Ultraviolet Radiation and Bio-optics in Crater Lake, Oregon, 2005
RESULTS
Proxy measurements for UV attenuation versus depth; Phytoplankton control of Kd,UV
Bio-optical signals provide several proxies for UV attenuation through the water column. The red beam transmissometer signal (from which cp660 is calculated) has been measured since 1987 in Crater Lake (Larson et al., 1996a) and provides an optical proxy particulate organic carbon (POC) and in the upper 80 m it can serve as a proxy measurement for UV attenuation. The rationale for using cp660 as a proxy for Kd,UV is described earlier. Depth profiles measured during 1999 for POC (calculated from cp660based upon POC analysis of discrete samples), total organic carbon (TOC), and dissolved organic carbon (DOC, calculated as TOC-POC) are shown in Figures 8A–C. The POC signal in the upper 130 m was much higher than at deeper depths and in this shallower range it increased slightly from July to September. TOC values showed similar trends but were more variable, especially the few measurements at deeper depths (not plotted below 150 m). The DOC values were of low accuracy because of combined errors in the POC and TOC measurements and difficulty of making low level measurements, but they show a similar seasonal trend to that for POC and TOC and a consistent small peak near a depth of 80 m.
On some occasions when there has been an influx of suspended inorganic particles the cp660 signal does reflect the concentration of POC. Figure 9A shows a series of cp660depth profiles during 1995, a summer that experienced a late snowmelt, and several large rain storms in June and July. Extra peaks are apparent in Figure 9A above 25 m and below 200 m. The change in this turbidity signal over time is plotted in Figure 9B for both the near-surface water (0–20 m) and representative deep water (>300 m) along with comparison data for the typical (dry) summer of 1994, when the deep cp660 remained low throughout the summer.
Chlorophyll-a concentration can also serve as a proxy for UVR attenuation. Figure 10A shows the average concentrations from 1984–2002 for shallow (0–30 m) and deep (40–140 m) chlorophyll-a extracted from phytoplankton retained on a 0.45 micron filter. Also plotted is average monthly rain in July and August for this period. Except for 1986–1987 the deep and shallow concentrations tended to change roughly in parallel, with a maximum during the late 1980s and low values since 1996. Figure 10B shows the variations in Kd,320 (estimated by our Secchi depth proxy) plotted against the chlorophyll-a concentration in the 0–30 m depth range. The regression line for the dry months or months where precipitation occurred as snow or where runoff would have entered cold surface waters (y = 015x + 0.08, r2=0.44) indicates that during periods with little runoff retained in the upper mixed layer about 44% of the variation in UV attenuation is explained by the absorption and scattering in UV wavelengths that co-varies with chlorophyll-a concentration.
