1. Introduction
The eastern part of the subtropical ocean basins is typically associated with persistent decks of stratocumulus clouds (Klein and Hartmann 1993). When air is advected equatorward by the trade winds, and is subjected to warmer ocean waters as well as weaker subsidence, the stratocumulus decks eventually break up and transform into trade cumulus. The stratocumulus to cumulus transition (SCT) is such a persistent feature of the subtropical environment that it is well reproduced in climatologically averaged fields (Sandu et al. 2010).
Considering that marine boundary layer clouds typically are optically thick and overlie a dark ocean, they constitute a large influence on the amount of solar radiation that is absorbed by the Earth system. The SCT is thus associated with a large contrast in the cloud radiative forcing both at the top of the atmosphere (e.g., Harrison et al. 1990) and at the surface (e.g., Gupta et al. 1993). Slingo (1990) demonstrated that the global radiative response to an absolute change in low clouds of approximately 4% would be comparable to the radiative forcing associated with a doubling of CO2. More recently, how the marine boundary layer clouds change has been identified as a key uncertainty in projected temperature responses due to a doubling of CO2, as simulated by GCMs (Bony and Dufresne 2005).
There have been both in situ (e.g., Albrecht et al. 1995) and remote sensing (Wood and Hartmann 2006; Lin et al. 2009; Kawai and Teixeira 2010; Karlsson et al. 2010; Mauger and Norris 2010; Sandu et al. 2010) observational efforts dedicated to characterizing the subtropical transition from stratocumulus to cumulus. The key processes and important structures responsible for the SCT have further been investigated theoretically and with modeling, using different levels of complexity. Deardorff (1980) and Randall (1980) built on ideas originally proposed by Lilly (1968) to hypothesize a cloud-top entrainment instability (CTEI) as a key mechanism for the breakup of stratocumulus into cumulus. The essence of the CTEI hypothesis is that the evaporative cooling that results when dry air is entrained and mixed into a cloud top, in certain conditions, makes that air negatively buoyant relative to its surrounding cloudy air. As a consequence of the downdrafts created by the buoyancy reversal, further entrainment follows and eventually this leads to the stratocumulus breakup. More recently, CTEI as the mechanism responsible for the cloud breakup has been questioned (e.g., Kuo and Schubert 1988; Yamaguchi and Randall 2008; Sandu et al. 2010).
Cloud-resolving modeling studies (e.g., Krueger et al. 1995; Wyant et al. 1997; Stevens 2000) show that SCT can be simulated by just considering advection over increasingly warmer sea surface temperatures. Decoupling of the cloud layer from the well-mixed boundary layer is suggested to be the precursor of the SCT (Wyant et al. 1997). The decoupling is attributed to gradually stronger surface latent heat fluxes associated with increasing sea surface temperatures (SSTs). As a consequence, the entrainment of dry and warm air per unit of cloud-top radiative cooling increases, which results in a deepening of the boundary layer as well as in an increasingly negative buoyancy flux below the cloud base (Bretherton and Wyant 1997). Eventually a pronounced two-layer structure develops, where the cloud layer is separated from the well-mixed boundary layer by a weakly stably stratified subcloud layer. According to Wyant et al. (1997), the diffusion of the stratus layer is a consequence of penetrative entrainment of dry free tropospheric air by increasingly stronger cumulus updrafts originating from the surface. Two recent studies (Mauger and Norris 2010; Sandu et al. 2010), combining satellite retrievals and reanalysis data, show the time scale of the SCT to be well correlated with the magnitude of the lower tropospheric stability (mainly governed by the SST), which gives some observational support for the transition hypothesis formulated by Wyant et al. (1997). Using Lagrangian sea surface temperature statistics along the SCT (Sandu et al. 2010) and a stochastic two-state model based on the results of Bretherton and Wyant (1997) that the extent of decoupling can be derived by considering the ratio of the surface flux of latent heat to the net cloud-top radiative cooling, Chung and Teixeira (2012) were able to simulate essential features of the climatological stratocumulus to cumulus transition.
This study investigates whether the stratocumulus to cumulus transition can be described by solely considering the surface energy flux along the transition. As will be discussed in section 2, in the marine subtropics the two major contributors to the surface energy budget are the absorbed solar radiation and the turbulent latent heat flux (e.g., Peixoto and Oort 1992; Kållberg et al. 2005). The amount of solar radiation absorbed at the surface is expected to be highly dependent on the cloudiness, while the turbulent latent heat flux, which to a first order is a function of SST, is expected to increase when the air is advected over an increasingly warmer ocean along the SCT. In this study we proceed from the climatological energy budget of the ocean surface layer, and the assumption that the net surface energy flux is small, to derive a simple model describing the climatologically important SCT along the trades.
2. Method
This presumed balance is analyzed along a Pacific Ocean transect off the coast of California (outlined in Fig. 1a) that traverses the region where the SCT typically occurs (Fig. 1c). The transect, which is roughly aligned with the trade winds, has previously been used in intercomparison studies as a test bed for the representation of boundary layer clouds in models (Siebesma et al. 2004; Karlsson et al. 2010; Teixeira et al. 2011). The assumption of a small net surface energy flux along the SCT is supported by the European Centre for Medium-Range Weather Forecasts (ECMWF) Interim Re-Analysis (ERA-Interim; Dee et al. 2011), which shows values of Fs within ±10 W m−2 in the transition region (Fig. 1a).
(a) The ERA-Interim (1989–2007) climatological net surface energy flux (Fs). (b) The average ICOADS and interannual range (1971–2000) sea surface temperature along the transect outlined in (a). (c) The average annual cycle and year-to-year range (2000–09) in Terra–MODIS total cloud cover and the annual average ISCCP D2 dataset (1983–2008) total cloud cover along the same transect. The gray shaded area in (b) and (c) indicates the part of the transect where the climatological SCT occurs (also associated with a small ERA-Interim net surface energy flux).
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
Figure 2 shows the climatological average and interannual range in radiative and turbulent fluxes along the transect according to ERA-Interim, the 40-yr ECMWF Re-Analysis (ERA-40; Uppala et al. 2005), the National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) reanalysis (NRA; Kalnay et al. 1996), and the objectively analyzed air–sea fluxes (OAFlux; Yu and Weller 2007) and International Satellite Cloud Climatology Project flux dataset (ISCCP FD; Zhang et al. 2004) datasets. The apparent disagreement between the datasets, which largely rely on the same observational foundation but use different assimilation techniques to obtain an atmospheric analysis, is an indication of the uncertainty regarding surface fluxes in these relatively remote regions. Some biases have, however, been identified. NRA shows the smallest amounts of absorbed solar radiation along the transect (Fig. 2d), which is likely associated with a too large ocean albedo in this reanalysis, as noted by Kalnay et al. (1996). De Szoeke et al. (2010) showed that the ISCCP FD net surface longwave flux has an approximate 15 W m−2 bias (i.e., not negative enough) compared to surface observations in the SCT region in the southeastern tropical Pacific Ocean. Considering the similar atmospheric conditions, similar biases may be expected in the Californian SCT regions that we analyze. The weak latitudinal gradient in the net longwave radiation and in the absorbed shortwave radiation in ERA-40 (Figs. 2c and 2d, respectively) is likely related to underestimation of both cloud cover and optical thickness in the region typically associated with stratocumulus and, at lower latitudes, to clouds that have unrealistically high reflectivity (e.g., Allan et al. 2004). As noted by Trenberth et al. (2010), the representation of the radiative influence of clouds on the surface energy budget is improved in ERA-Interim relative to ERA-40 in both the convective and the stratocumulus regions. These improvements likely explain the increased latitudinal gradient of the net longwave and absorbed solar radiation in ERA-Interim compared to ERA-40 (Figs. 2c,d).
Annual climatological averages (dots) and interannual range (bars) of the surface energy fluxes along the Pacific as given by ERA-40 (1980–2001), ERA-Interim (1989–2007), NRA (1984–2007), and combined OAFlux and ISCCP FD (1984–2007) datasets: (a) latent heat, (b) sensible heat, (c) net longwave, and (d) net shortwave. Net fluxes downward are defined as positive. Shaded gray area identifies the latitude range where the climatological SCT occurs.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
As discussed in Trenberth et al. (2009), all commonly used observationally based datasets show annually and globally averaged net ocean surface fluxes that are unphysically large. The authors argue that the surface downwelling longwave flux is the surface energy budget component least well characterized by observations, since it significantly depends on cloud-base height, which is not well determined by space-based measurements.
There are also empirical relationships to estimate the surface net longwave flux, developed for surface observers without direct radiance data available [see Fung et al. (1984) for a review]. Such empirical bulk formulas typically consist of an estimation of the clear-sky longwave flux that is adjusted with a cloudiness correction. The more sophisticated clear-sky components depend on the sea surface temperature and the near-surface water vapor and temperature (e.g., Anderson 1952), while the cloudiness correction is linearly or quadratically dependent on the total cloud fraction. However, as discussed in Fung et al. (1984), such parameterizations are not suitable in the presence of strong temperature inversions and humidity jumps aloft, which are conditions commonly present in the transition region. Furthermore, they conclude that the cloud correction is uncertain since no explicit consideration is taken to either the cloud-base temperature or the optical thickness, which are key variables in determining the cloud influence on the surface energy budget. Therefore, we will here utilize reanalysis data for the surface net longwave. More specifically the ERA-Interim net longwave is used and the sensitivity of this choice is discussed in section 3b.
Although the datasets are at odds in regards to the absolute values of the surface energy fluxes, they do agree that the absorbed solar radiation and the upward flux of latent heat are the two major contributors to the surface energy budget and the ones that show the largest changes along the transect and in the marine subtropics in general (Fig. 2). Apart from wind speed, the amount of surface energy lost in terms of latent heat strongly depends on SST and the near-surface RH. At the ocean surface, air is typically near saturation and as such its humidity depends strongly on the sea surface temperature through the Clausius–Clapeyron relationship. The near-surface air humidity also depends on the SST (although implicitly) since both the temperature offset from the SST and the near-surface RH are relatively invariant along the SCT. The observed equatorward increase in the latent heat flux (Fig. 2a) is mainly driven by the increasing SSTs along the transect (Fig. 1b). This can be qualitatively understood with Clausius–Clapeyron; a 5-K SST increase with a 1-K SST offset of the near-surface air at 80% relative humidity results in a 33% increase in the vertical humidity gradient. As a consequence, the latent heat flux would also show a relative increase of 33%.
While the amount of turbulent exchange of latent heat depends strongly on SST, the amount of solar radiation reaching the surface strongly relates to the cloud fraction. Much of the southwestern increase in the absorbed solar radiation along the transect (Fig. 2d) is thus inherently related to the SCT. Considering that the southwesterly gradient in temperature-driven latent heat loss is comparable in magnitude to the cloudiness-influenced increase in surface solar absorption, a simple cloud cover model aiming to describe the climatological stratocumulus to cumulus transition is proposed below.
Model description
The climatological net surface longwave flux is taken from the ERA-Interim reanalysis and it decreases (i.e., it becomes more negative) with decreasing latitude.
3. Results and discussion
a. Model solutions
A requisite for deriving the cloud transition using Eq. (8) is the climatologically averaged values of all variables on the right-hand side. Figure 3 shows the climatological average and interannual range of the sea surface temperature (To), the air temperature offset (Ta − To), the relative humidity, and the horizontal wind speed along the Pacific transect from ERA-Interim, OAFlux, and the enhanced version of the International Comprehensive Ocean–Atmosphere Data Set (ICOADS), release 2.5 (Woodruff et al. 2011). The gray-shaded area in the panels indicates the region associated with the climatological SCT, in which Eq. (8) will be solved. In terms of SST, the datasets show excellent agreement, depicting a close to linear temperature increase southward along the transect (Fig. 3a).
Annual climatological averages (dots) and interannual range (bars) of (a) SST, (b) air temperature offset, (c) relative humidity, and (d) horizontal wind speed along the transition transect as given by the ICOADS (1989–2007), ERA-Interim (1989–2007), and OAFlux (1984–2007) datasets. Shaded gray areas indicate the latitude range where the climatological SCT occurs.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
Although disagreeing on the absolute values of the RH and the horizontal wind speed, the datasets, more or less, agree on the spatial variability along the transect (Figs. 3c and 3d, respectively). Evident in all the datasets and somewhat surprising is the equatorward increase of near-surface RH in the region associated with the transition. Assuming a well-mixed subcloud layer, such an increase implies a lowering of the cloud base in the stratocumulus to cumulus transition. This result could be indicative of a decoupling of the stratocumulus layer and thus a decrease of the entrainment of relatively dry and warm free tropospheric air into the well-mixed subcloud layer. The observational datasets show least agreement in terms of the temperature offset (Fig. 3b). The ICOADS dataset distinguishes itself by a relatively small temperature offset and by showing an (weak) equatorward increase of the air–sea temperature offset, in contrast to ERA-Interim and OAFlux.
Since differences exist in the way the datasets are assembled, disagreements between them are inevitable. For example, opposed to the ERA-Interim and the OAFlux datasets, which have their near-surface variables adjusted to 10 m for horizontal wind and 2 m for humidity and air temperature, the ICOADS dataset consists of averages of the measured values, regardless of measurement height. This, together with less sampling, likely explain the relatively large interannual range compared to ERA-Interim and OAFlux.
The SZA, as a function of latitude, declination angle, and hour angle (e.g., Pielke 1984), is used to derive the climatological averaged sunlit SZA along the transition (Fig. 4a). These are in turn used to derive the SZA-dependent cloud and ocean surface albedos (αc and αo; Fig. 4b) in accordance with Cahalan et al. (1994) and Taylor et al. (1996), respectively. The cloud albedo is also dependent on liquid water path (LWP). As discussed by Turner et al. (2007), remote sensing of LWP is still associated with considerable uncertainty. In the derivation of the cloud fractions the average in-cloud LWP along the transition is assumed constant at 50 g m−2. Our choice of in-cloud LWP is somewhat arbitrary but comparable to what large-eddy simulations (LES) of stratocumulus (Stevens et al. 2005) and shallow cumulus (Siebesma et al. 2003) show. The very different distribution of LWP in different cloud regimes (Wood and Hartmann 2006), which potentially might be important for the average cloud albedo, is not considered in the derivation. However, observations indicate the mean cloud albedo to be fairly independent of cloud fraction in the transition region (Bender et al. 2011). The assumption of a constant in-cloud LWP, on average, is further justified by observed nonincreasing cloud-top heights (Karlsson et al. 2010) and boundary layer depths (von Engeln and Teixeira 2013) along the climatological transition. The model sensitivity to the choice of climatological LWP along the transect will be addressed later.
Annual climatological averages of (a) solar zenith angle, (b) cloud and ocean albedo (×10), (c) clear-sky atmospheric transmittance, and (d) correlation terms (see text for details) along the part of transect associated with the climatological SCT.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
The clear-sky atmospheric transmittance of solar radiation (Matm; Fig. 4c) is calculated from ERA-Interim data as the fraction of the surface downwelling clear-sky solar radiation to the insolation at the top of the atmosphere. ICOADS data are used to derive γUT and γUq that are introduced to account for temporal covariance on shorter than climatological time-scales in Eqs. (3) and (4), respectively. The results (Fig. 4d) indicate that the sensible and latent heat fluxes, on average, would be underestimated by 14% and 7%, respectively, if γUT and γUq were not included in Eqs. (3) and (4). For reasons that are not clear, both γUT and γUq are larger in the southern part of the transition.
The value of γSW [Eq. (5)] is calculated as the fraction of the annually averaged net surface shortwave flux when using a varying SZA (e.g., Pielke 1984) to the net surface shortwave flux when using the climatological values of the individual variables. The spatial variance of γSW (Fig. 4d) is derived by calculating γSW in the southwesterly and northeasterly part of the transect associated with the climatological SCT, using corresponding ISCCP climatological cloud fractions (Fig. 1c), and interpolating linearly in between these values. If the temporal covariance between the SZA-dependent variables is not accounted for, the surface absorbed solar radiation would be underestimated by about 10%.
In Fig. 5, the solutions of the cloud fraction model [Eq. (8)] derived with data from ICOADS, ERA-Interim, and the OAFlux dataset are shown together with the climatological total cloud fraction according to ISCCP and Moderate Resolution Imaging Spectroradiometer (MODIS) data and the ERA-Interim reanalysis. All three solutions manage to depict a general decrease in cloud cover with increasing SSTs. However, in terms of a monotonic cloud cover decrease, only the ERA-Interim and the ICOADS solutions show agreement with the observations and reanalysis. Both the ICOADS and the ERA-Interim solutions slightly overestimate the transition strength attributable to too much stratocumulus. It is worth noting that the cloud cover solution when using climatological ERA-Interim data shows at least as good agreement with the observations as the total cloud cover reported by the reanalysis product.
Solutions of the cloud fraction model using climatologically averaged surface variables from ICOADS, ERA-Interim, and OAFlux and the total cloud fraction as given by ISCCP and ERA-Interim, as a function of sea surface temperature along the transect associated with the climatological stratocumulus to cumulus transition.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
The discrepancy between the model solutions can mainly be attributed to differences in the derived latent heat fluxes [Eq. (4)] along the transition among the different datasets (Fig. 6). The association of lower cloud fractions with higher latent heat fluxes fits well with the deepening–warming decoupling process (Bretherton and Wyant 1997; Wyant et al. 1997). As previously discussed, according to this hypothesis the entrainment of dry and warm air per unit of cloud radiative cooling increases when SST (and the surface latent heat flux) increases. Eventually the downward flux of entrained warm and dry air will result in a conditionally unstable or stable subcloud layer. Thus, the increasing surface latent heat fluxes, via increased entrainment, decouple the cloud layer from the well-mixed layer. The increase of near-surface relative humidity along the transition in the observationally based datasets (Fig. 3c) is indicative of such a decoupling. The increase in relative humidity lowers the lifting condensation level and allows for cumulus development below the stratocumulus. In the conceptual model of Wyant et al. (1997) the more vigorous cumulus, which manage to penetrate through the stable subcloud layer and reach the cloudy layer, generate entrainment of dry free tropospheric air that finalizes the breakup of the decoupled stratocumulus layer.
Correlation between derived latent heat and cloud fraction based on ICOADS, ERA-Interim, and OAFlux data.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
b. Sensitivity studies
To investigate which variables are most important for the modeled cloud fraction transition, sensitivity studies are carried out. One at the time, the sensitivity to the choice of LWP, near-surface relative humidity, wind speed, and temperature offset relative to the SST and to the net surface longwave flux are investigated. Figure 7 shows these results. As a reference, the cloud fraction solution using the average values of the surface meteorological variables in the region associated with the climatological SCT, taken over the three observational datasets and a LWP of 50 g m−2, is included in the panels.
Cloud fraction sensitivity to (a) LWP, (b) air temperature offset (Ta − To), (c) RH, (d) horizontal wind speed, and (e) surface net longwave. The contours represent isolines of the variable the cloud fraction sensitivity is tested against. Black dashed line indicates cloud fraction solution using average values: LWP = 50 g m−2, Ta − To = −0.94 K, RH = 78%, U = 7.6 m s−1, and net LW = −50 W m−2.
Citation: Journal of Climate 27, 11; 10.1175/JCLI-D-13-00534.1
The cloud fraction depends on LWP via the cloud albedo [Eq. (8)]. This highly nonlinear dependence of cloud fraction on LWP can be attributed to the saturation of cloud albedo at high LWP values. When the cloud albedo increases, cloud fraction has to decrease (Fig. 7a) for the underlying assumption of local energy equilibrium to remain valid [Eq. (2)]. For the same reason, the amount of absorbed solar radiation at the surface has to increase to cancel the larger latent heat loss associated with increasing SSTs. This can be achieved by decreasing either the water content or the coverage of the cloud (Fig. 7a). For example, at a cloud coverage of 70%, an approximate LWP drop of 28 g m−2 from the colder to the warmer SSTs is necessary to increase the absorbed solar radiation at the surface to the amount that it balances the SST-driven increase in latent heat loss. To illustrate the nonlinearity, the corresponding LWP drop at a cloud coverage of 60% is about 42 g m−2.
To the first order, (in cloud) LWP is a function of cloud depth and the key implicit assumption of keeping LWP constant in the region of the transition is that mean cloud depth does not change much. As mentioned earlier, LES (Siebesma et al. 2003; Stevens et al. 2005), satellite observations (Bender et al. 2011; Karlsson et al. 2010), and reanalysis data (von Engeln and Teixeira 2013) have suggested this to be a reasonable assumption. If this is the case, then the precise value of LWP (as long as it is a realistic one) is less of an issue. Although we choose 50 g m−2, this value is fairly arbitrary (within reasonable limits) and the magnitude of the cloud cover decrease will not change significantly for constant values between 50 and 100 g m−2 (Fig. 7a).
The cloud fraction sensitivities to air–ocean temperature offset, relative humidity, and wind speed behave more linearly (Figs. 7b–d). The results can again be attributed to the assumption of local energy equilibrium (Fs ~ 0). A decreasing near-surface RH and increasing temperature offset and wind speed all tend to increase the latent heat loss. Lower cloud fractions, which increase the amount of absorbed solar radiation, are necessary to keep the equilibrium assumption valid. From Figs. 7b–d it is evident that the sensitivities to the near-surface variables are quite substantial. At a fixed 70% cloud fraction, the latent heat increase associated with a SST increase from 21° to 25°C is cancelled if accompanied by a 1.4 m s−1 decrease in the wind speed or by a decrease in RH or a temperature offset of 5% and 0.8°C, respectively.
Considering these sensitivities and that the observational datasets show considerable disagreement along the transition region, both in terms of the actual amount and the spatial variability (Fig. 3), the extent to which the model realizations agree is intriguing (Fig. 5).
As noted by Trenberth et al. (2009) and discussed in section 2, the surface energy component least well defined by observations is the downwelling longwave flux, which in turn will influence the uncertainty of the net surface longwave. Figure 7e shows the model sensitivity to the net surface longwave flux. In solving Eq. (8) along the transition, the climatological net surface longwave fluxes from ERA-Interim were used. The sensitivity of this choice can be addressed by noting that the range of surface net longwave radition between the different observational datasets is ≤15 W m−2 along the transition region (Fig. 2c). According to Fig. 7e a 15 W m−2 change in the longwave translates to a cloud fraction sensitivity of approximately 10%, regardless of sea surface temperature. Thus, if the model is solved along the transition using the net longwave from the combined OAFlux and ISCCP FD (Fig. 2c), this generates similar cloud fractions in the southern part of the transect (for higher SSTs) and increased cloud fractions (~10%) in the stratocumulus region (not shown). The modeled cloud fraction drop would in this case be somewhat overestimated compared to the observations. However, as mentioned earlier, ISCCP FD has been shown to underestimate the surface net longwave cooling in the stratocumulus region in the southeastern tropical Pacific (de Szoeke et al. 2010). Considering that the atmospheric settings are alike, it is reasonable to believe a similar bias exists in the Californian stratocumulus region.
4. Summary and conclusions
By considering the climatological surface energy budget off the coast of California, a model describing the cloud cover transition from stratocumulus to trade cumulus is developed. Air advected by the trade winds in the eastern parts of the subtropical ocean basins is subjected to increasingly warmer SSTs and is also associated with decreasing cloud cover. These gradients imply considerable differences in the surface energy fluxes, not least in the two key contributors to the surface energy budget in the marine subtropics; the absorbed solar radiation (due to changes in cloud cover) and the turbulent flux of latent heat (due to changes in SST).
Based on the assumption that the climatological net surface energy flux is small in the transition region (supported by the ECMWF reanalysis), parameterizations of the surface radiative and turbulent energy fluxes are applied and the set of equations are solved for the cloud fraction. The resulting cloud cover model [Eq. (8)] depicts decreasing cloud amounts associated with increasing latent heat fluxes, which supports the deepening–warming decoupling hypothesis formulated by Bretherton and Wyant (1997) and Wyant et al. (1997).
The model is solved using climatologically averaged surface meteorological variables from ship and buoy measurements (ICOADS) as well as from other observationally based datasets (ERA-Interim reanalysis and OAFlux). In terms of the climatological cloud cover transition, the solutions agree reasonably well with satellite observations. Noteworthy is that the model realization using surface meteorological data from the ERA-Interim dataset agrees at least as well with the observations as the cloud cover reported by the reanalysis.
The essential point that this paper illustrates is that in the subtropical region where the stratocumulus to cumulus transition is most prominent, there is a climatological surface energy balance between shortwave radiation and latent heat flux, which directly links the cloud cover to the sea surface temperature. This relation is confirmed by producing realistic cloud cover by using the observed SST, suggesting that the same relation can be used to predict the SST from the climatological cloud distribution in this region.
This is an important result in our quest to understand how low clouds and climate interact, by devising a simple model that directly and explicitly connects the subtropical sea surface temperature to the clouds, without the need for more complicated physics or dynamics.
Acknowledgments
The authors acknowledge the support provided by the Office of Naval Research, Marine Meteorology Program under Awards N0001411IP20087 and N0001411IP20069, by the NASA MAP Program, and by the NOAA MAPP/CPO Program. This research was mainly carried out at the Jet Propulsion Laboratory, California Institute of Technology, under a contract with the National Aeronautics and Space Administration. ICOADS data are provided by the NOAA/OAR/ESRL PSD, Boulder, Colorado, United States, from their website (available at http://www.esrl.noaa.gov/psd/). Global ocean heat flux and evaporation products were provided by the WHOI OAFlux project (http://oaflux.whoi.edu) funded by the NOAA Climate Observations and Monitoring (COM) program. ERA-Interim and ERA-40 reanalysis data were provided by ECMWF, Reading, UK, from their website at http://www.ecmwf.org.
REFERENCES
Albrecht, B. A., C. S. Bretherton, D. Johnson, W. H. Scubert, and A. Shelby Frisch, 1995: The Atlantic Stratocumulus Transition Experiment—ASTEX. Bull. Amer. Meteor. Soc., 76, 889–904, doi:10.1175/1520-0477(1995)076<0889:TASTE>2.0.CO;2.
Allan, R. P., M. A. Ringer, J. A. Pamment, and A. Slingo, 2004: Simulation of the Earth’s radiation budget by the European Centre for Medium-Range Weather Forecasts 40-year reanalysis (ERA40). J. Geophys. Res.,109, D18107, doi:10.1029/2004JD004816.
Anderson, E. R., 1952: Energy budget studies. U.S. Geol. Surv. Circ., 229, 71–119.
Bender, F. A.-M., R. J. Charlson, A. M.-L. Ekman, and L. Leahy, 2011: Quantification of monthly mean, regional-scale albedo of marine stratiform clouds in satellite observations and gcms. J. Appl. Meteor. Climatol., 50, 2139–2148, doi:10.1175/JAMC-D-11-049.1.
Bony, S., and J.-L. Dufresne, 2005: Marine boundary layer clouds at the heart of tropical cloud feedback uncertainties in climate models. Geophys. Res. Lett.,32, L20806, doi:10.1029/2005GL023851.
Bretherton, C. S., and M. C. Wyant, 1997: Moisture transport, lower-tropospheric stability, and decoupling of cloud-topped boundary layers. J. Atmos. Sci., 54, 148–167, doi:10.1175/1520-0469(1997)054<0148:MTLTSA>2.0.CO;2.
Cahalan, R. F., W. Ridgway, W. J. Wiscombe, T. L. Bell, and J. B. Snider, 1994: The albedo of fractal stratocumulus clouds. J. Atmos. Sci., 51, 2434–2455, doi:10.1175/1520-0469(1994)051<2434:TAOFSC>2.0.CO;2.
Chung, D., and J. Teixeira, 2012: A simple model for stratocumulus to shallow cumulus cloud transitions. J. Climate, 25, 2547–2554, doi:10.1175/JCLI-D-11-00105.1.
Deardorff, J. W., 1980: Cloud top entrainment instability. J. Atmos. Sci., 37, 131–147, doi:10.1175/1520-0469(1980)037<0131:CTEI>2.0.CO;2.
DeCosmo, J., K. B. Katsaros, S. D. Smith, R. J. Anderson, W. A. Oost, K. Bumke, and H. Chadwick, 1996: Air–sea exchange of water vapor and sensible heat: The Humidity Exchange Over the Sea (HEXOS) results. J. Geophys. Res., 101, 12 001–12 016, doi:10.1029/95JC03796.
Dee, D. P., and Coauthors, 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553–597, doi:10.1002/qj.828.
de Szoeke, S. P., C. W. Fairall, D. E. Wolfe, L. Bariteau, and P. Zuidema, 2010: Surface flux observations on the southeastern tropical Pacific Ocean and attribution of SST errors in coupled ocean–atmosphere models. J. Climate, 23, 4152–4174, doi:10.1175/2010JCLI3411.1.
Fu, C., R. Pyle, and H. Fan, 1994: A comparison study of the climatological air–sea heat fluxes estimated by different computational schemes of bulk formula. Adv. Atmos. Sci., 11, 189–200, doi:10.1007/BF02666545.
Fung, I. Y., D. E. Harrison, and A. A. Lacis, 1984: On the variability of the net longwave radiation at the ocean surface. Rev. Geophys. Space Phys., 22, 177–193, doi:10.1029/RG022i002p00177.
Gupta, S. K., W. F. Staylor, W. L. Darnell, A. C. Wilber, and N. A. Ritchey, 1993: Seasonal variation of surface and atmospheric cloud radiative forcing over the globe derived from satellite data. J. Geophys. Res.,98 (D11), 20 761–20 778, doi:10.1029/93JD01533.
Harrison, E. F., P. Minnis, B. R. Barkstrom, V. Ramanathan, R. D. Cess, and G. G. Gibson, 1990: Seasonal variation of cloud radiative forcing derived from the Earth Radiation Budget Experiment. J. Geophys. Res., 95, 18 687–18 703, doi:10.1029/JD095iD11p18687.
Kållberg, P., P. Berrisford, B. Hoskins, A. Simmmons, S. Uppala, S. Lamy-Thépaut, and R. Hine, 2005: ERA-40 Atlas. ERA-40 Project Rep. 19, ECMWF, 191 pp. [Available online at http://www.ecmwf.int/research/era/ERA-40_Atlas/.]
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Karlsson, J., G. Svensson, S. Cardoso, J. Teixeira, and S. Paradise, 2010: Subtropical cloud-regime transitions: Boundary layer depth and cloud-top height evolution in models and observations. J. Appl. Meteor. Climatol., 49, 1845–1858, doi:10.1175/2010JAMC2338.1.
Kawai, H., and J. Teixeira, 2010: Probability density functions of liquid water path and cloud amount of marine boundary layer clouds: Geographical and seasonal variations and controlling meteorological factors. J. Climate, 23, 2079–2092, doi:10.1175/2009JCLI3070.1.
Klein, S., and D. L. Hartmann, 1993: The seasonal cycle of low stratiform clouds. J. Climate, 6, 1587–1606, doi:10.1175/1520-0442(1993)006<1587:TSCOLS>2.0.CO;2.
Krueger, S. K., G. T. McLean, and Q. Fu, 1995: Numerical simulation of the stratus-to-cumulus transition in the subtropical marine boundary layer. Part I: Boundary-layer structure. J. Atmos. Sci., 52, 2839–2850, doi:10.1175/1520-0469(1995)052<2839:NSOTST>2.0.CO;2.
Kuo, H.-C., and W. H. Schubert, 1988: Stability of cloud-topped boundary layers. Quart. J. Roy. Meteor. Soc., 114, 887–916, doi:10.1002/qj.49711448204.
Lilly, D. K., 1968: Models of cloud-topped mixed layers under a strong inversion. Quart. J. Roy. Meteor. Soc., 94, 292–309, doi:10.1002/qj.49709440106.
Lin, W., M. Zhang, and N. G. Loeb, 2009: Seasonal variation of the physical properties of marine boundary layer clouds off the California coast. J. Climate, 22, 2624–2638, doi:10.1175/2008JCLI2478.1.
Mauger, G. S., and J. R. Norris, 2010: Assessing the impact of meteorological history on subtropical cloud fraction. J. Climate, 23, 2926–2940, doi:10.1175/2010JCLI3272.1.
Peixoto, J. P., and A. H. Oort, 1992: Physics of Climate. Springer-Verlag, 520 pp.
Pielke, R. A., 1984: Mesoscale Meteorological Modeling. International Geophysics Series, Vol. 78, Academic Press, 612 pp.
Randall, D. A., 1980: Conditional instability of the first kind upside-down. J. Atmos. Sci., 37, 125–130, doi:10.1175/1520-0469(1980)037<0125:CIOTFK>2.0.CO;2.
Sandu, I., B. Stevens, and R. Pincus, 2010: On the transitions in marine boundary layer cloudiness. Atmos. Chem. Phys., 10, 2377–2391, doi:10.5194/acp-10-2377-2010.
Siebesma, A. P., and Coauthors, 2003: A large eddy simulation intercomparison study of shallow cumulus convection. J. Atmos. Sci., 60, 1201–1219, doi:10.1175/1520-0469(2003)60<1201:ALESIS>2.0.CO;2.
Siebesma, A. P., and Coauthors, 2004: Cloud representation in general-circulation models over the northern Pacific Ocean: A EUROCS intercomparison study. Quart. J. Roy. Meteor. Soc., 130, 3245–3267, doi:10.1256/qj.03.146.
Slingo, A., 1990: Sensitivity of the Earth’s radiation budget to changes in low clouds. Nature, 343, 49–51, doi:10.1038/343049a0.
Stevens, B., 2000: Cloud transitions and decoupling in shear-free stratocumulus-topped boundary layers. Geophys. Res. Lett., 27, 2557–2560, doi:10.1029/1999GL011257.
Stevens, B., and Coauthors, 2005: Evaluation of large-eddy simulations via observations of nocturnal marine stratocumulus. Mon. Wea. Rev.,133, 1443–1462, doi:10.1175/MWR2930.1.
Taylor, J. P., J. M. Edwards, M. D. Glew, P. Hignett, and A. Slingo, 1996: Studies with a flexible new radiation code. II: Comparisons with aircraft short-wave observations. Quart. J. Roy. Meteor. Soc., 122, 839–861, doi:10.1002/qj.49712253204.
Teixeira, J., and Coauthors, 2011: Tropical and subtropical cloud transitions in weather and climate prediction models: The GCSS/WGNE Pacific Cross-Section Intercomparison (GPCI). J. Climate, 24, 5223–5256, doi:10.1175/2011JCLI3672.1.
Trenberth, K. E., J. T. Fasullo, and J. Kiehl, 2009: Earth’s global energy budget. Bull. Amer. Meteor. Soc.,90, 311–323, doi:10.1175/2008BAMS2634.1.
Trenberth, K. E., and Coauthors, 2010: Atmospheric reanalyses: A major resource for ocean product development and modeling. Proceedings of OceanObs’09: Sustained Ocean Observations and Information for Society, Vol. 2, J. Hall, D. Harrison, and D. Stammer, Eds., ESA WP-306, doi:10.5270/OceanObs09.cwp.90.
Turner, D. D., and Coauthors, 2007: Thin liquid water clouds: Their importance and our challenge. Bull. Amer. Meteor. Soc.,88, 177–190, doi:10.1175/BAMS-88-2-177.
Uppala, S. M., and Coauthors, 2005: The ERA-40 Re-Analysis. Quart. J. Roy. Meteor. Soc., 131, 2961–3012, doi:10.1256/qj.04.176.
von Engeln, A., and J. Teixeira, 2013: A planetary boundary layer height climatology derived from ECMWF reanalysis data. J. Climate, 26, 6575–6590, doi:10.1175/JCLI-D-12-00385.1.
Wood, R., and D. L. Hartmann, 2006: Spatial variability of liquid water path in marine low cloud: The importance of mesoscale cellular convection. J. Climate,19, 1748–1764, doi:10.1175/JCLI3702.1.
Woodruff, S., and Coauthors, 2011: ICOADS release 2.5: Extensions and enhancements to the surface marine meteorological archive. Int. J. Climatol., 31, 951–967, doi:10.1002/joc.2103.
Wyant, M. C., C. S. Bretherton, H. A. Rand, and D. E. Stevens, 1997: Numerical simulations and a conceptual model of the stratocumulus to trade cumulus transition. J. Atmos. Sci., 54, 168–192, doi:10.1175/1520-0469(1997)054<0168:NSAACM>2.0.CO;2.
Yamaguchi, T., and D. A. Randall, 2008: Large-eddy simulation of evaporatively driven entrainment in cloud-topped mixed layers. J. Atmos. Sci., 65, 1481–1504, doi:10.1175/2007JAS2438.1.
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, 527–539, doi:10.1175/BAMS-88-4-527.
Zhang, Y., W. B. Rossow, A. A. Lacis, V. Oinas, and M. I. Mishchenko, 2004: Calculation of radiative fluxes from the surface to top of atmosphere based on ISCCP and other global data sets: Refinements of the radiative transfer model and the input data. J. Geophys. Res.,109, D19105, doi:10.1029/2003JD004457.
