Year AD

Figure 8. Historical changes in concentrations of UVR-absorbing DOC (top panel) and the depth of penetration of UV-B (as 1 % surface irradiance; lower panel) for Clearwater (circles), George (triangle) and Whitepine (square) lakes, Ontario, Canada [23]. Filled symbols are inferred from analyses of fossil diatoms, open symbols are direct historical observations. Analyses demonstrate that lake acidification arising from regional smelting activities led to declines in DOC and up to 6-fold increases in the depth of UV-B penetration. In general, clear lakes were most susceptible to acidic precipitation, but were not most naturally variable in the past. Instead, forest fires are hypothesized to increase the variability in pre-mining DOC concentrations in impacted lakes. [Figure modified from [23] and reprinted with permission.]

showed that lake sensitivity to DOC loss, baseline variability (pre-emission), and degree of recovery all varied greatly among sites, with naturally-transparent lakes exhibiting the greatest sensitivity to humans, but not the highest natural variability. Instead, authors concluded that local fires may have selectively impacted certain lakes, leading to reductions in export of photo-protective DOM from watersheds [cf., 128].

Paleoecological analyses can also be used to quantify the relative impacts of multiple stressors on lake production and structure. For example, Leavitt et al. [26] used annual fossil pigment profiles, 19 year-long historical records, and multivariate statistics to measure unique algal responses to changes in irradiance and lake chemistry during a whole-lake acidification experiment. Their analysis showed that 80-83% of historical variance in the abundance of fossil chlorophylls and carotenoids could be explained statistically by measured changes in lake properties. This study further demonstrated that while effects of pH accounted for 50% of change in algal communities, irradiance (12%) and its interactions with lake chemistry (20%) were significant, substantial and independent determinants of algal community variability. Specifically, increased penetration of PAR during initial stages of acidification stimulated growth of benthic and metalimnetic algal populations, whereas 8-fold increases in UV-B penetration during severe acidification phases (pH<4.5) greatly suppressed growth of sensitive taxa (e.g., chrysophytes) and lake production. Fossil analyses further demonstrated that the deposition of photo-protective pigments similar to scytonemin (e.g., Ca) exhibited a strong linear correlation with the depth of UVR penetration (r2 = 0.70) when UVR-absorbing pigments were normalized to changes in total algal abundance [26]. Finally, by comparing fossil profiles among acidified lakes, the authors proposed a conceptual model to explain the high variability in ecosystem response to acidification; UVR suppressed primary production in clear lakes (<3 mg DOC l-1) or those with severe acidification (pH<4.5), whereas increased PAR transmission during acidification stimulated algal growth in other instances.

In contrast to impacts of drought and acidic precipitation, recent research has suggested that terrestrial disturbance associated with forest fires or timber harvest may increase DOM export to lakes and reduce UVR penetration [128] so long as soils are not totally destroyed [13,125]. Consistent with this view, paleoecological analysis of a chain of lakes in the Great Plains of central North America suggested that UVR penetration declined 5-fold in upstream lakes, but not those further downstream, concomitant with development of regional agriculture [129]. Although patterns are consistent with a 50% decline in soil C content and the ability of sequential lake basins to trap and remove organic matter through sedimentation, interpretations of changes in UVR index require analysis of mixing regime of these polymictic lakes in order to distinguish whether declines in UVR exposure represent increased water-column DOM, or reductions in mean algal exposure to UVR due to lake eutrophication.

0 0

Post a comment