UVR index

Figure 6. Historical change in past UVR exposure at Col Pond, Ellesmere Island, Canada (78°37'N, 74°42'W), during the past ~6000 years [Leavitt et al. unpublished data]. UVR exposure estimated as in Figure 3, using ratios of Ca: carotenoids. All pigment and derivative concentrations quantified using high-performance liquid chromatography [68]. Sediment age estimated by assuming constant mass accumulation rates since pond formation ca. 8500 yr BP [122]. No organic matter could be recovered for determinations of 14C activity. Analyses suggest that algal exposure to UVR was at least 3-fold greater prior to ~4000 yr BP than at present. Timing of declines in UVR exposure is similar to those seen in the Antarctic (Figure 5) and coincide with climatic cooling and UVR change at other latitudes [15,27].

ation through much of the core suggests that there has been no persistent change in either lake depth or UVR transparency. Because the pond lies in a rock basin, lacks submerged vegetation, is very shallow (~ 1 m) and has sediments with low organic matter content (~2% LOI) we infer that UVR has likely penetrated throughout the water column since pond inception.

Sedimentary pigment analyses showed that photoprotective compounds were abundant and UVR indices were elevated in the oldest sediments, but declined ~ 3-fold to modern values ca. 3000-4000 yr BP (Figure 6). As sedimentary organic matter content also declined at this time, we infer that reductions in UVR did not arise from increased DOC content within the lake (see methods above). Similarly, variations in UVR index did not reflect changes in cyanobac-terial abundance, as concentrations of myxoxanthophyll from those algae were constant through the period of inferred UVR decline. Unfortunately, chronological control is very poor at Col Pond, and sediment ages are estimated only approximately from published marine emergence curves for Cape Herschel [122] and by assuming that sediment accumulation rates were constant through time. With this caveat in mind, it is interesting to note that both the timing and direction of UVR change was similar at both poles, suggesting that fossil pigments may record changes in global irradiance regimes. Consistent with this hypothesis, timing of UVR declines corresponded to the onset of the modern, comparatively cool climate [123,124] and altered UVR regimes at mid-latitudes [15] and high elevations (Figure 2; [27]). Because our UVR indices are scaled by total algal production, they should be independent of changes in growing season duration arising from global cooling. We are presently conducting further research at other, less marine-influenced, sites to determine whether these historical patterns represent changes in regional irradiance due to cloud cover or whether sediments may be recording true variability in solar production or stratospheric transmission of UVR.

16.4.4 Rapid variation in UVR environments

In addition to long-term changes in UVR regimes arising from variations in global climate and carbon biogeochemistry, aquatic ecosystems are subject to extremely rapid alterations in UVR penetration due to natural and human-induced mechanisms. In particular, attention has focused on the combined impacts of ozone depletion, global warming and acidic precipitation, the so-called triple whammy of environmental disturbance [66,108]. Lake acidification by anthropogenic mineral acids both increases the rate of DOM removal from the water column [14], and reduces the specific attenuation of remaining DOM [42], leading to order-of-magnitude increases in UVR penetration [14]. In contrast, declines in stream flow during droughts reduce export of terrestrial and wetland DOM to lakes, while increasing water residence times, thereby causing more thorough DOM mineralization and precipitation and increased UVR penetration [13,125]. Further, mineral sulfur stored in shallow sediments can be oxidized during droughts to reform acids that deplete DOM and allow 3-fold increases in UV-B penetration [12]. Although lakes may recover from individual disturbances, concern is mounting that ecosystems may not be resilient to multiple concurrent stressors [108], perhaps resulting in rapid fundamental changes in lake organization (state change; [126]). Once again, we propose that sedimentary analyses can provide valuable insights into how human activities interact with other stressors to regulate the impact of irradiance on lakes.

Analyses of fossil pigments in the sediments of alpine lakes have shown that variation in UVR exposure arising from regional droughts can be greater than that attributed to human depletion of stratospheric ozone [24]. Prior paleoecological analyses have demonstrated the presence of droughts at low elevations and cool temperatures at treeline during ~ 1850-1900, both conditions which reduce terrestrial export of DOM to lakes [cf., 13]. As expected, deposition of photo-protective pigments was greater during drought intervals than at other times (Figure 7). When expressed relative to total algal abundance (as fossil carotenoids), UV-B exposure was found to be 5- to 10-fold greater than that in the most recent sediments, and as much as 4-fold greater than that seen in the most transparent Rocky Mountain lake. Extension of this analysis to other sites has demonstrated that this UVR event was widespread among regional lakes. Given that up to 200000 present-day North American lakes have DOM levels typical of montane lakes («2 mg DOC 1_1 [14]), that DOM-depleting continental-scale droughts have been more intense during the recent past [127], and that global warming will likely intensify droughts [108], these analyses suggest that UVR impacts on lakes may be widespread in the future.

Both pigment- and diatom-based reconstructions of past UVR environments have demonstrated that anthropogenic acidification alters the fundamental irra-diance regime of lakes [26,128]. For example, Dixit et al. [23] applied DOC-inference models to high-resolution cores from three central Canadian lakes receiving acidic precipitation and showed that DOC declined up to 75% and

2000

past UV-B exposure

Pipit L. Snowflake L.

S 1900

1800

Pipit L. Snowflake L.

S 1900

1800

1 1 1 1 1 L

C :

, i i i i i

droughts

0 300 0 150

UVR pigment / carotenoids (%)

Figure 7. Historical changes in UVR exposure in alpine Snowflake and Pipit lakes, western Canada [24]. Past UVR exposure estimated as in Figure 3, using ratios of Ca: carotenoids. All pigment and derivative concentrations quantified using high-performance liquid chromatography [68]. A total of four regional lakes demonstrate that UVR exposure increased ~ 10-fold during periods of lowland droughts and upland cool temperatures, both factors that reduce export of terrestrial DOM to lakes [24], In contrast, sites contain little evidence of increased UVR exposure arising from recent ozone depletion. Figure modified from [24], [Reprinted by permission from Nature [24], copyright 2000, Macmillan Magazines Ltd.]

penetration of UV-B increased up to 2.5-fold as a result of acid emissions from local smelters (Figure 8). In general, fossil inferences showed excellent agreement with historical measurements of DOC concentrations, greatly improving researcher's confidence in historical reconstructions. Interestingly, this study also

—•— Clearwater Inferred • o Clearwater Measured

▼ George Inferred —V" George Measured -9- Whltepine Inferred —Q ; Whltepine Measure^

—•— Clearwater Inferred • o Clearwater Measured

▼ George Inferred —V" George Measured -9- Whltepine Inferred —Q ; Whltepine Measure^

1840 1860 1880 1900 1920 1940 1960 1980 2000

-•— Clearwater Historical O - Clearwater Recent George Hletorioal - —George Recant -«- Whltepine Historical —a • • Whltepine Recent

1840 1860 1880 1900 1920 1940 1960 1980 2000

0 0

Post a comment