The impact of water turbidity on interannual sea surface temperature simulations in a layered global ocean model

The Naval Research Laboratory (NRL) Layered Ocean Model (NLOM) with an embedded bulk-type mixed layer model is used to examine the effects of ocean turbidity on sea surface temperature (SST) and ocean mixed layer depth (MLD) simulations over the global ocean. The model accounts for ocean turbidity through depth-dependent attenuation of solar radiation in the mixed layer formulation as determined from the diffusive attenuation coefficient at 490 nm (k(490)) obtained by the Sea-Viewing Wide Field-of-View Sensor (SeaWiFS). Interannual model simulations are used to assess the first-order effects of ocean turbidity on SST and MLD simulation. Results are reported from three model experiments performed using different values for the attenuation of photosynthetically available radiation (k(PAR)). It is shown that, although allowing incoming solar radiation to vary in time and space is desirable for predicting SST, in an OGCM use of a constant k(PAR) with a value of 0.06 m(-1) is generally sufficient in the deep ocean. The daily averaged SST time series from the three NLOM simulations are verified against daily in situ SSTs reported from 12 moored buoys in 1996 and 1997. Model results show that allowing the possibility of solar heating below the mixed layer reduces the root-mean-square error (rmse) difference between the daily yearlong model and buoy SST time series by up to 0.4degreesC and reduces the rmse at 11 of the 12 buoy locations. Although using spatially and temporally varying k(PAR) versus a constant k(PAR) 5 0.06 m(-1) (which is representative over most of the global ocean) had low impact overall, using it generally reduced the rmse at low latitudes, and using it can have a substantial impact locally in space and time. The model MLD results show low sensitivity to the k(PAR) value used.