A paleolimnological perspective on liming – implications for defining reference conditions in Swedish lakes

Sammanfattning: Using paleolimnological techniques, I have studied surface-water acidification and the effects of liming in Swedish lakes on a decadal to millennial time-scale. The overall objective was to contribute to the discussion on the fulfilment of goals within the Swedish liming program. One of the main goals of this program is to restore lakes to natural or nearly natural conditions, i.e. to a reference condition as termed in the EU Water Framework Directive. In this context, a key issue is to define reference conditions. This is a central theme of my thesis, as lake sediments offer a unique way to study past lake conditions.Past lake-water acidity of 12 reference lakes in the Swedish liming program (ISELAW) was determined using diatom analysis of sediment cores. Pollen, lead, and flyash from coal/oil combustion were used as indicators of impact from land use and atmospheric pollution. A general trend in these lakes is an initial decline in pH after lake formation due to natural soil processes, which was then followed by rather low pH values (pH 5.3-6.5). In six of the lakes pH increased as a result of expansion of agriculture (burning, forest grazing) 2000 to 1000 years ago. Local mining and long-range airborne pollution have also impacted the lakes since medieval time. These results show that the conditions of the study lakes were not natural prior to industrialization and recent (20th century) acidification.The ISELAW lakes were selected on the basis of representing typical limed lakes, and they have been limed and monitored since at least the 1980s. A comparison of chemical/biological monitoring data and the paleolimnological data gives somewhat diverging results. Most of the monitoring data suggest that the lakes were subjected to acidification during the 20th century, but the paleolimnological data can only identify clear evidence of acidification in five of the 12 lakes, hence, all lakes were probably not recently acidified. According to conclusions from monitoring the lakes have recovered following liming. The paleolimnological data give a more complex picture and three different responses have been identified: 1) a return to a diatom composition found in the lake one hundred to several thousand years ago; 2) very small shifts in the diatom composition; or 3) a diatom composition previously not found in the lake. The latter response raised the question whether liming can cause an unnatural diatom community. A comparison of diatoms in surface sediment samples of 31 limed lakes with pre-industrial reference samples from 291 lakes showed that liming does not create an unnatural diatom composition. These results illustrate that the goals for liming were not reached in all of the limed lakes, and that paleolimnology can play an important role for assessments of acidification and liming. The comparative study also highlights the importance of designing monitoring programs that can produce reliable and long data series.Given the results of the paleolimnological investigations, it is obvious that we cannot assume that the 19th century represented a natural or near natural state, and thus is a realistic reference conditions. Natural long-term lake development and previous land-use impacts need to be considered in defining reference conditions. Neither can we disregard the fact that humans always will impact nature. Although paleolimnological studies are time consuming, I believe that they could be simplified to the extent that paleolimnology could become a routine method for environmental management.