DIEL CYCLES OF THE PARTICULATE BEAM ATTENUATION COEFFICIENT UNDER VARYING TROPHIC CONDITIONS IN THE NW MEDITERRANEAN SEA: OBSERVATIONS AND MODELING.

# IT45J-02

Pierre Gernez, [email protected], Scripps Institution Oceanography, UC San Diego, CA David Antoine

1) Introduction Photosynthesis shows a diel cycle driven by the variability in solar radiation (Doty and Oguri 1957). Diel scale measurements, usually made from ships or in cultures, are extremely labor intensive, of limited duration and difficult to generalize. To assess variability in these diel cycles thus require new approaches The particulate beam attenuation cp (m-1) is the sum of the scattering and absorption by particles from ~0.5 to 20 µm and can be measured routinely in situ and non-intrusively. Repeated measurements in the world oceans (e.g. Siegel et al. 1989) have revealed that a daily cycle in cp is a nearly ubiquitous feature, characterized by a diurnal increase and a nocturnal decrease. It has provided an alternative and nonintrusive method to investigate in-situ the growth and production of particle assemblages (Cullen et al. 1992; Claustre et al. 2007). The development of instrumented moorings and autonomous profiling platforms provides new opportunities to assess the persistence of the diurnal cycles under a wide range of environmental conditions and to characterize the diel variability over long periods. A 2-year (2006-2007) time series of near continuous of cp , chlorophyll fluorescence and irradiance measurements taken from the BOUSSOLE optical buoy (Antoine et al. 2006, 2008) is analyzed here. The site, in the northeastern Mediterranean sea is characterized by a seasonal cycle in the physical conditions, which results in seasonal changes in nutrient concentrations, particle abundance, size distribution, composition and optical properties. Based on this unique dataset, the objectives are 1) to characterize the diel cp variability under varying environmental conditions and 2) to estimate biogeochemical information about the mechanisms underlying the cp diel variability.

2) Methods The parameters used to characterize the diel variability are:

Laboratoire Océanographie Villefranche, CNRS-INSU-UPMC, France

Yannick Huot

Département Géomatique Appliquée, Université Sherbrooke, Canada

Fig. 3: cp versus [Chl] (open and plain symbols: 2006 and 2007). The time-series of cp and [Chl] (divided by 3) is inserted in the top-left box. The seasons are indicated by different colors: winter mixing, Spring phytoplankton bloom, Bloom collapse, Summer oligotrophy.

The two-year time series of daily averages 3) Time series and segmentation is displayed in Fig. 3. In order to “examine” the seasonal trend, the dataset has been segmented in four distinct seasons: 1) winter mixing, 2) spring phytoplankton bloom, 3) its collapse and 4) summer oligotrophy. Each season is characterized by different physical (i.e. mixing or stratification), trophic (i.e. high or low biomass), and bio-optical (i.e. position on the cpvs. [Chl] plot, Fig. 3) conditions.

Correlation between cp and [Chl] is observed all year long, but the relationship is not the same during mixing, bloom, collapse or oligotrophy.

Fig. 2: Temporal variation of cp (daily means in thick line, instantaneous measurements in grey points), daily means of [Chl] (green line; divided by 3) and monthly cruises values of zm (dashed line; filled circles). The vertical lines and the letters indicate the position of the examples displayed in Fig. 4.

4) Diel variablity A. Examples Winter mixing: In March 2006 (Fig. 4a), in spite of extremely low cp ~ 0.02 m-1, diel cycles occur. The daily variation (~0.004 m-1) is as large as the 6 days changes. In January 2007, cp is higher (~0.09 m-1 ; Fig. 4b). Diel cycles are again observed, with amplitude of 0.01 m-1. Winter µcp is ~ 0.3 d-1. Spring bloom: Irregular diel cycles are superimposed on an overall increase. In 2006 (Fig. 4c), cp increases by five from 0.03 to 0.15 m-1. The diel amplitude is ~ 0.05 m-1 and µcp is ~ 1 d-1. At the bloom apogee in 2007 (Fig. 4d), cp increases from 0.1 to 1 m-1. The diel amplitude reaches 0.8 m-1 and µcp 2.5 d-1. Collapse: In May 2006 (Fig. 4e) and June 2007 (Fig. 4f), cp decreases by ~ 2. Diel cycles are observed with amplitude below 0.07 m-1 and µcp ~ 0.3 d-1. Summer oligotrophy: Very regular diel cycles are observed (Figs. 5g, h). The diel amplitude is ~ 0.01 m-1 and µcp is ~ 0.3 d-1.

Fig. 4: Temporal evolution of cp (thick line) and PAR (thin line). Panels a) to g) show the time-intervals indicated in Fig. 2.

B. Diurnal variability by season

% variation from sunrise : where cp1 is cp at sunrise, k is a fraction of a day equals to 0, 0.5 and 1 at sunrise, sunset and the next sunrise. Fractions of day are used to compare varying day lengths. Instantaneous rate of variation: (in d-1) Diurnal rate of variation : (in d-1), with t1 and t2 : sunrise and sunset. Fig. 1: example showing the parameters defined above: cp(t) (thick line), cp1, 2 (grey points), r(t) (thin line), µcp (grey area).

The daily averaged % variation from sunrise is shown at each season (Fig. 5). A diel cycle clearly appears. ∆cp varies with the seasons: 10 - 20% during mixing, collapse and oligotrophy (Fig. 5a, c, d), and 40% - 60% during bloom (Fig. 5b). Standard deviation is maximal during bloom (>100%). In contrast, the diel cycles are the most regular during oligotrophy. The cycles are balanced except during the bloom (> 0) and collapse (< 0). The timing is nearly the same at all seasons: cp starts increasing at dawn and starts decreasing before sunset. The rate of variation r(t) shows a similar trend: it is ~0.6 d-1 during mixing, collapse and oligotrophy (Fig. 6a, c, d) and ~1.6 d-1 during bloom (Fig. 6b). It is striking that r(t) does not follow the PAR but is positively skewed to the morning. Night-time variations are small and almost constantly < 0.

Fig. 5 : Averaged ∆cp and PAR during mixing (a), bloom (b), collapse (c) and oligotrophy (d).

C. Rate of variation: not in phase w/ solar illumination With a morning maximum and a decrease before sunset, r is in advance on PAR (Fig. 6). This is consistent with results of laboratory studies, where maximal increases in cell size, refractive index, carbon per cell and bulk cp all occurred early in the morning (Stramski and Reynolds 1993).

5) Modeling the rate of variation A lag between PAR and photosynthetic growth has previously been observed in cultures (Harding et al. 1981; Bruyant et al. 2005). It is proposed that the variation of r(t) has the same origins: the phytoplankton photosynthetic parameters are varying during the day.

A. Model description As in Siegel et al. (1989), r(t) is the sum of a light-dependent growth term rG(t) and of a constant loss term rL(t). The loss term is the averaged r at night. The growth term is modeled like a photosynthetic growth (Jassby and Platt 1976): and where rmax is the maximum rate of variation and Ek is the saturation irradiance. The growth efficiency α = rmax / Ek. The variation of α(t) and Ek(t) has been parameterized after Bruyant et al. (2005): and Fig. 7: a) diel cycle of r: measurements (dashed line) and model output (solid line) ; b) model output of rmax (solid lines) and α (dashed line); c) model output of Ek (solid line) and measured PAR (dashed line).

Fig. 6 : As in Fig.5 but for the rate of variation r.

B. Model results The model compares favorably with the observations (Fig. 7a). Results are consistent with Bruyant et al. (2005): α varies by ~2, rmax by ~ 4 and Ek by ~ 3. α and of rmax have morning maxima whereas Ek is synchronized with PAR. Averaged parameters have been computed during mixing, bloom and collapse (not shown). The same seasonal trend is observed for α and rmax, with maxima during the bloom. Ek increases from winter to summer, likely as result of photo-adaptation.

6) Conclusion In-situ high-frequency cp measurements could be used to investigate the growth rate and carbon accumulation of particles assemblages. The accuracy of these bulk estimations would be greatly fostered by parallel measurements of individual particulate optical properties (such as automated flow cytometry and imagery).

C. Model application The net community production ∆POC is estimated from cp after Siegel et al. (1989) and Claustre et al. (2007). The results are compared with a standard [Chl] primary production model PP (Morel 1991). During oligotrophy, ∆POC is slightly lower than PP, but the difference vanishes when the mixing starts (Fig. 8). During the bloom, ∆POC is three times higher than PP. References Antoine, D. et al.(2006). BOUSSOLE: A Joint CNRS-INSU, ESA, CNES and NASA Ocean Color Calibration et Validation Activity, p. 59. NASA Tech. Memo. Antoine, D., et al.(2008). The "BOUSSOLE" buoy. A new transparent-to-swell taut mooring dedicated to marine optics: design, tests and performance at sea. J. Atmos. Ocean. Technol., 25:968:989. Bruyant, F. et al. (2005). Diel variations in the photosynthetic parameters of Prochlorococcus : Combined effects of light and cell cycle. Limnol. Oceanogr., 50: 850-863. Claustre, H., et al.(2007). Gross community production and metabolic balance in the South Pacific Gyre, using a non intrusive bio-optical method. Biogeosc., 5: 463-474. Cullen, J.J., et al. (1992). Photosynthetic characteristics and estimated growth rates indicate grazing is the proximate control of primary production in the Equatorial Pacific. J. Geophys. Res., 97: 639-654. Doty, M. S., and M. Oguri (1957). Evidence for a photosynthetic daily periodicity. Limnol.Oceanogr., 2: 37-40. Harding, L.W., et al. (1981). Diel periodicity of photosynthesis in marine phytoplankton. Mar. Biol., 61:95-105. Jassby, A.D., Platt, T. (1976). Mathematical formulation of the relationship between photosynthesis and light for phytoplankton. Limnol. Oceanogr., 21: 540-547. Loisel, H., and A. Morel (1998). Light scattering and chlorophyll concentration in case 1 waters: A reexamination. Limnol. Oceanogr., 43: 847-858. Morel, A. (1991). Light and marine photosynthesis: A spectral model with geochemical and climatological implications. Progr. Oceanogr., 26(3):263-306. Stramski, D., and R. A. Reynolds (1993). Diel variations in the optical-properties of a marine diatom. Limnol. Oceanogr., 38: 1347-1364.

Fig. 8: Seasonal variation of ∆POC (black) and PP (grey) in 2006 and 2007. Points indicate daily values; lines a 3-days running mean.