Influence of a New Turbulence Regime on the Global Air–Sea Heat Fluxes

Erik Sahlée Rosenstiel School of Marine and Atmospheric Science, University of Miami, Miami, Florida

Search for other papers by Erik Sahlée in
Current site
Google Scholar
PubMed
Close
,
Ann-Sofi Smedman Department of Earth Sciences, Meteorology, Uppsala University, Uppsala, Sweden

Search for other papers by Ann-Sofi Smedman in
Current site
Google Scholar
PubMed
Close
,
Anna Rutgersson Department of Earth Sciences, Meteorology, Uppsala University, Uppsala, Sweden

Search for other papers by Anna Rutgersson in
Current site
Google Scholar
PubMed
Close
, and
Ulf Högström Department of Earth Sciences, Meteorology, Uppsala University, Uppsala, Sweden

Search for other papers by Ulf Högström in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Recent research has found that boundary layer turbulence changes its organization as the stratification approaches neutral from the unstable side. When the thermal forcing weakens in combination with wind speed above approximately 10 m s−1, detached eddies are formed in the upper part of the surface layer. These eddies effectively transport drier and colder air from aloft to the surface as they move downward, thereby enhancing the surface fluxes of sensible and latent heat. This effect has been observed over both land and sea; that is, it is not dependent on the nature of the underlying surface. Here the authors perform a sensitivity study of how this reorganization of the turbulence structure influences the global air–sea heat fluxes. Using modified bulk formulations incorporating this effect, the magnitude of the enhancement in a climatic sense was estimated by the use of 40-yr ECMWF Re-Analysis (ERA-40) data in the bulk formulas. It is found that for the 1979–2001 period, the global increase of the latent and sensible heat fluxes over the ice-free oceans is 3.6 and 1.2 W m−2, respectively. These numbers suggest that this effect is of some significance. The results also indicate that the regional and seasonal variability may be large. The largest annual increases are found over the southern oceans between 30° and 60°S where the sensible heat flux increases by 2.3 W m−2 and the latent heat flux by 6.5 W m−2. Ocean areas close to the equator experience almost no increase, whereas the latent heat flux from the Arabian Sea during the monsoon period is enhanced by 11.5 W m−2.

Corresponding author address: Erik Sahlée, Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149-1098. Email: esahlee@rsmas.miami.edu

Abstract

Recent research has found that boundary layer turbulence changes its organization as the stratification approaches neutral from the unstable side. When the thermal forcing weakens in combination with wind speed above approximately 10 m s−1, detached eddies are formed in the upper part of the surface layer. These eddies effectively transport drier and colder air from aloft to the surface as they move downward, thereby enhancing the surface fluxes of sensible and latent heat. This effect has been observed over both land and sea; that is, it is not dependent on the nature of the underlying surface. Here the authors perform a sensitivity study of how this reorganization of the turbulence structure influences the global air–sea heat fluxes. Using modified bulk formulations incorporating this effect, the magnitude of the enhancement in a climatic sense was estimated by the use of 40-yr ECMWF Re-Analysis (ERA-40) data in the bulk formulas. It is found that for the 1979–2001 period, the global increase of the latent and sensible heat fluxes over the ice-free oceans is 3.6 and 1.2 W m−2, respectively. These numbers suggest that this effect is of some significance. The results also indicate that the regional and seasonal variability may be large. The largest annual increases are found over the southern oceans between 30° and 60°S where the sensible heat flux increases by 2.3 W m−2 and the latent heat flux by 6.5 W m−2. Ocean areas close to the equator experience almost no increase, whereas the latent heat flux from the Arabian Sea during the monsoon period is enhanced by 11.5 W m−2.

Corresponding author address: Erik Sahlée, Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149-1098. Email: esahlee@rsmas.miami.edu

1. Introduction

The single most important source of atmospheric water vapor is evaporation from the global oceans (Peixoto and Oort 1992). Thus, it is easy to realize the importance of understanding the processes governing the air–sea flux of water vapor for predicting and understanding climate and climate change. Compared to the latent heat flux, the air–sea flux of sensible heat per area unit is generally much smaller over the ocean. However, since the global oceans cover about 70% of the earth’s area, correct understanding of the air–sea sensible heat flux is needed as well for accurate predictions of the future climate.

Models for weather forecasting and climate predictions commonly use a bulk approach when calculating the fluxes of sensible and latent heat to and from the sea surface. This is a way to parameterize the flux by the use of bulk variables and exchange coefficients. The success of this method primarily relies on a correct description of the exchange coefficients.

Numerous field experiments have been conducted to study the air–sea heat fluxes with the goal to construct accurate bulk formulations. However, although great achievements have been made, not the least with the development of the Coupled Ocean–Atmosphere Response Experiment (COARE) 3.0 algorithm (Fairall et al. 2003), there is still a need for improvements. For instance Zhang et al. (2006) have shown improved model predictions of storm intensity when including effects of sea spray in the calculations. Swells, that is, waves moving faster than the wind, have been shown to influence the boundary layer and the wind gradients (Smedman et al. 1999; Rutgersson et al. 2001a; Smedman et al. 2003; Sullivan et al. 2008) and thus indirectly affect the exchange coefficients (Sahlée et al. 2008; Smedman et al. 2007b).

Very recently, it has been found that the boundary layer turbulence changes its structure during near neutral, slightly unstable stratification. The turbulence regime that ensues during such conditions has been termed the unstable very close to neutral (UVCN) regime. The mechanism and theory behind this phenomenon have been described in Smedman et al. (2007a, b) and Sahlée et al. (2008) along with the effects on the sensible and latent heat fluxes. In short, the UVCN regime can be described as a top-down controlled boundary layer where detached eddies at the upper part of the surface layer move toward the surface, enhancing the exchange of sensible and latent heat by bringing down drier and colder air. New parameterizations incorporating the UVCN effects were presented in Rutgersson et al. (2007). These parameterizations were tested in two numerical models, one 3D climate model covering northern Europe and one ocean model for the Baltic Sea. It was found that the latent heat flux during episodes of several days could be increased by as much as 100 W m−2 for regions where the UVCN regime was active. The effect on the sensible heat flux was significantly smaller. The long-term effect on the latent heat flux over the Baltic Sea was also found to be influenced by the UVCN regime by several W m−2. This suggests that the UVCN regime is a very important process that needs to be implemented in the current models when predicting the future weather and climate.

This work is a sensitivity study where the global impact of the UVCN regime on the air–sea heat fluxes in a climatic sense is investigated. This is done by using bulk variables from the 40-yr European Centre for Medium-Range Forecasts (ECMWF) Re-Analysis (ERA-40) database in the parameterizations proposed by Rutgersson et al. (2007). The bulk parameterizations and the UVCN regime are described in section 2. The experimental setup is described in section 3. The results are presented in section 5 and discussion and conclusions in section 6.

2. Theory

a. Bulk formulations

The vertical turbulent fluxes of sensible and latent heat can be described using bulk formulas:
i1520-0442-21-22-5925-e1
i1520-0442-21-22-5925-e2
where
i1520-0442-21-22-5925-eq1

Subscripts s refer to the sea surface, and subscripts 10 refer to the 10-m height.

Applying the Monin–Obukhov similarity theory by expressing the wind and scalar gradients through nondimensional profile functions, the exchange coefficients can be expressed as
i1520-0442-21-22-5925-e3
i1520-0442-21-22-5925-e4
where k is the von Kármán constant (0.4); z is the measurement height (here 10 m); z0, z0q, and z0t are the roughness lengths for momentum, humidity, and temperature, respectively (m). Here ψm, ψq, and ψh are the integrated nondimensional profile functions:
i1520-0442-21-22-5925-e5
where x = q, h, and m, and ζ = z/L. The nondimensional profile functions are expressed as
i1520-0442-21-22-5925-e6
i1520-0442-21-22-5925-e7
i1520-0442-21-22-5925-e8
where u* is the friction velocity (m s−1), q* = (/u*) and θ* = (/u*) are the scaling parameters for humidity and temperature. Here z/L is a stability parameter where L is the Obukhov length (m):
i1520-0442-21-22-5925-e9
where θ0 is the potential temperature of the surface layer (K), g is the acceleration of gravity (m s−2), and is the kinematic flux of virtual potential temperature.
Most investigations remove the effects due to the atmospheric stratification from CE and CH to enable comparison with other experiments. These are called the neutral exchange coefficients, CEN and CHN, and have the form
i1520-0442-21-22-5925-e10
i1520-0442-21-22-5925-e11

b. The UVCN regime

The unstable boundary layer in moderate winds is characterized by the longitudinal roll type of eddy structure; see, for example, Atlas et al. (1986) and Mason and Sykes (1982) for observations and model studies of these structures. However, in two recent studies (Smedman et al. 2007b; Sahlée et al. 2008) observations from the Östergarnsholm field station in the Baltic Sea show that the roll structures break down as the thermal forcing weakens, that is, the boundary layer approaches neutral stratification. By analysis of temperature spectra, cospectra of vertical velocity–temperature and vertical velocity–humidity, it was found that as the instability decreases a second peak at a higher frequency develops. This peak represents an increase in the energy of the turbulence at scales an order of magnitude smaller. Cases exist where the smaller-scale turbulence dominates; that is, the spectral peak shifts completely. This new turbulence structure was termed the unstable very close to neutral regime, the UVCN regime. As shown in Smedman et al. (2007a), these results were observed at both a land site and a marine site; that is, the formation of these structures is not dependent on the nature of the underlying surface.

Quadrant analysis (Lu and Willmarth 1973; Willmarth and Lu 1974; Raupach 1981) showed that during UVCN conditions a large part of the upward latent and sensible heat fluxes originated from cold and dry air brought down to the surface from layers aloft. This is quite different from the moderately unstable boundary layer in which the quadrant analysis showed that the heat fluxes were dominated by warm–humid air moving upward, as intuitively expected.

A theoretical framework explaining the experimental results can be found in the studies of the very high Reynolds number boundary layers by Hunt and Morrison (2000) and Hunt and Carlotti (2001). These theories have been confirmed by measurements in the neutral boundary layer by Högström et al. (2002). In the neutral boundary layer detached eddies are created in the shear at the upper part of the surface layer. As these eddies move downward they become distorted because of blocking by the surface and stretching in the presence of the strong shear. In other words, the eddies are compressed and elongated. The horizontal length scale is of the boundary layer height but the vertical length scale only approximately 1/30 of the boundary layer height. The strong shear near the surface also initiates a replication of the eddies, a feature often called cat paws or honami waves. A series of photographs in Hunt and Morrison (2000) illustrates the effects on a sea surface due to these structures (their Fig. 6). A sketch of the above described mechanism is shown in Högström et al. (2002; their Fig. 1).

The theory of the detached eddies was linked with the horizontal roll structures by Smedman et al. (2007a). It was suggested that they are two parts, one steady and one unsteady, of the same dynamical system. As the thermal forcing decreases, the system bifurcates into two branches, the horizontal rolls being the steady branch and the detached eddies the unsteady. The likelihood of formation of detached eddies increases with decreasing buoyancy force. The observed high-frequency peak of the spectra was interpreted as a signature of the detached eddies, thus the double-peaked (camel shaped) spectra and cospectra could be explained as a transition regime with both the steady and unsteady branch present.

The UVCN regime was observed to ensue for L values <−150 m. Expressed in bulk variables the limit is probably close to conditions with U > 10 m s−1 and an air–sea temperature difference ΔT < 2 K. These are approximate limits, and in Smedman et al. (2007a) it is suggested that a correct threshold should instead be expressed in terms of h/L where h is the height of the boundary layer. However, in the absence of continuous measurements of the boundary layer height, no threshold value has been presented.

An important effect of the UVCN regime is the enhancement of the heat fluxes. In Smedman et al. (2007b) CHN was found to increase with decreasing ΔT for wind speeds above 10 m s−1. A similar result was obtained for CEN in Sahlée et al. (2008).

A possible explanation for this observed increase of the heat transport efficiency is sea spray, which, for example, Andreas and DeCosmo (2002) has shown is able to influence both the humidity and sensible heat flux. However, Sahlée et al. (2008) investigated the influence of the sea spray mediated fluxes on CEN using the model of Andreas (2004), which incorporates sea spray in the flux calculations. It was found that sea spray only could explain a 10% increase of CEN whereas the observed increase was 56%. Instead, it is hypothesized that the reason for the increase is linked to the detached eddies (Smedman et al. 2007b; Sahlée et al. 2008). As the eddies move downward they bring down drier and colder air as shown by the quadrant analysis, thus enhancing the surface fluxes. This is also in agreement with the results by Maitani and Ohtaki (1989) who observed that downdrafts of dry air were a more efficient mechanism for the upward humidity flux compared to updrafts of humid air.

These new results are based solely on measurements from the Östergarnsholm tower in the Baltic Sea. However, it can be assumed that these measurements indeed represent open-ocean conditions. This issue has been thoroughly investigated in Högström et al. (2008). During the fall of 2003, a field experiment was conducted in the waters around Östergarnsholm. Two wave rider buoys were deployed (run and maintained by the Finnish Institute of Marine Research) as well as an Air–Sea Interaction Spar (ASIS) buoy [run and maintained by the Rosenstiel School of Marine and Atmospheric Science (RSMAS) University of Miami]. The ASIS buoy was deployed in 36-m-deep water ∼4 km southeast of the tower and instrumented with both slow response sensors for wind and temperature measurements and turbulence instrumentation. Comparison between the measurements at the tower and at the ASIS buoy showed very good correlation, as did the comparison between the different wave field measurements. Thus, it was concluded that measurements made at the tower certainly represent the upwind open-ocean conditions and that the results from the Östergarnsholm measurements are valid also for the global oceans.

c. Parameterization of the UVCN regime

To investigate the importance of including the UVCN effect in the global climate models, the suggested parameterization of Rutgersson et al. (2007) will be employed. This parameterization is based on the results of Smedman et al. (2007b) and Sahlée et al. (2008). For unstable, non-UVCN conditions, that is, 0 > L > −150 m, they suggest the following parameterization:
i1520-0442-21-22-5925-e12
i1520-0442-21-22-5925-e13
As shown in Smedman et al. (2007b) and Sahlée et al. (2008) these values agree well with those obtained from the COARE 3.0 algorithm.
For UVCN conditions, defined as L < −150 m and U10 > 9 m s−1 in Rutgersson et al. (2007), the exchange coefficients were found to be functions of both wind speed and air–sea temperature difference. They suggest the following expressions:
i1520-0442-21-22-5925-e14
i1520-0442-21-22-5925-e15
The experimental data only allowed the exchange coefficients to be evaluated for wind speeds up to 14 m s−1. For higher wind speeds, we make the conservative assumption of no further increase of CEN and CHN with wind speed; that is, U10 is set to 14 m s−1 in Eqs. (12)(15).

Stability effects on CE and CH can be calculated by solving for the roughness lengths in Eqs. (10) and (11), which then are put into Eqs. (3) and (4) together with the integrated stability functions. These functions may lead to unrealistic values at very stable and very unstable situations because of lack of measurements at these stabilities when the relations where established. Limits have been set at z/L = −5 and z/L = 0.5. The parameterization could also have been done using the roughness lengths directly as suggested by Rutgersson et al. (2007). However Rutgersson et al. (2007) used a 3D climate model where parameterizations of the roughness lengths are more suitable. In the present context the choice is a matter of taste. Parameterizing the exchange coefficients are more closely related to the original work in Smedman et al. (2007a, b) and Sahlée et al. (2008), which is why this approach was chosen here.

d. ERA-40

ERA-40 is a 45-yr-long reanalysis of meteorological observations, from September 1957 to August 2002 (Uppala et al. 2005). Essentially, a reanalysis is a description of the state of the atmosphere created using all available observations. These observations are used as input to a numerical model together with the latest forecast as a first guess. The numerical model then calculates a complete description of the state of the atmosphere, a procedure that can be done as far back in time as there are enough observations available. Thus, a complete description of the atmosphere can be created for long time period. Since the same forecast model is used for the entire time period, the reanalysis is unaffected by modifications in the model. However, the data will be influenced by changes in the observation systems, for example, changes in time and spatial data coverage and changes of instruments.

The model used in ERA-40, the integrated forecast system (IFS), is a version of the ECMWF model operational in June 2001. It has a horizontal resolution of approximately 125 km × 125 km (strictly speaking, a T159 spectral resolution), with 60 vertical levels, the lowest being very close to 10-m height. This is a coarser resolution than the operational model at the time (which had a T511 resolution), but the resolution needed to be lowered in order to constrain the computational time. The choice of data assimilation system was also restricted by the computational cost, which is why a three-dimensional variational (3DVAR) data assimilation system was used instead of the available 4DVAR system. (A description of the IFS can be found online at http://www.ecmwf.int/research/ifsdocs/index.html.)

In the IFS the atmospheric model is coupled with an ocean wave analysis model (WAM; Komen et al. 1994). Thus, the air–sea system is allowed to interact; for example, the wind creates ocean waves, which in turn change the surface roughness influencing the fluxes of moisture, sensible heat, and momentum.

3. Experimental setup

The global oceanic surface fluxes of latent and sensible heat were calculated from Eqs. (1) and (2) using bulk variables from ERA-40. Sea ice coverage is stored in ERA-40, which made it possible to exclude cases when the ocean surface was covered with ice from the calculations. Analyses and 6-h forecast fields every sixth hour for the time period 1979–2001 were used to calculate the long-term averages. The year 1979 was chosen as a start year because of the improvements in the observation system in the beginning of this year. Among those were increased amount of aircraft data, improved instrumentation on new satellites, the possibility of acquiring good quality wind information from geostationary satellites and the deployment of drifting ocean buoys in the Southern Hemisphere. In addition, prior to 1979 there were problems with some of the data handling and missing or limited data for certain regions which lower the quality of the reanalysis in the early period (Uppala et al. 2005).

The parameterizations of CHN and CEN described in Eqs. (12)(15) were used in Eqs. (1) and (2) to perform a sensitivity analysis of the climatic importance of the UVCN regime on the global air–sea heat fluxes. Hence, two calculations of the fluxes were made: one using only Eqs. (12) and (13), that is, no UVCN effects on the heat fluxes (reference run), and one using also Eqs. (14) and (15). The UVCN regime was assumed to be present for wind speeds >10 m s−1 in combination with ΔT < 2 K. During stable stratification, it is assumed that CEN = CHN = 0.77 × 10−3 (Rutgersson et al. 2001b).

CE and CH were calculated from the parameterized values of CEN and CHN by inserting the values of z0q and z0t in [solved from Eqs. (10) and (11)] in Eqs. (3) and (4). The ψh were calculated from the formulation of Högström (1988). Thus, by assuming ϕq = ϕh (Edson et al. 2004),
i1520-0442-21-22-5925-e16
The integrated ϕm function was calculated using the ϕm function presented in Högström (1996):
i1520-0442-21-22-5925-e17
The surface roughness length z0 is available directly from the ERA-40 data. However it was found that these data were erroneous. The ERA-40 surface roughness length was stored using 16-bit coding, but this proved to be insufficient to cover the large range of z0 values (P. Kållberg 2005, personal communication). The z0 fields over the sea surface were restored using the same equation as in the IFS (i.e., Beljaars 1995):
i1520-0442-21-22-5925-e18
where αM is a constant set to 0.11, ν is the kinematic viscosity (m2 s−1), and αCh is the Charnock coefficient calculated from the wave model and stored in ERA-40. The friction velocity in Eq. (18) was calculated according to
i1520-0442-21-22-5925-e19
where τ is the surface stress (obtained from ERA-40).

All fields were interpolated to a 1.125° × 1.125° latitude–longitude grid when retrieved from the ERA-40 database.

4. Results

Figure 1a shows the annual average frequency of occurrence of UVCN conditions, calculated for the period 1979–2001. The most prominent maximum is found in the southern oceans where some areas experience UVCN condition during 25% of the time. In the Northern Hemisphere (NH) oceans, UVCN conditions are less frequent, existing between 10% and 15% of the time. In Fig. 1b the frequency of occurrence of the UVCN regime is shown for different seasons. A clear seasonal cycle is seen for the NH oceans, where UVCN conditions are most frequent during the winter (December–February).

Figures 2a and 2b show the relative frequency of occurrence for wind speeds above 10 m s−1 as an annual average and averaged over the different seasons, respectively. The relative frequency of occurrence for ΔT < 2 K is shown in Figs. 3a,b. These four figures can help to explain the results presented in Figs. 1a,b. Over the tropics it is seen that although small air–sea temperature differences are common, the wind speed rarely exceeds 10 m s−1, thus UVCN conditions are very uncommon in these regions. Over the southern oceans, on the other hand, the wind speed often exceeds 10 m s−1; however, a ΔT < 2 K is not as common, which makes ΔT the controlling parameter at these latitudes.

Over NH oceans, both wind speed and ΔT limits the UVCN regime depending on season. During winter, the wind speeds frequently exceed 10 m s−1 but ΔT < 2 K is not as frequent, especially in the west and northwest parts of the oceans. This is due to cold air outbreaks from the continents over the relatively warm oceans which create a large air–sea temperature difference, preventing formation of the UVCN regime. In the summer on the other hand, the wind speeds are much lower, exceeding 10 m s−1 only during 20%–30% in certain regions. The air–sea temperature difference is more frequently <2 K, that is, during summer over NH oceans wind speed limits the onset of the UVCN regime.

Distinct maxima of UVCN conditions are also seen in the Indian Ocean, Arabian Sea, and Bay of Bengal during June, July, and August in Fig. 1b. The frequency of occurrence of the UVCN regime has a peak of 50% of the time in the Arabian Sea. This is most likely due to the strong wind speeds associated with the monsoon circulation, present in the region during these months.

The global distribution of the mean annual increase of the latent heat flux due to the UVCN regime is shown in Fig. 4a (positive numbers indicate upward fluxes). In the North Atlantic and North Pacific, the UVCN effect increases the latent heat flux in a range from 2.5 W m−2 up to almost 10 W m−2. The maximum increase over the southern oceans is of similar magnitude; however, larger areas experience an increase compared to the North Atlantic and North Pacific. When the mean seasonal fluxes are studied, as shown in Fig. 4b, a few local areas display very strong increases. Over the Arabian Sea during June–August, the increase of the latent heat flux has a maximum of 45 W m−2. Just north of Madagascar the increase is even stronger, 67 W m−2. Although these areas are only represented by a few grid points, the increase can be explained by the strong monsoon winds.

For the months December–February two small areas show a very strong increase. North of Colombia in the Caribbean Sea and in the South China Sea the UVCN regime enhances the latent heat flux by almost 40 W m−2. The southern part of the Caribbean Sea displays almost the same increase during June–August. These increases seem to be due to local wind maxima present in the regions during these months.

Since the UVCN regime is most frequent over the North Atlantic and North Pacific during Northern Hemisphere winter, December–February are the months when the largest increase of the latent heat flux is observed. During NH summer, the increase is quite weak in these regions, exceeding 5 W m−2 only over a few small areas.

In Fig. 5a, the mean relative increase of the latent heat flux is shown. In the North Atlantic, the increase ranges from 5% up to 15%, where the maxima are located over the North Sea and a small area east of Newfoundland where the relative increase is slightly larger than 20%. The relative increase is of similar magnitude over the North Pacific. Over the southern oceans the relative increase of the latent heat flux is greater compared to similar latitudes in the Northern Hemisphere. Large areas display an increase well over 15% with local maxima exceeding 45%. The seasonal variation of the relative increase of the latent heat flux is shown in Fig. 5b. This figure displays almost the same pattern as the absolute increase shown in Fig. 4b with some minor exceptions. Some areas show a strong relative increase of the flux although the absolute increase is small. This result is obviously due to small mean fluxes.

The increase of the mean annual sensible heat flux due to the UVCN regime, shown in Fig. 6a, is observed to be much smaller than the increase of the latent heat flux. The maximum is found over the Southern Ocean and represents an increase of approximately 6 W m−2. The seasonal variation of the increase, shown in Fig. 6b, follows the same pattern as the seasonal variation of the increase of the latent heat flux (Fig. 5b) although the magnitude of the increase is considerably smaller.

The mean annual global increase of the sensible heat flux due to the UVCN regime is only 0.8 W m−2 or a relative increase of 8%. The increase of the latent heat flux is 3 times larger, 2.4 W m−2; however, the relative increase is smaller compared to the sensible heat flux, 4.2%. If the fluxes are averaged over the global ice-free oceans instead of averaging over the entire globe, the increase of the sensible heat flux is 1.2 W m−2 and the increase of the latent heat flux is 3.6 W m−2. However, as indicated by Figs. 4 –6 the increase varies strongly between regions and seasons.

In Table 1, four different regions are studied in detail: North Atlantic (between 30° and 60°N), southern oceans (30°–60°S), the North Pacific Ocean (30°–60°N), and the Arabian Sea. The table is divided into three parts showing mean values (1979–2001) of the surface fluxes calculated with and without the UVCN effect for three different time periods (a) the entire year, (b) December–February, and (c) June–August. Examining, for instance, the Southern Ocean, there is some seasonal variation of the increase. During December–February the increase of the latent heat flux is 4.7 W m−2 compared to the maximum during June–August, 8.0 W m−2.

Over the Arabian Sea we notice that the mean yearly increase of the latent heat flux due to the UVCN effect is 3.2 W m−2. The seasonal variation for this region is larger compared to that of the southern oceans. Minimum in NH winter with an increase of 0.3 W m−2 and a maximum during NH summer during which the mean increase is 11.5 W m−2. As discussed earlier, this is a result of the strong monsoon winds.

Generally the sensible heat flux is small over the global oceans. This explains why the relative increase is as large as 28% over the Arabian Sea during June–August despite the absolute increase is 10 times smaller compared to the increase of the latent heat flux. This is also the case for the large relative increase during December–February over the Southern Ocean. Large relative increases are also present in the some parts of the ocean close to Antarctica (not shown). However, during the winter months outbreaks of cold air over the relatively warm oceans significantly enhances the sensible heat flux. This feature is more prominent in the NH compared to the SH (Table 1) because of the larger land areas, which act as source regions for cold air masses.

Figure 7a shows zonal averages of the air–sea latent heat flux, calculated as annual mean for the 1979–2001 period. The flux has been calculated both with and without inclusion of the UVCN effect, the lower figure showing the difference between the two. In agreement with the previous results we notice that the largest increase of the latent heat flux when including the UVCN regime in the calculation is observed at the southern oceans with a maximum of 8 W m−2 close to 50°S. Figure 7b shows the seasonal variation of the zonally averaged latent heat flux. As in Fig. 7a, the figure also shows the increase of the flux when the UVCN effect is included in the calculations. Because of relatively low wind speeds at higher latitudes during summers, the UVCN increase of the flux is relatively small. This is particular seen in the NH above 30°N where the UVCN effect is reduced to just 1–2 W m−2 during NH summer but well above 5 W m−2 during NH winter when stronger wind speeds prevail. Such a strong variation is not seen for the Southern Hemisphere, although the interannual variability of the wind speed is equally large. As discussed earlier, this is due to the strong mean wind speed at the southern oceans. Thus, although the wind speed is slightly reduced during SH summer, wind speeds above 10 m s−1 are still frequent.

As expected, the UVCN effect for the regions close to the equator is very close to zero because of the weak mean wind speeds.

5. Discussion and conclusions

A sensitivity study of the UVCN turbulence regime on the long-term heat fluxes was performed using output from the ERA-40 reanalysis. By inclusion of UVCN effects in bulk formulations of the air–sea heat fluxes it was found that the global air–sea latent heat flux was increased by 2.4 W m−2 or 4.2% and the global air–sea sensible heat flux was increased by 0.8 W m−2, or 8%. The air–sea temperature difference in the near neutral boundary layer is relatively small, which is why the UVCN increase of the sensible heat flux is 3 times smaller compared to the increase of the latent heat flux. In a global context, the UVCN enhancements of the heat fluxes are of similar magnitude as the increase of the radiative forcing due to the anthropogenic emissions of greenhouse gases, which, according to the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4), amounts to 2.63 ± 0.26 W m−2 (Forster et al. 2007). However, the increase of the radiative forcing is evenly distributed over the globe, whereas the UVCN effect shows large regional and seasonal variation. The enhancement of the latent heat flux for some areas exceeds 10 W m−2, for instance during June–August over the Arabian Sea.

The exact increase of the surface fluxes due to the UVCN regime is, of course, subject to uncertainty. Uncertainty is introduced through the use of reanalysis bulk variables. These bulk variables, such as the 10-m humidity and the 10-m temperature, over the open oceans are to some extent influenced by what we believe are erroneous parameterizations of the air–sea heat fluxes. Thus, feedbacks connected to the flux parameterization are likely to be influenced when the bulk formulations are modified. For instance, an increased latent heat flux would moisten the boundary layer, which in time would reduce humidity gradient and hence reduce the flux. Increased surface fluxes would also lead to energy loss from the sea surface, which in turn would reduce the SST. As a consequence this would most likely influence the convection in the ocean and thus affect the ocean mixing and the depth of the thermocline. This investigation has not incorporated any feedbacks, which probably would have an impact on the results.

Feedback effects due to the UVCN regime were investigated for the Baltic Sea in Rutgersson et al. (2007). Results from a simulation with an ocean model [the Program for Boundary Layers in the Environment model for the Baltic Sea (PROBE-Baltic); see Omstedt and Nyberg 1996 and Omstedt and Axell 2003 for a description of the model] indeed showed cooling of the SST, which reduced heat fluxes and created a shallower and colder mixed layer. Including the feedbacks effects in the simulation reduced the UVCN effects slightly. Although, these feedback effects were quite small in the Baltic Sea they most likely are of a larger magnitude in oceans where the UVCN regime is more frequent.

Using a regional climate model, the Rossby Center Atmospheric Model (RCA; Jones et al. 2004), Rutgersson et al. (2007) showed that feedback effects connected to the UVCN regime also had impacts on the atmosphere over the Baltic Sea. In mean the surface layer got slightly warmer (0.1 K at 2-m height) and slightly more humid (0.1 g kg−1 at 2-m height).

ERA-40 output may be subject to some bias, which also would add to the uncertainty of the results. Such bias would most likely be present over areas where observations are sparse, for example, the Southern Ocean. However, after the introduction of satellite measurements the reanalysis output seems to be of good quality. Bromwich and Fogt (2004) compared the 2-m temperature and mean sea level pressure between observation and reanalyses over the mid–high latitudes on the Southern Hemisphere. The conclusions were made that the bias in these parameters were small and reanalyses output showed good correlation with measurements. Caires et al. (2004) evaluated the 10-m wind speed from different reanalysis including ERA-40 and came to similar results. Of course, the absolute values of the fluxes, both in this study and in the reanalysis, are very uncertain. However, we do not draw any conclusions from these; our main interest was to see if the UVCN regime may have any global significance, which it seems to have.

Improved description of the air–sea heat fluxes in the reanalysis dataset has been suggested also by others. Bengtsson et al. (2007) acknowledge the poor quality of the ERA-40 fluxes, which, together with other problems such as the cloud and precipitation parameterizations, can lead to imbalance between the global evaporation and precipitation.

Attempts are currently being made to create a more accurate database for the global air–sea heat fluxes (Yu and Weller 2007). The improvement is accomplished by a better description of the bulk variables through an objective analysis technique, which better synthesize observations and model outputs. The COARE 3.0 algorithm is then used to calculate the fluxes. This is a step forward compared to the parameterization used in the ERA-40 reanalysis. However, since the UVCN effect is not included in the COARE 3.0 algorithm, it underestimates the heat fluxes during conditions with high wind speeds and small air–sea temperature differences, as shown in Smedman et al. (2007b) and Sahlée et al. (2008). This study illustrates that this might lead to errors for some regions on climatic time scales.

The results presented here show that the enhancements of the heat fluxes due to the formation of the UVCN regime are of a relatively large magnitude. However, the exact numbers are uncertain; they are probably overestimated. As a sensitivity study the conclusion can be made that the UVCN regime is potentially globally significant and may have an influence on the predictions of the future climate. We want to stress that this work is not an attempt on making a new climatology for surface fluxes. Rather the results presented shows that it is probably worthwhile pursuing further studies investigating the specific impacts on the climate and to study possible feedbacks due to the UVCN regime, which can be done using a global climate model. New field experiments might reveal more details regarding the mechanisms controlling the UVCN regime. This may resolve some of the uncertainty concerning the exact life cycle of an UVCN event. For instance, it is unclear what importance the boundary layer height has on the onset of these events, as discussed in section 2b.

The current parameterization is based on tower measurements at the Östergarnsholm Island in the Baltic Sea. An extensive field experiment has shown that this site indeed represents maritime conditions, especially during strong winds and small air–sea temperature differences (Högström et al. 2008). Thus, conclusions drawn from the Östergarnsholm measurements are expected to be valid also for the global oceans.

The heat flux parameterizations of Rutgersson et al. (2007) were based on measurements during which the wind speed did not exceed 14 m s−1. Thus, to increase the accuracy of the bulk formulations more measurements of air–sea fluxes during UVCN conditions are needed, especially during high wind speeds. Preferably, measurements in other oceans should also be made. The potentially large effects of the UVCN regime also support revisiting old field experiments for new analyses of the data, that is, performing analyses in a similar way as in Smedman et al. (2007b) and Sahlée et al. (2008).

Acknowledgments

We thank Per Kållberg for the assistance with ERA-40 and also ECMWF for making the ERA-40 data available. We are grateful to Hans Bergström for assisting with the field measurements and data analyses. This work was supported by the Swedish Research Council, Grant 621-2005-4293.

REFERENCES

  • Andreas, E. L., 2004: A bulk air–sea flux algorithm for high-wind, spray conditions, version 2.0. Preprints, 13th Conf. on Interactions of the Sea and Atmosphere, Portland, ME, Amer. Meteor. Soc., 8.2A.

  • Andreas, E. L., and J. DeCosmo, 2002: The signature of spray in the HEXOS turbulent heat flux data. Bound.-Layer Meteor., 103 , 303333.

    • Search Google Scholar
    • Export Citation
  • Atlas, D., B. Walter, S-H. Chou, and P. J. Sheu, 1986: The structure of the unstable marine boundary layer viewed by lidar and aircraft observations. J. Atmos. Sci., 43 , 13011318.

    • Search Google Scholar
    • Export Citation
  • Beljaars, A. C. M., 1995: The parameterization of surface fluxes in large-scale models under free convection. Quart. J. Roy. Meteor. Soc., 121 , 255270.

    • Search Google Scholar
    • Export Citation
  • Bengtsson, L., and Coauthors, 2007: The need for a dynamical climate reanalysis. Bull. Amer. Meteor. Soc., 88 , 495501.

  • Bromwich, D., and R. L. Fogt, 2004: Strong trends in the skill of the ERA-40 and NCEP–NCAR reanalyses in the high and midlatitudes of the Southern Hemisphere, 1958–2001. J. Climate, 17 , 46034619.

    • Search Google Scholar
    • Export Citation
  • Caires, S., A. Sterl, J-R. Bidlot, N. Graham, and V. Swail, 2004: Intercomparison of different reanalyses. J. Climate, 17 , 18931913.

  • Edson, J. B., C. J. Zappa, J. A. Ware, and W. R. McGillis, 2004: Scalar flux profile relationship over the open ocean. J. Geophys. Res., 109 .C08S09, doi:10.1029/2003JC001960.

    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., E. F. Bradley, J. E. Hare, A. A. Grachev, and J. B. Edson, 2003: Bulk Parameterization of air–sea fluxes: Updates and verification for the COARE algorithm. J. Climate, 16 , 571591.

    • Search Google Scholar
    • Export Citation
  • Forster, P., and Coauthors, 2007: Changes in atmospheric constituents and in radiative forcing. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 129–234.

    • Search Google Scholar
    • Export Citation
  • Högström, U., 1988: Non-dimensional wind and temperature profiles in the atmospheric surface layer. Bound.-Layer Meteor., 42 , 263270.

    • Search Google Scholar
    • Export Citation
  • Högström, U., 1996: Review of some basic characteristics of the atmospheric surface layer. Bound.-Layer Meteor., 78 , 215246.

  • Högström, U., J. C. R. Hunt, and A. Smedman, 2002: Theory and measurements for turbulence spectra and variances in the atmospheric neutral surface layer. Bound.-Layer Meteor., 103 , 101124.

    • Search Google Scholar
    • Export Citation
  • Högström, U., and Coauthors, 2008: Momentum fluxes and wind gradients in the marine boundary layer—A multi-platform study. Boreal Environ. Res., in press.

    • Search Google Scholar
    • Export Citation
  • Hunt, J. C. R., and J. F. Morrison, 2000: Eddy structure in turbulent boundary layers. Eur. J. Mech., 19 , 673694.

  • Hunt, J. C. R., and P. Carlotti, 2001: Statistical structure of the high Reynolds number turbulent boundary layer. Flow Turb. Combust., 66 , 453475.

    • Search Google Scholar
    • Export Citation
  • Jones, C. G., U. Willén, A. Ullerstig, and U. Hansson, 2004: The Rossby centre regional atmospheric climate model. Part I: Model climatology and performance for the present climate over Europe. Ambio, 33 , 199210.

    • Search Google Scholar
    • Export Citation
  • Komen, G. J., L. Cavaleri, M. Donelan, K. Hasselmann, S. Hasselmann, and P. A. E. M. Janssen, 1994: Dynamics and Modelling of Ocean Waves. Cambridge University Press, 532 pp.

    • Search Google Scholar
    • Export Citation
  • Lu, S. S., and W. W. Willmarth, 1973: Measurements of the structure of the Reynolds stress in a turbulent boundary layer. J. Fluid Mech., 60 , 481511.

    • Search Google Scholar
    • Export Citation
  • Maitani, T., and E. Ohtaki, 1989: Turbulent transport processes of carbon dioxide and water vapor in the surface layer over a paddy field. J. Meteor. Soc. Japan, 67 , 809815.

    • Search Google Scholar
    • Export Citation
  • Mason, P. J., and R. I. Sykes, 1982: A two-dimensional numerical study of horizontal roll vortices in an inversion capped planetary boundary layer. Quart. J. Roy. Meteor. Soc., 108 , 801823.

    • Search Google Scholar
    • Export Citation
  • Omstedt, A., and L. Nyberg, 1996: Response of Baltic Sea ice to seasonal, interannual forcing and climate change. Tellus, 48 , 644662.

    • Search Google Scholar
    • Export Citation
  • Omstedt, A., and L. B. Axell, 2003: Modeling the variations of salinity and temperature in the large gulfs of the Baltic Sea. Cont. Shelf Res., 23 , 265294.

    • Search Google Scholar
    • Export Citation
  • Peixoto, J. P., and A. H. Oort, 1992: Physics of Climate. American Institute of Physics, 520 pp.

  • Raupach, M. R., 1981: Conditional statistics of Reynolds stress in rough-wall and smooth-wall turbulent boundary layers. J. Fluid Mech., 108 , 363382.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., A. Smedman, and U. Högström, 2001a: Use of conventional stability parameters during swell. J. Geophys. Res., 106 , 2711727134.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., A. Smedman, and A. Omstedt, 2001b: Measured and simulated sensible and latent heat fluxes at two marine sites in the Baltic Sea. Bound.-Layer Meteor., 99 , 5384.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., B. Carlsson, and A. Smedman, 2007: Modelling sensible and latent heat fluxes over sea during unstable, very close to neutral conditions. Bound.-Layer Meteor., 123 , 395415.

    • Search Google Scholar
    • Export Citation
  • Sahlée, E., A. Smedman, U. Högström, and A. Rutgersson, 2008: Re-evaluation of the bulk exchange coefficient for humidity at sea during unstable and near-neutral conditions. J. Phys. Oceanogr., 38 , 257272.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, H. Bergström, A. Rutgersson, K. Kahma, and H. Pettersson, 1999: A case study of air–sea interaction during swell conditions. J. Geophys. Res., 104 , 2583325851.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., X. Guo-Larsén, and U. Högström, 2003: Effects of sea state on the momentum exchange over the sea during neutral conditions. J. Geophys. Res., 108 .3367, doi:10.1029/2002JC001526.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, J. C. R. Hunt, and E. Sahlée, 2007a: Heat/mass transfer in the slightly unstable atmospheric surface layer. Quart. J. Roy. Meteor. Soc., 133 , 3751.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, E. Sahlée, and C. Johansson, 2007b: Critical re-evaluation of the bulk transfer coefficient for sensible and latent heat over the ocean during unstable and neutral conditions. Quart. J. Roy. Meteor. Soc., 133 , 227250.

    • Search Google Scholar
    • Export Citation
  • Sullivan, P. P., J. B. Edson, T. Hristov, and J. C. McWilliams, 2008: Large-eddy simulations and observations of atmospheric marine boundary layers above non-equilibrium surface waves. J. Atmos. Sci., 65 , 12251245.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Willmarth, W. W., and S. S. Lu, 1974: Structure of the Reynolds stress and the occurrence of bursts in the turbulent boundary layer. Advances in Geophysics, Vol. 18, Academic Press, 287–314.

    • Search Google Scholar
    • Export Citation
  • Yu, L., and R. A. Weller, 2007: Objectively analyzed air–sea heat fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, W., W. Perrie, and W. Li, 2006: Impact of waves and sea spray on midlatitude storm structure and intensity. Mon. Wea. Rev., 134 , 24182424.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

(a) Annual relative frequency of occurrence of UVCN conditions average over the entire period (1979–2001). (b) Relative frequency of occurrence of UVCN conditions for different seasons, averaged over the 1979–2001 period.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 2.
Fig. 2.

(a) Relative frequency of occurrence for wind speeds >10 m s−1 over the global oceans; annual average calculated for the 1979–2001 period. (b) Same as (a), but showing the seasonal average.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 3.
Fig. 3.

(a) Annual average of the relative frequency of occurrence for air–sea temperature difference <2 K. (b) Same as (a), but showing the seasonal average.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 4.
Fig. 4.

(a) The enhancement of the mean annual latent heat flux due to UVCN effects, averaged over the 1979–2001 period. (b) Same as (a), but showing the seasonal average.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 5.
Fig. 5.

As in Fig. 4, but relative enhancement.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 6.
Fig. 6.

As in Fig. 4, but for sensible heat flux.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 7.
Fig. 7.

(a) (top) Zonal average of the annual global ocean latent heat flux, averaged over the 1979–2001 period, and (bottom) difference between zonal averaged UVCN affected fluxes and the unaffected fluxes (reference). (b) As in Fig. 7a, but showing the results for the different seasons.

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Fig. 7.
Fig. 7.

(Continued)

Citation: Journal of Climate 21, 22; 10.1175/2008JCLI2279.1

Table 1.

Regional effects of the UVCN regime on the air–sea heat fluxes. N. Atlantic = North Atlantic; N. Pac. O. = North Pacific Ocean; S. oceans = southern oceans; H = sensible heat flux; E = laten heat flux; ref = reference calculation. (a) The yearly averages and the averages for (b) December, January, February (DJF) and (c) June, July, August (JJA).

Table 1.
Save
  • Andreas, E. L., 2004: A bulk air–sea flux algorithm for high-wind, spray conditions, version 2.0. Preprints, 13th Conf. on Interactions of the Sea and Atmosphere, Portland, ME, Amer. Meteor. Soc., 8.2A.

  • Andreas, E. L., and J. DeCosmo, 2002: The signature of spray in the HEXOS turbulent heat flux data. Bound.-Layer Meteor., 103 , 303333.

    • Search Google Scholar
    • Export Citation
  • Atlas, D., B. Walter, S-H. Chou, and P. J. Sheu, 1986: The structure of the unstable marine boundary layer viewed by lidar and aircraft observations. J. Atmos. Sci., 43 , 13011318.

    • Search Google Scholar
    • Export Citation
  • Beljaars, A. C. M., 1995: The parameterization of surface fluxes in large-scale models under free convection. Quart. J. Roy. Meteor. Soc., 121 , 255270.

    • Search Google Scholar
    • Export Citation
  • Bengtsson, L., and Coauthors, 2007: The need for a dynamical climate reanalysis. Bull. Amer. Meteor. Soc., 88 , 495501.

  • Bromwich, D., and R. L. Fogt, 2004: Strong trends in the skill of the ERA-40 and NCEP–NCAR reanalyses in the high and midlatitudes of the Southern Hemisphere, 1958–2001. J. Climate, 17 , 46034619.

    • Search Google Scholar
    • Export Citation
  • Caires, S., A. Sterl, J-R. Bidlot, N. Graham, and V. Swail, 2004: Intercomparison of different reanalyses. J. Climate, 17 , 18931913.

  • Edson, J. B., C. J. Zappa, J. A. Ware, and W. R. McGillis, 2004: Scalar flux profile relationship over the open ocean. J. Geophys. Res., 109 .C08S09, doi:10.1029/2003JC001960.

    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., E. F. Bradley, J. E. Hare, A. A. Grachev, and J. B. Edson, 2003: Bulk Parameterization of air–sea fluxes: Updates and verification for the COARE algorithm. J. Climate, 16 , 571591.

    • Search Google Scholar
    • Export Citation
  • Forster, P., and Coauthors, 2007: Changes in atmospheric constituents and in radiative forcing. Climate Change 2007: The Physical Science Basis, S. Solomon et al., Eds., Cambridge University Press, 129–234.

    • Search Google Scholar
    • Export Citation
  • Högström, U., 1988: Non-dimensional wind and temperature profiles in the atmospheric surface layer. Bound.-Layer Meteor., 42 , 263270.

    • Search Google Scholar
    • Export Citation
  • Högström, U., 1996: Review of some basic characteristics of the atmospheric surface layer. Bound.-Layer Meteor., 78 , 215246.

  • Högström, U., J. C. R. Hunt, and A. Smedman, 2002: Theory and measurements for turbulence spectra and variances in the atmospheric neutral surface layer. Bound.-Layer Meteor., 103 , 101124.

    • Search Google Scholar
    • Export Citation
  • Högström, U., and Coauthors, 2008: Momentum fluxes and wind gradients in the marine boundary layer—A multi-platform study. Boreal Environ. Res., in press.

    • Search Google Scholar
    • Export Citation
  • Hunt, J. C. R., and J. F. Morrison, 2000: Eddy structure in turbulent boundary layers. Eur. J. Mech., 19 , 673694.

  • Hunt, J. C. R., and P. Carlotti, 2001: Statistical structure of the high Reynolds number turbulent boundary layer. Flow Turb. Combust., 66 , 453475.

    • Search Google Scholar
    • Export Citation
  • Jones, C. G., U. Willén, A. Ullerstig, and U. Hansson, 2004: The Rossby centre regional atmospheric climate model. Part I: Model climatology and performance for the present climate over Europe. Ambio, 33 , 199210.

    • Search Google Scholar
    • Export Citation
  • Komen, G. J., L. Cavaleri, M. Donelan, K. Hasselmann, S. Hasselmann, and P. A. E. M. Janssen, 1994: Dynamics and Modelling of Ocean Waves. Cambridge University Press, 532 pp.

    • Search Google Scholar
    • Export Citation
  • Lu, S. S., and W. W. Willmarth, 1973: Measurements of the structure of the Reynolds stress in a turbulent boundary layer. J. Fluid Mech., 60 , 481511.

    • Search Google Scholar
    • Export Citation
  • Maitani, T., and E. Ohtaki, 1989: Turbulent transport processes of carbon dioxide and water vapor in the surface layer over a paddy field. J. Meteor. Soc. Japan, 67 , 809815.

    • Search Google Scholar
    • Export Citation
  • Mason, P. J., and R. I. Sykes, 1982: A two-dimensional numerical study of horizontal roll vortices in an inversion capped planetary boundary layer. Quart. J. Roy. Meteor. Soc., 108 , 801823.

    • Search Google Scholar
    • Export Citation
  • Omstedt, A., and L. Nyberg, 1996: Response of Baltic Sea ice to seasonal, interannual forcing and climate change. Tellus, 48 , 644662.

    • Search Google Scholar
    • Export Citation
  • Omstedt, A., and L. B. Axell, 2003: Modeling the variations of salinity and temperature in the large gulfs of the Baltic Sea. Cont. Shelf Res., 23 , 265294.

    • Search Google Scholar
    • Export Citation
  • Peixoto, J. P., and A. H. Oort, 1992: Physics of Climate. American Institute of Physics, 520 pp.

  • Raupach, M. R., 1981: Conditional statistics of Reynolds stress in rough-wall and smooth-wall turbulent boundary layers. J. Fluid Mech., 108 , 363382.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., A. Smedman, and U. Högström, 2001a: Use of conventional stability parameters during swell. J. Geophys. Res., 106 , 2711727134.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., A. Smedman, and A. Omstedt, 2001b: Measured and simulated sensible and latent heat fluxes at two marine sites in the Baltic Sea. Bound.-Layer Meteor., 99 , 5384.

    • Search Google Scholar
    • Export Citation
  • Rutgersson, A., B. Carlsson, and A. Smedman, 2007: Modelling sensible and latent heat fluxes over sea during unstable, very close to neutral conditions. Bound.-Layer Meteor., 123 , 395415.

    • Search Google Scholar
    • Export Citation
  • Sahlée, E., A. Smedman, U. Högström, and A. Rutgersson, 2008: Re-evaluation of the bulk exchange coefficient for humidity at sea during unstable and near-neutral conditions. J. Phys. Oceanogr., 38 , 257272.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, H. Bergström, A. Rutgersson, K. Kahma, and H. Pettersson, 1999: A case study of air–sea interaction during swell conditions. J. Geophys. Res., 104 , 2583325851.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., X. Guo-Larsén, and U. Högström, 2003: Effects of sea state on the momentum exchange over the sea during neutral conditions. J. Geophys. Res., 108 .3367, doi:10.1029/2002JC001526.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, J. C. R. Hunt, and E. Sahlée, 2007a: Heat/mass transfer in the slightly unstable atmospheric surface layer. Quart. J. Roy. Meteor. Soc., 133 , 3751.

    • Search Google Scholar
    • Export Citation
  • Smedman, A., U. Högström, E. Sahlée, and C. Johansson, 2007b: Critical re-evaluation of the bulk transfer coefficient for sensible and latent heat over the ocean during unstable and neutral conditions. Quart. J. Roy. Meteor. Soc., 133 , 227250.

    • Search Google Scholar
    • Export Citation
  • Sullivan, P. P., J. B. Edson, T. Hristov, and J. C. McWilliams, 2008: Large-eddy simulations and observations of atmospheric marine boundary layers above non-equilibrium surface waves. J. Atmos. Sci., 65 , 12251245.

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012.

  • Willmarth, W. W., and S. S. Lu, 1974: Structure of the Reynolds stress and the occurrence of bursts in the turbulent boundary layer. Advances in Geophysics, Vol. 18, Academic Press, 287–314.

    • Search Google Scholar
    • Export Citation
  • Yu, L., and R. A. Weller, 2007: Objectively analyzed air–sea heat fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88 , 527539.

    • Search Google Scholar
    • Export Citation
  • Zhang, W., W. Perrie, and W. Li, 2006: Impact of waves and sea spray on midlatitude storm structure and intensity. Mon. Wea. Rev., 134 , 24182424.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    (a) Annual relative frequency of occurrence of UVCN conditions average over the entire period (1979–2001). (b) Relative frequency of occurrence of UVCN conditions for different seasons, averaged over the 1979–2001 period.

  • Fig. 2.

    (a) Relative frequency of occurrence for wind speeds >10 m s−1 over the global oceans; annual average calculated for the 1979–2001 period. (b) Same as (a), but showing the seasonal average.

  • Fig. 3.

    (a) Annual average of the relative frequency of occurrence for air–sea temperature difference <2 K. (b) Same as (a), but showing the seasonal average.

  • Fig. 4.

    (a) The enhancement of the mean annual latent heat flux due to UVCN effects, averaged over the 1979–2001 period. (b) Same as (a), but showing the seasonal average.

  • Fig. 5.

    As in Fig. 4, but relative enhancement.

  • Fig. 6.

    As in Fig. 4, but for sensible heat flux.

  • Fig. 7.

    (a) (top) Zonal average of the annual global ocean latent heat flux, averaged over the 1979–2001 period, and (bottom) difference between zonal averaged UVCN affected fluxes and the unaffected fluxes (reference). (b) As in Fig. 7a, but showing the results for the different seasons.

  • Fig. 7.

    (Continued)

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 155 32 2
PDF Downloads 81 20 1