Fossil evidence of past UVR environments in lakes

Present concern about the interactive effects of ozone depletion, DOM-degrading acidic precipitation and global warming has led several investigators to conclude that extreme variations in the biogeochemistry of DOM and its impacts on UVR penetration may be the most significant challenge to aquatic ecosystem integrity and function [e.g., 66,108]. Because the range of historical variance in DOM flux and UVR attenuation is often greater than that arising from modern processes [15,24], paleoecological reconstructions of past UVR and its impacts on lake ecosystems may provide essential insights into the role of high energy irradiance in structuring aquatic ecosystems. Here we use a case study approach to demonstrate the value of such retrospective analyses, particularly in the case of long-term environmental change, early lake evolution and human impacts. As our intent is to stimulate research in this area, our examples include both research with well-documented mechanistic explanations, and provocative new studies which, although less substantiated, have the potential to greatly improve our understanding of past UVR environments and their impacts on lakes.

16.4.1 Holocene climate change

To address the potential impact of long-term climate change relative to that of ozone depletion, Pienitz and Vincent [15] combined paleolimnological analyses with bio-optical models based on present-day conditions in lakes of northern Canada. Specifically, they estimated past underwater light conditions from DOC concentrations that were inferred from fossil diatom assemblages preserved in Holocene sedimentary deposits from a lake near the treeline (Queen's Lake) in the central Northwest Territories, Canada (Figure 2). Analysis of fossil pollen records indicate that this region of the continent underwent deglaciation ca. 8000

% planktonic species 0 10 20

% periphytic species Inferred DOC (mg I"1)

% planktonic species 0 10 20

Inferred T320 or TPAR (m) 0.1 1 10

Inferred UVR/PAR 0.05 0.1 0.5

Inferred T320 or TPAR (m) 0.1 1 10

Inferred UVR/PAR 0.05 0.1 0.5

Figure 2. Changes in fossil diatom community structure and inferred optical conditions in Queen's Lake, Northwest Territories (Canada)[15]. Diatom data are expressed as a percentage of the total number of valves in each sample associated with planktonic or benthic taxa. Diatom species data were used to infer DOC concentration (mg 1_1), biologically weighted UV exposure (7"*PI or r*DNA) and underwater spectral balance (water column transparency for 320 nm UVR [J,320], PAR [7pAR], and the ratio between the two [UVR/PAR]) over the last 6000 years. The dotted lines delimit the period of mid-Holocene maximum forest cover. This analysis demonstrates that biotic exposure to UVR varies substantially due to changes in catchment vegetation and DOC supply, and that, during mid-Holcene climatic warm periods, UVR exposure declined two-orders of magnitude. Climatic cooling at ~3000 yr BP reduced DOC inputs by reducing soil development and DOC supply, and led to increases in UVR penetration that are up to 4000-fold more significant than those expected to arise from a moderate (30%) ozone depletion. [Figure reprinted by permission from Nature [15], copyright 2000, Macmillan

Magazines Ltd.]

Figure 2. Changes in fossil diatom community structure and inferred optical conditions in Queen's Lake, Northwest Territories (Canada)[15]. Diatom data are expressed as a percentage of the total number of valves in each sample associated with planktonic or benthic taxa. Diatom species data were used to infer DOC concentration (mg 1_1), biologically weighted UV exposure (7"*PI or r*DNA) and underwater spectral balance (water column transparency for 320 nm UVR [J,320], PAR [7pAR], and the ratio between the two [UVR/PAR]) over the last 6000 years. The dotted lines delimit the period of mid-Holocene maximum forest cover. This analysis demonstrates that biotic exposure to UVR varies substantially due to changes in catchment vegetation and DOC supply, and that, during mid-Holcene climatic warm periods, UVR exposure declined two-orders of magnitude. Climatic cooling at ~3000 yr BP reduced DOC inputs by reducing soil development and DOC supply, and led to increases in UVR penetration that are up to 4000-fold more significant than those expected to arise from a moderate (30%) ozone depletion. [Figure reprinted by permission from Nature [15], copyright 2000, Macmillan

Magazines Ltd.]

yr BP, and that terrestrial vegetation was sparse and tundra-like for the first 3000 years of lake existence [109].

Diatom community structure and inferred DOC levels showed three distinct and abrupt changes during the history of Queen's Lake. First, analyses showed that both diatom biomass and inferred DOC concentrations were low (< 2 mg DOC l-1) during the first ~3000 years of lake existence, with particularly few fossils recovered from sediments > 6500 yr BP. This initial period was followed by a major shift in fossil species composition and inferred chemical conditions ca. 5000 yr BP, with increased ratios of periphytic : planktonic taxa to >70% total diatom assemblage. This second period also corresponded to a major increase in algal production, recorded as the sediment mass-specific concentration of diatom valves, as well as a three-fold increase in inferred DOC levels. Based on fossil pollen analyses, Pienitz and Vincent [15] argued that changes in lake chemistry and production resulted from climatic warming that stimulated treeline advance and increased forest density for about 2000 years. Finally, diatom-based reconstructions indicated that DOC concentrations declined > 85% after 3000 yr BP, concomitant with climatic cooling and a southward retreat of treeline [109].

The large and rapid changes in DOC imply that Queen's Lake experienced major shifts in the underwater optical environment over the last 6000 years as a consequence of climate-induced variation in forest development [89]. Consistent with this hypothesis, application of bio-optical models derived from measurements in high-latitude waters showed that the inferred DOC shifts were equivalent to a two order-of-magnitude decrease in exposure to biologically-effective UVR between 6000 yr BP and the mid-Holocene vegetation maximum (Figure 2). In contrast, the most recent 3000 years were characterised by a > 50-fold increase in levels of damaging UVR, with recent inferences agreeing closely with present-day estimates of UVR exposure. Overall, changes in DOC concentrations arising from climatic variability increased exposure to photosynthetically damaging UVR 4000-fold more than did a moderate (30%) decline in stratospheric ozone levels [15].

Climatic control of past UVR exposure has also been identified as a key factor regulating lake production and algal community composition in montane lakes at treeline [27]. For example, geochemical and palynological analysis of sediments demonstrates that Crowfoot Lake, Alberta (51 °26'N, 116°31'W), lay above treeline during the Younger Dryas (ca. 11 100-10100 14C yr BP), was a subalpine lake for the next ~ 6000 years, then returned to its present position near timberline following regional climatic cooling ca. 4000 yr BP [110,111]. Analyses of fossil pigments confirmed that algal abundance was reduced 10- to 25-fold during periods of high UVR exposure, inferred from both fossil pigment- and bulk organic matter-based estimates of irradiance penetration (Figure 3). Through the use of coupled DOC-UVR optical models, it was shown that algal abundance was reduced whenever the depth of UVR penetration (as 1 % surface irradiance) exceeded mean lake depth, and deepwater refugia were lost, especially early in the lakes history (>10500 yr BP) and following climatic cooling at ~ 3500 yr BP [27]. The authors argued provocatively that lake production was suppressed by a combination of low DOM and high UVR rather than by variations in mineral nutrient flux because patterns of fossil pollen, modern lake evolution [28], and terrestrial nutrient cycling were inconsistent with regulation of lake production by N and P, the major mineral nutrients [27]. Further, as changes in UVR exposure were linked to long-term climatic variability, and because increased UVR exposure occurred despite substantial pools of terrestrial DOM, both diatom- and pigment-based analyses suggest that future global warming may increase UVR penetration, alter gross community composition, and strongly suppress the primary production of many boreal lakes.

16.4.2 Early lake evolution

Recently, several lines of evidence have combined to suggest that biotic exposure to UVR should be greatest immediately after deglaciation, prior to the develop-

UV-B exposure depth of UV-B algal biomass forest penetration b.

development d.

0 200

UV compound. Ca Carotenoids fo/

0 20

nmol B-carotene g dry weight

10 20

needles dm3

0 200

UV compound. Ca Carotenoids fo/

0 20

nmol B-carotene g dry weight

10 20

needles dm3

Figure 3. Change in past UVR exposure inferred from pigments (a), depth of UV-B penetration inferred from sedimentary organic matter content (b), and total algal abundance (c) in sediments from Crowfoot Lake, Alberta, Canada [27], Forest development (d) was estimated from changes in concentration of locally-derived tree needles [from 111]. Algal exposure was measured as the ratio of UVR-absorbing pigment Ca: carotenoids (alloxanthin, diatoxanthin, lutein). Depth of UVR penetration (as 1% surface irradiance, in m) was inferred from historical changes in the organic matter content of Crowfoot Lake sediments. Total algal abundance was estimated from concentrations of the ubiquitous carotenoid ^-carotene. This analysis demonstrates that algal abundance was low whenever UVR penetration was great during both the recent and early Holocene periods, when local forests were absent or poorly developed. Production of photo-protective pigments was greatest when the depth of UVR penetration exceeded the mean depth of Crowfoot Lake (vertical dashed line) during ~ 11 300-10050 14C yr BP and -4000 14C yr BP-present. Rates of algal decline ca. 4000 yr BP reflect reduced DOC supply at high elevations and are similar to those expected to occur as a result of DOC declines arising because of global warming at low elevations [108]. See text for details. [Figure modified from 27.]

ment of terrestrial sources of UVR-absorbing DOM. First, optical models show that DOM from terrestrial sources is the single most important factor regulating UVR penetration within a lake [10,11]. Second, whole-lake experiments and empirical studies show that variations in terrestrial DOM supply and mineralization are more significant factors regulating UVR exposure than is modern stratospheric ozone depletion [12,14]. Third, analysis of lake chronoseries from Glacier Bay, Alaska, demonstrate that DOC content is low for at least the first century following modern lake formation [28] and that this initial high UVR exposure can structure biotic communities [2,112]. Finally, the observations that diatoms are particularly sensitive to changes in UVR exposure [e.g., 49,57], and that these taxa are rare in early post-glacial sediments [89], suggest together that extreme UVR transparency is a common mechanism directing the early evolution of glacial lake ecosystems.

Quantification of past UVR environments using fossil pigments has been used to document that exposure to UVR is greatest early in the lake's history, prior to development of regional forests (Figure 4). Here, analysis of the complete postglacial history of three lakes in sub-humid central British Columbia, Canada, showed that UVR-absorbing algal pigments were present for at least 1000 years following deglaciation, but were absent from sediments at all other times during the past 12000 years [27]. Presently, UVR penetration is inconsequential at all sites (<10 cm) due to high levels of DOM (>10 mg DOC 1_1). In addition, reconstructions showed that algal biomass was 10-fold lower during the period of elevated UVR penetration than at any other time in lake history. At all sites, sharp reductions in UVR penetration and increased lake production occurred concomitant with the development of terrestrial carbon sources ca. 10 700 yr BP, consistent with the hypothesis that terrestrial carbon is the key factor regulating

0 0

Post a comment