Search

20th century eutrophication and global spread of anoxia is a threat for freshwater ecosystems. Little is known about Holocene anoxia and meromixis events when anthropogenic impacts were weaker and natural ecosystem variability played the dominant role. In this study, we examine the relationship between lake mixing and lake production, climate variability, vegetation cover, catchment erosion and (pre)historic anthropogenic impacts in Moossee (Switzerland), over the last 15,000 years. We use sub-annually resolved calibrated hyperspectral imaging data (total chlorophyll for paleoproduction, bacteriopheophytin for anoxia and meromixis) combined with X-ray fluorescence and pollen data. Production shows a first increase at 14,500 cal yr BP, a further increase after 7500 cal BP, relative maxima in the late Bronze, Iron and Middle Ages, and the unprecedented peak in the 20th Century. Until 7500 cal BP, the lake was well mixed with only scarce phases of seasonal to multiannual anoxia. Repeated meromixis events occurred between 7500 and 2500 cal BP when temperatures were high, forests closed, and lake production was already enhanced. After the forests were cleared (2500 cal BP) the lake remained mostly holomictic. Holocene meromixis events were systematically terminated by local deforestation related to Neolithic and Bronze Age lakeshore settlements: charcoal peaked, tree pollen dropped below a threshold of 80%, soil erosion and lake production increased and bacteriopheophytin disappeared. Meromixis re-established after the termination of lakeshore settlements and the onset of afforestation with tree pollen exceeding 80%. These repeated cycles unambiguously document how even early human societies affected the mixing regime and biogeochemical cycling in this lake.

Chironomids preserved in varved Lake Żabinskie (54°07′54.5″ N; 21°59′01.1″ E; 120 m a.s.l), northeastern Poland, were used to reconstruct mean-August air temperature since 1896 AD at annual (1949–2011 AD) and at 3–4 year resolutions (1896–1948 AD). This is one of very few studies using chironomids at such high temporal resolution, for a total of 130 sediment samples analyzed. To infer temperature a combined (Northeastern Canada and Poland) transfer function was developed. This transfer function had 112 lakes (50 from Poland, 72 from Canada) and 95 taxa. The mean-August-air temperature gradient was 23.5 °C. The coefficient of determination (r2jack) was 0.88, the root-mean-square error of prediction (RMSEP) was 1.30 °C and the maximum bias was 2.08 °C. The average mean August air temperature measured during the 1961–1990 AD reference period was 16.3 °C and the chironomid-inferred average temperature for the same period was 16.6 °C. For the 1981–2010 AD reference period, the temperatures were 1 °C warmer (i.e. measured = 17.3 °C; inferred = 17.6 °C). The differences of only 0.3 °C between the measured and the inferred temperatures during the reference periods were a first indication that the merged transfer function could provide accurate estimates of mean August air temperature. Chironomid-inferred mean August air temperatures since 1896 AD had strong relationships with instrumental data at near-annual resolution (rPearson = 0.74, pcorr Although the RMSEP of the transfer function was higher than most changes recorded, the average difference between the instrumental and the inferred mean August air temperatures was 0.75 °C. The results of this study suggest that not only do chironomids reconstruct the right pattern in temperature changes (83% of the inferences recorded the same temperature increase or decrease than the instrumental data) they can also infer changes with the right amplitude (61% of the inferences had differences with the instrumental data below 0.7 °C). This study is one more example of the possible accuracy of chironomids to reconstruct climate, at least for the past 100 years and at this site.

This study presents a proxy-based, quantitative reconstruction of cold-season (mean October to May, TOct–May) air temperatures covering nearly the entire last millennium (AD 1060–2003, some hiatuses). The reconstruction was based on subfossil chrysophyte stomatocyst remains in the varved sediments of high-Alpine Lake Silvaplana, eastern Swiss Alps (46°27’N, 9°48′W, 1791 m a.s.l.). Previous studies have demonstrated the reliability of this proxy by comparison to meteorological data. Cold-season air temperatures could therefore be reconstructed quantitatively, at a high resolution (5-yr) and with high chronological accuracy. Spatial correlation analysis suggests that the reconstruction reflects cold season climate variability over the high- Alpine region and substantial parts of central and western Europe. Cold-season temperatures were characterized by a relatively stable first part of the millennium until AD 1440 (2σ of 5-yr mean values = 0.7 °C) and highly variable TOct–May after that (AD 1440–1900, 2σ of 5-yr mean values = 1.3 °C). Recent decades (AD, 1991-present) were unusually warm in the context of the last millennium (exceeding the 2σ-range of the mean decadal TOct–May) but this warmth was not unprecedented. The coolest decades occurred from AD 1510–1520 and AD 1880–1890. The timing of extremely warm and cold decades is generally in good agreement with documentary data representing Switzerland and central European lowlands. The transition from relatively stable to highly variable TOct–May coincided with large changes in atmospheric circulation patterns in the North Atlantic region. Comparison of reconstructed cold season temperatures to the North Atlantic Oscillation index (NAO) during the past 1000 years showed that the relatively stable and warm conditions at the study site until AD 1440 coincided with a persistent positive mode of the NAO. We propose that the transition to large TOct–May variability around AD 1440 was linked to the subsequent absence of this persistent zonal flow pattern, which would allow other climatic drivers to gain importance in the study area. From AD 1440–1900, the similarity of reconstructed TOct–May to reconstructed air pressure in the Siberian High suggests a relatively strong influence of continental anticyclonic systems on Alpine cold season climate parameters during periods when westerly airflow was subdued. A more continental type of atmospheric circulation thus seems to be characteristic for the Little Ice Age in Europe. Comparison of Toct–May to summer temperature reconstructions from the same study site shows that, as expected, summer and cold season temperature trends and variability differed completely throughout nearly the entire last 1000 years. Since AD 1980, however, summer and cold season temperatures show a simultaneous, strong increase, which is unprecedented in the context of the last millennium. We suggest that the most likely explanation for this recent trend is anthropogenic greenhouse gas (GHG) forcing.

Determining if temperatures of the last century exceed natural variability necessitates the use of high-resolution paleo-reconstructions extending beyond the instrumental data (i.e. >150 years). Although syntheses using tree-ring, stalagmite, and borehole based reconstructions are available, biological proxies preserved in lake sediments have been neglected as a source of high-resolution information on climate change. Here, we present a decadal-scale mean July air temperature reconstruction covering the past millennium from varved Seebergsee, Switzerland. This reconstruction is compared to instrumental data at local, regional and European scales, to another high-resolution chironomid-inferred temperature reconstruction from Lake Silvaplana (Switzerland), to a composite of paleo-climate reconstructions from the Greater Alpine Region and to various millennial scale climate reconstructions from the northern hemisphere. When compared to local and regional instrumental records since ca 1760 AD significant (p 0.5) are obtained, suggesting that chironomids accurately register the changes in temperature for the past ca 250 years. At European scale, the Seebergsee reconstruction correlates (rPearson = 0.40; p < 0.01) with instrumental and early instrumental data back to 1500 AD. On the millennial time scale, the chironomid reconstruction of Seebergsee provides a pattern of temperature changes mirrored by the chironomid reconstruction from Silvaplana (rPearson = 0.44; p < 0.01) and the Greater Alpine Region composite of reconstructions (rPearson = 0.40; p < 0.01). This includes warmer-than-the-last-century mean July air temperatures (+1.2 °C on average) during the end of the “Medieval Climate Anomaly” (MCA) and colder-than-the-last-century temperatures (−0.5 °C on average) during the “Little Ice Age”. Both chironomid reconstructions inferred a warming during the last decades, but this chironomid-inferred warming does not exceed the MCA temperatures. This result is not singular as many millennial temperature reconstructions in the northern Hemisphere do not show unprecedented warming of the last century at local/regional scale. However, the chironomid assemblages found in Seebergsee and Silvaplana since ca 1950 AD seem to be unique (i.e. they show unprecedented assemblage compositions) for the past 1000 years.

The present interglacial, the Holocene, spans the period of the last 11,700 years. It has sustained the growth and development of modern society. The millennial-scale decreasing solar insolation in the Northern Hemisphere summer lead to Northern Hemisphere cooling, a southern shift of the Intertropical Convergence Zone (ITCZ) and a weakening of the Northern Hemisphere summer monsoon systems. On the multidecadal to multicentury-scale, periods of more stable and warmer climate were interrupted by several cold relapses, at least in the Northern Hemisphere extra-tropical area. Based on carefully selected 10,000-year-long time series of temperature and humidity/precipitation, as well as reconstructions of glacier advances, the spatiotemporal pattern of six cold relapses during the last 10,000 years was analysed and presented in form of a Holocene Climate Atlas (HOCLAT; see http://www.oeschger.unibe.ch/research/projects/holocene_atlas/ ). A clear cyclicity was not found, and the spatiotemporal variability of temperature and humidity/precipitation during the six specific cold events (8200, 6300, 4700, 2700, 1550 and 550 years BP) was very high. Different dynamical processes such as meltwater flux into the North Atlantic, low solar activity, explosive volcanic eruptions, and fluctuations of the thermohaline circulation likely played a major role. In addition, internal dynamics in the North Atlantic and Pacific area (including their complex interaction) were likely involved.

Chironomids (non-biting midges) can provide accurate climate reconstructions from the Late Glacial to the present. Until now, anoxic lakes have been avoided for temperature reconstructions since chironomid assemblages are sensitive to changes in oxygen concentrations in the hypolimnion. However, anoxic lakes may have varved sediments, providing the possibility for near-annual climate reconstructions. Here, we tested the applicability of two calibration methods to reconstruct mean July air temperatures from chironomid assemblages preserved in the sediments of the anoxic Seebergsee located in the northern Swiss Alps: a calibration-in-space approach and a calibration-in-time approach. The calibration-in-space approach (i.e. chironomid assemblages from surficial lake sediments (0–1 cm) calibrated against meteorological data) provided accurate inferences (i.e. similar temperature changes as measured at the closest meteorological station, and at regional stations) in the Seebergsee stratigraphy until anoxia increased in the lake. With the increase of anoxia, the chironomid-inferred temperatures were generally colder than measured temperatures. A calibration-in-time approach (i.e. calibration of chironomid assemblages in a time series against instrumental data from the closest meteorological station) provided accurate reconstructions (i.e. similar to the regional records) for the past 100 years, including the time period of inferred anoxia. However, its applicability should be further tested on longer temporal scales.

High-resolution quantitative temperature records are needed for placing the recent warming into the context of long-term natural climate variability. In this study we present a quantitative high-resolution (9-year) summer (June–August) temperature reconstruction back to AD 1177 for the south-eastern Swiss Alps. This region is a good predictor for summer temperatures in large parts of western and central Europe. Our reconstruction is based on a combination of the high-frequency component of annually resolved biogenic silica (bSi flux) data and the low-frequency component of decadal chironomid-inferred temperatures from annually laminated well dated sediments (varves) from proglacial Lake Silvaplana, eastern Swiss Alps. For the calibration (period AD 1760–1949) we assess systematically the effects of six different regression methods (Type I regressions: Inverse Regression IR, Inverse Prediction IP, Generalised Least Squares GLS; Type II regressions: Major Axis MA, Ranged Major Axis RMA and Standard Major Axis SMA) with regard to the predicted amplitude and the calibration statistics such as root-mean-square error of prediction (RMSEP), reduction of error (RE) and coefficient of efficiency (CE). We found a trade-off in the regression model choice between a good representation of the amplitude and good calibration statistics. The band-pass filtered bSi flux record is in close agreement both in the structure and the amplitude with two fully independent reconstructions spanning back to AD 1500 and AD 1177, respectively. All known pulses of negative volcanic forcing are represented as cold anomalies in the bSi flux record. Volcanic pulses combined with low solar activity (Sporer and Maunder Minimum) are seen as particularly cold episodes around AD 1460 and AD 1690. The combined chironomid and bSi flux temperature record (RMSEP = 0.57 °C) is in good agreement with the glacier history of the Alps. The warmest (AD 1190) and coldest decades (17th century; 1680–1700) of our reconstruction coincide with the largest anomalies in the Alpine tree-ring based reconstruction; both records show in the decadal variability an amplitude of 2.8 °C between AD 1180 and 1950, which is substantially higher than the amplitude of hemispheric reconstructions. Our record suggests that the current decade is slightly warmer than the warmest decade in the pre-industrial time of the past 800 years.

At this critical moment of Earth’s history, as we move from a century with rapidly growing human impacts on all the different ecosystems of our planet, to a century with a probable further acceleration in the pace of environmental change, resource use, and vulnerability for societies and economies, we have to rethink the changing relationship between nature and human beings from the past, through the present, towards a future full of uncertainty. It becomes more and more evident that major natural processes from the local to the global level are influenced by human activities, creating a much higher degree of complexity through the interaction of processes which are within the domain of both the natural and social sciences. This implies a need to bridge the gulf between the two cultures of science in order to advance our understanding of contemporary driving forces and their rapidly growing impact on earth’s ecosystems. “The biggest changes happened in our century, more precisely in the last 50 years, with a rate unknown before in Earth’s history” (Pfister 1995a). In light of continued population growth, economic development, urbanisation, industrialisation and resource use, it is clear that human impacts on ecosystems world-wide will continue to increase in the next century.