SEASONAL VARIABILITY OF LIGHT AVAILABILITY AND UTILIZATION IN THE SARGASSO SEA

A 2 year time series of optical biogeochemical, and physical parameter s, taken near the island of Bermuda, is used to evaluate the sources o f temporal variability in light availability and utilization in the Sa rgasso Sea. Integrated assessments of light availability are made by e xamining the depth of Constant percent incident photosynthetically ava ilable radiation (%PAR) isolumes. To first order, changes in the depth of %PAR isolumes were caused by physical precesses: deep convective m ixing in the winter which led to the spring bloom and concurrent shado wing of %PAR depths and the occurrence of anomalous thermohaline and l ater masses during the summer and fall seasons. Spectral light availab ility variations are assessed using determinations of diffuse attenuat ion coefficient spectra which illustrate a significant seasonal cycle in colored detrital particulate and/or dissolved materials that is unr elated to changes in chlorophyll pigment concentrations. Temporal vari ations in the photosynthetic light utilization index Psi are used to a ssess vertically integrated light utilization variations. Values of Ps i are highly variable and show no apparent seasonal pattern which indi cates that Psi is not simply a ''biogeochemical constant.'' Determinat ions of in situ primary production rates and daily mean PAR fluxes ate used to diagnose the relative role of light limitation in determining vertically integrated rates of primary production integral PP. The me an depth of the light-saturated zone (the vertical region where the da ily mean PAR flux was greater than or equal to the saturation irradian ce I-k) is only similar to 40 m, although mole than one half of integr al PP occurred within this zone. Production model results illustrate t hat accurate predictions of integral PP are dependent upon rates of li ght-saturated production rather than upon indices of light limitation. It seems unlikely that significant improvements in simple primary pro duction models will come from the partitioning of the Earth's seas int o biogeochemical provinces.