[A-CG52] [JJ] Biogeochemical linkages between the ocean and the atmosphere during phytoplankton blooms

[ACG52-06] Effect of optical properties variability on retrieval of chlorophyll a from ocean color data in Oyashio and coastal Oyashio waters in early spring

*Toru Hirawake¹, Hisatomo Waga², Takuro Kaneko², Koji Suzuki³, Youhei Yamashita³, Jun Nishioka⁴ (1.Faculty of Fisheries Sciences, Hokkaido University, 2.Graduate school of Fisheries Sciences, Hokkaido Univeristy, 3.Faculty of Environmental Earth Science, Hokkaido University, 4.Institute of Low Temperature Science, Hokkaido University)

Keywords:optical property, ocean color remote sensing, Oyashio

Massive phytoplankton bloom occurs in Oyahsio and coastal Oyashio waters off Kushiro during spring. Ocean color images also illustrates the spring bloom as high chlorophyll a (chl.a) concentration. While a validation result showed the estimated chl.a concentrations in these waters are within a range of +/-35% of in situ values, effects of Tokachi river plume on chl.a estimation were anticipated. However, little is known about optical properties in this region. We measured remote sensing reflectance (R_rs), absorption coefficient, backscattering coefficient (b_bp) and chl.a of the waters during the cruise of R/V Hakuho-maru in March 2015 and investigated the effects of the optical properties on chl.a estimation from satellite data.
In the coastal stations of study area where was highly turbid for a few days due to passing of a low pressure during 10–13 March, absorption coefficient of non-algal particles (a_d) accounted for >60% of that of particulate matters (a_p) and b_bp was also higher than those in offshore stations. These optical properties induced high R_rs and resulted in overestimation of chl.a concentration by 2-3 folds when the standard ocean color chl.a algorithm was applied to the R_rs. On March 20, however, ratio of ad and b_bp decreased and estimated chl.a concentration using the algorithm coincided with in situ data.
Phytoplankton absorption coefficient at 443 nm (a_ph(443)) and b_bp(555) were linearly correlated with chl.a concentration and a_d(443), respectively. Estimated values of chl.a concentration and a_CDOM(443) applying these relationship to satellite ocean color data indicated that relatively higher ocean color chl.a in offshore region had interferences by non-algal particles and CDOM advected from coastal region. In this study region, satellite chl.a images should be carefully used and retrieval of chl.a from a_ph is better than using standard band-ratio algorithm because spatio-temporal variability in optical properties are dynamic and complicated not only in coastal region but also in offshore.