JEQ Grow Your Career With ASA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 4 January 2008
Published in J Environ Qual 37:22-29 (2008)
DOI: 10.2134/jeq2007.0142
© 2008 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Agricola
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Related Collections
Right arrow Ecosystem Restoration
Right arrow Surface Water Quality

TECHNICAL REPORTS

Ecological Effects after an Alum Treatment in Spring Lake, Michigan

Alan D. Steinman* and Mary Ogdahl

Annis Water Resources Inst., Grand Valley State Univ., 740 West Shoreline Drive, Muskegon, MI 49441

* Corresponding author (steinmaa{at}gvsu.edu).

Received for publication March 21, 2007.

   ABSTRACT
TOP
 NOTES
 ABSTRACT
INTRODUCTION
 Materials and Methods
 Results
 Discussion
 REFERENCES
 
A whole-lake alum treatment was applied to eutrophic Spring Lake during October and November 2005. Eight months later, an ecological assessment of the lake was performed and compared with data collected in 2003 and 2004. Field measurements showed reduced soluble reactive phosphorus (SRP) and total phosphorus (TP) concentrations in the water column the summer after the alum application, but chlorophyll levels and irradiance profiles were not significantly affected. Total macroinvertebrate density declined significantly in 2006 compared with 2004, with chaoborids and oligochaetes experiencing the greatest reductions. Internal phosphorus release rates, measured using sediment cores incubated in the laboratory, ranged from –0.052 to 0.877 mg TP m–2 d–1 under anaerobic conditions. These internal loading rates were significantly lower than those measured in 2003 at three out of four sites. Mean porewater SRP concentrations were lower in 2006 than in 2003, but this difference was statistically significant only under aerobic conditions. The NaOH-extractable SRP fraction in the sediment was also significantly lower in 2006 compared with 2003, whereas the HCl-extractable SRP sediment fraction showed the opposite pattern. Overall, these results indicate that the alum treatment effectively reduced internal P loading in Spring Lake. However, water column phosphorus concentrations remain high in this system, presumably due to high external loading levels, and may account for the high chlorophyll levels. An integrated watershed management approach that includes reducing internal and external inputs of P is necessary to address the cultural eutrophication of Spring Lake.

Abbreviations: SRP, soluble reactive phosphorus • TP, total phosphorus


   INTRODUCTION
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
Materials and Methods
 Results
 Discussion
 REFERENCES
 
INTERNAL phosphorus (P) loading is a frequent phenomenon in shallow, eutrophic lakes throughout the world and may prevent lake water quality from recovering even after external P loads are reduced (Sas, 1989). Phosphorus release from the sediments can occur via at least two different mechanisms: (i) release at the sediment–water interface during periods of anoxia or hypoxia and the subsequent diffusion of dissolved phosphate into the water column; and (ii) wind-induced resuspension and bioturbation at the sediment surface, whereby the sediment pore water P can be released into the water column or the P adsorbed to sediment particles can desorb into the water column (Selig, 2003). In eutrophic lakes, internal loading can account for a substantial amount of the total P load (Moore et al., 1998). Indeed, many studies have shown that reductions in external loading, to levels where water quality improvement should be detected, do not have the desired effect because of the counteracting release of P from sediments (Björk, 1985; Graneli, 1999; Steinman et al., 1999).

Although many sediment management technologies exist to deal with internal loading, one of the most common practices is chemical treatment (Cooke et al., 1993). Chemical applications are intended to bind P and usually include aluminum sulfate (alum), lime, or iron (Cooke et al., 1993). Alum is particularly effective due to its dual mode of action for P removal. Alum reacts with soluble P to form an insoluble precipitate (Stumm and Morgan, 1996) and also forms an insoluble aluminum hydroxide floc at pH 6 to 8, which has a high capacity to adsorb large amounts of inorganic P (Kennedy and Cooke, 1982). By these two mechanisms, an alum application can irreversibly bind P and inhibit diffusive flux from sediments.

Spring Lake is a eutrophic system with surface total phosphorus (TP) levels that ranged from 60 to 120 µg L–1 during June and July of 2003. Phosphorus release rate studies, using laboratory-incubated sediment cores, indicated that internal loading accounted for between 55 and 65% of the TP load entering the lake water column on an annual basis (Steinman et al., 2004). This same study found that an alum application of 24 mg Al L–1 was extremely effective at reducing TP release from the sediments (Steinman et al., 2004). Additional experiments showed that P release rates at alum concentrations ≥10 mg L–1 were no different from release rates at concentrations of 25 mg L–1 and that resuspension of sediments substantially increased TP concentrations, even at high alum concentrations, but total soluble P concentrations remained low in the water as long as alum was present (Steinman et al., 2006a). As a consequence, it was concluded that alum application may be an effective tool to reduce P flux from sediments in Spring Lake, but it also was noted that external P load reduction must accompany alum application to address the long-term impacts associated with cultural eutrophication (Hansson et al., 1998; Steinman et al., 2004, 2006a).

In the fall of 2005, an alum treatment of ~80 g Al m–2 was applied as a liquid slurry to the surface of Spring Lake in locations where depths were ≥4.6 m (~47% of the lake's surface area). This study was conducted to assess the ecological influence of this field application.


   Materials and Methods
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
Results
 Discussion
 REFERENCES
 
Site Description
Spring Lake is located in west-central Michigan and drains to the Grand River, approximately 1 km east of Lake Michigan (Fig. 1 ). This drowned river mouth lake has a surface area of 5.25 km2, with mean and maximum depths of 6 and 13 m, respectively. Water residence time in the lake is approximately 5 mo in the winter and 11 mo in the summer. The lake pH is generally circumneutral (Lauber, 1999). The Spring Lake watershed covers 134 km2, with the major land use/land cover categories being forest/undeveloped (57%), urban (19%), agriculture (18%), and wetland (6%). The lake's shoreline is densely populated with primary residences.


Figure 1
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 1. Site map of Spring Lake, showing location of four sampling sites. Inset: Outline of Michigan's lower peninsula, with encircled dot showing location of Spring Lake.

 
Field Sampling
All samples were collected in July 2006 from the same four locations that were sampled in 2003 and 2004 (Fig. 1) (Steinman et al., 2004). At each site, vertical profiles of dissolved O2, pH, temperature, specific conductance, redox potential, turbidity, chlorophyll a, and total dissolved solids were measured using a Hydrolab DataSonde 4a (2003, 2004) or a YSI 660 multi-parameter sonde (2006). Photosynthetically active radiation (PAR) profiles were measured using a Li-Cor LI-193SA spherical quantum sensor. Secchi disk depth was also measured to estimate water clarity. Water samples for P analysis were collected with a Van Dorn bottle. Water for soluble reactive phosphorus (SRP) analysis was syringe filtered immediately through 0.45-µm-membrane filters into scintillation vials. Samples were stored on ice until transported to the laboratory, always within 5 h of collection. Total phosphorus samples were stored at 4°C, and SRP samples were frozen until analysis. Soluble reactive phosphorus and TP were analyzed on a BRAN+LUEBBE Autoanalyzer (USEPA, 1983). Soluble reactive phosphorus detection values were 5 µg L–1.

In 2004 and 2006, three replicate benthic samples were collected for invertebrate analysis from each of the four Spring Lake coring sites using a petite Ponar dredge (sample area, 152 x 152 cm; volume, 2.4 L). Upon collection, the benthic samples were washed through a 500-µm sieve under gentle pressure. Each sample was saved in its entirety and preserved in 95% ethanol. Rose Bengal stain was added to the ethanol to aid in sorting invertebrates from organic debris, and samples were stored until identification in the laboratory.

Sediment core sampling and laboratory incubation followed the procedures of Steinman et al. (2004). Sediment cores were collected in July 2006 from the same four sites as the field samples (Fig. 1). Six sediment cores were collected from each site using a piston corer (Fisher et al., 1992; Steinman et al., 2004). The corer was constructed of a graduated 0.6-m-long polycarbonate core tube (7-cm inner diameter) and a polyvinyl chloride attachment assembly for coupling to aluminum drive rods. The piston was advanced 20 to 25 cm before deployment to maintain a water layer on top of the core during collection. After collection, the core was brought to the surface, and the bottom was sealed with a rubber stopper before removal from the water, resulting in an intact sediment core that was ~20 cm in length, with a 25-cm overlaying water column. The piston was bolted to the top of the core tube to keep it stationary during transit. Core tubes were placed in a vertical rack and maintained at ambient temperature during transit. An additional core was collected from each site, and the top 5 cm was removed for sediment chemistry analyses in the laboratory.

Laboratory Methods
Benthic samples were placed in a shallow white pan for sorting of invertebrates. Using a stereomicroscope, all organisms were identified to the family level, with the exception of worms, which were identified to class level of Oligochaeta and Nematoda.

The 24 sediment cores (six per site) were placed in a Revco environmental growth chamber in the dark, with the temperature maintained to match ambient bottom-water conditions in Spring Lake at the time of collection. The water column in three of the cores from each site was bubbled with N2 (with 330 ppm CO2) to create buffered anaerobic conditions, and the remaining three were bubbled with oxygen to create aerobic conditions.

Internal load estimates were made using the methods outlined in Moore et al. (1998), with minor modifications (Steinman et al., 2004). A 40-mL water sample was removed by syringe through the sampling port of each sediment core tube at 12 h and at 1, 2, 4, 8, 12, 16, 19, and 22 d after time 0. The 40-mL subsample was replaced with filtered water (collected at the same time as the cores were removed) from the corresponding site in the lake; this maintained the original volume in the core tubes. Immediately after removal, a 20-mL subsample was refrigerated for analysis of TP, and a 20-mL subsample was filtered through a 0.45-µm membrane filter and frozen for analysis of SRP.

Flux (P release rate) calculations were based on the increase in water column TP or SRP using the following equation (Steinman et al., 2004):

Formula 1[1]

where Prr is the net P release rate or retention per unit surface area of sediments, Ct is the TP or SRP concentration in the water column at time t, C0 is the TP or SRP concentration in the water column at time 0, V is the volume of water overlaying the sediment cores, and A is the planar surface area of the sediment cores. Apparent maximum P release rates were calculated based on changes in concentration from the linear portion of each curve with the caveat that the initial and final samplings could not be consecutive dates to avoid potential short-term bias. For this study, only the TP internal loading data are presented; SRP concentrations from the first 12 d of the incubations were often below detection, thereby limiting the ability to make definitive conclusions.

After the incubations, the top 5 cm of sediment was removed from each core. The sediment was homogenized and subsampled for metals (Fe, Ca, Mg, Al) analysis and ash-free dry mass. The ashed material was analyzed for TP as described previously. Another subsample (5 g) from the wet sediment was centrifuged to remove excess porewater and sequentially fractionated (Moore and Reddy, 1994) to determine the fraction of phosphorus bound to iron and calcium minerals in the sediments. Porewater was filtered, frozen, and analyzed for SRP as described previously. Residual sediment was shaken for 17 h with 0.1M NaOH and centrifuged. The supernatant was drawn off, filtered, frozen, and analyzed for SRP. This fraction refers to the Al- and Fe-bound phosphorus and represents a mineral association that can become soluble under anoxic conditions. After this extraction, the sediment was extracted for 24 h with 0.5M HCl, and the supernatant was centrifuged, filtered, frozen, and analyzed for SRP. This fraction refers to the Ca- and Mg-bound phosphorus and represents a stable mineral association.


   Results
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
Discussion
 REFERENCES
 
Field Results
Depth, Temperature, Dissolved Oxygen, and Chlorophyll a
Water depths were slightly greater in 2006 than 2003, with Sites 1 and 2 deeper than Sites 3 and 4 (Fig. 2 , Table 1 ). Temperatures were relatively constant throughout the water column at Sites 1, 3, and 4, with some evidence of stratification at Site 2 (Fig. 2). Concentrations of dissolved oxygen declined with depth at all sites, with Sites 1 and 2 having concentrations <1 mg L–1; in 2003, only Site 2 had a dissolved oxygen concentration <1 mg L–1 (Table 1).


Figure 2
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 2. Selected limnological characteristics (temperature, dissolved oxygen, and chlorophyll a) at Sites 1 through 4 in July 2006 post-alum treatment.

 

View this table:
[in this window]
[in a new window]

  		Table 1. Selected limnological characteristics of sampling sites in Spring Lake. Data from 2003 (pre-alum treatment) are located above data from 2006 (post-alum treatment) within each cell. In 2003, collections were made from Sites 1 and 2 on 10 and 11 June, respectively, and from Sites 3 and 4 on 16 July. Data from all sites in 2006 were collected on 1 Aug.

 
Chlorophyll a concentrations showed similar profiles at Sites 1 and 2, with concentrations between 15 and 20 µg L–1 through much of the water column and declining to near 5 µg L–1 at the lake bottom (Fig. 2). Chlorophyll a concentrations increased with depth at Site 3 and were considerably higher at Site 4 than at the other sites (Fig. 2). At Sites 1 and 2, the 2006 near-surface chlorophyll a concentrations were somewhat lower than in 2003, but near the bottom of the water column chlorophyll a concentrations were greater in 2006 than in 2003 (Table 1). Comparisons of chlorophyll a at Sites 3 and 4 cannot be made because an algicide treatment was applied at those sites shortly before the samples were collected in 2003, which likely accounts for the low concentrations at that time.

Secchi Depth, Light Extinction Coefficient, and Total Dissolved Solids
In 2006, secchi disk depths at all sites were <1 m and showed no apparent relationship with chlorophyll concentration. The light extinction coefficient data tracked the chlorophyll data better than secchi depth, as indicated at Site 4, where the high chlorophyll concentrations corresponded to the higher extinction coefficient (Table 1). Secchi disk depths were lower at all sites in 2006 than in 2003; extinction coefficients generally were similar between the 2 yr. Total dissolved solids were slightly greater at Sites 1 and 2 than at Sites 3 and 4 (Table 1) and did not show any obvious relationship to secchi disk or light extinction data. Total dissolved solids levels were lower in 2006 than in 2003 at all sites and all depths (Table 1).

SRP and TP
Soluble reactive phosphorus concentration in the water column was below detection (detection limit = 5 µg L–1) at all sites except near-bottom at Site 2, where it was still low at 6 µg L–1 (Table 1). The SRP concentrations at Site 1 and Site 2 surface were below detection limits in 2003 and 2006. As a consequence, the apparent reduction in 2006 (from <0.01 to <0.005 µg L–1) merely reflects a more sensitive detection limit than was available in 2003. However, the lower SRP concentrations at the other sites in 2006 are a reflection of a real reduction in SRP after the alum application. Total phosphorus concentrations in 2006 were ≤50 µg L–1 at all sites (Table 1); TP concentrations were significantly lower in 2006 than 2003 at the near-surface (F = 24.03; p = 0.003) and the near-bottom (F = 8.65; p = 0.026).

Invertebrates
Five major groups of benthic invertebrates were identified from the Spring Lake sediments. In general, oligochaetes were the dominant invertebrate, followed by chironomids and chaoborids, with very sparse numbers of ceratopogonids and nematomorphs. The last two groups were excluded from statistical analysis because of their low abundances. Chironomidae density showed no significant difference between 2004 and 2006 at Sites 1, 2, and 4 and were significantly greater (F = 13.50; p < 0.025) in 2006 than in 2004 at Site 3 (Fig. 3 ). Mean density of Chaoboridae was significantly greater in 2006 than in 2004 at Site 1 (F = 14.256; p < 0.02) but declined at the three remaining sites (Fig. 3); differences were not significant between 2006 and 2004 at Sites 2 and 4 but were significant in 2006 at Site 3 (F = 113.83; p < 0.001). The mean density of Oligochaeta declined at all sites in 2006 compared with 2004 (Fig. 3), but these declines were statistically significant only at Sites 3 and 4 (Site 3: F = 25.392; p < 0.01; Site 4: F = 107.668; p < 0.001). Mean total invertebrate density declined between 2004 and 2006 at all sites (Fig. 4 ), with the decline being statistically significant at Sites 3 (F = 44.474; p < 0.01) and 4 (F = 86.016; p < 0.001).


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Fig. 3. Mean + 1 SE invertebrate densities (organisms m–2) of major groups in Spring Lake in 2004 and 2006.

 

Figure 4
View larger version (10K):
[in this window]
[in a new window]

  		Fig. 4. Total invertebrate density (mean + 1 SE) (organisms m–2) in Spring Lake in 2004 and 2006.

 
Laboratory Results
Redox state had no significant effect on mean TP release rates in 2006 (p > 0.2) (Fig. 5 ). This is in contrast to the 2003 data, when the flux rates in the anaerobic treatment were significantly greater than in the aerobic treatment (Steinman et al., 2004). Mean maximum TP release rates in 2006 ranged from –0.05 to 0.88 mg TP m–2 d–1 under anaerobic conditions (Table 2 ). The negative values suggest that the sediments may act as a sink for TP. Mean TP release rates were reduced at least one order of magnitude in 2006 compared with 2003 at all sites, with the reductions statistically significant at three out of four sites (Table 2).


Figure 5
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 5. Total phosphorus (TP) concentrations measured in water column above sediment cores taken from Sites 1 through 4 exposed to two different redox conditions. Data are means ± 1 SD (n = 3).

 

View this table:
[in this window]
[in a new window]

  		Table 2. Mean (±SD) maximum flux rates of total phosphorus (TP) from Spring Lake sediment cores incubated under anaerobic conditions collected in summer 2006 compared with those collected in summer 2003 (data from Steinman et al., 2004).

 
In general, TP concentrations from the 2006 sediment cores quickly declined once placed in the incubation chamber, regardless of site or treatment (Fig. 5). In most cases, TP concentrations stayed flat after the initial decline, although one replicate core from Sites 2 and 3 showed an increase in the anaerobic treatments; however, even those increases were relatively modest (maximum concentrations of 100 and 290 µg L–1, respectively) compared with the maximum concentrations of >1000 µg L–1 measured in 2003.

The TP concentration in the 2006 sediment cores (as function of dry weight) before incubation ranged from 80 (Site 4) to 1217 mg kg–1 (Site 2) (Fig. 6 ). No inferential statistics were applied to these data because replicate cores were not sampled at each site. With the exception of Site 4, these numbers are similar to those measured in Spring Lake in 2004 (1135–1592 mg kg–1). The TP concentration in the initial core at Site 4 was anomalously low, and the value should be viewed with caution (Fig. 6). Redox state had no significant effect on postincubation sediment TP (p > 0.17), although site was statistically significant (F = 5.615; p < 0.01) because the sediment TP at Site 4 was significantly lower than at Sites 1 and 2, but not compared with Site 3 (Tukey's HSD). The interaction between site and redox was not statistically significant (p = 0.711).


Figure 6
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 6. Total phosphorus (TP) concentration in dry sediment (mg kg–1) from summer 2006 sediment cores analyzed before and at the end of the laboratory incubations.

 
Soluble reactive phosphorus in the porewater at the end of the incubations ranged from below detection (<5 µg L–1) to 150 µg L–1 in one core from Site 2, with mean values of 30 and 6 µg L–1 in anaerobic and aerobic treatments, respectively (Fig. 7 ; Table 3 ). Overall, SRP porewater concentration was significantly influenced by site (F = 43.20; p < 0.001), with Site 1 being significantly lower than Site 3, and redox, with porewater SRP significantly greater in anaerobic than aerobic treatments (F = 4.924; p = 0.014; Fig. 7). The site x redox interaction term was marginally significant (F = 2.578; p = 0.092).


Figure 7
View larger version (21K):
[in this window]
[in a new window]

  		Fig. 7. (A) Soluble reactive phosphorus (SRP) concentrations in sediment porewater of cores at the end of the incubation period. (B) Sodium hydroxide– and HCl-extractable SRP concentrations in core sediment at the end of the incubation period.

 

View this table:
[in this window]
[in a new window]

  		Table 3. Comparison of mean porewater soluble reactive phosphorus (SRP) (±SD) concentrations (mg L–1) and NaOH- and HCl-extractable SRP concentrations (µg g dry wt–1) from Spring Lake sediments pre- (2003) and post-alum (2006) treatment under aerobic and anaerobic conditions.

 
Extractable SRP was analyzed in a three-way ANOVA to examine the effects of site (Sites 1–4), redox state (aerobic vs. anaerobic), and extractant (NaOH vs. HCl). There were clear differences between the NaOH-extractable and HCl-extractable fractions of SRP regardless of site (Fig. 7), with SRP from the NaOH extraction significantly lower than SRP from the HCl extraction (F = 1058.25; p < 0.001). Site was also statistically significant (F = 31.062; p < 0.001), with all sites being significantly different from each other except Sites 1 vs. 2 (Fig. 7). However, the interaction term between site and extraction type was highly significant (F = 11.395; p < 0.001) because the effect of site was influenced by the type of extraction; there was not a significant site effect in the NaOH fraction (p = 0.886), but there was a significant site effect in the HCl fraction (p = 0.001; Fig. 7). Redox had no significant effect on extractable SRP (p = 1.00). The redox x extraction type interaction term was marginally significant (p = 0.092), and the site x redox x extraction type interaction term was significant (F = 3.903; p = 0.017).

There were distinct differences in sediment fractions of SRP between the pre-alum and post-alum sediment cores. Mean concentrations of porewater SRP were lower after the alum treatment in the aerobic and anaerobic treatments, but these differences were not statistically significant because of the high variance in the replicate samples (Table 3). Sodium hydroxide–extractable SRP declined significantly in both redox treatments, but HCl-extractable SRP increased significantly in both redox treatments (Table 3).


   Discussion
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
REFERENCES
 
Internal P loading can be a significant source of nutrients in shallow, eutrophic lakes and can result in serious impairment to water quality (Welch and Cooke, 1995, 1999; Steinman et al., 1999, 2004; Søndergaard et al., 2001; Nürnberg and LaZerte, 2004). This process has ecological and societal implications. Even when external loading rates are relatively low, high internal loading rates can help trigger or sustain algal blooms. Hence, costly attempts to reduce external loading (via best management practices in the watershed) may not improve water quality, at least in the short-term, although improvements should occur eventually, assuming internal loading is suppressed.

Prior studies have shown that alum treatments usually have clear short-term benefits (Cooke et al., 1993; Welch and Schrieve, 1994; Welch and Cooke, 1999), but the question of long-term effectiveness is less clear (Welch and Cooke, 1995, 1999); the degree to which external loads have been reduced after alum treatment has been implicated as a critical factor affecting the longevity of alum treatments (e.g., Hansson et al., 1998, Lewandowski et al., 2003). Continued inputs of high phosphorus will fuel the production of new biomass, which becomes the basis for future internal loads to the system (Carpenter, 2005).

The water column phosphorus data from our snapshot samples in Spring Lake showed significant declines in 2006 compared with 2003, which was presumably related to the alum treatment in 2005 (see below). However, the light and chlorophyll data did not reveal a marked improvement in water quality between 2003 and 2006. The substantial algal biomass in the 2006 Spring Lake samples suggests that even though TP was reduced, there is sufficient phosphorus in the water column (20–50 µg L–1) to stimulate algal growth. It is unclear if the source of this phosphorus is release from the sediments or external loading from point and nonpoint sources in the watershed. Nonpoint sources of phosphorus, such as tributaries, storm drain runoff, septic systems, waterfowl, fertilizer application, and atmospheric deposition are likely contributing significant amounts of phosphorus to Spring Lake. Lauber (1999) identified tributary inflow, septic systems, and lawn fertilizer as the three major sources of external loading to Spring Lake.

Benthic invertebrate community composition and density are effective metrics for assessing environmental conditions in streams (Barbour et al., 1999) and lakes (Canfield et al., 1996). Although genus- and species-level information is preferable for bioassessment work, family-level information has value (e.g., King and Richardson, 2002). In general, oligochaetes tend to be more tolerant of nutrient-enriched conditions, especially compared with chironomids, which are more sensitive to nutrient enrichment (e.g., Wiederholm, 1980), although this generality is dependent on the specific taxa comprising each family and local conditions. Doke et al. (1995) reported no effect of alum on chironomid or oligochaete populations in a Washington lake but reported a doubling in chaoborid density. They attributed this unexpected result to a change in trophic structure, which led to more food resources. Narf (1990) observed a general increase in chaoborid and chironomid populations in five Wisconsin lakes after alum treatments. We also observed a general increase in mean density of Chironomidae but a decline in the mean densities of Chaoboridae and Oligochaeta in Spring Lake after alum application. Our results are consistent with improved water quality conditions, but longer-term sampling is needed to determine if these changes will be sustained. The mean chironomid densities in Spring Lake were generally similar to those found in Muskegon Lake, a drowned river mouth system ~20 km north of Spring Lake (Carter et al., 2006).

Benthic invertebrate density responses to alum application have been variable, including no effect (Narf 1990), increases (Narf 1990), and declines (Smeltzer et al., 1999). In this last study, invertebrate species density and richness declined the year after alum application, but recovery to pretreatment levels occurred within 2 yr, and significant increases above pretreatment levels were evident after 10 yr. Smeltzer et al. (1999) speculated that chronic toxicity from the alum treatment may have been responsible for the year 1 invertebrate decline, but they ruled out acute toxicity as a factor, given the high tolerances of Chaoborus and Chironomus to aluminum (e.g., Havas and Likens 1984). Our year 1 results in Spring Lake were similar to those of Smeltzer et al. (1999). Most of the macroinvertebrate decline was associated with a loss in oligochaete numbers. Chronic toxicity may have affected the oligochaete populations in Spring Lake, although the water chemistry argues against aluminum toxicity as a likely factor. The pH of Spring Lake ranges from ~7.5 to ~8.5, which is considered high enough to prevent aluminum toxicity (Kennedy and Cooke 1982).

Our comparisons of field data between 2003/2004 and 2006 must be viewed with caution for several reasons: (i) The 2003 data were collected earlier in the summer than the 2006 data, which directly influences some parameters (e.g., temperature) and indirectly influences other parameters (e.g., chlorophyll a and dissolved oxygen); (ii) as snapshots, these one-time samples (of physical and chemical conditions) may not be representative of average ambient conditions because antecedent events (e.g., storms, algicide applications) may have a strong but ephemeral influence on the data; and (iii) it may take years of monitoring to detect significant changes in the populations of longer-lived organisms, such as benthic invertebrates (e.g., Smeltzer et al., 1999). However, alum applications are intended to have strong and detectable effects on lake systems, so we believe our post-alum application data provide a conservative comparison of treatment effectiveness. We intentionally measured a variety of lake system responses to provide a holistic approach to the study. Finally, spatial variability in Spring Lake also influences our ability to generalize the findings of this study. The number of samples taken from each site for invertebrate analysis and internal P loading measurements was limited, so our data may not capture the full range of conditions throughout the lake. However, our sampling sites covered as much of the geographic range in Spring Lake as possible; in addition, invertebrate density variability was relatively low (coefficients of variation were usually less than 20%) within and among sites, suggesting the distribution of invertebrate populations was relatively uniform throughout Spring Lake.

The 2006 anaerobic TP release rates were substantially lower than those measured in 2003 and were in the same general range as those measured in oligotrophic systems (Nürnberg and LaZerte, 2004). This provides evidence that the alum treatment effectively reduced the release rate of TP from the sediments. The differential responses of the sediment fractions were somewhat surprising. The NaOH-extractable SRP in Spring Lake declined after the alum treatment. Because this fraction contains the Al- and Fe-bound phosphorus, one might expect this fraction to increase after an alum application. Increases in the NaOH-extractable P fraction after alum dosing have been observed in other lakes (Rydin and Welch, 1999; Reitzel et al., 2005). At the pH of Spring Lake water, alum dissociates to give trivalent Al3+ ions, which hydrolyze rapidly to form soluble monomeric and polymeric species and an amorphous Al(OH)3 floc. The monomeric species can precipitate soluble P as Al(PO4), whereas the floc can remove soluble and particulate forms of P by adsorption or physical entrapment (Bottero et al., 1980; Galarneau and Gehr, 1997; Omoike and Valoon, 1999). One possible explanation for our results is that P that was loosely bound to the alum floc may have become exchanged with soluble calcium, forming a relatively stable mineral association and thereby accounting for the increase in the HCl-extractable SRP.

In summary, our data indicate that the Spring Lake alum treatment has resulted in a decrease in phosphorus concentrations, reductions in overall invertebrate density and some populations, and reduced rates of internal phosphorus loading. However, mean TP concentrations still exceed eutrophic thresholds and are likely to be responsible, at least in part, for the relatively high chlorophyll concentrations in Spring Lake. On an annual basis, the three main external sources of TP to Spring Lake are tributary inflow, septic tank systems, and inorganic fertilizer applied to lawns and agricultural land (Lauber 1999). Hence, any long-term solution to the phosphorus problem in Spring Lake will require approaches that address these sources entering this system. As noted by Steinman et al. (2006a), alum application is a short-term solution to the longer-term problem of internal P loading. Welch and Cooke (1999) concluded that a reasonable expectation of longevity of benefits from alum treatments is 10 to 15 yr. However, impacts associated with nonpoint sources of P in impaired lakes can last for hundreds or thousands of years (Carpenter, 2005). Hence, it is critical that we address the underlying reasons for impaired water quality. External load reduction must complement any chemical addition (Hansson et al., 1998) regardless of the long-term effectiveness of alum treatment. Implementation of nutrient management plans on farms in the upper watershed, conversion of septic systems to sewers, use of low- or no-P fertilizer, reducing stormwater discharge, and implementation of other best management practices should be emphasized and implemented wherever possible in the Spring Lake watershed (Walsh et al., 2005; Steinman et al., 2006b).


   ACKNOWLEDGMENTS
 
We are grateful to Scott Kendall, Jennifer Cymbola-Gray, and Kelly Wessell for their help with field and/or laboratory sampling, Gail Smythe for assistance with laboratory analysis, and Rick Rediske for assistance with data analysis, interpretation, and reporting. We are grateful to Tony Groves and Pam Tyning for their collaborative spirit and to the Spring Lake–Lake Board for funding.


   NOTES
TOP
 NOTES
ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 REFERENCES
 
All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher.


   REFERENCES
TOP
 NOTES
 ABSTRACT
 INTRODUCTION
 Materials and Methods
 Results
 Discussion
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Agricola
Right arrow Articles by Steinman, A. D.
Right arrow Articles by Ogdahl, M.
Related Collections
Right arrow Ecosystem Restoration
Right arrow Surface Water Quality

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome