Climatic signals and frequencies in the Swedish Time Scale, River Ångermanälven, Central Sweden

Abstract: Any future climate variation forced by human activities will be superimposed on the background of natural climate variation. Therefore, before interpreting the present climate and addressing future climate scenarios some knowledge of past climate is necessary. This thesis offers a rare glimpse into a long record of fluvial activity in Central Sweden and illuminates some of the possible forcing agent behind past (and future) discharge variation. Along the Swedish East Coast varved deposits of sand silt and clay couplets make up a chronology, which extend from the present into the Late Glacial. This chronology is known as the Swedish Time Scale (STS) and the c. 8000 varves were deposited in River Ångermanälven, Central Sweden. Of these varves, the last c. 2000 years are considered secure in terms of coherent chronology and internal thickness variation. A 2000 year long geometric mean varve thickness series was calculated to account for the internal thickness variation, which is postulated to form a proxy for fluvial sediment transport. Geometric mean varve thickness was compared to observed maximum daily annual discharge Qmax (1909-1971 AD) and the relationship expressed in a power equation. Thus, a reconstruction of past discharge for the last 2000 years could be produced. Extreme reconstructed discharge events were shown to be reasonable, considering the range of the observed discharge. Observed Qmax normally occurs during the snowmelt flood. Thus it is reasonable to attribute the variation in reconstructed Qmax to the snowmelt flood and, therefore, to melt water generation. Accumulated observed winter precipitation data from eleven meteorological stations from within and in the vicinity of the Ångermanälven catchment were compared to Qmax. Nine time series shared variation with Qmax and were complied into an average accumulated winter precipitation series. This series shares c. 40% of its variation with Qmax (observed and reconstructed) and it is reasonable to attribute some of the variation in the reconstructed Qmax to snow accumulation changes. The reconstructed Qmax series holds a centennial length waveform, which is interpreted in terms of the late Holocene climate succession, i.e. the Roman Warm Period, Dark Ages Cold Period, Medieval Warm Period, Little Ice Age and Contemporary Warm Period. Non-stationary frequencies in the sub decadal, decadal and centennial bands have been detected in the 2000 year long geometric mean varve series. Sub decadal frequencies are tentatively interpreted as forced by the North Atlantic Oscillation.. Decadal frequencies are interpreted as sunspot forcing. Centennial frequencies are tentative interpreted as Suess cycle forcing. Missing varves in the STS (minimum 500 years) have previously been postulated on the grounds of an independent calibrated radiocarbon calendar age obtained from varve deposits and its correlation to the GRIP Ice Core. However, an evaluation of the STS by cross correlation does not support the presence of a centennial error in the post glacial part of STS. Additional analyses on the relationship between reconstructed discharge and other climate (-derived) proxies have been tested, including temperature, tree rings, atmospheric indices, sea-ice extent, C-14 production variation, global temperature reconstructions and volcanic eruptions. These analyses were mostly internally contradictive and non-conclusive.