Lakes in the Netherlands, their origin, eutrophication and restoration: state-of-the-art review<Superscript>*</Superscript>

This article starts with a brief description of the origin and eutrophication of shallow Dutch lakes, followed by a review of the various lake restoration techniques in use and the results obtained. Most freshwater lakes in the Netherlands are very shallow (<2 m), and owe their origins to large-scale dredging and removal of peat during the early 17th century. They vary in area from a few hectares to a few thousand hectares, and are generally found in the northern and western part of the country. Most of them lie in the catchment areas of the major rivers: the Rhine, the Meuse and the Schelde. Because of their natural and aesthetic value, these lakes fulfil a recreational function. The lakes are important to the hydrology, water balance and agriculture in the surrounding polder country. The external input to the lakes of phosphorus (P) and nitrogen (N) and of polluted waters from the rivers and canals have been the major cause of eutrophication, which began during the 1950s. In addition, more recently climate changes, habitat fragmentation and biotic exploitation of many of these waters have probably led to loss of resilience and thus to accelerated eutrophication. Lake eutrophication is manifested essentially in the poor under-water light climate with high turbidity (Secchi-disc, 20–40 cm) caused usually by cyanobacterial blooms (e.g. Oscillatoria sp.), and loss of littoral vegetation. Despite recent perceptible reductions in external P inputs, non-point sources, especially of N from agriculture, still remain high and constitute a major challenge to the lake restorers. Lake recovery is also invariably afflicted by in-lake nutrient sources. These include P loading from the P-rich sediments, mineralization in the water and release by the foraging and metabolic activities of the abundant benthivorous and planktivorous fish, mainly bream (Abramis brama). A variety of restoration techniques have been employed in the Dutch lakes: hydrological management, reduction of P in the external loads, in-lake reduction or immobilisation of P, and complementary ecological management. This last involves biomanipulation, or the top-down control of the food web. Hydrological management has resulted in an improvement in the lake water quality only in a few cases. The failure of lake restoration measures (e.g. in the Loosdrecht lakes, described as a case study) has led water managers to use biomanipulation in other lakes under restoration. Lake biomanipulation principally involves reducing the existing planktivore population, bream in most cases, and introducing piscivores such as northern pike (Esox lucius). Lake Zwemlust is discussed as a case study, with brief mention of some other small lakes which have been biomanipulated. The restoration studies reveal that decrease of P to low levels is no guarantee that cyanobacterial populations will also follow suit. This is because cyanobacteria can withstand great variation in their P content and thus in their C:P ratios. Thus, for a unit weight of P, the Cyanobacteria can yield relatively more biomass and cause greater turbidity than, for example, green algae, which have relatively lower C:P ratios. This is possibly an explanation for the success of these filamentous Cyanobacteria in many Dutch lakes, and the failure of restoration endeavours. In addition, a falling trend in chlorophyll-a content in these shallow lakes does not set off an immediate increase in ∗ NIOO Publication no. 2991.