Modeling the Near‐Surface Diurnal Cycle of Sea Surface Temperature in the Mediterranean Sea

Abstract

The diurnal cycle of sea surface temperature (SST) is an important component of the ocean-atmosphere system and is necessary for accurately computing air-sea heat fluxes. Ocean temperatures in the near-surface are highly sensitive to atmospheric conditions and can vary significantly depending on time of day. Ocean general circulation models are unable to fully capture the near-surface diurnal SST variability, because they do not possess the necessary vertical structure and resolution. Furthermore, SST observations come from a number of sources that represent the temperature at various near-surface depths. This presents difficulties when assimilating SST observations as well as constructing robust climate records of SST. In this study we model the fine-scale near-surface structure allowing SST comparisons between foundation SST, SST at depth, subskin SST, and skin SST. Hourly model results, forced and initialized using readily available reanalysis data, are from a 2-year period, 2013-2014, over the Mediterranean Sea. Various solar absorption parameterizations are examined, and the resulting SSTs are compared to Spinning Enhanced Visible and InfraRed Imager-derived observations of the skin temperature. Plain Language SummaryThe sea surface temperature (SST) is an important variable of the climate system. It governs the exchange of energy between the ocean and atmosphere. Accurate knowledge of SST is essential for weather and ocean forecasting. The vast majority of our observations of SST are remotely sensed and derived from satellite instruments that record the temperature within the top 1 mm. These observations are complemented by a mixture of ocean moorings and gliders that measure the temperature at various near-surface depths (∼20 cm to 1 m). SSTs are highly sensitive to atmospheric conditions and can vary significantly depending on time of day and near-surface depth. This presents difficulties when combining different observations to construct robust climate records of SST. In addition, large-scale computational models of the ocean typically resolve temperature in a surface layer ∼1 m thick and are unable to fully capture SST variability. It is therefore challenging to properly compare and/or merge modeled SSTs with observed SSTs. To address these challenges, we present a computational model of the fine-scale near-surface ocean structure. The model is tested in the Mediterranean Sea and used to produce a 2-year data set that makes possible SST comparisons at various near-surface depths at any hour of the day.

Evaluation of Ocean Color Remote Sensing Algorithms for Diffuse Attenuation Coefficients and Optical Depths with Data Collected on BGC-Argo Floats

“…For example, for photosynthesis, a euphotic or isolume depth is defined based on a relative light level (e.g., 1% [3]) or an absolute intensity (e.g., 0.415 mol photons m −2 d −1 [4]) below which photosynthesis is assumed as zero. Moreover, accurately quantifying the attenuation of sunlight within the upper ocean is essential for physical and biogeochemical models, affecting the modeled upper-ocean temperature (e.g., [5,6]) and ecosystem dynamics (e.g., [7]). Sea-surface sunlight is globally available from space agencies ( [8]), and a variety of models to describe its attenuation through the water have been devised (e.g., [9,10]).…”

A new global gridded sea surface temperature data product based on multisource data

“…The standard deviation obtained in a data comparison is approximately 0.68 • C, as determined by a comparison of the AVHRR products with coincident ferry observations (Barton and Pearce, 2006). The data also provided a quality index for each pixel based on the evaluation test results stored in the pathfinder_quality_level metric, which allows the identification of cloudy pixels and/or suspicious observations, with the quality level 0 representing the worst quality and the quality level 7 being the best (Pisano et al, 2016). In our data processing method, we only considered values with quality flags 4-7.…”

“…Subset 2 accounts for 20 % of the total number of in situ observations, which were used to verify the accuracy of the reconstruction results. The spatially coincident criterion restricts the maximum distance between in situ measurements and the center of the satellite image grid cells to within 2.3 km, which is approximately half the spatial resolution of MODIS, so that the in situ observations always fall within the MODIS SST pixels (Minnett, 1991;Pisano et al, 2016).…”

“…In addition, due to the difference of the inversion algorithm parameters, the sea temperature retrieved from the same type of sensor may also be different. For example, although the AVHRR sensor is an infrared remote sensor and its brightness temperatures represent the sea surface skin temperature, AVHRR SSTs correspond to subsurface SSTs because they are statistically regressed to coincident in situ buoy SSTs (Chao et al, 2009a;Kilpatrick et al, 2001;Pisano et al, 2016). Starting with the AVHRR Pathfinder Version 5.3, an average skin-subsurface temperature difference of 0.17 K, determined from Marine Atmospheric Emitted Radiance Interferometer (M-AERI) matchups, was used to eliminate the subsurface bias so that the SSTs were more closely tuned to the sea surface skin temperatures (Sea Surface Temperature-Pathfinder C-ATBD).…”

Using Canonical Correlation Analysis to produce dynamically-based highly-efficient statistical observation operators

“…column below. While the dataset ofPimentel et al (2018) uses a fine vertical resolution to calculate the SST observations, 30 the CCA OO will consider only the levels of a typical OGCM. Within the SOSSTA project this OGCM is the CMEMS 5 Ocean Sci.…”