The release of dissolved phosphorus from lake sediments

Boers, P.C.M.


Chapter 1. Introduction: Eutrophication is one of the world's major water quality problems. Attempts to alleviate eutrophication of lakes have involved the control of phosphorus loadings. In such cases, an internal loading of phosphorus from the sediments may retard an improvement of the water quality. Chapters 2 - 5 deal specifically with the sediments of Lake Loosdrecht.

Chapter 2. Distribution and forms of phosphorus in the sediments of the Loosdrecht Lakes (the Netherlands): The phosphorus loading entering the Dutch Lake Loosdrecht was reduced in 1984. A survey of the sediments of this lake indicated that an enrichment of these sediments with phosphorus was limited to the area near the former inlet of polluted water from the River Vecht. In most other areas the phosphorus accumulation in the upper 18 cm. was 10 - 15 g P m -2, equal to the external P loading of approximately one decade. This is probably due to the transport of phosphorus from the surficial sediments with the percolating. seepage water. A rough estimation indicates that the amount of phosphorus carried away with this seepage compensates for the external loading. Results of selective extractions, desorption experiments and bio-assays all indicate that 5 % or less of the phosphorus in the upper sediment layers is "available" for algal growth.

Chapter 3. Unmasking the particulate organic matter in a lake ecosystem: origin and fate of POM in the shallow eutrophic Loosdrecht Lakes: The sediments of Lake Loosdrecht consist for over 50 % of organic matter. Pyrolysis mass spectroscopical analysis indicated that over 95% of this organic matter is old macrophyte debris, probably originating from the peat deposits in the lake area. In winter, this material also contributes significantly to the seston in the water column. In summer the seston consists almost entirely of algal material. However, only traces of this latter material could be detected even in the upper mm, of the sediments, suggesting that the influx of algal matter in the sediments is small or that a fast degradation of the material reaching the sediments takes place.

Chapter 4. Phosphorus release from the peaty sediments of the Loosdrecht Lakes (the Netherlands): The release of phosphorus from the sediments of Lake Loosdrecht was studied in some detail using continuous flow reactors. The highest release rates, up to about 4 mg P m -2d -1, were found in summer and the lowest, less than 0.5 mg P m -2d -1, were measured in winter. Temperature and downward seepage are the two most important factors controlling release rates. pH of the water column, which many other researchers found to be an important controlling factor, did not play any role in Lake Loosdrecht. The influence of temperature and seepage were more pronounced in summer than in winter. The magnitude of the release rate and the influence of temperature and seepage could be largely, explained by a simple steady-state model based upon mineralization and precipitation processes. According to this model, the phosphorus release from the sediments of Lake Loosdrecht is mainly governed by mineralization processes. Retardation of phosphorus by precipitation plays only a minor role.

Chapter 5. Ion concentrations in interstitial waters as indicators for phosphorus release processes and reactions: Processes governing the release of phosphorus in the sediments of Lake Loosdrecht were studied in more detail using pore water concentrations of phosphorus, ammonium, iron and DOC and pH. The local differences in pore water composition were much larger than those in sediment composition. So, pore water concentrations are much more sensitive indicators for processes in the sediments than sediment composition. From the close correlation between phosphorus and ammonium concentrations it was concluded that mineralization is the dominant process regulating phosphorus concentrations in the pore waters. Possibly, also precipitation of vivianite plays a role. Finally, several indications were found that the large amounts of refractory organic matter in the sediments of Lake Loosdrecht limit the availability of iron for binding of phosphorus.

Chapter 6. The influence of pH on phosphate release from lake sediments: The release of phosphorus from the sediments is often believed to be increased by a high pH in the water column. This is usually attributed to a decreased adsorption capacity for phosphate of iron(III)oxyhydroxydes at the sediment-water interface. This mechanism is commonly studied in laboratory experiments in which the pH of the water column is manipulated by titration with NaOH. In nature, however, pH is increased by a reduced partial CO 2 pressure caused by uptake of CO 2 , by the phytoplankton. This process was simulated in the laboratory in continuous flow reactors by stripping the CO 2 , from the air which was lead through the water column. In the latter case the phosphorus release rate increased much less than when using the traditional experimental setup with NaOH. Measurements of pH gradients near the sediment-water interface showed that when using NaOH additions the high pH in the water column penetrated much deeper in the sediments than when using CO 2 , stripping. Addition of NaOH increases the alkalinity, and the transport of alkalinity from the sediments to the water reduces or even reverses in direction. By removing CO 2 , the alkalinity is not changed or eventually reduced by precipitation of calcite.

Chapter 7. Lake restoration: estimation of internal phosphorus loading after reduction of external loading from sediment data: From literature data on a number of lakes with reduced external phosphorus loadings it was concluded that in most cases an annual net loading from the sediments takes place during several years. For the limited number of lakes for which sediment data were available, the internal loading rate was found to be highly significantly related to the phosphorus content and the P/Fe ratio in the sediment. These relationships can help to predict the internal loading rate, to be expected after reduction of the external loading.

Chapter 8. Phosphorus fixation with iron (III)chloride: a new method to combat internal phosphorus loading in shallow lakes?: The traditional method to speed up the recovery of a lake with a significant internal loading is to remove the phosphorus rich upper sediment layers. However, this method is costly and time consuming. Treatment of the sediments with phosphorus binding compounds might be an alternative. Iron(III)chloride is a candidate, as it is a natural phosphorus-binding agent and non-toxic. Laboratory experiments indicated that addition of at least 100 g Fe m -2to lake sediments will strongly decrease the phosphorus release rate. The sediments of the shallow lake Groot Vogelenzang were treated with 100 g Fe m -2resulting in an immediate improvement of the water quality. This improvement lasted for only three months. A large external phosphorus loading, caused by rapid flushing with water from the surrounding lakes is held responsible for this failure. Nevertheless, release experiments and phosphorus budget calculations both indicated that the phosphorus retention in the sediments of lake Groot Vogelenzang had increased with about 3 mgP m -2d -1,

Chapter 9. Changes in phosphorus cycling in a shallow lake due to foodweb manipulations: Biomanipulation is a recently developed method to improve water quality in lakes by removing undesired fish species and reintroducing other, desired but lacking species. 'Me goal of this lake restoration technique is to improve the predation on whitefish and thereby to decrease the predation of whitefish on zooplankton. Application of biomanipulation in small, shallow lakes usually results in a highly increased Secchi depth and strongly reduced algal biomass. In this chapter, the changes in the phosphorus cycling of the shallow lake Wolderwijd/Nuldernauw after biomanipulation were quantified. Before biomanipulation, most of the phosphorus in the food web is accumulated in fish, but a small portion is recycled rapidly in the water column. Sedimentation of detritus is the most important loss process. No evidence could be found for the hypothesis that the bottom stirring activities of fish enhance the phosphorus release.

The availability of phosphorus for phytoplankton may decrease after biomanipulation for two reasons. The first is that the detritus formed from green algae is more refractory than the detritus from cyanobacteria. Therefore, the phosphorus content of cyanobacteria is recycled more rapidly. Cyanobacteria. will probably disappear after biomanipulation due to the improved light climate. The second reason is that the increased bottom area covered with submerged macrophytes enhances the net sedimentation of detritus from the water column and concomittantly the regeneration rate of phosphorus in the water column decreases. These changes in the phosphorus cycling in the lake may reinforce the success of biomanipulation.


Undoubtedly, the release of phosphorus from sediments forms an important part of the phosphorus cycling in shallow lakes. In the Loosdrecht Lakes, this release was found to be of the same magnitude as the external loading, although only a small proportion of the sediment phosphorus is "exchangeable". Mineralization of organic phosphorus reaching the sediments is the most important process for the phosphorus release. The availability of iron in the sediments of these lakes appeared to be an important factor in the retention of the phosphorus that is released by mineralization. The presence of large amounts of refractory organic matter in the sediments of the Loosdrecht Lakes seems to limit the potential of iron to bind phosphorus. The sources and fate of iron in aquatic sediments need further research. Probably, the availability of iron for the binding of phosphate is related to the cycling of other elements, such as sulfur, in the sediments.

Just as important for the phosphorus cycling in a shallow lake is the settling to the sediments. The characterization of the particulate organic matter in the sediments of the Loosdrecht Lakes indicated that either the influx of algal matter into the sediments is very small or that this material is mineralized very rapidly. Possibly, the slow recovery of Lake Loosdrecht and other shallow, wind-exposed lakes is not only caused by internal phosphorus loading, but also by a poor sedimentation of particulate phosphorus from the water column.

A quantification of the cycling of phosphorus through the food web revealed that changes in the sedimentation process may affect the availability of phosphorus to algae. In turn, the sedimentation is influenced by the presence of submersed macrophytes.

Attention for the environmental conditions is necessary for a proper quantification of the internal loading rates. Some of the relevant conditions were discussed in this thesis. Resuspension of bottom material by benthivorous fish is not of any importance for the release of phosphorus from the sediments. The influence of the pH in the water column was found to be less important than reported by other researchers. Temperature and the sedimentation of particulate organic matter are two reasons for release maxima in summer.

One of the most important questions in lake restoration is whether the improvement of the water quality will be delayed, following a reduction of the external loading. Although the results of the studies in Lake Loosdrecht indicate that the internal loading is of the same order of magnitude as the external loading, no direct proof of a delay of the recovery of the lake due to this internal loading could be found.

From literature data on a number of shallow lakes with reduced external phosphorus loadings it could be concluded that indeed initially a net internal loading may take place upon restoration. The magnitude of this internal loading is related to the phosphorus content of the sediments and to the P/Fe ratio in the sediment. This latter may be a measure of the residual phosphorus binding capacity of the sediments. As long as no other methods to predict the occurrence of internal loading are available, the relations found between internal loading and sediment composition may be used. The question to what extent chemical extractions, bio-assays or laboratory release experiments can be used to predict the behavior of a lake after a reduction of the external loading is an important but still unanswered one.

According to this study, upon restoration a net internal loading can be expected to take place already at a phosphorus content of the sediment of 1.4 mg P g -1. So, internal loading will take place in most of the Dutch lakes after a decrease of the external loading. To what extent this internal loading will delay the improvement of the water quality, depends on other characteristics of the lake, as the residual external loading and the residence time. Although internal loading will not last forever, in a number of cases additional "sediment restoration techniques" will be needed to recover lakes within an acceptable space of time. As dredging of the upper sediment layers is extremely costly and time consuming, alternatives for this sediment restoration technique are needed. Fixation of phosphorus in the sediments with phosphorus binding chemicals may be an alternative. In this study, the availability of iron was found to be an important factor in the immobilization of phosphorus in freshwater sediments and addition of iron compounds to the sediments may strengthen this natural Process. Although the method is promising, its suitability to reduce internal loading must still be proven. An important unanswered question is to what extent the added iron remains available for phosphorus binding.