwater runoff volumes carrying extremely high concentrations of TP in the coves and lake. However, this further points to the long-term solution for water quality management of GLSM being intensive watershed management.
State Park Bay – peroxide / buffered liquid alum
Runoff from the large April 4th storm did not affect this cove nearly as dramatically as Kozy Cove (In large part due to the land-use and relative size of the drainage area. Total Phosphorus and chl were 113 and 81 µg/L on March 31st before treatment and transparency was 0.45±0.03 m, similar to the open lake. Also, peroxide had little or no effect on TP or chl; transparency averaged 0.30±0.02 m before and 0.29±0.03 m after the peroxide treatment at the four sites in the cove. Total Phosphorus averaged 104±37 µg/L and chl 126±4 µg/L 1 hour and 1 day after peroxide treatment.
Alum had an immediate effect on TP for only two days, reducing concentrations 81%, before severe storms (extremely windy conditions) caused lake water to breach the barrier between the open lake and the cove. That TP reduction in the water column was the direct level of response expected from alum. The response to alum as average values for time and composite samples from all four sampling sites is shown below. Chlorophyll was reduced 86% and transparency increased three fold initially.
Transparency (m) TP (µg/L) chl (µg/L)
April 7th – 9th
Two days after alum 1.33±0.5
(n = 12) 21±8
(n = 3)
April 13th – May 4th
1-4 weeks after alum 0.40±0.13
(n = 7) 83±11
(n = 4) 66±25
(n = 4)
Windy Point Marina – buffered liquid alum
This cove was also unaffected by runoff from the April 4th storm. Total Phosphorus and chl were 89 and 87 µg/L on March 31st before treatment and transparency averaged 0.46±0.06 m, similar to conditions in the open lake and Otterbein control. Alum had an immediate effect on water quality, as at SPB, reducing TP in the water column by at least 90%. Chlorophyll was reduced 97% and transparency increased three fold. However, the effect was short-lived, similar to SPB, due to stormy and windy conditions that caused breaching of the barrier and contamination by open lake water. The short-and longer-term response to alum as average values for time and composite samples from all four sites as shown below:
Transparency (m) TP (µg/L) chl (µg/L)
April 8th – 9th
Two days after alum 1.45±0.62
(n = 8) <10
(n = 2) 2.7±0.5
(n = 2)
April 13th – May 4th
1-4 weeks after alum 0.62±0.11
(n = 7) 160±56
(n = 3) 136±82
(n = 3)
1. Effectiveness of granular alum could not be evaluated due to excessive runoff of TP and suspended solids from a large storm the day the experimental treatment began.
2. Peroxide had no effect on transparency, TP or chl. The treatment failed to oxidize the algal mass and solubilize particulate P to a definable degree.
3. The two alum treatments were highly successful for the first day or two, reducing TP and chl by an order of magnitude and increasing transparency three fold. The alum treatment in SPB removed 8 pounds of TP from the water column. In WPM, alum reduced water column TP by 5 pounds. These responses were expected and reflect the effect of a higher alum dose (86 mg Al/L) as compared to the dose used in 2010. Unfortunately, stormy and windy conditions caused the isolation barriers to be breached contaminating the inner coves with lake water. Nevertheless, the higher alum dose resulted in much lower TP and chl initially than in 2010 demonstration.
4. The alum dose used in the 2011 demonstration treatments (86 mg Al/L) removed an average of 86 µg/L TP in the two coves (SPB and WPM), or a ratio of 1000:1 of Al added to TP removed. This ratio is high because it considers only water column TP, while the dose was based on water column plus sediment mobile and organic phosphorus. Post treatment sediment core data will show how much sediment phosphorus was inactivated by the demonstration treatments.
5. With an average percent TP removal efficiency of 85% (as shown in the two 2011 demonstration treatments), a whole lake alum treatment of Grand Lake St. Marys at the full alum dose required to inactivate lake sediments (86 mg Al/L) and at the right time of year (early spring) would result in a mass reduction of approximately 14,500 pounds of phosphorus and 15,900,000 pounds of wet algae removed from the water column of the lake.