A model for predicting reduction in mobile phosphorus of lake sediment by aluminum drinking water treatment residuals.

Drinking water treatment residual (DWTR) derived from flocculation and sedimentation of raw water using aluminum coagulants is a valuable environmental remediation byproduct capable of inactivating phosphorus (P). However, no generalizable model exists in the literature to describe reduction of releasable (mobile) P in lake sediment as a result of DWTR addition. The reduction of mobile P (sum of labile P and reductant soluble P) was investigated in over 100 sub-samples using five sediment samples from two lakes and three DWTRs from different water treatment plants. A consistent relationship was determined across a range of mobile P contents (0.23 g/m2/cm to 0.92 g/m2/cm, or 15.8 to 186.1 µg/g DW) and DWTRs. The relationship was best described as a function of the mobile P content of the sediment and the oxalate-extractable aluminum content of the DWTR. An empirical model was developed to predict the immediate reduction in mobile P following the addition of DWTR containing aluminum. This model was validated using two additional lake sediments and one additional DWTR (R² = 0.995). Thus, the immediate inactivation of P in lake sediment following DWTR addition can be predicted with this model, which can be used with internal P loading or other water quality goals to determine an appropriate DWTR dose. Further recommendations were made about dosing DWTRs for lake restoration, allowing practitioners to use DWTR to inactivate P in lake sediment without conducting individual sorption experiments.

Drinking water treatment residual as a ballast to sink Microcystis cyanobacteria and inactivate phosphorus in tropical lake water.

The combination of a low dose of coagulant with a ballast that can inactive phosphorus (P) in lake sediment-a technique known as "flock and lock"-is one method for restoration of eutrophic lakes. The effectiveness of a drinking water treatment residual (DWTR) as a ballast in flock and lock was assessed using assays of eutrophic lake water from Thailand dominated by Microcystis aeruginosa cyanobacteria colonies by measuring changes in chlorophyll-a, pH, and zeta potential. P sorption isotherms were developed from long-term batch equilibrium experiments; desorption of nutrients and metals was assessed via leaching experiments; and morphological changes to cellular structure were assessed using scanning electron microscopy. Results showed that combining DWTR with a low dose of aluminum sulfate (0.6-4.0 mg Al/L) effectively sank 74-96% of Microcystis, with DWTR dose (50-400 mg/L), initial chlorophyll-a concentration (92-976 µg/L), pH (7.4-9.3), and alkalinity (99-108 ppm CaCO3) identified as factors significantly associated with sinking efficacy. P sorption capacity of the DWTR (7.12 mg/g) was significantly higher than a local soil (0.33 mg/g), enabling the DWTR to inactivate P in lake sediment. Desorption of Al, Fe, Ca and N from the DWTR was estimated to contribute to a marginal increase in concentrations of those compounds in the water column of a small shallow lake (1.2, 0.66, 53.4, and 0.07 µg/L, respectively) following a simulated application. Therefore, pre-treated DWTRs may be a viable alternative ballast in the flock and lock approach to lake restoration, supplementing or replacing modified local soils or lanthanum modified clays.

A comparison of aluminum dosing methods for reducing sediment phosphorus release in lakes.

Aluminum (Al) treatment is one of the most commonly used approaches to reduce internal phosphorus (P) loading in lakes. However, the adequate amount of Al that should be added to permanently inactivate mobile (releasable) sediment P can be determined using many different methods. These methods differ substantially in their specified design sediment depth, targeted P pool(s), and expected binding ratio. In this study, Al doses for inactivating sediment P in Beung Gii Lake of Thailand were determined using the most commonly used methods reported in literature and then compared. Experimental procedures included sediment P fractionation, Al assay experiments, and a geochemical model. Mobile P was detected in the lake's sediment at 2.52, 5.42, and 7.65 g/m2 in the upper 4, 10, and 15 cm, respectively, with additional P contained in labile organic form. Comparing the resulting Al doses for the lake, it was found they varied by nearly an order of magnitude (45-306 g Al/m2). This result highlights the importance of choosing a dosing method, because such a range of Al doses would likely result in highly variable levels of effectiveness and longevity, including both under- and overdosing. Based on the results of this study and a review of literature, a conservative, fixed ratio between Al and mobile plus labile organic sediment P (11:1) is recommended. All potentially releasable P (both mobile organic and inorganic forms) within the active sediment layer should be used to determine the total Al dose. Finally, the calculated Al dose in most cases will need to be split into sub-doses, based on lake morphology and total Al dose, to ensure maximum binding efficiency. Al dosing strategy should seek to minimize the risk for overdosing, maximize binding efficiency, and ensure all potentially releasable P forms are targeted during treatment.

Persistent and widespread long-term phosphorus declines in Boreal lakes in Sweden.

We present an analysis of long-term (1988-2013; 26years) total phosphorus (TP) concentration trends in 81 Swedish boreal lakes subject to minimal anthropogenic disturbance. Near universal increases in dissolved organic carbon (DOC) concentrations and a widespread but hitherto unexplained decline in TP were observed. Over 50% of the lakes (n=42) had significant declining TP trends over the past quarter century (Sen's slope=2.5%y-1). These declines were linked to catchment processes related to changes in climate, recovery from acidification, and catchment soil properties, but were unrelated to trends in P deposition. Increasing DOC concentrations appear to be masking in-lake TP declines. When the effect of increasing DOC was removed, the small number of positive TP trends (N=5) turned negative and the average decline in TP increased to 3.9%y-1. The greatest relative TP declines occurred in already nutrient poor, oligotrophic systems and TP concentrations have reached the analytical detection limit (1µgL-1) in some lakes. In addition, ongoing oligotrophication may be exacerbated by increased reliance on renewable energy from forest biomass and hydropower. It is a cause of significant concern that potential impairments to lake ecosystem functioning associated with oligotrophication are not well handled by a management paradigm focused exclusively on the negative consequences of increasing phosphorus concentrations.

Ecological resilience in lakes and the conjunction fallacy.

There is a pressing need to apply stability and resilience theory to environmental management to restore degraded ecosystems effectively and to mitigate the effects of impending environmental change. Lakes represent excellent model case studies in this respect and have been used widely to demonstrate theories of ecological stability and resilience that are needed to underpin preventative management approaches. However, we argue that this approach is not yet fully developed because the pursuit of empirical evidence to underpin such theoretically grounded management continues in the absence of an objective probability framework. This has blurred the lines between intuitive logic (based on the elementary principles of probability) and extensional logic (based on assumption and belief) in this field.

Longevity and effectiveness of aluminum addition to reduce sediment phosphorus release and restore lake water quality.

114 lakes treated with aluminum (Al) salts to reduce internal phosphorus (P) loading were analyzed to identify factors driving longevity of post-treatment water quality improvements. Lakes varied greatly in morphology, applied Al dose, and other factors that may have affected overall treatment effectiveness. Treatment longevity based on declines in epilimnetic total P (TP) concentration averaged 11 years for all lakes (range of 0-45 years). When longevity estimates were used for lakes with improved conditions through the end of measurements, average longevity increased to 15 years. Significant differences in treatment longevity between deeper, stratified lakes (mean 21 years) and shallow, polymictic lakes (mean 5.7 years) were detected, indicating factors related to lake morphology are important for treatment success. A decision tree developed using a partition model suggested Al dose, Osgood index (OI, a morphological index), and watershed to lake area ratio (related to hydraulic residence time, WA:LA) were the most important variables determining treatment longevity. Multiple linear regression showed that Al dose, WA:LA, and OI explained 47, 32 and 3% respectively of the variation in treatment longevity. Other variables (too data limited to include in the analysis) also appeared to be of importance, including sediment P content to Al dose ratios and the presence of benthic feeding fish in shallow, polymictic lakes.

In-lake measures for phosphorus control: The most feasible and cost-effective solution for long-term management of water quality in urban lakes.

Both in-lake and catchment measures designed to reduce phosphorus (P) loading were implemented as part of a 12.3 million USD restoration project for the Minneapolis Chain of lakes in Minnesota (USA). Treatment wetlands, 'in-pipe' measures, and in-lake aluminum sulfate (alum) treatment were applied to restore water quality in the four urban lakes. Different alum dosing methods led to between 4 and 20+ (modeled) years of water quality improvements in these lakes after treatment and only one of the four lakes continues to meet water quality goals approximately 25 years after the project started. Due to limited space and poor performance, reduction of total external loads was low (1-13%) for three lakes. Changes to internal P sediment release rates after application of alum correlated well with epilimnetic total P (TP) concentrations in these lakes, indicating that improvements in water quality were mainly driven by reduced internal loading via in-lake measures. Substantial reductions to external P loading were only achieved at Cedar Lake (49%) via conversion of an existing natural area to a treatment wetland, but even Cedar Lake is no longer meeting management goals. When expressed in terms of dollars spent per unit P removed, in lake alum treatment was on average 50 times more effective than in-catchment measures. The results of this study indicate that substantial external nutrient reductions may not be adequate to sustainably maintain water quality in urban lakes and that continued in-lake management of P accumulated in lake sediment will not only be necessary, but will also be more cost efficient relative to in-catchment measures.

Anthropogenic oligotrophication via liming: Long-term phosphorus trends in acidified, limed, and neutral reference lakes in Sweden.

Restoration of acidified lakes by liming does not, in many cases, improve productivity to a pre-acidified state. We hypothesize that the poor recovery detected in many of these lakes is due to constrained in-lake phosphorous (P) cycling caused by enhanced precipitation of metals in higher pH, limed waters. Long-term (1990-2012) data for 65 limed, circum-neutral (pH 6-8), and acidified lakes in Sweden were analyzed to determine trends for P and potential drivers of these trends. Limed lakes not only had lower mean values and stronger decreasing trends for total P than non-limed lakes, but they also had the highest percentage of decreasing trends (85 %). A P release factor (Hypolimnetic P/Epilimnetic P) was developed to elucidate differences in internal P cycling between lake groups. Consistently, lower P release factors in limed lakes show limitation of internal P cycling during summer months that may be a factor limiting P bioavailability and thus productivity of these systems.

A simple model for predicting aluminum bound phosphorus formation and internal loading reduction in lakes after aluminum addition to lake sediment.

The conversion of mobile phosphorus (P) to aluminum bound P (Al-P) after addition of Al to over 300 sub-samples from 35 sediment cores collected from 20 lakes in the upper Midwest, United States was investigated in this study. Consistent relationships between mobile P reduction and Al-P formation were detected across a broad range of mobile sediment P contents (0.04-2.8 g P m(-2) cm(-1) or 0.083-2.8 mg P g(-1)DW) and lake types. The conversion of mobile P to Al-P was dependent on the initial mobile sediment P content and the amount of Al added to the sediment. An empirical model was then developed to predict the formation of Al-P based on the amount of Al added relative to the initial mass of mobile P in the sediment. The results were compared to sediment collected from an Al treated lake and good agreement was found between the model and in-situ changes to sediment P fractions caused by Al treatment. The model developed in this study, unlike previous models with extreme, singular endpoints, allows for a continuum of estimates for mobile P conversion to Al-P, along with efficiency of P binding by Al, as Al dose varies. Model results can be used in conjunction with mobile sediment P based predictions for internal P loading to calculate an Al dose required to meet internal phosphorus loading goals for lake management and restoration without the need for expensive, time consuming Al additions to sediment.

Prediction of reference phosphorus concentrations in Swedish lakes.

The relationship between total phosphorus (TP) and chemical, climatic, morphological, and geographic variables was examined for over a thousand reference lakes across Sweden. A significant relationship was found between TP and both absorbance of irradiance (at 420 nm, filtered) and altitude for all lakes. These two variables alone, however, were not able to adequately predict TP concentrations in naturally turbid lakes. A natural particulate matter factor (PM(n)) was developed as part of this study to incorporate the effect of natural suspended matter on lake phosphorus concentration. Variability in TP concentration was well explained with the addition of PM(n) to the model (R(2) = 0.71) even though conditions external and internal to the lakes varied greatly. The ability of the three parameter model to identify culturally eutrophic systems was then successfully tested using a data set of lakes with known anthropogenic phosphorus loads. Thus, the model allows for estimation of reference TP concentration and by extension can be used to identify when a reference concentration has been exceeded due to anthropogenic phosphorus loading. The model output also provides a realistic end point to which phosphorus concentrations should be reduced to achieve a natural trophic state in both clear water and naturally turbid lakes.

Variability in phosphorus binding by aluminum in alum treated lakes explained by lake morphology and aluminum dose.

Sediment cores from six aluminum sulfate treated lakes in Minneapolis, MN were analyzed to determine the effectiveness of phosphorus (P) binding by aluminum (Al). Two of the study lakes are polymictic and the remaining four are dimictic. Above background concentrations of Al and Al-bound-P (P(Al)) were detected in all six lakes at varying sediment depths. In contrast to previous studies, however, the binding relationship between Al and P was not consistent between lakes and substantial variation was also detected within each sediment profile. Average lake sediment Al:P(Al) ratios ranged from 5.6 to 15 (molar) with higher ratios, or less efficient P binding, generally being detected in deep, dimictic lakes with high sediment Al content due to treatment. Multiple linear regression was used to explain the variability among average Al:P(Al) ratios detected in each core and a lake morphometry index (Al Depth Index, core collection depth divided by the square root of lake area) along with Al dose described most of the variation (92%). Even though P bound to the added Al appears to be permanently removed from the internal P cycle in each lake (as evidenced by burial with new sediment), the differences in binding efficiency may indicate lower P inactivation, on a per unit Al basis, when elevated amounts of Al are added to the sediment, especially in deeper areas of lakes where sediment focusing may cause elevated Al accumulation to occur.

A method for comparative evaluation of whole-lake and inflow alum treatment.

A method is described to evaluate two methods of phosphorus (P) management in lakes using aluminum sulfate (alum)--in-lake and tributary (inflow) treatment--and compare the resulting in-lake P levels. For in-lake treatment, a technique is described to calculate the optimum alum dose based on measurement of "mobile P" in lake sediments. Mobile P is defined as loosely sorbed and Fe-P, the fraction of sediment P subject to release under anoxic conditions. A linear relationship (r2 = 0.90) was found between P-release rate and the mobile-P content in sediment cores. Addition of alum to aliquots of sediment showed predictable relationships between (i) alum dose and aluminum-bound P (Al-P) formed and (ii) mobile-P loss and Al-P formation. The decrease in sediment P release that would result from in-lake alum treatment was estimated from the residual mobile P after treatment. A method also is presented to estimate the amount of alum needed to bind potentially mineralizable sediment organic P. For inflow treatment, jar tests with urban runoff in metropolitan St. Paul and Minneapolis, Minnesota (USA) were used to study effects of alum dose on P removal from water. With sufficient mixing, a dose of 8 mg AlL(-1) reduced total P (TP) and soluble reactive P to low levels regardless of pH, TSS, and TOC, but doses