Many mechanisms, such as mixing, provide UV-B protection by reducing
radiant energy to which cells are exposed, but they also reduce the energy
available for photosynthesis.
The several photoprocesses that are simultaneously active within
the cell (cf Smith et al., 1992) have different wavelength sensitivities,
operate with different time constants, respond differently,
and probably non-linearly, to changing irradiance levels as they are
mixed within the water column.
Thus physical mixing significantly complicates attempts to resolve the
biological balance between damage (usually shorter less penetrating
wavelengths) and repair (usually longer more penetrating wavelengths) processes.
Early analytic models [Morowitz 1950, Zepp Cline 1977, Ray Smith Baker
assessment 1982], often overlooked by more recent workers, show that that
balance between `protection' to near surface organisms and `added exposure'
of otherwise deep organisms by vertical mixing is a balance between depth
and rate of mixing and optical properties of the water column.
Smith [1989]
[Ray Smith 1989 PCPB]
recently reviewed this subject and pointed out that
an important consideration, with respect to the dose response of
phytoplankton, is the ratio of biological dose to the photosynthetically
available radiation,
, versus biological damage.
If the time scale for a biological response (e.g., UV damage, photoadaptation) is shorter than that for vertical mixing, phytoplankton will exhibit a vertical gradient of this response [Lewis Cullen Platt 1984 vertical mixing, Cullen Lewis 1988]. On the other hand, if mixing occurs with a time scale shorter than that of the biological response, no such gradient will be observed. The rates of various photoprocesses (which are strong functions of wavelength), as compared with that of vertical mixing (which rapidly alters the in-water spectral irradiance), will determine the overall effect on the populations at depth. It is also important to recognize that transient stratification, often associated with phytoplankton `bloom' events [Riley 1942, Sverdrup 1953, Brown Evans Brown Gordon Smith Baker 1985], can often coincide with periods of high irradiance, thus maximizing UV exposure.
While it has long been known that vertical mixing is a major complication in attempting to quantify UV-B effects on phytoplankton [Kullenberg 1982, Ray Smith Baker assessment 1982], recent work suggests that linking the rates of the various photoprocesses to a physical mixing model may be more complex than previously thought. Work by Prezelin and co-workers [Prezelin Putt Glover 1986, Prezelin Boucher Smith 1992 daytime, Prezelin Boucher Smith 1994] shows strong diurnal patterns in the photosynthetic capacity and depth-dependent photosynthesis-irradiance relationships and also strong diurnal effects in UV-inhibition of photosynthesis. They note that the problem is also confounded by wavelength changes in irradiance with depth. Cullen and Lesser [1991] [Cullen Lesser 1991] have shown that photoinhibition of phytoplankton production is dependent on irradiance only for time scales longer than the induction period for photoinhibition, i.e., > approximately 1 hr. On shorter time scales relative photoinhibition is time (dose) dependent. They conclude that conventional productivity incubations (typically 2 to 24 hr.) will adequately represent conditions in the water column only if this column is relatively stable over these same periods.
Helbling et al. [1994], [Helbling Villafane Holm-Hansen 1994] using rotating incubator experiments to simulate vertical mixing, studied the effects of UV radiation on Antarctic marine phytoplankton. These workers found that the magnitude and sign of the difference in column integrated production between rotating samples (exposed to varying irradiance levels with a 6 hr time scale) and fixed samples depended upon the mean irradiance level. This demonstrated a failure of reciprocity under the variable irradiance treatment, consistent with a disproportionate amount of inhibition resulting from short periods of high irradiance experienced by the rotating samples.