VAMOS Ocean-Cloud-Atmosphere-Land Study (VOCALS)

"Radiator fin" (RF) proposed field program.

Scientific overview and tentative plans

SCIENTIFIC BACKGROUND

Interactions between the South American continent and the Southeast Pacific (SEP) Ocean are extremely important for both the regional and global climate system. The great height and length of the Andes Cordillera forms a sharp barrier to zonal flow, resulting in strong winds (coastal jet) parallel to the coasts of Chile and Peru. This, in turn, drives intense oceanic upwelling along these coasts, bringing cold, deep, nutrient/biota rich waters to the surface. As a result, the coastal SEP sea-surface temperatures (SSTs) are colder along the Chilean and Peruvian coasts than at any comparable latitude elsewhere. The cold surface, in combination with warm, dry air aloft is ideal for the formation of marine stratocumulus clouds, and supports the largest and most persistent, yet poorly observed, subtropical stratocumulus deck in the world (Klein and Hartmann 1993).

Three fundamental issues impeding our understanding of the weather and climate system are (a) our current lack of understanding and quantification of the indirect effect of aerosols upon cloud radiative properties; (b) biases in tropical rainfall, SST, and winds that repeatedly occur in coupled ocean-atmosphere models, which several studies have traced in part to errors simulating ocean dynamics in the low-latitude coastal upwelling zones, and errors in simulating of boundary layer clouds and their radiative properties (e.g. Mechoso et al. 1995, Ma et al. 1996); (c) our inability to make consistently accurate regional weather predictions, especially in coastal areas dominated by low cloud.

VOCALS (VAMOS Ocean-Cloud-Atmosphere-Land Study) is an international program in which modeling, extended-time observations (including a wealth of new satellite sensors, buoy, island and coastal measurements), and intensive field observations (including annual buoy maintenance cruises) in the SEP are coordinated to address these issues over the period 2003-2010. While extended-time observations are rapidly improving our understanding of this sparsely-traveled region, they have raised a set of interconnected scientific questions better addressed by an intensive field campaign.

The following sections describe four key areas of science that we hope to address with such a campaign which we propose for a four week period during October/November 2007. We call this the VOCALS "radiator fin" (RF) experiment in recognition of the proposed cruise track and all Pierrehumbert's (1995) characterization of the subtropical subsidence regions as the Ôradiator fins' of the tropics in which the free troposphere is very dry and infrared radiation efficiently cools the earth system.

1. Aerosol-cloud-drizzle interactions

Observations in the SEP made during the EPIC 2001 field campaign (Bretherton et al. 2004) suggest that drizzle production is modulated by cloud droplet number concentration, which is directly related to aerosol concentration (e.g. Twomey and Warner 1967, Martin et al. 1994, Breon et al. 2002). Tantalizing evidence has been presented (Stevens et al. 2004) suggesting a direct link between drizzle and cloudiness in MBL clouds that is manifest through regions of broken cloud organized into roughly polygonal lattices, called "open cellular convection", embedded within otherwise overcast stratocumulus. These regions have been termed POCs (Figure 1), or "pockets of open cells" (Stevens et al. 2004), and measurements suggest that POCs tend to be associated with low aerosol concentration (Petters et al. 2004), and intense drizzle production (Stevens et al. 2004). This link between drizzle production and cloudiness is central to the hypothesis of Albrecht (1989), namely that increases in anthropogenic aerosol may lead to a reduction in precipitation and a corresponding increase in global cloud cover and thickness. There have been attempts to test this hypothesis in GCMs (e.g. Lohmann and Feichter 1997, see also review paper by Haywood and Boucher 2001), with sensitivities to increases in anthropogenic aerosol varying widely between models. This is hardly surprising given (a) the huge quantitative differences in the sensitivity of the parameterizations of drizzle production to cloud microphysics (Wood 2004, Pawlowska and Brenguier 2004); (b) the inadequate representation of the turbulent structure and entrainment characteristics of cloud-topped boundary layers in GCMs (e.g. see Bretherton et al. 2004) which strongly interacts with their cloud (Bretherton and Wyant 1997) and precipitation (Nicholls 1987, Baker 1993) characteristics; (c) inadequate understanding of how to parameterize mechanisms modulating subgrid variability of cloud optical depth and precipitation. POCs are relevant to all three of these model uncertainties. There is a strong need for detailed observational studies of POCs, particularly studies with collocated aircraft in-situ measurements and ground/shipborne remote sensing, to determine whether POCs do indeed evince a fundamental mechanism whereby aerosols can influence MBL cloudiness.

Figure 1: GOES Visible satellite image showing extensive stratocumulus cloud cover over the SEP, punctuated by pockets of open cells (POCs).

A two-month "climatology" of the frequency of occurrence of open cellular convection during September/October 2000 has been constructed using a neural network method applied to MODIS data (Wood and Hartmann 2004). This clearly shows that open cells occur almost twice as frequently in the accessible regions (<1000 km from the coast) of the SEP than in the NEP (Figure 2). In addition to demonstrating the climatological significance of open cellular convection in regions dominated by marine stratocumulus, it also highlights the suitability of the SEP as a location for a field program to examine POCs. During September/October 2000, open cells were present almost 40% of the time at the IMET Buoy (85W, 20S) and around 20% of the time at San Felix Island (80W, 27S). However, there are almost no direct observations of the variability of aerosol or cloud droplet size distributions that we think may be associated with POCs. These observations are crucial to interpreting satellite observations that we do have, so that the link between aerosol variability, drizzle, and cloud organization can be convincingly made, and the role of anthropogenic vs. natural aerosol can be understood. In particular, open cell organization is also associated with largely non-precipitating shallow clouds, e.g. the climatological transition from stratocumulus to trade cumulus cloud regimes. These open cell convective clouds are usually seen further offshore, between 1500-2000 km from the coast in both the NEP and SEP, or in midlatitude cold air outbreaks. How much aerosol variability is affecting the cloud fraction in either region is an important question, and one which we currently do not know how to answer with satellite observations alone.

NE Pacific SE Pacific

Figure 2: "Climatology" of the frequency of occurrence of open cellular convection in the NE Pacific (left) and SE Pacific (right) subtropical regions during September/October 2000 (Wood and Hartmann, 2004).

2. Aerosol production and variability

We believe POCs are associated with spatiotemporal patchiness in boundary layer aerosol, but our understanding of the latter is quite scant, especially in the SEP. Chemical transport models suggest that there are significant sources of both anthropogenic and natural aerosol that can influence the near-coastal SEP (Chin et al. 1996), with an extremely strong fall-off in anthropogenic influence westwards. Satellite retrievals of the cloud droplet effective radius over the SEP mirror these strong zonal gradients (Han et al. 1994, see also Fig. 3 below), with an increase in effective radius away from the coast. The optical thickness of a cloud is inversely proportional to its effective radius and increases linearly with liquid water path (vertically integrated liquid water content). Therefore, the causes of variability in effective radius and liquid water path need to be understood in order to accurately determine the radiative properties of clouds.

Variability in the cloud effective radius is largely controlled by variability in the aerosol concentration, size distribution, and chemical composition, which are generally thought to play a more important role in marine boundary layer clouds than variations in updraft speed and thermodynamics. Aerosols that can act as nucleation sites for cloud droplets are termed cloud condensation nuclei (CCN) In the marine boundary layer, the concentration of CCN is strongly correlated with the concentration of aerosols with diameters larger than approximately 50 nm (often called the accumulation mode). Smaller aerosols are too small to efficiently act as CCN, and do not typically form cloud droplets. The effective radius tends to be large in a cloud growing in a low CCN boundary layer, because the available liquid water is shared among fewer droplets. It is therefore crucial to understand the sources and sinks of these particles if we are to understand the variability in cloud droplet sizes.

While CCN concentrations range from <10 to >200 cm^-3 in remote oceanic MBLs, the total aerosol concentration (including the Aitken, accumulation, and coarse modes) is remarkably stable (~300 cm^-3). This implies that there must be a source of gas-to-particle conversion (GPC) that continually restores the population (Clarke et al. 1998). There is considerable debate about whether the major source is actually within in the MBL. This stems from the idea that the MBL typically has a large aerosol surface area that condensable vapors will favor over the energetic barrier of new nucleation (Covert et al. 1996). However, a number of observational studies do show compelling evidence indicating that nucleation of new particles can occur in the MBL (Clarke et al. 2002, Kollias et al. 2004, Petters et al. 2004), especially when background aerosol concentrations have been lowered by precipitation. Indeed, the Kollias et al. study used data from the PACS 2003 cruise in the SE Pacific (part of VOCALS), indicating that this region is prone to nucleation events. Coastal regions are particularly prone to nucleation events (O'Dowd et al. 2002).

New nucleation results in very small particles that require several days to grow and coagulate to a size sufficient to be effective CCN (Hoell et al. 2000). Processes by which these particles are produced and lost through precipitation scavenging require investigation, though this is also being studied in other ongoing projects in the NEP.

Models of new nucleation generally do not favor new particle formation in the MBL (Raes et al. 1995, Clarke et al. 1998), but these have attempted to simulate only the GPC of sulphuric acid produced via the oxidation of dimethylsulfide (DMS) which is considered to be the main oceanic aerosol precursor species (Charlson et al. 1987). Mass spectrometry of recently nucleated particles in the MBL show large fractions of iodine-containing species (O'Dowd et al. 2002) that have been shown in smog-chambers to successfully nucleate even at high relative humidities that are unfavorable for sulphuric acid GPC. Iodine production over the coastal zone is related to algae, but this source alone is insufficient to account for the observed concentrations of certain iodine species over the open ocean.

Recent inventories of sulfur production from ships (Capaldo et al. 1999) suggest that even over the remote ocean, a significant fraction of the sulfate aerosol production can be anthropogenically produced. In the NE Atlantic and Pacific where several field programs to investigate aerosol-cloud interactions have been conducted (e.g. ASTEX, Albrecht et al. 1995; ACE-2, Johnson et al. 2000; DYCOMS-II, Stevens et al. 2003), estimates suggest that 50-80% of the atmospheric sulfur dioxide is ship-produced (Capaldo et al. 1999). Although subsequent research (Davis et al. 2001) has cast some doubt on the quantitative aspects of the Capaldo et al. findings, it seems likely that only in the remote regions of the Southern Hemisphere, such as the SEP, that a truly unpolluted aerosol environment can be found which provides an analogue to pre-industrial conditions over the oceans of the Northern Hemisphere.

Figure 3: Estimated median cloud droplet concentration N_d from MODIS data (Sept/Oct 2000). Estimates are deduced from cloud top effective radius and optical depth retrievals (King et al. 1997) using transformation to N_d of Szczodrak et al. (2001).

3. Ocean heat transport

The Peru current drives one of the strongest and most extensive regions of coastal upwelling which strongly depresses coastal SSTs. Instead of a uniform upwelling along the coast, with a surface offshore Ekman flow and a deeper return flow, it is now understood that much of the transport is turbulent and time-varying, with mesoscale eddies, Rossby, and coastally-trapped Kelvin waves playing an important role as in the NE Pacific (e.g. Kelly et al. 1998). The complexity of the processes involved in transporting heat from the coastal upwelling regions to regions further offshore has hindered a complete understanding the role of coastal upwelling in determining the climatological distribution of SST and its subseasonal variability over the SEP. It is also unclear whether other processes such as coastal wave-trapping are also of importance. Deployment of an IMET Buoy at 85W, 20S is in part motivated by the need to understand these ocean heat and momentum transport processes in the SEP, a region that has been much less intensively studied than the NE Pacific. However, the scope of a single measurement location is limited, especially as the processes involved are varying strongly in time and space. A spatial survey of the vertical structure of the oceanic wave and eddy field in this region, e.g. via towed SeaSoars (Pollard 1986) or other underwater gliders, would provide a rich context for understanding the buoy time series and altimetry data, as a well as a comparison with coupled and regional ocean model simulations.

4. Diurnal cycle

Clouds over the SEP exhibit a much stronger diurnal cycle of cloud cover (Rozendaal et al. 1995) and cloud liquid water path (Wood et al. 2002) than the MBL clouds at comparable latitudes in the northern hemisphere. Regional model simulations (Garreaud and Mu–oz, 2004) suggest that a large-scale subsidence wave formed by the interaction of the coastal jet along the Chilean coast with dry convective heating over the western Andean slopes can travel at least 1000 km over the SEP and lead to a strong diurnal cycle of subsidence. The phase of the wave may be such that it can strengthen the existing diurnal cycle of MBL depth and cloud liquid water path. Recent satellite measurements of surface divergence from Seawinds (Robert Wood, personal communication) suggest that the peak-to-peak amplitude of the divergence over much of the SEP can be 40-60% of the mean, with reduced subsidence during the night over the region of climatological maximum low cloud cover (Bretherton et al. 2004). This has a rectified effect on the clouds and their radiative properties because it allows the MBL depth to deepen more rapidly at night resulting in thicker, more water laden clouds (Garreaud and Mu–oz, 2004) compared with MBLs without the diurnal cycle of subsidence. These clouds are more prone to early-morning drizzle, and are then followed by rapid thinning aided by the increased daytime subsidence that reduces the efficiency with which the clouds can reflect sunlight back to space.

Further modeling work must be carried out to examine its vertical structure, phase speed, seasonal cycle and climatological significance. However, it is also important to observationally test these model predictions. Although the satellite divergence measurements provide one means to examine the diurnal cycle over the SEP, radiosonde measurements of the thermodynamic structure of the lower free-troposphere in the SEP will be invaluable in providing information concerning the wave's depth and phase speed that can be compared to mesoscale model simulations like those of Garreaud and Mu–oz (2004).

Specific questions to be addressed during VOCALS-RF

1. What factors influence the formation of drizzle in the remote and coastal SEP?

2. What are the physical processes responsible for generating pockets of open cells (POCs) within otherwise overcast stratocumulus cloud regimes? Does a POC cause aerosol depletion through scavenging, or is depleted aerosol a POC precursor?

3. What are the aerosol characteristics in both the coastal and remote SEP region? How do the contrasts affect cloud microphysics, and how are they related to POCs? What are the major sources and sinks of aerosol over the SEP, both natural and anthropogenic?

4. What are the major transport processes for coastally upwelled water to regions further offshore? Do mesoscale eddies play a dominant role? With current coupled ocean+atmospheric models at sufficient horizontal resolution, can the mean SST distribution in the SEP be successfully reproduced?

5. Do the depth, phase speed, and vertical structure, of the diurnal subsidence wave ("upsidence wave") originating in northern Chile/southern Peru agree with Garreaud and Muñoz simulations?  
   

6. Are current numerical models able to accurately simulate the synoptic variability in cloud cover and albedo in the SEP on daily timescales? If not, can we deduce whether model initialization data, boundary layer physics, or aerosol-cloud microphysics is chiefly to responsible?

Logistics of planned VOCALS 2007 field program

An effective study of the processes above cannot possibly be achieved with a single measurement platform, and so we will combine the capabilities of a ship, at least one instrumented aircraft, and ground based in-situ and remote sensing measurements at San Felix Island. Such a combination of observations has not been assembled in the NEP despite many coastally-focused stratocumulus studies off the Californian coast that rely mainly upon aircraft observations. The long, well-calibrated IMET buoy record has no counterpart in the NEP stratocumulus regime, nor is there an island in the NEP that is both well offshore but still in the stratocumulus region. Table 1 summarizes the chief measurement platforms, their coverage and the broad scientific goals to be addressed.

VOCALS 2007 Study Region

The VOCALS study region is shown in Fig. A1, and comprises that part of the SEP east of 90W and between 15S and 30S. All field campaign operations will be conducted within the study region. San Felix Island is at 80W, 27S. The IMET Buoy is at 85W, 20S. These locations will form the NW and SE extrema of a "radiator fin" pattern (Fig. A1) to be carried out over a four week period by the NOAA R/V Ronald H Brown. This pattern is designed to maximize the sampling of ocean currents and ocean surface properties, particularly its mesoscale variability, in the region between the S. American coast and the remote SEP to the west of the Peru current. A schematic cross section from coastal northern Chile to the IMET Buoy is given in Fig A2.

Table 1: Measurement platforms, their coverage, and main science goals

Measurement Platform	Coverage/sampling	Main science goals
NOAA R/V Ronald H Brown (ship)	"Radiator fin" pattern originating close to the S American coast and moving to the WSW over a period of 4 weeks (Fig A1)	• Cloud-aerosol-drizzle interactions • MBL structural properties and energy/moisure budgets • Ocean heat transport, SST and ocean current mapping • Diurnal cycle of MBL and free-troposphere
NSF NCAR C-130 (aircraft)	(a) Lagrangian-type flights to follow and map out POCs and surrounding stratocumulus (Fig A3) (b) Cross sectional flights between Arica (Chile) and either the IMET Buoy or San Felix Island (Figs A1 and A4)	• Cloud-aerosol-drizzle interactions • Aerosol characterization, sources/nucleation • MBL and cloud structure between the S American coast and the remote SEP
IMET Buoy (85°W, 20°S)	Fixed location, surface sampling of MBL properties, radiation, SST, ocean currents (Fig A1)	• Daily to interannual variability of SST, cloud and MBL properties. • Ocean heat transport
San Felix Island (80°W, 27°S)	Fixed location sampling of aerosol, cloud, and drizzle properties. Possible deployment of the ARM Mobile Facility (AMF) for period around field program.	• Cloud-aerosol-drizzle interactions • POCs structure and initiation • Aerosol sources/sinks in the remote SEP
Coastal Chile/Peru	TBD	• Coastal meteorology and cloud cover

Figure A1: Map of the southeast Pacific (SEP), and South American orography. The VOCALS 2007 study region comprises the region between latitudes 15S and 30S, and all ocean east of 90W. The proposed track of the Ronald H Brown is shown, along with the locations of the IMET Buoy, San Felix Island, and the flight tracks for the cross-section flights with the NCAR C-130.

Figure A2: Schematic showing an E-W cross section from the IMET Buoy (85W, 20S) to the S. American coast, detailing the lower atmospheric and upper oceanic structure and some important processes.

Aircraft flight plans (tentative)

The NCAR C-130 will be the primary aircraft involved in VOCALS 2007. There is some possibility that other aircraft will participate in the field phase (e.g. The Met Office BAe-146), but these are not crucial to the success of the program. Two flight plans (Table 1) have been proposed for C-130 activities:

(a) Cross-section missions (Fig A3: These flights are designed to examine contrasts between the MBL and lower atmosphere close to the S. American coast and that in the remote SEP. Specific focus will be placed on examination of aerosol characteristics, MBL structure/depth, cloud morphology, microphysics, and drizzle production (using the University of Wyoming 94 GHz cloud radar, WCR, on the C-130). Other interests include cloud remote sensing, and there will be efforts made to coordinate some of these missions with satellite overpasses, particularly those of Terra and Aqua/A-train at roughly 10:30 am/pm and 1:30 am/pm local time. The essential idea is to fly along the 20S parallel out from the coast to the IMET Buoy. On the way out, the flight will comprise a set of straight and level runs below (30 m AMSL daytime, 150 m AMSL nighttime) and above (1800 m AMSL) cloud, and to sample the cloud layer itself using slant profiles ascending at some 150 m/minute. The precise set of runs used will be dependent upon the needs of the key participants. The above cloud runs will be used to sample the drizzle structure using the MMCR, to characterize the free-tropospheric aerosol and thermodynamic structure, and to use the shortwave imaging radiometer to remotely sense the cloud bulk and microphysical properties. Below cloud runs will aim to determine physicochemical properties of aerosols, and to determine the lower boundary conditions (SST, surface fluxes, winds). The return leg back from the IMET buoy will be carried out at approximately 6km with dropsondes being launched at regular intervals. Because drizzle is expected to be more prevalent during the night, flights will be flown both in the day and at night. This type of flight plan was implemented very successfully during EPIC 2001 for studying southerly cross-equatorial boundary layer inflow into the East Pacific ITCZ (e.g. Raymond et al. 2004, deSzoeke and Bretherton 2004).

(b) POC missions (Fig A4: These flights are designed specifically to examine microphysical and dynamical processes occuring both in pockets of open cells (POCs) and in the surrounding cloud. POCs that are completely surrounded by overcast stratocumulus clouds are of the most interest, but boundaries between open and closed cellular convection will also be a focus of these missions. Of particular importance will be a characterization of the aerosol and cloud microphysics in the two regions. If possible, these flights will be coordinated with the ship which will be able to provide a scanning radar capability (C-band on the Ronald H Brown) as well as aerosol and cloud characterization. The idea is to use geostationary satellite imagery to locate POCs, and then to target missions accordingly. Once a POC boundary has been reached, the aim is to carry out across-wind stacks of five straight and level runs below, in, and above cloud (with additional porpoising runs to characterize the cloud top and inversion layers). The aircraft will be allowed to drift with the MBL mean wind (i.e. with the advecting POC).

Figure A3: Flight plan for C-130 cross-sectional missions. The flight will consist of a set of slant profiles and straight and level runs from the coast out to either the IMET Buoy, or to San Felix Island. Runs above cloud will be used to remotely sense the cloud and drizzle structure. Profiles will be used for in-situ characterization of the cloud and drizzle. On the return leg at approximately 6 km altitude, a series of dropsondes will be launched to map out the vertical structure of the MBL and lower free troposphere across the region.

Figure A4: Lagrangian-type C-130 flight plan to sample POCs and surrounding overcast stratocumulus. Flight will consist of 4-6 crosswind stacks drifting with the mean MBL wind (left). Each stack will consist of straight and level runs close to the surface, below mean cloud base, just above cloud base, below cloud top, and above cloud. Each run will be approximately 120 km in length, with roughly equal fractions inside and outside of the POC region.

References

Baker, M. B., 1993: Variability in concentrations of cloud condensation nuclei in the marine cloud-topped boundary layer. Tellus, 45B, 458-472.

Breon, F. M., D. Tanre, and S. Generoso, 2002: Aerosol effect on cloud droplet size monitored from satellite. Science, 295, 834-838.

Bretherton, C.S, and M.C. Wyant, 1997: Moisture transport, lower tropospheric stability and de coupling of cloud-topped boundary layers. J. Atmos. Sci., 54, 148-167.

Charlson, R. J., J. E. Lovelock, M. O. Andreae, and S. G. Warren, 1987: Oceanic phytoplankton, atmospheric sulphur, cloud albedo and climate. Nature, 326, 655-661.

Chin, M., D. J. Jacob, G. M. Gardner, P. A. Spiro, M. Foreman-Fowler and D. L. Savoie, 1996: A global three-dimensional model of tropospheric sulfate, J. Geophys. Res., 101, 18,667-18,690.

Clarke A. D., D. Davis, V. N. Kapustin, F. Eisele, G. Chen, I. Paluch, D. Lenschow, A. R. Bandy, D. Thornton, K. Moore, L. Mauldin, D. Tanner, M. Litchy, M. A. Carroll, J. Collins, C. Albercook, 1998: Particle nucleation in the tropical boundary layer and its coupling to marine sulfur sources, Science, 282, 89-92.

Covert, D. S., V. N. Kapustin, T. S. Bates, and P. K. Quinn, 1996: Physical properties of marine boundary layer aerosol particles of the mid-Pacific in relation to sources and meteorological transport. J. Geophys. Res., 101 , 6919-6930.

Davis, D. D., G. Grodinsky, P. Kasibhatla, J. Crawford, G. Chen, S. Liu, A. Bandy, D. Thornton, H. Guan, and S. Sandholm, 2001: Impact of emissions on marine boundary layer NO_x and SO₂ distributions over the Pacific basin, Geophys. Res. Lett., 28, 235-238.

Han, Q, W. B. Rossow, and A. A. Lacis, 1994: Near-global survey of effective droplet radii in liquid water clouds using ISCCP Data. J. Clim., 7, 465-497.

Hoell, C., C. O'Dowd, S. R. Osborne, D. Johnson, 2000: Time-scale analysis of marine boundary layer aerosol evolution: Lagrangian case studies under clean and polluted cloudy conditions, Tellus, 52B, 423-438.

Kelly, K. A., R. C. Beardsley, R. Limeburner, K. H. Brink, J. Paduan, and T. K. Chereskin, 1998: Variability of the near-surface eddy kinetic energy in the California Current based on altimetric, drifter, and moored current data. J. Geophys. Res., 103, 13067-13083.

Lohmann, U. and J. Feichter, 1997: Impact of sulfate aerosols on albedo and lifetime of clouds: A sensitivity study with the ECHAM4 GCM. J. Geophys. Res., 102, 13685-13700.

Ma, C.-C., C. R. Mechoso, A. W. Robertson and A. Arakawa, 1996: Peruvian stratus clouds and the tropical Pacific circulation - a coupled ocean-atmosphere GCM study. J. Clim., 9, 1635-1645.

Pollard, R., 1986: Frontal surveys with a towed profiling conductivity temperature depth measurement package (SeaSoar), Nature, 323, 433-435.

Mechoso, C. R., A. W. Robertson, N. Barth, M. K. Davey, P. Delecluse, P. R. Gent, S. Ineson, B. Kirtman, M. Latif, H. Le Treut, T. Nagai, J. D. Neelin, S. G. H. Philander, J. Polcher, P. S. Schopf, T. Stockdale, M. J. Suarez, L. Terray, O. Thual and J. J. Tribbia, 1995: The seasonal cycle over the Tropical Pacific in general circulation models. Mon. Wea. Rev., 123, 2825-2838.

Nicholls, S., 1987: A model of drizzle growth in warm, turbulent, stratiform clouds. Quart. J. Roy. Meteorol. Soc., 113, 1141-1170.

O'Dowd C.D. , J. L. Jimenez, R. Bahreini, R. C. Flagan, J. H. Seinfeld, K. Hameri, L. Pirjola, M. Kulmala, S. G. Jennings, and T. Hoffmann, 2002: Marine aerosol formation from biogenic iodine emissions, Nature, 417, 632-636.

Pierrehumbert, R. T. 1995: Thermostats, Radiator Fins, and the Local Runaway Greenhouse. J. Atmos. Sci. 52, 1784-1806.

Raes, F., 1995: Entrainment of free tropospheric aerosols as a regulating mechanism for cloud condensation nuclei in the remote marine boundary layer. J. Geophys. Res., 100, 2893-2903.

Raymond, D. J., S. K. Esbensen, M. Gregg, and C. S. Bretherton, 2004: EPIC2001 and the coupled ocean-atmosphere system of the tropical east Pacific. Bull. Am. Meteor. Soc., in press.

Szczodrak, M., P. H. Austin, and P. B. Krummel, 2001: Variability of optical depth and effective radius in marine stratocumulus clouds. J. Atmos. Sci., 58, 2912-2926.

de Szoeke, S.P. and C.S. Bretherton, 2004: Quasi-Lagrangian large eddy simulations of cross-equatorial flow in the east Pacific atmospheric boundary layer. J. Atmos. Sci., 61, 1837-1858.