High-energy, short wavelength ultraviolet-B (UVB: 280-320 nm) and ultraviolet A (UVA: 320-400 nm) radiation initiate photochemical reactions and are biologically harmful on the molecular, cellular, and organismal levels. Global increases in the flux of UVB, the most damaging solar radiation reaching the earth’s surface, is a serious consequence of the measured declines in stratospheric ozone concentrations. At present, although there is abundant evidence of the negative effects of UV on various marine organisms, there are few studies characterizing the spectral attenuation of UV in the sea and no information available on how the vertical distribution of specific UV wavelengths change over space and time. These data are essential to understanding the ecological and biogeochemical significance of UV in the marine environment. This work is the first to examine the latitudinal and seasonal variability of solar UV penetration relative to photosynthetically active radiation (PAR: 400-700 nm) into productive coastal and open ocean waters.

The cross calibration of instruments for measuring underwater UV spectral irradiance was a critical component of this study. Radiometric instruments produced by four independent commercial manufactures were compared viewing a 1000W standard lamp, new and aged UVB lamps, and natural sunlight. Substantial differences in wavelength accuracy, absolute responsivity to a particular light source, and suitability for deployment in the marine environment were observed. A multi-channel filter radiometer was the best performer and used to profile PAR and spectrally specific UV at 305, 320, 340, and 380 nm. Wavelength specific diffuse attenuation coefficients (Kd) were calculated from the field data and used to define the 1% and 10% attenuation depths (Z).

Three optically distinct regions of UV, but not PAR, attenuation were identified in Narragansett Bay, Rhode Island. Seasonally, Kd values were highest in summer and lowest in winter. Variation in KdUV was a negative, linear function of salinity along the 31 km estuarine transect, suggesting that quasi-conservative constituents (e.g., DOM) primarily determined UV attenuation. The Z10% for Kd305, an index of biologically damaging UV levels, was less than 1 m. In the tropical Pacific study region between 0-15 N and 140-149 W, penetration of UV and PAR co-varied. Average Kd values from 0-6 N were higher than values from 9-15 N. The Z10% for Kd305 in these latitudinal regions was 16 and 20 m, respectively. Kd correlated with patterns in Chl a and beam attenuation, indicating that phytoplankton biomass (including population related DOM and POM) controlled oceanic UV and PAR attenuation. Phytoplankton in Narragansett Bay are unlikely to be affected by increases in the flux of UVB due to stratospheric ozone depletion whereas populations in the tropical and subtropical waters of the Pacific are presently UV stressed.

The photoinhibitory effects of ambient solar UV on primary producers were investigated in the tropical open ocean. UV exclusion incubations were employed to differentiate the effects of UV and PAR on the growth and PSII photochemistry of natural phytoplankton assemblages and to demonstrate the importance of exposure to full spectrum solar irradiance in studies designed to examine iron limitation. Phytoplankton exposed to full spectrum sunlight had a 20% greater decrease in PSII photochemistry. UV was also shown to negatively affect growth of Synechococcus spp. Effects were mainly due to UVA, but were intensified by UVB. Iron clearly stimulated growth and photosynthesis, but did not result in the radical shifts to larger species that have been consistently observed in similar container experiments that exclude UV.