Ultraviolet Radiation and Bio-optics in Crater Lake, Oregon, 2005
Spectral measurements of incident and underwater solar radiation
Spectral measurements of downwelling irradiance (Figure 1A), upwelling irradiance (Figure 2A), spectral Kd (Figure 3) in August 2001 were nearly identical to the 1960s data of Smith et al 1973 except that the recent measurements included a greater range of wavelengths, responded to lower light levels, and in the case of Eu, resolved algal fluorescence distinctly at 683 nm. The LI-1800uw has been shown previously to record solar irradiance accurately within the constraints of its 8 nm bandwidth except at wavelengths below 308 nm (Kirk et al., 1994). Figure 1 demonstrates similar performance: matching within 96% on average and -13% and +9% at all wavelengths from 308 to 400 nm in comparing a scan of incident irradiance with a spectrum generated by a high resolution radiative transfer model adjusted for local conditions and bandwidth.
Figures 2A and 2B show signs of “internal light sources” including chlorophyll-a fluorescence (683 nm), Raman scattering (impact at shorter wavelengths from 700–490 as depth increases), and possibly phycoerythrin fluorescence (589 nm, Hewes et al., 1998). The fluorescence signals indicate the presence of phytoplankton but cannot be directly used as indices of concentration because of photoacclimation and quenching near the surface. The detection of any internal source should serve as a warning because it invalidates attempts to measure diffuse attenuation at the wavelength and depth where it is detected.
While the conventional method of characterizing potential exposure to UVR is to measure diffuse attenuation, another factor which strongly influences exposure is incident irradiance. Our discovery of a positive correlation between phytoplankton chlorophyll-a and stratospheric ozone means that variations in incident UV-B irradiance could be playing an important role and thus UV-B irradiance should be monitored continuously on or near the lake.
The upwelling irradiance (Figure 2A) and radiance spectra (not shown) help to explain the unusual color of Crater Lake. The low concentration of absorbing substances (e.g. phytoplankton and CDOM) near the surface clearly accounts for part of the phenomenon, thereby allowing for a greater optical role for water molecules to transmit and scatter light of certain wavelengths. Spectral reflectance models that predict the optical behavior of pure water (e.g. Morel & Prieur 1977) indicate that while Crater Lake has extremely low levels of absorbing substances near the surface, it also contains particles that increase backscatter (for more on scattering, see Boss et al., this issue). The optical role of abundant glass-like suspended particles reported in the surface waters by Utterback et al. (1942) should be investigated.
Comparison of spectral measurements from 1969 in Crater Lake with similar measurements in Lake Tahoe (Smith et al., 1973) showed that Crater Lake had a greater proportion of short wavelengths in its upwelling spectrum and in the deep downwelling spectra. The differences between these two lakes appear to have increased since 1969 because of rising levels of nutrients, phytoplankton, and suspended mineral particles in L. Tahoe (Jassby et al., 1999).