Elsevier

Remote Sensing of Environment

Volume 226, 1 June 2019, Pages 38-50
Remote Sensing of Environment

Cool skin signals observed from Advanced Along-Track Scanning Radiometer (AATSR) and in situ SST measurements

https://doi.org/10.1016/j.rse.2019.03.035Get rights and content

Highlights

  • Characterization of cool skin signals on a global scale for nearly 10 years

  • Investigation of cool skin dependencies on different meteorological variables

  • Spatial and seasonal patterns of cool skin signals

  • Initial investigation into relationships between skin effect and wave variables

Abstract

Nighttime cool skin sea surface temperature (SST) signals, defined in this study as the differences between the SSTskin from the Advanced Along-Track Scanning Radiometer (AATSR) onboard Envisat satellite and in situ SSTs from drifting buoys and moorings, ΔT = SSTskin − SSTinsitu, are investigated on a global scale from July 2002 to April 2012. Global mean ΔT, averaged over the full study period, is −0.13 K, with most values falling between −0.1 and −0.2 K. The dominant role of wind speed on the ΔT is shown, with weaker winds usually corresponding to a cooler skin. The effect of air-sea temperature difference is also significant: warm skin (ΔT > 0 K) can be observed under large positive air-sea temperature differences. Other geophysical variables, such as the total column water vapor, in situ SST, and net heat flux, also affect ΔT, but to a lesser degree. Significant increase of ΔT size with SSTinsitu is observed when SSTinsitu is >28 °C. Tropical waters, such as the tropical Indian Ocean and the tropical warm pool (western Pacific and eastern Indian Ocean), are more frequently covered with a cool skin, largely due to the calm winds, very warm waters (especially for SSTinsitu >28 °C), and other environmental conditions supporting the development of large cool skin events. The ΔT seasonal pattern in the southern hemisphere is more regular, compared to the northern hemisphere. In both hemispheres, larger cool skin signals are seen during the local summer, mainly due to weaker winds. According to several previous cool skin models, higher winds tend to result in thinner cool skin layer depths, and hence in smaller ΔT amplitudes, regardless of stronger evaporation and heat loss. Given that wind is closely coupled with waves and turbulent mixing with wave breaking, the dependencies of ΔT on a few wave parameters are also investigated. A strong (moderate) dependency of ΔT on wave height (wave steepness) is identified, while the dependency of ΔT on wave breaking probability is less discernible.

Introduction

It has been long known that the skin sea surface temperature (SST) is usually slightly cooler than the temperature immediately below, referred to as the cool skin effect or “skin effect”, for short (e.g., Woodcock, 1941, Woodcock, 1947). According to the Group for High Resolution SST (GHRSST; Donlon et al., 2007) practical definitions, the skin SST refers to the temperature at around 10–20 μm depth, measured by an infrared (IR) radiometer typically operating at 3.7–12 μm wavelengths. Conductive diffusion is dominant in this layer. Cool skin exists because of the combined cooling effects of the longwave radiation and the latent and sensible heat fluxes (e.g., Saunders, 1967; Fairall et al., 1996). Since under most circumstances the net heat flux is from the ocean to the atmosphere, the cool skin is usually present with an amplitude of a few tenths of a degree. In the daytime, when the wind is calm and solar insolation strong, diurnal warming can have amplitudes of several Kelvins, which may offset the cool skin effect (e.g., Fairall et al., 1996; Gentemann et al., 2003). Diurnal warm layer tends to vanish at night, when a near constant temperature profile is restored in the upper few metres up to the bottom of the skin layer, which is referred to as the subskin SST, SSTsubskin, measured by a microwave (MW) sensor at ~1 mm depth. Incorporating a diurnal warm and cool skin layers' schemes in a numerical weather prediction or a climate model has been shown to improve the model's accuracy, due to more accurate estimation of air-sea interactions (e.g., Robertson and Watson, 1992; Zeng and Beljaars, 2005; Brunke et al., 2008; Masson et al., 2012; Clayson and Bogdanoff, 2013; Akella et al., 2017).

Numerous studies have been conducted to observe and/or model the cool skin layer (e.g., Saunders, 1967; Hasse, 1971; Brutsaert, 1975; Liu et al., 1979; Hepplewhite, 1989; Schlüssel et al., 1990; Soloviev and Schlüssel, 1994, Soloviev and Schlüssel, 1996; Fairall et al., 1996; Wick et al., 1996; Wick and Jessup, 1998; Artale et al., 2002; Donlon et al., 2002; Castro et al., 2003; Minnett, 2003; Tu and Tsuang, 2005; Ward, 2006; Minnett et al., 2011; Alappattu et al., 2017). Most of these studies used shipborne IR skin SSTs and coincident depth SSTs at a few centimetres to metres depths. The spatial and temporal scales are therefore restricted to the ships' duration and routes. While some studies used both daytime and nighttime data (e.g., Kent et al., 1996; Minnett, 2003; Minnett et al., 2011), it is not uncommon for some authors to adopt nighttime data only, to minimize the complication caused by diurnal warming (e.g., Horrocks et al., 2003). With data of high accuracy but sometimes limited in amount, a series of physical or empirical cool skin models have been developed, which can be generally divided into two groups. The first group, represented by the model proposed in Saunders (1967), considers two essential mechanisms controlling the heat fluxes across the molecular skin layer: free convection, caused by the thermal instability, and the salinity gradient across the cool skin itself under very calm winds (<2 m s−1), and forced convection driven by the surface shear stress. Many studies followed with foci on determining the Saunders' proportionality constant λ and then the thickness of the cool skin layer (e.g., Paulson and Simpson, 1981; Robinson et al., 1984; Wu, 1985; Fairall et al., 1996; Artale et al., 2002; Tu and Tsuang, 2005). The other group of parameterizations was developed based on the surface renewal theory, which assumes that a part of the surface layer is removed and replaced by water from beneath (e.g., Brutsaert, 1975; Liu et al., 1979; Schlüssel et al., 1990; Soloviev and Schlüssel, 1994; Wick et al., 1996; Castro et al., 2003). In addition to the physical models, empirical parameterizations have been also proposed in more recent studies, relating the cool skin layer amplitude to environmental variables such as wind speed (e.g., Donlon et al., 2002; Minnett et al., 2011; Alappattu et al., 2017). Different models have been intercompared with each other in several studies (e.g., Kent et al., 1996; Castro et al., 2003; Horrocks et al., 2003; Tu and Tsuang, 2005).

Although spaceborne radiometers have been retrieving the skin SST on a global scale for almost four decades since early 1980s, satellite data have been scarcely used in cool skin investigation, likely due to several reasons. First, satellite IR and in situ SSTs represent waters of spatially and temporally different scales. Infrared SSTs are almost instantaneous area averages, whereas in situ SSTs are point data, which may be measured either instantaneously or averaged in time. Such systematic differences exist regardless of the sizes of the temporal and spatial windows employed in their collocation. Second, traditionally, the uncertainties in IR SST retrievals are considered too large (with a typical standard deviation, SD ~0.5 K, when validated against drifting or mooring buoy measurements), making it challenging to analyse the cool skin signal, whose amplitude is typically much smaller. Several factors contribute to this large uncertainty, related to the instrument (spectral response, radiometric noise, in-flight calibration, etc.) and retrieval algorithms (including cloud screening, aerosol detection, correction for the effects of water vapor absorption, etc.) (e.g., Kilpatrick et al., 2015).

These issues have been presumably minimized in the (Advanced) Along Track Scanning Radiometers, (A)ATSRs, SSTskin data sets produced by the (A)ATSR Reprocessing for Climate (ARC) project. The (A)ATSRs are characterized by more accurate in-flight calibration and dual-view technique, which in turn offer an improved potential for accurate atmospheric correction (Llewellyn-Jones et al., 2001). In particular, the ARC project retrieves SSTskin using coefficients based on radiative transfer models (RTM), independently of in situ measurements. Recall that many operational SST retrieval algorithms are empirically regressed against in situ measurements, and therefore may not be fully independent from those. Although not without its own limitations, the ARC SSTskin data may be thus better suited for skin effect studies (Murray et al., 2000; see more detail in 2.1 Data sets, 3 Quality of AATSR and in situ SSTs).

This paper characterizes the cool skin behaviours on a global scale, using nearly ten-year nighttime AATSR SSTskin data, in conjunction with collocated in situ SSTs, measured by drifting, coastal and tropical mooring buoys. The structure of the paper is as follows. Section 2 introduces the data sets and methods. Section 3 briefly describes AATSR and in situ data. Section 4 characterizes the cool skin signals, including their statistics and relationships with different environmental variables. Discussion and conclusions are presented in 5 Discussion, 6 Conclusions, respectively.

Section snippets

AATSR SSTskin data

The AATSR sensor was flown onboard ESA's Envisat satellite, launched in March 2002 as a successor to ATSRs −1 and −2 (launched in July 1991 and April 1995, respectively). Compared to previous missions, several improvements have been made to this family of instruments, including: 1) the dual-view (nadir and forward views ~ 55° from zenith) geometry, within a few minutes of each other, allowing for more effective atmospheric correction; 2) a rigorous pre-launch calibration programme and

Quality of AATSR and in situ SSTs

The high quality of AATSR SSTs have been illustrated in a number of studies (e.g., Corlett et al., 2006; Noyes et al., 2006; O'Carroll et al., 2006; O'Carroll et al., 2008; Reynolds et al., 2010; Kennedy et al., 2012; Merchant et al., 2012; Merchant et al., 2014; Xu and Ignatov, 2016). Several publications used a triple collocation method (TCM) and showed that AATSR SSTs are as precise as in situ observations, or better. For instance, O'Carroll et al. (2008) showed that the spatially averaged

General statistics

The overall statistics of AATSR satellite SSTskin minus SSTinsitu are shown in Fig. 1. The mean ΔT (defined as SSTskin – SSTinsitu) is −0.13 K, which is comparable to the findings in previous studies such as Donlon et al. (2002) and Minnett et al. (2011). The 0.46 K SD is obviously larger than the SDs of ~ 0.16 or 0.14 K in O'Carroll et al. (2008) and Merchant et al. (2012), which were based on a TCM, which accounts for the uncertainties in SSTinsitu. The RSD (robust SD, calculated as 1.5 times

Discussion

The AATSR dual-view three channel (D3) skin SST, in conjunction with high-quality in situ SSTs from the NOAA iQuam system, are employed here for cool skin analyses, thanks to its high quality (attributable to innovative sensor features, such as the dual-view geometry and stable calibration), and independency from in situ measurements (due to the physics-based retrieval algorithm). Several prior studies have touched on the cool skin effect using ATSRs data. For example, Murray et al. (2000)

Conclusions

Cool skin signals revealed from nighttime AATSR skin SST data and NOAA iQuam in situ SST measurements (defined as ΔT = SSTskin − SSTinsitu), have been described in detail on a global scale for nearly ten years from July 2002 to April 2012. Foci are on skin effects' overall statistics, dependencies on different meteorological variables, spatial distribution, and seasonal patterns. So far, the cool skin has not been systematically analysed on a global scale and over a long period, using a

Acknowledgements

This work was supported by the Australian Research Council Discovery under Project DP170101328. The work of A. V. Babanin was supported by the DISI Australia-China Centre under Grant ACSRF48199. We would like to thank Helen Beggs and Chris Griffin for their useful comments on the manuscript. The views, opinions, and findings contained in this paper are those of the authors and should not be construed as an official NOAA or U.S. Government position, policy, or decision.

References (75)

  • E. Noyes et al.

    The accuracy of the AATSR sea surface temperatures in the Caribbean

    Remote Sens. Environ.

    (2006)
  • H. Zhang et al.

    Investigating sea surface temperature diurnal variation over the tropical warm pool using MTSAT-1R data

    Remote Sens. Environ.

    (2016)
  • S. Zieger et al.

    Observation-based source terms in the third-generation wave model WAVEWATCH

    Ocean Model.

    (2015)
  • S. Akella et al.

    Assimilation for skin SST in the NASA GEOS atmospheric data assimilation system

    Q. J. R. Meteorol. Soc.

    (2017)
  • D.P. Alappattu et al.

    Warm layer and cool skin corrections for bulk water temperature measurements for air-sea interaction studies

    Journal of Geophysical Research: Oceans

    (2017)
  • F. Ardhuin et al.

    Semiempirical dissipation source functions for ocean waves. Part I: definition, calibration, and validation

    J. Phys. Oceanogr.

    (2010)
  • V. Artale et al.

    Role of surface fluxes in ocean general circulation models using satellite sea surface temperature: validation of and sensitivity to the forcing frequency of the Mediterranean thermohaline circulation

    Journal of Geophysical Research: Oceans

    (2002)
  • A.V. Babanin

    Breaking and Dissipation of Ocean Surface Waves

    (2011)
  • A.V. Babanin et al.

    Breaking probabilities for dominant surface waves on water of finite constant depth

    Journal of Geophysical Research: Oceans

    (2001)
  • A.V. Babanin et al.

    Numerical investigation of spectral evolution of wind waves. Part II: dissipation term and evolution tests

    J. Phys. Oceanogr.

    (2010)
  • M.A. Brunke et al.

    Integration of a prognostic sea surface skin temperature scheme into weather and climate models

    Journal of Geophysical Research: Atmospheres

    (2008)
  • M.A. Brunke et al.

    An assessment of the uncertainties in ocean surface turbulent fluxes in 11 reanalysis, satellite-derived, and combined global datasets

    J. Clim.

    (2011)
  • W. Brutsaert

    A theory for local evaporation (or heat transfer) from rough and smooth surfaces at ground level

    Water Resour. Res.

    (1975)
  • S.L. Castro et al.

    Further refinements to models for the bulk-skin sea surface temperature difference

    Journal of Geophysical Research: Oceans

    (2003)
  • S.L. Castro et al.

    Error characterization of infrared and microwave satellite sea surface temperature products for merging and analysis

    Journal of Geophysical Research: Oceans

    (2008)
  • A.H. Chaudhuri et al.

    A comparison of atmospheric reanalysis surface products over the ocean and implications for uncertainties in air–sea boundary forcing

    J. Clim.

    (2013)
  • C.A. Clayson et al.

    The effect of diurnal sea surface temperature warming on climatological air–sea fluxes

    J. Clim.

    (2013)
  • D.P. Dee et al.

    The ERA-interim reanalysis: configuration and performance of the data assimilation system

    Q. J. R. Meteorol. Soc.

    (2011)
  • M.A. Donelan et al.

    Wave-follower field measurements of the wind-input spectral function. Part II: parameterization of the wind input

    J. Phys. Oceanogr.

    (2006)
  • C.J. Donlon et al.

    Observations of the oceanic thermal skin in the Atlantic Ocean

    Journal of Geophysical Research: Oceans

    (1997)
  • C. Donlon et al.

    Toward improved validation of satellite sea surface skin temperature measurements for climate research

    J. Clim.

    (2002)
  • C. Donlon et al.

    The global ocean data assimilation experiment high-resolution sea surface temperature pilot project

    Bull. Am. Meteorol. Soc.

    (2007)
  • C. Fairall et al.

    Cool-skin and warm-layer effects on sea surface temperature

    Journal of Geophysical Research: Oceans

    (1996)
  • C.L. Gentemann et al.

    Radiometric measurements of ocean surface thermal variability

    Journal of Geophysical Research: Oceans

    (2008)
  • C.L. Gentemann et al.

    Diurnal signals in satellite sea surface temperature measurements

    Geophys. Res. Lett.

    (2003)
  • L. Hasse

    The sea surface temperature deviation and the heat flow at the sea-air interface

    Bound.-Layer Meteorol.

    (1971)
  • C. Hepplewhite

    Remote observation of the sea surface and atmosphere the oceanic skin effect

    Int. J. Remote Sens.

    (1989)
  • Cited by (11)

    • A full physics algorithm to retrieve nighttime sea surface temperature with IASI: Toward an independent homogeneous long time-series for climate studies

      2022, Remote Sensing of Environment
      Citation Excerpt :

      For (T2m-Tdepth) > 2.5 K, ΔT becomes positive for ERA5 and IASI-L2, contrary to the other products for which ΔT approaches 0 K but remains negative. A continuous positive increase of ΔT with increasing (T2m-Tdepth) was also observed with AATSR SST data as shown in Fig. 4 of (Zhang et al., 2019), but with a larger amplitude of variation, from about −0.25 K for (T2m-Tdepth) ≈ −5 K to +0.5 K for (T2m-Tdepth) ≈ 4 K. Sections 3.3.1–3.3.3 demonstrate that our retrieved values closely follow the predictions made using (Fairall et al., 1996), as well as the empirical relationship of (Zhang et al., 2020) for wind speed, implying a good accuracy, with a very small bias.

    • Ocean Waves in Large-Scale Air-Sea Weather and Climate Systems

      2023, Journal of Geophysical Research: Oceans
    View all citing articles on Scopus
    View full text