No abstract

The influence of temperature on light transmission in the spectral range from 400 to 760 nm has been determined in a two-cell instrument constructed especially for this purpose. Light transmission was measured over a 1-m path length in both a photometric and a spectral mode in double-ion-exchanged fresh water and filtered seawater with a salinity of approximately 25%?. For both groups of samples the temperature-dependence coefficient of the absorption was found to be -0.00091 ± 0.00006 m(-1) K(-1) in the range from 400 to 550 nm, in contrast to earlier findings. Reproducible signals could be observed only when the samples were left undisturbed for long periods of time between shifts in temperature. The temperature was scanned in alternating directions between 6 and 30 °C in steps of ± 2 °C. The time for recording a set of data was between 2 and 3 weeks. Finally, the temperature dependences of the absorption spectra were recorded in the range 400 to 760 nm. These measurements are only partially in agreement with earlier measurements.

Abstract. A primary data set consisting of 70 series of angular radiance distributions observed in clear blue western Mediterranean water and a secondary set of 12 series from the more green and turbid Lake Pend Oreille, Idaho, have been analyzed. The results demonstrate that the main variation of the shape of the downward radiance distribution occurs within the Snell cone. Outside the cone the variation of the shape decreases with increasing zenith angle. The most important shape changes of the upward radiance appear within the zenith angle range 90ø-130 ø . The variation in shape reaches its minimum around nadir, where an almost constant upward radiance distribution implies that a flat sea surface acts like a Lambert emitter within _+8% in the zenith angle interval 140ø-180 ø in air. The ratio Q of upward irradiance and nadir radiance, as well as the average cosines /•a and/•, for downward and upward radiance, respectively, have rather small standard deviations, _<10%, within the local water type. In contrast, the irradiance reflectance R has been observed to change up to 400% with depth in the western Mediterranean, while the maximum observed change of Q with depth is only 40%. The dependence of Q on the solar elevation for blue light at 5 m depth in the Mediterranean coincides with observations from the central Atlantic as well as with model computations. The corresponding dependence of/•a shows that diffuse light may have a significant influence on its value. Two simple functions describing the observed angular radiance distributions are proposed, and both functions can be determined by two field observations as input parameters. The e function approximates the azimuthal means of downward radiance with an average error _<7% and of upward radiance with an error of-1%. The a function describes the zenith angle dependence of the azimuthal means of upward radiance with an average error _<7% in clear ocean water, increasing to _<20% in turbid lake water. The a function suggests that the range of variation for/•, falls between 0 and 1/2, and for Q it is between rr and 2rr. The limits of both ranges are confirmed by observations. By combining the e and • functions, a complete angular description of the upward radiance field is achieved. IntroductionRadiance is the single most important quantity from which various irradiances and vertical attenuation coefficients can be derived. In addition, irradiance distribution functions, the light absorption coefficient, and, in principle (by inverse methods), the light-scattering function can be determined for a completely specified radiance field. The total radiance field gives us all the pertinent inherent and apparent parameters describing the optical characteristics of natural waters. Underwater measurements of angular radiance distributions are generally very few. Except for one recent example [Voss, 1988], it is noteworthy that one has to go 20 years or more back in time in order to find such distributions from lakes, fjords, and ocean waters [Jerlov, 1951[Jerlov, , 1976

The light absorption coefficient of a medium such as seawater is an inherent optical property that can be determined in absolute values by means of the uncalibrated in situ light absorption meter described here. The method is based on the assumptions that the optical conditions remain steady within the measuring period, that the horizontal divergence of the vector equals zero, and that the natural fluorescent light energy within the visible part of the spectrum can be ignored. The light absorption meter measures simultaneously the sum and the difference of the scalar and vector irradiance.

Measurements of the radiance and degree of polarization made in 1971 in the Mediterranean Sea are presented along with the simulation of all observed quantities by a Monte Carlo technique. It is shown that our independent scattering treatment utilizing a Stokes vector formalism to describe the polarization state of the light field produces remarkably good agreement with those values measured in situ.