Ultraviolet Radiation – 11 RESULTS Spectral diffuse attenuation, Validation, and Variations over Space and Time

Ultraviolet Radiation and Bio-optics in Crater Lake, Oregon, 2005



Spectral diffuse attenuation, Validation, and Variations over Space and Time

Spectral diffuse attenuation (Kd,?) is shown in Figure 3 calculated from the spectral irradiance data shown in Figure 1 (Kd was averaged from multiple scans for specific depth ranges and the standard errors of the mean for these are indicated by error bars). One additional depth range (10–18 m) not appearing in Fig 1A was calculated from Biospherical Instruments, Inc. (BSI) PRR-800 and PUV-2500 profiling radiometers. In the UV and blue wavebands Kd spectra were lowest near the surface and rose with increasing depth down to the deep chlorophyll maximum (DCM=130 m). The Kd pattern at wavelengths longer than 550 nm varied little with depth when measured near the surface but at depths deeper than 20–30 m and shallower at the longest visible wavelengths the Kd values declined with increasing depth (not plotted). LI-COR Kd spectra for shallow depths (not plotted) were similar to the plotted Kd spectrum from the BSI radiometers but were noisier because fewer measurements were used in the calculations.

Variations of Kd with depth for specific wavelengths on 20 August 2001 appear in Figure 4. The LI-COR UV wavelengths were selected to match the filter wavebands (320, 340, 380 nm and PAR) of two BSI radiometers (all three were deployed within about 1 hour of solar noon). The curves show a region of uniform and low Kd,UV from 0–10 m (the thermally mixed layer was also 0–10 m), then a pattern of increasing Kd,UV with depth to a peak at the DCM near 130 m. For broadband visible irradiance (PAR, 400–700 nm) the Kd,PAR rose rapidly above 15 m but paralleled the other sensors at depths below 60 m. The three instruments agreed closely for most depths over which the same wavelengths were measured, with the greatest variation occurring near the surface. Figure 5 shows the same type of pattern but in this case averaged for the period 1996–2002. There are parallel changes with depth in Kd,320, Kd,380, and Kd,blue (summarized also in Table 1). Figure 6 shows that the ratio of Kd320:Kd380 varied regularly with depth (Kd320:Kd380 =-0.0031*depth + 2.32, r2=0.68) for the 1996–2002 summer data. In contrast, the ratios Kd,320:Kd,blue and Kd,380:Kd,blue were relatively constant for two depth ranges: 0–10 m and below 10 m. Related optical changes with depth based on particle absorption coefficients are shown in Figure 12B (described below).

Seasonal and interannual changes in UV Kd near for different depth ranges are summarized in Figure 7. July and August were typically the months with the lowest Kd,320 averaged over 0–40 m, the depth range where maximal UVR impact on organisms was expected. If the single measurement in January is typical then winter values are much higher than midsummer values for this depth range (Figure 7A). Over the period from 1996–2002 the lowest summer average Kd,320 was observed in 2001; the highest were in 1998 and 1999. When July and August data are averaged over this period the greatest interannual variations in Kd,380 occurred in the depths from 20–40 m and the least variations occurred from 100–140 m. The range from 0–20 m changed in parallel with 20–40 m except for 1999 (Figure 7B).