Portal de Periódicos da CAPES

From controversy to consensus: making the case for recent climate change in the Arctic using lake sediments

2007; Wiley; Volume: 5; Issue: 9 Linguagem: Inglês

10.1890/1540-9295(2007)5[466

1540-9309

John P. Smol, Marianne S. V. Douglas,

Geological Studies and Exploration

Frontiers in Ecology and the EnvironmentVolume 5, Issue 9 p. 466-474 Paleoecology ReviewFree Access From controversy to consensus: making the case for recent climate change in the Arctic using lake sediments John P. Smol, Corresponding Author John P. Smol Paleoecological Environmental Assessment and Research Lab (PEARL), Department of Biology, Queen's University, Kingston, ON, Canada K7L 3N6* (E-mail: smolj@queensu.ca)Search for more papers by this authorMarianne SV Douglas, Marianne SV Douglas Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, AB, Canada T6G 2E3Search for more papers by this author John P. Smol, Corresponding Author John P. Smol Paleoecological Environmental Assessment and Research Lab (PEARL), Department of Biology, Queen's University, Kingston, ON, Canada K7L 3N6* (E-mail: smolj@queensu.ca)Search for more papers by this authorMarianne SV Douglas, Marianne SV Douglas Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, AB, Canada T6G 2E3Search for more papers by this author First published: 01 October 2007 https://doi.org/10.1890/060162Citations: 210AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract We live in a constantly changing environment, yet tracking ecological change is often very difficult. Long-term monitoring data are frequently lacking and are especially sparse from Arctic ecosystems, where logistical difficulties limit most monitoring programs. Fortunately, lake and pond sediments contain important archives of past limnological communities that can be used to reconstruct environmental change. Here, we summarize some of the paleolimnological studies that have documented recent climate warming in Arctic lakes and ponds. Several hypotheses have been evaluated to determine if warming, resulting in changes in ice cover and related variables (eg increased habitat availability), was the factor most strongly influencing recent diatom and other biotic changes. Striking and often unprecedented community changes were evident in post-1850 sediments, and could be linked to ecological shifts consistent with warming. Because future temperature increases are predicted to be greatly amplified in polar regions, the ecological integrity of these sensitive ecosystems will be further imperiled. Ecologists and environmental scientists are constantly faced with the challenge of working at the most appropriate temporal and spatial scales to reach defendable conclusions. Few recent questions have elicited more discussion and debate than those concerning the nature, magnitude, and effects of global climate change. Although there is no longer any serious scientific debate about the influence of greenhouse-gas emissions on climate, many questions remain concerning the relative importance of natural versus anthropogenic causes of climate change, as well as the spatial and temporal scales of climate-related effects on ecosystems. Critical questions include: What are the natural modes of climate change? Has climate changed during the period of accelerated release of greenhouse gases? If so, when and to what degree? Have different regions and ecosystems responded similarly to climatic forcing? What future changes can we expect? In a nutshell: Ecologists, environmental scientists, and policy makers require long-term monitoring data to make effective decisions; however, direct observational data are often lacking and are especially sparse for polar regions, where environmental and climatic changes are amplified due to various feedback mechanisms Advances in the field of paleolimnology – the study of lake and river histories – allow scientists to reconstruct past environmental conditions based on proxy data preserved in dated sediment profiles Paleolimnological studies throughout much of the Arctic have documented striking and often unprecedented ecological changes in recent lake sediments that can be linked to warming Many Arctic lakes and ponds have already passed critical ecological thresholds due to recent warming Each of these questions has a temporal component, yet long-term monitoring data, which will help us to find the answers, are often scarce or non-existent. Fortunately, a variety of natural archives preserve proxy data of past environmental and ecological changes, some examples of which are highlighted in this issue of Frontiers. Here, we focus on paleolimnology – the multi-disciplinary science that uses the physical, chemical, and biological information preserved in lake and river sediments to reconstruct past environmental conditions (Cohen 2003; Smol in press). We also summarize the development of ideas (and associated challenges) following an earlier paper (Douglas et al. 1994), which proposed that a worrisome record of recent climate warming is preserved in Arctic sediments. The lack of long-term instrumental data in polar regions is especially disconcerting, because Arctic ecosystems, due in part to a variety of positive-feedback mechanisms, are warming at a faster rate than are other parts of the planet (ACIA 2004). In fact, based on the brief temporal windows of available instrumental meteorological data, average temperatures in the Arctic have risen at twice the rate of the rest of the world over the past few decades (ACIA 2004). Not surprisingly, polar regions are therefore often referred to as the "miners' canary" of the planet. Using a variety of different scientific approaches, it is now widely established that greenhouse-gas emissions can be linked to recent climate warming. However, it is important to reemphasize the brevity and poor spatial coverage of the instrumental record for most high latitude regions. Despite the vastness of the Canadian High Arctic, just five weather stations have collected meteorological records dating back to the late 1940s or early 1950s. It is not possible to address many key questions related to climate and environmental change in the Arctic based on such a temporally and spatially limited dataset. Paleolimnological assessments of long-term environmental change A characteristic feature of much of the north is the large number of ponds and lakes that dot the landscape. Because lakes and ponds are important sentinels of environmental change, and are often "biological hotspots" in tundra ecosystems (Schindler and Smol 2006), each of these water bodies potentially contains a record of past environmental change archived in its sediment sequence (Pienitz et al. 2004). Lacustrine sediments accumulate slowly over time, incorporating a vast library of information (ie morphological and biogeochemical markers) about the biological communities that lived within the water body (eg siliceous remains of diatoms, chitinous body parts of invertebrates; Figure 1), as well as from the surrounding catchment (eg pollen grains and spores, soil particles from erosion). Figure 1Open in figure viewerPowerPoint Light micrographs of some paleolimnological indicators commonly used in Arctic studies. (a) Valve of a Diploneis diatom. Diatoms are widely used in paleolimnological studies to track past changes in limnological variables, many of which can be linked directly or more often indirectly to climatic changes. (b) Head shield and shell of the cladoceran Acroperus harpae. These indicators represent important food-web components in lake ecosystems. (c) Head capsule of a Sergentia chironomid larva. Chironomids are used extensively in paleolimnological studies for a variety of applications, but have been most widely used to reconstruct past temperatures and deep-water oxygen levels. The job of the paleolimnologist is to reconstruct environmental conditions by using the physical, chemical, and biological information stored in these sequentially deposited sediments. Once the chronology of sediment deposition is established (typically done using radiometric techniques, such as 210Pb and 14C dating), the timing, nature, and magnitude of past ecological changes can be deciphered. Paleolimnology has seen tremendous progress over the past two decades, with major advances in methodology, approaches, and interpretation of data using statistically and ecologically robust methods (Smol in press). This science has special relevance to high latitude regions, where other important sources of paleoenvironmental data (eg annual growth rings from trees) are often limited or unavailable due to the short growing season, and the paucity or absence of trees. Paleolimnological approaches have been used in a wide spectrum of investigations, ranging from studies of lake acidification and eutrophication to other water-quality and environmental issues (Cohen 2003; Smol in press). Because climate exerts an important influence on biological communities, it is perhaps not surprising that a substantial portion of paleolimnological research has recently focused on studies of long-term climate change. For example, the chitinous head capsules of chironomid larvae (Heegard et al. 2006), and cladoceran exoskeletons (Sarmaja-Korjonen et al. 2006), as well as a suite of approaches using algal micro-fossils (Smol and Cumming 2000), have been developed to track (either directly or indirectly) past climate changes, including those in high latitude regions (Pienitz et al. 2004). Climate change, lake ice cover, and changing aquatic communities Although some attempts have been made to use proxy indicators to directly infer climate variables, such as temperature (eg Pienitz et al. 1995; Barley et al. 2006), most paleoenvironmental inferences from the Arctic are indirect, and use changes in biotic communities to track an environmental variable that is indirectly linked to climate. Because ice and snow cover are characteristic features of Arctic lakes, and because the length of the ice-free growing season and thus habitat availability are closely linked to climate (Figure 2), Smol (1983, 1988) proposed that paleoindicators that have reasonably well-defined microhabitat requirements (such as diatom algae) could be used as climate proxies in these regions. For example, during colder periods, deep lakes will retain more extensive ice covers that may persist throughout the short Arctic summer (Figure 2a), so that only a small, shallow moat of open water will thaw in the littoral zone. With warmer temperatures, more and more of the central raft of snow and ice will melt, exposing additional, deeper open-water habitats for algal and other biological growth (Figure 2b). With continued warming, the entire float of ice may melt, resulting in a dramatically altered physical, chemical, and biological environment, including different regimes of light penetration, mixing, gas exchange, and other characteristics (Figure 2c). Because different species of algae and invertebrates characterize different regions and microhabitats of lake and pond ecosystems (eg aerophilic, shallow-water periphytic, planktonic), changes in past assemblages can be used to infer past ice cover, from which qualitative climate inferences can be made. Figure 2Open in figure viewerPowerPoint A variety of climate-related factors influence the limnological conditions (and hence biota) of high-latitude lakes, but changes in ice and snow cover are often overriding controllers because they control light quantity and quality (as well as other limnological variables). This drawing shows changing ice and snow conditions on an Arctic lake during relatively (a) cold, (b) moderate, and (c) warm conditions. During colder years, a permanent raft of ice may persist throughout the short summer, precluding the development of large populations of planktonic diatoms, and restricting much of the primary production to the shallow, open-water moat. Many other physical, chemical, and biological changes occur in lakes that are either directly or indirectly affected by snow and ice cover (Douglas and Smol 1999). Modified from Smol (1988). The central argument is that an extensive float of ice and snow would dramatically affect limnological conditions, resulting in marked changes in biological communities (Smol 1983, 1988; Douglas and Smol 1999). Although light penetration and perhaps mixing are probably the most important primary drivers of biological communities in these lakes, many related limnological variables are also important and can be reconstructed using paleolimnological approaches (Table 1). Lotter and Bigler (2000) used a similar approach involving ice-cover changes and altered habitat availability to track climate changes in alpine lakes, and Thompson et al. (2005) subsequently developed a transfer function to infer duration of ice cover based on diatom assemblage species composition. Table 1. Overall trends that are often recorded in high Arctic ponds and lakes with different temperatures The Cape Herschel study ponds Although diatom-based approaches were being used for a wide spectrum of paleolimnological applications by the 1990s, very little research had been done on environmental change in high Arctic lakes and ponds following an initial study by Smol (1983). As Blake (1978) had initiated a sediment coring program in the region, the availability of cores prompted Douglas et al. (1994) to apply high-resolution paleolimnological techniques to study the environmental history of a series of ponds on Cape Herschel, Ellesmere Island, in the Canadian High Arctic. Shallow ponds were chosen because, due to their low volumes, they were expected to be especially sensitive bellwethers of environmental change. The results were surprising: following several millennia of relative ecological stability, episodes of nearly complete species turnover in diatom taxa were noted, beginning in the 19th century and becoming especially evident during the 20th century (Figure 3a). Due to the timing and nature of these remarkable diatom shifts, we were able to rule out as causative factors the atmospheric transport of pollutants or nutrients and the potential effects of increased UV radiation, as well as the possible confounding effects of coring artifacts and diatom dissolution. On the other hand, the shifts in assemblages were consistent with those expected to occur with a warming climate (Table 1). Although planktonic diatoms could not develop in these shallow ponds, the species changes indicated a proliferation of littoral taxa, previously only present at trace levels, indicating far more complex periphytic diatom communities, as well as increases in percentages of moss epiphytes, all of which require a longer ice-free season to develop in this harsh environment. We concluded that diatoms indicated marked environmental change in the Cape Herschel ponds, which could best be explained by warming. These conclusions were immediately and repeatedly challenged at conferences and workshops, as well as in some publications. Of course, some of these challenges were mounted by climate-change skeptics. (In the early 1990s, there was far less evidence for climate warming and its links to greenhouse-gas emissions.) However, because diatoms can be influenced by a wide spectrum of environmental variables, it was also argued (eg Anderson 2000) that the effects of other potential stressors would have to be addressed more fully before the Douglas et al. (1994) conclusions could be widely accepted. Figure 3Open in figure viewerPowerPoint (a) Stratigraphic profiles of the major diatom species changes that have occurred over the past few millennia in Col Pond (Cape Herschel, Ellesmere Island, Canada), one of the original study ponds in the Douglas et al. (1994) study. The figure highlights the striking species assemblage changes that have occurred in the post-1850 sediments, indicating decreased ice cover and other climate-related changes in this pond. (b) Stratigraphic profile of major diatom species changes in Slipper Lake (NWT, Canada); note the increase in Cyclotella diatom taxa in the more recent sediments, which may indicate decreased ice cover and, possibly, increased thermal stratification with recent warming. Approximate dates are shown to the right of the profiles. Modified from Rühland and Smol (2005). We use the Douglas et al. (1994) study as a point of departure for this review, and show how paleolimnological approaches were then applied throughout the Arctic, to address challenges to the conclusions of this initial study. We also summarize the developing body of paleolimnological literature linking changes in biological communities to climate warming in Arctic freshwater environments. Tracking the development of planktonic communities in deeper lakes As one might expect in a warming climate (Table 1), diatom assemblages in shallow ponds at Cape Herschel and in other high Arctic areas (eg Antoniades et al. 2005; Keatley et al. 2006) shifted toward those characteristic of extended open-water periods, with more habitat differentiation and complexity in the littoral zone. However, if changing climate and ice cover are major factors in influencing diatom assemblages, then an increase in planktonic taxa should be observed in deeper, ice-covered lakes, concurrent with supposed warming (Figure 2c). Sorvari and Korhola (1998) and Sorvari et al. (2002) explored some of these relationships by studying fossil diatom assemblages in lake sediment cores from the Lapland region of Finland. They found that planktonic diatoms, such as small euplanktonic Cyclotella taxa, increased in the more recent sediments, with compensatory decreases in small, benthic diatoms. Such species shifts would be expected with a longer ice-free growing season, and, potentially, with increased thermal stratification in these sub-Arctic lakes. Working on a suite of 50 lakes across the Canadian Arctic treeline region, Rühland et al. (2003) and Rühland and Smol (2005) showed similarly marked ecological shifts in diatom assemblages, from benthic Fragilaria taxa in pre-20th century sediments to planktonic Cyclotella taxa in the more recent sediments from a wide spectrum of Arctic lakes (Figure 3b). These diatom assemblage changes documented the development of communities fundamentally different from those that existed prior to the mid-19th century. A variety of potential driving factors for these species changes were investigated (eg anthropogenic acidification, nutrient enrichment, atmospheric deposition of contaminants); however, none of these influences satisfactorily explained the shifts in diatom assemblages. Algal composition can also be influenced by shifts in top-down controls (eg planktivory) causing changes in food-web structure in major grazing groups, such as Cladocera. However, a subsequent analysis of cladoceran microfossils (Sweetman et al. in press), from the identical sediments used in the Rühland et al. (2003) lake survey, found no links between the reported diatom species changes and shifts in grazing pressure. Recent increases in euplanktonic chrysophyte algae have also been recorded in the high Arctic (Wolfe and Perren 2001). Climatic warming, resulting in longer ice-free growing seasons and/or increased thermal stratification, seem to be the most plausible explanation for the recorded species changes. Diatom species changes in lakes that still support extensive ice cover Lakes that continue to support extensive ice cover would be predicted to display muted responses in their past diatom assemblages if, in fact, climate and ice cover were the major factors driving assemblage changes in the shallow ponds and the deeper lakes described above. Such ice-covered lakes still exist in the high Arctic if they are of sufficiently high latitude, elevation, and depth (ie thermal inertia) to maintain their ice cover for extended periods of time, in some cases throughout the short Arctic summer (see Figure 2a). One such site is Char Lake, in the hamlet of Resolute Bay (Cornwallis Island, Nunavut, Canada). Michelutti et al. (2003) undertook a detailed analysis of diatoms in the recent sediments of this well-studied lake. Due to its depth and morphometry, Char Lake has often maintained an ice cover for extended periods during the short summer, and, in many years, a central float of ice persists on the lake. If the primary factor precluding the development of planktonic diatoms in Arctic lakes were indeed ice cover, one would expect relatively minimal species changes in such a lake. The results of Michelutti et al. (2003) supported the validity of this hypothesis. As expected, the entire stratigraphic profile was overwhelmingly dominated by small, benthic Fragilaria diatoms, typical of cold, ice-covered Arctic lakes, with no development of planktonic communities in recent sediments. However, a subtle yet distinct assemblage shift in littoral taxa occurred in the uppermost sediments, indicating modest increases in habitat availability that were suggestive of decreased ice cover. These changes were consistent with nearby meteorological measurements showing warmer conditions over this time period. Because Char Lake was well studied between 1968 and 1972, as part of the Inter-national Biological Programme, followed by our additional limnological sampling since the early 1990s, it was also possible to reject changes in water chemistry as an alternative explanation for the most recent diatom species changes. Paleolimnological studies from far northern locations, such as Upper Dumbell Lake near Alert (northern Ellesmere Island), provide further evidence of species changes that may occur in lakes characterized by extended periods of ice cover. In this deep, ice-covered lake, very few diatoms were recorded in sediments before ca 1950, presumably due to the fact that there was such extensive ice cover that even shallow-water, littoral taxa were limited by the harsh ice conditions (Doubleday et al. 1995). Only with accelerated warming over the past few decades did enough of the shallow water moat thaw in summer to allow small, benthic taxa to increase in abundance. Still further north, on Ward Hunt Island at North America's northern limit, Antoniades et al. (2007) recorded a similar appearance of diatoms, as well as dramatic production increases (recorded by fossil algal pigments), in the recent sediments of Ward Hunt Lake. Rejecting nutrients as primary causative factors for the recent development of planktonic diatom communities Climate-related variation in ice cover was invoked as a key explanatory variable for all the paleolimnological interpretations of recent warming described thus far. However, the distribution of algae and other indicators is also influenced by a wide spectrum of environmental variables, with nutrients, such as phosphorus and nitrogen, recognized as major controllers of algal communities. Nutrients are typically very low in these high Arctic water bodies. Because some nutrient additions can occur due to atmospheric deposition, which can be related to human activities such as long-range atmospheric transport of pollutants to remote locations, it was important to assess the role that any recent nutrient enrichment may have played in the observed increases in planktonic diatoms. The role of nutrient fertilization as the causative factor for increased planktonic diatom growth was assessed by undertaking a paleolimnological study from a culturally eutrophied high Arctic lake. Although eutrophic lakes in the high Arctic are rare, Meretta Lake, just a few kilometers from Char Lake, is one such site. Meretta Lake was used as a sewage lake, receiving effluent from the Resolute Bay airport and surrounding buildings, beginning in the 1940s and continuing until 1998. We undertook a detailed diatom-based paleolimnological analysis of this lake's recent sediments and, although we recorded species changes in benthic taxa coincident with fertilization in the 1940s, no planktonic diatoms developed in this system (Douglas and Smol 2000). Moreover, the subtle species changes noted in the shallow-water taxa were much smaller than shifts in diatom assemblages from temperate regions receiving comparable nutrient inputs. In a similar study, Michelutti et al. (in press) investigated the diatom species response to over 25 years of increasing and persistent sewage additions to Annak Lake, located on the Belcher Islands in Arctic Canada. As in Meretta Lake, there was no development of planktonic taxa following nutrient increases, and the diatoms showed a delayed and subdued response to sewage inputs. The similarly muted diatom species responses to marked eutrophication in both Meretta and Annak lakes lends further support to the overriding role of physical factors, such as lake ice cover, in determining the nature of these freshwater species assemblages. Control sites If the main factor affecting species assemblages in Arctic freshwaters is in fact climate and not aerial transport of nutrients or pollutants, one would expect biotic changes to be less pronounced in sediment cores taken from the few areas in the Arctic believed to have experienced little warming over recent decades. Instrumental records and modeling (eg Kattenberg et al. 1996; Serreze et al. 2000; ACIA 2004), as well as other proxy indicators (eg D'Arrigo et al. 2003), suggest that recent warming has been muted in parts of northern Quebec and Labrador. Extensive paleolimnological studies from these regions (led mostly by R Pienitz of the Université Laval, Quebec City, Canada) have consistently recorded near "straight-line" profiles for bioindicators in these important control sites (eg Laing et al. 2002; Ponader et al. 2002; Smol et al. 2005), further supporting the primary role of climate warming in causing the recent species shifts recorded in other Arctic regions. The control regions described above also provide important reference areas for other types of investigations. For example, similar to atmospherically transported nutrients, it is well known that Arctic ecosystems have been affected by the long-range transport of a variety of other pollutants, such as PCBs. Because these potential stressors would have impacted Arctic ecosystems over the same time frame as purported warming, it could be argued that the species changes recorded by paleolimnologists in recent sediments were actually tracking pollutants, and not decreased ice cover and other climate-related variables. Paterson et al. (2003) specifically assessed the role that persistent organic pollutants, such as PCBs, may have on diatoms and chrysophyte assemblages by tracking community changes in Labrador lake cores that showed PCB contamination, most likely as a result of seepage from nearby military installations. They found that, although geochemical analyses clearly recorded striking increases in pollution concentrations, there were no concurrent changes in diatom or chrysophyte profiles. Changes in persistent organic pollutants do not explain the recent algal species changes recorded in other Arctic sediment cores. Tracking diatom species changes in two linked lakes exhibiting different ice-cover regimes Although it was becoming clear that atmospherically transported nutrients and persistent organic pollutants, such as PCBs, could not explain the species changes recorded in the post-1850 sediments of many Arctic lakes, the role of several unmeasured variables, including some less-studied, aerially transported chemicals, was yet to be assessed. The challenge remained to find lakes in close proximity with nearly identical limnological features, but differing in local climate and therefore the extent of ice cover. This is not an easy task, but two such sites were identified by Keatley et al. (in press) from northern Ellesmere Island. Because of the proximity of the two lakes (in fact, one drained into the other), both had similar limnological characteristics and were subjected to nearly identical aerial deposition of contaminants. However, due to shading from a nearby hill, one of the paired lakes maintains a more extensive ice cover during the summer. Keatley et al. hypothesized that if ice cover (and related limnological variables) were the key controller of diatom community structure in Arctic lakes, then the sedimentary profile in the lake with extended ice cover should record more muted post-1850 diatom changes than the profile from the nearby lake with little ice cover. The sedimentary record of these two lakes strongly supported this hypothesis (Keatley et al. in press). The lakes recorded strikingly different diatom assemblage changes, with the ice-covered lake tracking little ecological change, whereas the less-shaded lake recorded changes in its recent sediments consistent with warming. Tracking past changes in total algal production in lakes with decreased ice cover The paleolimnological studies discussed thus far were based primarily on species assemblage changes using relative frequency data. If ice and snow cover have been decreasing in lakes, as illustrated in Figure 1, not only should there be changes in the relative frequencies of different taxa, but overall primary production in the lake should also have increased. Michelutti et al. (2005) investigated this question using a new paleolimnological application of reflectance spectroscopy that allowed researchers to infer historical chlorophyll-a concentrations, and hence past trends in primary production from the recent sediments of six Baffin Island