Umeå University's logo

Disentangling the process information contained in a diatom sediment signature is crucial for reliable future predictions based on paleolimnological records. The aim of this study was to identify the hierarchical importance of seasonal processes for the formation of an annual diatom sediment record. We combined annual vertical water column chemistry, lake thermal structure, sequential sediment trap samples and annually resolved varved sediment diatom records from a boreal lake in northern Sweden. Whole-year water column monitoring indicates that the entire annual diatom sediment signal in different years can be driven by winter air temperature-induced timing of ice and snow melt and persistent under-ice stratification, which promote an early diatom bloom under the ice before spring lake over-turn. In a year of late ice thinning and development of a chlorophyll maximum after spring lake over-turn, an annually integrated diatom sediment signal was built by a continuous diatom flux, with important implications for diatom-based model predictions.

Sediment trap studies are of great importance to develop a deeper understanding of how environmental processes are influencing sediment deposition. We analyzed diatom assemblages from 2002 to 2015 in a sediment-trap record from a boreal lake (Nylandssjön, Sweden) together with environmental parameters, and compared the diatom record in the trap with the corresponding annually laminated sediments. In general, our data set indicates a large year-to-year variability of diatom succession and abundance patterns, which is well reflected in the varved sediments. The relative abundance of major taxa seems primarily controlled by seasonal characteristics, such as above-average winter air temperature and autumn runoff. Specifically, Cyclotella glomerata dominated the annual sediment trap record (and also the corresponding sediment varves) in years with warmer air temperatures in March-April, and Asterionella formosa dominated the annual sediment assemblages in years characterized by higher runoff before autumn lake over-turn. This study highlights that short-term environmental conditions and extreme events are of major importance for interpreting annual sediment signals.

Assessing ecosystem dynamics beyond experimental time-scales is a prerequisite to understand effects of ongoing environmental change and essential to improve predictions of future conditions. In this study, we analyze a time series of an annually resolved sediment record from a boreal lake (Nylandssjön) in northern Sweden, consisting of diatom data since AD 1931 and geochemistry data since AD 1953. The aim of this paper is to quantify to which extent the main primary producers (i.e., diatom assemblage) in a typical boreal lake are related to processes in the lake catchment, sediment properties, and climate variables (temperature, precipitation, circulation patterns). The diatom assemblage consists mainly of Cyclotella taxa (C. glomerata, C. radiosa, and C. rossii), Asterionella formosa, Fragilaria delicatissima and other small Fragilaria taxa. Although the dominant taxa remain the same over the entire sediment record, their relative abundances exhibit some distinct shifts. We use ordination techniques and variance partitioning to quantify the explanatory power of environmental variables to explain shifts in the diatom assemblage composition. A series of redundancy analyses (RDA) using one sole constraining variable showed that both geochemical proxies and climate variables explained a statistically significant amount of the variance in the diatom data. The most powerful sediment proxy variables were Pb (ppm, 17.9%), C (%, 16.8%), N (%, 15.7%) and sedimentation rate (g cm-2 yr-1, 13.2%). For climate variables, temperature during winter/spring (April 6.4%, May 5.0%, February 4.6%) and the Northern Atlantic Oscillation index (NAO 6.4%) showed the highest predictive power, whereas precipitation data did not turn out to be a powerful predictor. Variance partitioning showed that sediment proxies (sedimentation rate, C, N; 20.8%), variables associated with pollution (Pb, Cu, P; 23.1%) and climate (winter/spring temperature, NOA; 10.4%) explained significant amounts of the variance in the diatom data. This quantification indicates to which extent sediment properties and climate conditions are jointly responsible for shaping the ecosystem properties in boreal lakes, highlighting in particular the importance of winter climate.

To assess the rates of compaction in recent, varved (annually laminated) lake sediments, we used a collection of 13 freeze cores sampled from 1979 to 2012 in Nylandssjon (northern Sweden). This unique series of stored freeze cores allowed us to measure how the thickness of individual varves changed when they were overlain by new varves. The compaction rate was greatest during the first few years after deposition; varve thickness decreased by 60% after 5 years (sediment depth 5cm). Thereafter, the compaction rate declined, but after 33 years (the study period) there was still a weak trend of continued compaction (sediment depth 12cm). The rate of compaction is clearly linked to loss of water along with an increase in dry-bulk density of the sediment. Despite compaction causing considerable varve thickness changes over time, the year-to-year variation in varve thickness is preserved while sediment ages. Understanding compaction processes is of fundamental importance to paleolimnology, for example in studies involving calibration of varve thicknesses of recent varves against environmental parameters.