The large lakes of Sweden, Mälaren, Hjälmaren, Vättern and Vänern, have been subjected to water-quality monitoring for almost four decades. Physicochemical variables, plankton and benthic invertebrates have been regularly assessed. Hydrological and sediment conditions, macrophytes, fish, primary production, bacteria and attached algae have been periodically investigated. The human impact, including industrial activities, was reflected in excessive amounts of organic matter, nutrients, metals and persistent organic compounds. From the late 1960s all municipal sewage works in the catchments of the lakes were upgraded to the highest technical standard, including chemical precipitation of phosphorus, and phosphorus discharge from the sewage works was thereby reduced by 90–95%. In addition, industries were obliged to restrict discharge of harmful substances. The reactions of the lakes to the remedial measures are discussed as well as the value of various indicators. The studies were instrumental in designing a national lake monitoring program. Additionally, results from large lake monitoring have contributed to the establishment of national water-quality criteria including, physical, chemical, and biological indicators.

The four largest Swedish lakes Vänern, Vättern, Mälaren and Hjälmaren all have a tectonic origin with multidirectional fault fissures, which influence the topography of the bottoms and the creation of subbasins. All the four lakes have been subject to several glacials, the latest of which started to recede from central Sweden c. 10 000 years ago. The many differences between the four lakes can be explained by their shapes, sizes and the relation between the sizes of the drainage basins and the size of the lakes. This paper focuses on processes considered relevant to aquatic organisms of the lakes: water balance, water residence time, water temperature, water exchange between subbasins within the lakes, and other internal water movements. Links between pollution, water exchange and retention of nutrients are discussed with an example from Lake Vänern.

In-lake concentration changes of phosphorus (P) and nitrogen (N) in lakes Vättern, Vänern, Mälaren, and Hjälmaren in response to diminished input has been examined from the mid-1960s onwards. In the former two deep and oligotrophic lakes with slow water renewal, drastic reductions in P-input from the middle of 1970s caused just minor reductions in P-concentration over a very long time. At the same time accumulation occurred in the water mass of inorganic N and possible reasons are discussed. In the latter two mesotrophic to hypertrophic lakes, two shallow basins in L. Hjälmaren showed slow recovery due to release of P from sediments. The same basins and two basins in L. Mälaren have suffered from N-deficiency, particularly during the pre-phosphorus reduction years, and nitrogen fixation was indicated. In two L. Mälaren basins recovery of in-lake P concentrations was better than expected in comparison to the so-called IMSA-model for lake recovery from nutrient pollution. In the other five lakes/basins chlorophyll concentrations after 20 years were similar compared to those modeled.

Riverine phosphorus (P) concentration and P-transport to Lake Mälaren, the third largest lake in Sweden, has been monitored for 35 years in 12 major tributaries. During a period of 15 months, complementary assessments of particulate P, suspended matter and dissolved reactive P were made. Particulate P comprised 64%, dissolved unreactive P 23% and dissolved reactive P 13% as flow-weighted means, with high seasonal variability. “Background” or “reference” P-transports in the streams were estimated by different methods and the anthropogenic contribution to P-transport was shown to be reduced over time. Potential algal availability of particulate P showed a mean availability of c. 45% for water draining arable and forested land, while the availability was higher for sewage discharge particulate P and algal P. Calculated total bioavailable P in tributaries was shown to equal reactive P measured on coarsely filtered water and it was indicated that the proportion of bioavailable P was higher during the period with high anthropogenic P-contribution than with a lower contribution.

Lake Mälaren is the water supply and recreation area for more than 1 million people in central Sweden and subject to considerable environmental concern. To establish background data for assessments of contemporary levels of trophy and heavy metal pollution, sediment cores from the lake were analyzed. Diatom–inferred lake–water phosphorus concentrations suggest that pre-20^th century nutrient levels in Södra Björkfjärden, a basin in the eastern part of Mälaren, were higher (c. 10–20 µg TP L⁻¹) than previously assumed (c. 6 µg TP L⁻¹). Stable lead isotope and lead concentration analyses from 3 basins (S. Björkfjärden, Gisselfjärden and Asköfjärden) show that the lake was polluted in the 19^th century and earlier from extensive metal production and processing in the catchment, particularly in the Bergslagen region. The lake has experienced a substantial improvement of the lead pollution situation in the 20^th century following closure of the mining and metal industry. The lead pollution from the old mining industry was large compared to late-20^th century pollution from car emissions, burning of fossil fuels and modern industries.

The four largest Swedish lakes, Vänern, Vättern, Hjälmaren, and Mälaren, were surveyed by means of aerial IR-color photography in the 1970s. Along small slope gradients in all lakes the emergent communities extended to a width of hundreds of meters, while the extent on steeper slopes was a few meters. Because of the fractionate shape, L. Mälaren has a very long shoreline, which promotes vegetation growth in the transition zone between land and water. Sheltered conditions are numerous and water lilies and other floating leaved and freely floating plants are abundant. Emergent as well as submerged macrophytes grow vigorously in the northernmost part of the oligotrophic lake Vättern. In L. Vänern exposed shores predominate, but sheltered bays and archipelagos are overgrown with reeds. In the shallow lake Hjälmaren monospecific stands of Phragmites australis are dominant. Recurrent surveys of L. Mälaren and L. Vänern in the 1990s showed changes that could be connected to the watertable fluctuations, which have been stabilized since the first investigation. Reeds and water lilies expanded in sheltered parts of the lakes, while the distribution of reed stands growing at their deep limit decreased.

Twenty alien species have become established in the lakes Mälaren, Vänern, Vättern and Hjälmaren. Intentional introductions include fish and the signal crayfish from North America, ornamental plants, and the Canada goose. Unintentional introductions include the crayfish plague introduced with infected crayfish, the zebra mussel, and Chinese mitten crab introduced with ballast water. The introduction of pathogens and parasites, in particular the crayfish plague, to the lakes has had the greatest environmental and socioeconomic effects and has contributed to the decimation of the indigenous noble crayfish. The stocking of brown trout and salmon with origins from different biogeographical regions has contributed to the extinction of relict indigenous fish species in L. Vänern. Although major ecosystem damage caused by the introduction of alien species, with the exception of the crayfish plague, has not occurred in the four large Swedish lakes, local problems of considerable dignity occur occasionally.

The four largest Swedish lakes, Vänern, Vättern, Mälaren, Hjälmaren, host important commercial fisheries for char, salmon, trout, whitefish, vendace (cisco), perch, pike-perch, pike and eel, i.e. highly diverse biological resources. Case studies illustrate physical, chemical and biological impacts on some of these commercial species caused by constructions of dams and ship canals, eutrophication, and overexploitation. Although some original species have been lost and a few new species have been added, the recent human interference has basically caused major shifts in dominance of the fish community structures because of eutrophication, alterations in the abundance of eel or crayfish, and due to overfishing. The latter is in some cases caused by the Great Lake Fishery Paradox—in an environment with several predators and competitors, but with ample food resources, especially salmonid fish but also species like pike-perch may adapt a life history favoring growth over sexual maturation. If harvested at a conventional size these populations will decline rapidly due to too small spawning stocks.

Phytoplankton and environmental variables have been monitored in the large Swedish lakes Mälaren, Hjälmaren, Vättern and Vänern since the 1960s. Measures to reduce phosphorus input and industrial waste products were taken during the 1970s. The phosphorus loading was then reduced by 90–95% resulting in a halving of the phosphorus concentrations in the most affected basins. The phytoplankton community reacted rapidly with decreased biomasses of cyanobacteria in summer as well as decreased biomasses of spring diatoms and cryptophycean flagellates. Other reactions were a contracted period of waterbloom, an increased taxon richness, an increased evenness in the biomass over the growth season, and a change in the species size structure within the phytoplankton community. Furthermore, the species richness in the large lakes is compared in relation to lake characteristics. A presentation of the occurrence of toxic cyanobacteria in the lakes is also given. Maximum–minimum values of 13–0.1 µg microcystin L⁻¹ are established in connection with waterblooms in Hjälmaren and Mälaren. The use of phytoplankton as a monitoring variable to detect water-quality changes is outlined and assessment criteria are presented.

The past effluents of mercury (Hg) into Lake Vänern were considerable. The consequences of, and recovery from these have been monitored through continuous measurements of mercury in sediment and fish. Mercury levels in lake sediments in the vicinity of the main source of mercury, a chloralkali plant on the northern shore, have only decreased by slightly more than a half since the mid-1970s, despite a radical decrease in effluents from the source, already during the 1960s. The mercury levels in pike (Esox lucius) have decreased to a similar extent during this time period. They are now about 30% higher in the worst affected parts of the lake compared to the least affected parts. Lower levels have been measured in perch (Perca fluviatilis) and salmonoid fish in the lake. Despite the increased presence of mercury in the sediment of Lake Vänern, the mercury levels in the fish of the lake are relatively low compared to fish in lakes situated in the same region, but not affected by any local mercury effluents. As calculated, the total fish biomass of L. Vänern holds less than 1000th of the amount of mercury contained in the upper, biologically active layers of the bottom sediment of the lake. This demonstrates the potential influence of various environmental factors and motivates continued monitoring of mercury levels in the lake in the future.

Lake Vättern in southern Sweden is a large oligotrophic lake with high surface to catchment area ratio (ca. 0.4) and a water residence time of 60 years. The lake combines sensitivity to atmospheric POP-pollution with general susceptibility for effects of POPs and slow concentration decline. Time series, from the 1960s until 1996, of PCBs and DDT in fish and data on TCDD in sediment of L. Vättern are presented and compared to other large lakes. The long time dataset of POPs in Arctic char (Salvelinus salvelinus) shows a significant annual decline of about 5% for PCBs and 13% for DDT. These rates are in agreement with other studies. Nutrients as well as biological effects and factors affecting the fate of POPs in oligotrophic lakes with long residence times are discussed. We propose that oligotrophic clear lakes are important ecosystems for monitoring POPs in biota.

Factors affecting long-term (1982–2000) population densities of the glacial relict amphipod Monoporeia affinis were studied in Sweden's three largest lakes. Monoporeia showed large population fluctuations in all three lakes, with conspicuous peaks in density occurring in Lakes Vänern and Mälaren. In Lake Vänern, amphipod densities showed highly significant relationships with spring maximum diatom biovolume at a 1-yr lag. The lack of relationship between diatom biovolumes and Monoporeia densities in L. Vättern is likely due to the larger depth and the lower nutrient content of this lake. In eutrophic L. Mälaren, summer hypoxia (< 4 mg O₂ L⁻¹) is likely an important regulating factor. Hypolimnetic temperature showed a clear periodicity with relatively warm deep water occurring between 1989 and 1994. Hypolimnetic temperatures in Vänern and Vättern were correlated with total solar irradiance. However, neither hypolimnetic water temperature nor diatom biovolumes correlated with the North Atlantic Oscillation winter index. We speculate that variations in temperature and nearbottom oxygen concentrations negatively affect population densities by acting on recruitment success (reproduction) and juvenile (young-of-the-year) survival.

Pelagic fish population biology was studied in the large Swedish lakes Vänern, Vättern, Mälaren and Hjälmaren. It is crucial for fish fry in temperate regions to hatch early in the growth season to survive, and achieve large size before winter, and it is suggested that the key factors are to match the spring development of phyto- and zooplankton, but to avoid predation. This is more easily accomplished by the studied spring spawners smelt (Osmerus eperlanus) and pike-perch (Stizostedion lucioperca) than autumn spawners, such as vendace (Coregonus albula). It is shown that hatching of vendace fry shortly after ice-break-up is beneficial for year-class strength. In oligotrophic large lakes with few predatory species a rapid increase in water temperature after ice-break is also promoting recruitment, whereas this is not the case in eutrophic lakes where predation pressure from other species may become too high. The results indicate that autumn spawners will have difficulties in adapting to global warming and it is also suggested that the life history can explain the large variations observed in year-class strength between years.

Winters in Sweden have become warmer in the 1990s, and as a consequence the timing of ice break-up and the growth and decline of spring phytoplankton has shifted, starting earlier. Even spring temperatures have become warmer, leading to an earlier beginning of the summer phytoplankton growth. The spring-ward shift in phytoplankton population growth has resulted in an extension of the growing season by at least one month. Although mean total phytoplankton biomass from May to October has not increased, the spring and early summer biomass of temperature-sensitive phytoplankton groups, such as cyanobacteria and chlorophytes, has increased in the 1990s. No increase was noted for other phytoplankton groups. Considering that some species of cyanobacteria that commonly occur during a summer bloom, such as Anabaena, Aphanizomenon, and Microcystis, can be toxic, the effect of warmer winters on aquatic ecosystems is potentially far-reaching.