SciELO - Scientific Electronic Library Online

vol.68 número2  suppl.TIProcMAPPING OIL SPILLS IN THE CASPIAN SEA USING ERS-1/ERS-2 SAR IMAGE QUICK-LOOKS AND GISESTIMATING POLYNYA CLOUDINESS índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Gayana (Concepción)

versión impresa ISSN 0717-652Xversión On-line ISSN 0717-6538

Gayana (Concepc.) v.68 n.2 supl.TIProc Concepción  2004 


Gayana 68(2) supl. t.I. Proc. : 305-310, 2004 ISSN 0717-652X



Kristina B. Katsaros

P.O. Box 772, Freeland, WA 98249, U.S.A. (email:
Alexander V. Soloviev Oceanographic Center, Nova Southeastern University, 8000 North Ocean Drive
Dania Beach, FL 33004-3078, U.S.A. (email:


During days with strong insolation and low wind speed, there may be uneven net heating of the water layer near the surface of the ocean caused by variations in horizontal temperature at the sea surface. The heat loss from the water caused by evaporation, sensible, or longwave radiation is proportional to the sea surface temperature and is, therefore, greater from warm water compared to that from the relatively colder water. As a result, under low wind speed conditions and clear skies, the horizontal SST discontinuities, occurring at fronts, eddies, or in storm wakes, may diminish or even vanish. This phenomenon is illustrated here with some field and modeling results. The time dependence is important for the impact on remote sensing of SST, and it is found to be short enough that substantial masking of SST gradients can occur during the first six hours of the diurnal heating cycle, but the effect would continue to grow if calm and solar heating persist for several subsequent days. An integrated effect of this uneven net heating is seen in the seasonal masking of subsurface temperature gradients in the Gulf of Mexico and Florida Straits.



The sea surface temperature (SST) is modulated due to interplay of many factors. Among the most important ones are the short and longwave radiation, latent and sensible heat fluxes, freshwater input, convection, and mixing caused by wind and waves. The wind-wave mixing and convection produce a surface mixed layer. The surface mixed layer participates in the large-scale air-sea interaction, has substantial heat capacity, and is an important element of the weather and climate system. How the SST is connected to the temperature structure of the surface mixed layer is of interest for many practical applications, including the development of remote sensing techniques for monitoring the climate and global carbon cycles. The SST is determined by both oceanic and atmospheric processes. Here we discuss the particular situation of low wind speed and strong solar heating. which leads to an upper ocean warm layer (e.g., Stramma et al., 1986).

The SST can differ from the bulk due to a cool skin layer, which has been well studied and discussed (Katsaros, 1980). This cool layer can exist on top of a warm layer due to cooling by surface fluxes, which can be quite large. We do not discuss the cool skin contributions to variability in air-sea heat fluxes, since the phenomenon of interest in this article would only be slightly changed in magnitude by inclusion of the cool skin effects in our calculations. These calculations are meant to be representative rather than exact.

Lukas (1990) and Soloviev and Lukas (1997) found that diurnal warming and precipitation effects might increase the sensitivity of SST to atmospheric fluxes due to re-stratification of the surface mixed layer. A shallow stable layer that is not destroyed by negative buoyancy flux during the night may keep all cooling or heating confined near the sea surface and lead to large temperature modulations. It is also well known that in the Gulf of Mexico, the summer heating masks the existence of eddies and the value of the remotely sensed SST from the whole Gulf reaches a uniform high temperature, typically 30oC at the height of the summer season. Thus, this equalizing process occurs both on shorter diurnal to weekly and seasonal time scales. This contribution is built on more comprehensive articles by Katsaros and Soloviev (2004) and by Katsaros et al. ( 2005).

Several aspects of the upper ocean temperature structure are examined here by modeling the air-sea fluxes for conditions typical of the Florida Straits and Caribbean area. We also present data from the Florida Straits, which support the general conclusions of the calculations. The results show when one might expect this feedback phenomenon to smooth out temperature gradients in a boundary layer near the surface of the ocean and give misleading information about the presence of temperature fronts and eddies. They may well continue to exist just below this rather uniform, warm surface layer. Such knowledge is important for the fishing industry, for predicting hurricane intensity development (Shay et al., 2000) and many other applications. A wide range of combinations of meteorological parameters are possible over the world oceans, so the examples presented here are only to be thought of as indicative. We have not included any effects of oceanic advection or mixing due to swell, wave breaking, or internal waves and tides.


Equations for the heat, salinity, and momentum balance in the upper layer of the ocean are as follows (e.g., Katsaros and Buettner, 1969; Soloviev et al., 2001a):





where is the turbulent heat flux; QR is the solar radiative heat flux, a function of depth (z, positive downwards); and F is the turbulent salt flux, with S = salinity; and are the components of the turbulent shear stress, all in the upper ocean; r is the density of sea water; cp is the specific heat of sea water; is the Richardson number; and f is the Coriolis parameter. All fluxes of heat and momentum are taken positive downwards.

Km, Kt, and Ks are the turbulent exchange coefficients for momentum, heat, and salt determined from the Tropical Ocean Global Atmosphere-Coupled Ocean Atmosphere Response Experiment (TOGA-COARE) mixing parameterization (Soloviev et al., 2001b):



where k = 0.4 is the Von Karman constant, u* is the friction velocity in water, z the depth, Ri the gradient Richardson number, Rim = -0.20, Ris = -1.0, Ricr = 0.25, a = 16, am = 1.26, as = -28.86, cm = 8.38, and cs =98.96. Kmt is the thermocline value of the mixing coefficient determined according to Peters et al. (1988), and Ks = Kt.

The absorption of solar radiation with depth (QR) is parameterized by an exponential sum,

according to Paulson and Simpson (1981), modified by Soloviev and Schluessel (1996), where QR0 is the short wave solar radiation flux arriving at the interface, αi and βi the empirical coefficients. To calculate the albedo, A, for the short-wave radiation forcing, we used the Fortran program written by Peter A. Coppin (CSIRO Centre for Environmental Mechanics, Australia), employing the Payne (1972) model.

The surface boundary conditions are as follows:



where QT, QE, and QL are the sensible, latent, and net long wave radiative flux densities, positive downwards (note that the solar radiation does not enter the surface boundary condition because it is treated as the volume source of heat), and S0 is the surface salinity.


where tx0 and ty0 are the surface east- and northward components of the momentum flux.

Momentum, sensible, and latent fluxes are calculated from TOGA COARE 2.6 bulk-flux algorithm (Bradley et al., 2000). The net long wave radiation flux is parameterized as follows:

QL = εW Ea W s TW ,

where σ = 5.67 x 10_ 8W . m _ 2K _ 4

is the Stefan-Boltzmann constant, TW is the sea surface temperature, εW » 0.97 is the infrared emissivity of water, and Ea is the long wave irradiance from the sky (calculated for clear sky conditions using the algorithm of Girdyuk and Malevskiy-Malevich, 1973). We did not attempt to account for atmospheric diurnal variations or tropospheric thermal structure, since the exact value of QIR is not important for this study.

A grid with 40 evenly spaced points within the top 10 m of the ocean is used for the calculation. The initial temperature and salinity profiles are homogeneous. The initial velocity profile has a very small constant vertical gradient 2 10-4 s-1 (to avoid discontinuities on the Richardson number at the first step).


Figure 1 shows the calculated evolution of the temperature difference between the uppermost grid point at 0.25 m and 5 m depth, representing "bulk" SST, due to diurnal warming under low wind speed conditions, which is representative of conditions in the Gulf of Mexico or Florida Straits.

Five cases of atmospheric forcing are shown on these plots. From these figures it follows that the smaller the upward heat flux, the greater is the increase in SST. This suggests that the sea-to-air heat flux controls the interface temperature changes during weak wind and net heating. In this example, representing the Florida Straits region, the diurnal heat gain is substantially greater than the heat losses (by a factor of almost 3, when integrated over 24 hours), as illustrated by the panels for the insolation and the heat losses in Figures 1a and 1c. Figure 1d shows the change in wind stress for five starting scenarios. The interesting, dramatic change occurs when the air temperature is warmer than the sea surface temperature at the start of the heating cycle, but this stable stratification vanishes over time, and the wind stress is then more effectively transmitted to the sea surface. Our example in Figure 1 gives just some representative scenarios, typical for the Straits of Florida region and with wind speed constant at 2 m s-1.

Figure 2 illustrates the consequences for the SST for the same range in initial water temperatures and the same atmospheric fixed conditions as seen in Figure 1. The dependence of the change in SST, i.e., SSTmax, the maximum temperature reached during the day, SSTmax minus the initial temperature, SSTinit, as a function of the initial temperature is seen in this figure. It also includes graphs for the same initial water temperatures and atmospheric conditions, but with greater wind speeds. Even at a mean 10 m height wind speed of 6 m s-1 substantial warming occurs at the surface during maximum solar radiation. Many other combinations of the input variables will occur in a given month and particularly over the course of a year, but these calculated values are representative of the diurnal heating process, which leads to a weakening of horizontal temperature gradients. The ÄT is largest in the initially colder water (illustrated by the negative slope of the curve) by an amount that depends on the mean wind speed, all other factors remaining equal. This suggests that, if a front exists in the water, the surface-temperature gradient has been reduced after such a day masking the front from remote sensing.

Figure 1. Modeling the effect on SST due to atmospheric regulation. Parameters are representative of the calculated Straits of Florida cases (wind speed 2ms-1, specific humidity 16 g kg-1, and air temperature 28°C). (a) March of the insolation rate; (b) model calculation of the temperature difference between sea surface (T0) and 5 m depth (T5) over 22 hours; (c) associated values of the net heat loss term; and (d) wind stress change. Symbols in (c) represent the following: QT, sensible heat loss; QE, longwave; and QL, latent heat loss. Curves in the plot are labeled with the difference between initial temperature, Tw, and fixed air temperature, Ta.

Figure 2. Dependence of the change in SST on initial water temperature. Modeling parameters are representative of the Gulf of Mexico and Florida Straits (wind speed at 10 m height from 2 m s-1 to 6m s-1, specific humidity 16gkg-1, and air temperature 28°C).


Description of theField Site

As a part of the South Florida Ocean Measurement Center (SFOMC), Nova Southeastern University (NSU) and the University of South Florida (USF) deployed buoys in the Florida Straits. Measurements collected at the site relevant to this paper are from MicroCat SBE-37SM instruments measuring sea temperature and conductivity at 0.5, 5, 10, and 15 m on the central buoy in the array and meteorological variables collected by a Coastal Climate Weather Package on the same buoy. (The accuracy of the temperature sensor on SBE-37SM is better than 0.002oC.) The meteorological variables include wind speed and direction, surface pressure, air and near surface water temperature, relative humidity, solar radiation, longwave radiation, and the cumulative rain (measured every 5 min).

Data Used

Figure 3 gives a time series of the upper ocean temperature evolution at two depths and the nearby meteorological measurements for June 2-27, 2000, sampled every 30 minutes. We have selected this period for analysis because of the prolonged interval of low wind speeds in the middle of the month. The vertical temperature, salinity, and density profiles (not shown) at the central mooring location, averaged over the 25 days, indicate that some stratification was present. This ensured that the diurnal cycle of solar radiation rather than barotropic tidal motions was the main cause of the observed variability of sea surface temperature.

Result from Florida Straits

Figure 4a shows a plot of the ÄT in the water calculated as the difference between the 0.5 and 5 m measurements. The data are from the same period as in Figure 3. They represent data for all 25 days, night as well as daytime. There were, however, the following selection criteria: (1) Salinity difference in the upper 5 m less than 0.5 psu (practical salinity units); (2) Wind speed less than 4 m s-1. In order to select the cases with an expected diurnal thermocline an additional, selection criterion was used: (3) Only daytime hours from 1100 through 1700 local standard time (LST) and conditions of the water-air temperature difference, Tw - Ta < 1.5oC.

After applying selection criteria 1, 2 and 3, a total of 30 points remained (Figure 4b). The remaining points are less scattered than those in Figure 4a, since the cases of strong convective cooling (nighttime or large water-air temperature differences) have been filtered out.

Figure 3. Hydro-meteorological conditions during the June 2000 experiment on the southeast Florida shelf: (a) temperature difference between 0.5 and 5m depth; (b) wind speed Ua (bold line) and wind direction á; (c) water temperature Tw (bold line) and air temperature Ta; (d) rain rate R (bold line) and relative humidity r; (e) shortwave solar radiation QR (bold line) and downwelling longwave radiation QLD.

Figure 4. (a) Dependence of the temperature difference, ÄT, between 0.5 and 5 m on the temperature at 5m depth for 25 days in June 2000 under conditions of no rain and wind speeds <4 m s-1.


For calculation of the exact amount of increase in SST over a 24 hour period and the associated reduction in SST gradients in any particular situation, one must perform this calculation with the appropriate upper ocean mixing, the complete stratification effects and variations in the atmospheric fluxes as air flows from warm to cold, cold to warm water, or parallel to the SST gradients, etc. Any advection in the water should also be accounted for, although it is typically not very large over one day. Our case is not the most extreme as Gentemann et al. (2003) and Stuart-Menteth et al. (2003) in extensive analyses of satellite data reported even larger day to night SST differences.

This example is only indicative of the situations that one would encounter in summertime at low latitudes, whenever the wind speed is relatively weak and the diurnal heat fluxes result in net heating. The important finding of this work and that of Katsaros and Soloviev (2004) and Katsaros et al. (2005) is that the diurnal heating is not uniform, but will typically heat cold water more than nearby warmer water, thereby masking existing temperature gradients. The fishing industry may be particularly interested in this limitation of remotely sensed SST to locate front. Katsaros and Soloviev (2004) explained the effect of atmospheric regulation by dependence of air-sea fluxes on air-sea temperature difference, but dependence of the wind stress on air-sea temperature difference also contributes to the atmospheric regulation of SST. The wind stress feedback is important mainly when the SST is originally below the air temperature and increases during the diurnal warming to a temperature above the air temperature.


K. Katsaros acknowledges support by the National Oceanic and Atmospheric Administration, while she was at the Atlantic Oceanographic and Meteorological Laboratory in Miami at the beginning of this work and valuable assistance by Ms. Gail Derr. The data on the Florida shelf were collected as a part of the Nova Southeastern University and University of South Florida cooperative agreement under ONR Grant N00014-98-1-0861 (via a subcontract to Florida Atlantic University) and ONR Grant N00014-02-1-0950.


Bradley, E. F., Fairall, C. W., Hare, J. E., & Grachev, A. A.: 2000. 'An old and improved bulk algorithm for air-sea fluxes', in preprint, 14th Symposium on Boundary Layer and Turbulence, Aspen, CO, August 7-11, 2000, American Meteorological Society, 45 Beacon St., Boston, MA, pp. 294-296. [         [ Links ]1]

Gentemann, C., Donlon, C. J., Stuart-Menteth, A., & Wentz, F. J.: 2003, 'Diurnal signals in satellite sea surface temperature measurements', Geophysical Research. Letters, 30 (3), 1140-1143. [         [ Links ]2]

Girdyuk, G. B., & Malevskiy-Malevich, S. P.: 1973, 'Method for calculation of the net long-wave radiation at the ocean surface', GGO Proceedings, No. 277 (in Russian). [         [ Links ]3]

Katsaros, K. B., & Buettner, K. J.: 1969, Influence of rainfall on temperature and salinity of the ocean surface. Journal of Applied Meteorology, 8, 15-18. [         [ Links ]4]

Katsaros, K. B.: 1980, 'The aqueous thermal boundary layer', Boundary-Layer Meteorology., 18, 107-127. [         [ Links ]5]

Katsaros, K. B., & Soloviev, A. V.: 2004, 'Vanishing horizontal sea surface temperature gradients at low wind speeds', Boundary-Layer Meteorology, 112 (2), 381-396. [         [ Links ]6]

Katsaros, K. B., Soloviev, A. V., Weisberg, R. E., & Luther, M. E.: 2005, 'Reduced horizontal sea surface temperature gradients under conditions of clear sky and weak winds, submitted to Boundary Layer Meteorology. [         [ Links ]7]

Lukas, R.: 1990, 'The Role of salinity in the dynamics and thermodynamics of the western Pacific warm pool', in preprint, Proceedings, International TOGA Scientific Conference, WCRP-43, WMO/TD No. 379, pp. 73-81 (available from A. Soloviev). [         [ Links ]8]

Paulson, C. A., & Simpson, J. J.: 1981, 'The temperature difference across the cool skin of the ocean, Journal of Geophysical Research, 86, 11,044-11,054. [         [ Links ]9]

Payne, R. E.: 1972, 'Albedo of the sea surface', Journal of the Atmospheric Sciences, 29, 959-970. [         [ Links ]10]

Peters, H., Gregg, M. C., & Toole, J. M.: 1988, 'On the parameterization of equatorial turbulence', Journal of Geophysical Research, 93, 1199-1218. [         [ Links ]11]

Shay, L. K., Goni, G. J., & Black, P. G.: 2000, 'Effects of a warm oceanic feature on Hurricane Opal', Monthly Weather Review, 128, 1366-1383. [         [ Links ]12]

Soloviev, A. V., & Schluessel, P.: 1996, 'Evolution of cool skin and direct air-sea gas transfer coefficient during daytime', Boundary-Layer Meteorology, 77, 45-68. [         [ Links ]13]

Soloviev, A. V., & Lukas, R.: 1997, 'Large diurnal warming events in the near-surface layer of the western equatorial Pacific warm pool', Deep-Sea Research, 44, 1055-1076. [         [ Links ]14]

Soloviev, A. V., Edson, J., McGillis, W., Schluessel, P., & Wanninkhof, R.: 2001a 'Fine thermohaline structure and gas exchange in the near-surface layer of the ocean during GasEx-98', in Gas Transfer at Water Surfaces, M. A. Donelan, W. M. Drennan, E. S. Saltzman, and R. Wanninkhof (eds.), AGU Geophysical Union Monograph, Vol. 127, pp. 181-185. [         [ Links ]15]

Soloviev, A. V., Lukas, R., & Hacker, P.: 2001b, 'An approach to parameterization of the oceanic turbulent boundary layer in the western Pacific warm pool, Journal of Geophysical Research, 106, 4421-4435. [         [ Links ]16]

Soloviev, A. V., Walker, R., Weisberg, R., & Luther, M.: 2003, 'Coastal observatory investigates energetic current oscillations on the southeast Florida Shelf', EOS, Transactions, AGU, 84 (42), 441-450. [         [ Links ]17]

Stramma, L., Cornillon, P., Weller, R. W., Price, J. F., & Briscoe, M. G.: 1986, 'Large diurnal sea surface temperature variability: Satellite and in-situ measurements', Journal of Physical Oceanography, 16, 827-837. [         [ Links ]18]

Stuart-Menteth, A. C., Robinson, I. S., & Challenor, P. G.: 2003, 'A global study of diurnal warming using satellite-derived sea surface temperature', Journal of Geophysical Research, 108 (C5), 3155-3170. [         [ Links ]19]


Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons