Observational Strategies from the Micro to the Meso Scale

Jean-Louis Brenguier and Robert Wood

Abstract

1. Introduction (RW)

Changes in clouds represent the most significant source of uncertainty in our ability to predict the response of the climate to anthropogenic emissions (IPCC 2007). Our current understanding is that two distinct types of response are likely to dominate the cloud radiative response: (a) cloud feedbacks associated with the response of clouds to changes in the large-scale circulation induced by increasing greenhouse gas concentrations (Cess et al. 1989, Colman 2003, Soden and Held 2006); (b) changes in cloud optical properties due to the indirect effects of anthropogenic aerosols (Haywood and Boucher 2000, Lohmann and Feichter 2005). Both types of cloud response necessarily involve processes occurring over a wide range of scales, but the spatiotemporal scale of the perturbation itself is fundamentally different because of the great difference in residence time for greenhouse gases (few years) and anthropogenic aerosol emissions (few days). This scale difference presents important constraints on strategies for understanding anthropogenically-induced cloud changes. Thus, while it would be futile to conduct plume experiments to examine the cloud response to greenhouse gases, this is not the case for anthropogenic aerosols.

One of the biggest and often unappreciated challenges we face when attempting to understand the cloud radiative response, is that the anthropogenic perturbations are small in comparison with natural variability. Over much of the globe the mean shortwave cloud radiative forcing is 30-90 W/m2 (Hartmann and Michelsen 1993) whereas the global mean perturbations due to cloud feedbacks and anthropogenic aerosols are estimated to be of order 1-3 W/m2 (Colman 2003, IPCC 2007). Coupled with the large spatiotemporal variability in cloud fields, this makes the attribution of cloud changes to particular anthropogenic perturbations very difficult to detect using observations. While there is likely to be some regional enhancement of the cloud response (particularly for aerosols), this 30-fold difference is seldom addressed when planning observational studies. For the case of the cloud response to changing greenhouse gases, this constraint effectively means that observational strategies must focus upon regional and interannual variability, together with a better understanding of the connections between the large scale meteorological fields and the small-scale cloud response. For the response to anthropogenic aerosols the picture is perhaps clearer owing to the greater regional enhancement of the effect due to localized aerosol sources, and the greater degree of day-to-day temporal and mesoscale spatial variability in aerosol concentrations due to modulation by variability in the large-scale flow.

The cloud radiative response to anthropogenic aerosols originates from changes in the microphysical properties of the clouds (Twomey 1977) that subsequently affect cloud macrophysical properties by altering the energetics of the cloud system (Haywood and Boucher 2000). For low clouds, the chief energetic impacts of changing microphysical properties are not directly radiative but occur through modifications to precipitation and entrainment processes (Albrecht 1989, Wang et al. 2003, Ackerman et al. 2004, Wood 2007, Bretherton et al. 2007). However, changes to the surface radiative balance may also be important (Jiang and Feingold 2006), but the timescale of the response of surface fluxes would be expected to be quite long over the oceans.

(e.g. diagnostics such as lower tropospheric variability and estimated inversion strength that have been

For aerosols, our ability to detect unequivocal is perhaps more hopeful - test changes

In a changing climate, clouds are likely to be perturbed, in fact the whole hydrological cycle will be perturbed, with significant changes in the general circulation, cloud cover and spatial distribution and intensity of precipitation. Cloud are thus sensitive to CO2 concentration changes via the large scale. However, they are also sensitive to anthropogenic aerosol, but via the microscale. Overall, one can see clouds responding to climate change, since clouds are an intrinsic part of climate, but their response will be modulated by anthropogenic aerosol. The modulation is the topic of this paper.

2. HOW AEROSOL IMPACT CLOUD MICROPHYSICS

2.1 Condensation and ice nuclei

Aerosol serve as a support for the formation of cloud primary particles, both liquid and ice.

2.1.1 CCN (JLB)

Briefly describe the role of CCN, what they are, from where they come, how anthropogenic aerosol contributes to the CCN population

2.1.2 IN (RW)

The formation of ice particles in clouds at temperatures warmer than approximately -35 C requires heterogeneous ice nuclei (IN). At colder temperatures homogeneous freezing of haze particles is the likely dominant mechanism for ice particle formation, but heterogeneous freezing may also play a role (Karcher and Lohmann 2002, .....). In-situ observational studies that clearly demonstrate a link between aerosol physicochemical properties and ice formation are lacking. Both theoretical treatments (Karcher and Lohmann 2002) and observations (Strom...) suggest a weak dependence of ice crystal concentration upon the aerosol size distribution at cold temperatures. It is unlikely that a significant effect analogous to the Twomey effect in warm clouds exists in clouds formed by homogeneous freezing.

At warmer temperatures a host of potential heterogeneous ice nucleation mechanisms exist but few detectors exist that can accurately distinguish between them. The only widely used ice nucleus counter (Rogers 1993) only measures the ability of aerosols to form ice by heterogeneous deposition, whereas several studies observe that ice particles often form in large numbers only after liquid water droplets have formed (see Field et al. 2001). Observed ice crystal concentrations frequently exceed the concentrations of deposition IN measured (Verlinde et al. 2007) effectively refuting the hypothesis that deposition nucleation is of major importance in the atmosphere. There are no in-situ observational techniques to measure heterogeneous nucleation by contact freezing, immersion freezing, and ice formed by freezing of evaporating droplets. Without investment in instrumental techniques to make such measurements it will be difficult to make progess in understanding the importance of these processes in the atmosphere and to improve parameterizations for either process-based or large-scale numerical models.

References:

Discrimination of micrometre-sized ice and super-cooled droplets in mixed-phase cloud. Hirst E., Kaye P H., Greenaway R S., Field P., and Johnson D W. Atmospheric Environment 35, 1, 33-47, 2001.

Rogers, D.C., Measurements of natural ice nuclei with a continuous flow diffusion chamber. Atmospheric Research, 29, 209-228, 1993.

2.2 Primary cloud particle formation

2.2.1 CCN activation (JLB)

How to move from CCN to CDNC, adding w. Evidence of correlation between CCN and CDNC conc

2.2.2 Homogeneous freezing and IN activation (RW)

The key problems hampering observational studies of both homogeneous freezing and heterogeneous IN activation are the lack of accurate information on vertical velocity (ref...) for upper tropospheric clouds, the poor differentiation between liquid and ice for particles in the 1-10 micron size range (ref...), and ongoing difficulties in measuring supersaturations at very low temperatures (Jensen et al. 2005).

Vertical velocity measurements from aircraft have an absolute accuracy of approximate 20-30 cm/s (Khelif et al. 1999) which exceeds the typical mean ascent speed for many upper tropospheric clouds. This hampers closure studies of ice formation. Improvements in aircraft positioning and air motion detection have been made in the past ten years but real improvements on the vertical wind accuracy have yet to be demonstrated on commonly used research aircraft.

The measurement of supersaturation, especially for the low moisture contents common in the upper troposphere, has improved dramatically over the last decade, due to more sensitive measurements of water vapor using fluorescence techniques and tunable laser diodes (see Jensen et al. 2005 and references therein). However, uncertainties still exist regarding the absolute accuracy of the supersaturation measurements which leaves the implications for ice formation open to debate (Peter et al. 2007). Still, it is likely that improvements will continue to be made which makes this area potentially fruitful.

New instruments that can better distinguish ice from liquid droplets in the 1-10 micron range are being developed (Hirst et al. 2002) which should provide important information regarding the possible microphysical controls on ice formation. These measurements are able to determine the sphericity of individual particles (rather than simply the bulk properties of a large sample) and are already providing important information for determining the properties of mixed phase clouds (ref???).

Since the realization in the mid 1980s (Gardiner and Hallett 1986) that ice particle measurements from aircraft can be affected by shattering on probe inlets, there has been a vigorous debate regarding the concentrations of ice crystals in cold clouds. High ice crystal concentrations are frequently inferred from measurements using forward scattering probes in cirrus clouds (e.g. ???) but the use of particle inter-arrival time information from modified probes has reinvigorated the debate, and results suggest that ice particle concentrations might be biased high by a factor of 2-5 by shattering (Field et al. 2003, 2006). These issues must be taken into account when designing new probes.

References:

Khelif, D., S.P. Burns, and C.A. Friehe, 1999: Improved Wind Measurements on Research Aircraft. J. Atmos. Oceanic Technol., 16, 860–875.

Jensen, E.J., J.B. Smith, L. Pfister, J.V. Pittman, E.M. Weinstock, D.S. Sayres, R.L. Herman, R.F. Troy, K. Rosenlof, T.L. Thompson, A.M. Fridlind, P.K. Hudson, D.J. Cziczo, A.J. Heymsfield, C. Schmitt, and J.C. Wilson, 2005: Ice supersaturations exceeding 100% at the cold tropical tropopause: Implications for cirrus formation and dehydration. Atmos. Chem. Phys., 5, 851-862.

When dry air is too humid Author(s): Peter T (Peter, Thomas), Marcolli C (Marcolli, Claudia), Spichtinger P (Spichtinger, Peter), Corti T (Corti, Thierry), Baker MB (Baker, Marcia B.), Koop T (Koop, Thomas) Source: SCIENCE 314 (5804): 1399-+ DEC 1 2006

2.3 How cloud microphysics develop depending on primary cloud particles

2.3.1 Liquid phase (JLB)

Higher CDNC, smaller droplets, less precipitation embryos

2.3.2 Ice clouds (RW)

Do we see a change in cirrus microphysics related to IN?

2.3.3 Mixed phase clouds (JLB)

Apparently mixed phase clouds depends more on CCN than on IN via ice multiplication

2.4 How cloud radiative properties are modulated by the aerosol

The Twomey effect

2.4.1 Ship tracks (RW)

Shiptracks are linear perturbations of stratus and stratocumulus clouds which form in response to the emissions of submicron sulfate particles and their precursors from diesel powered ships (Durkee et al. 2000). The Monterey Areas Shiptracks experiment in 2000 (Durkee et al. 2000) was an excellent example of a hypothesis-driven field experiment (rare in the atmospheric sciences) from which it became clear that aerosols are the dominant cause of the tracks. Shiptracks are therefore a unique laboratory to test hypotheses about how aerosols impact cloud microphysical, macrophysical and radiative properties.

Analysis of many shiptracks using satellite data have revealed that although the cloud droplet effective radius decreases in shiptracks, the cloud optical thickness may increase or decrease, with a mean response that is close to zero (Coakley and Walsh 2002). The results constitute direct refutation of the Twomey hypothesis whereby cloud optical thickness would be expected to increase in response to increased CCN, and suggest that liquid water content in ship-influenced clouds are reduced compared with the surrounding cloud. This surprising result can be readily understood, however, to be a response whereby the increased aerosol concentration reduces the coalescence rate, invigorates the boundary layer eddies, and drives a stronger entrainment of warm, dry air into the MBL (Ackerman et al. 2004, Wood 2007). This raises the lifting condensation level and thins the cloud.

Given the wealth of data from satellites, especially those in the A-Train constellation (ref??), it seems logical that these data should be used to further probe the role that aerosols play in cloud perturbations due to shiptracks. The combination of the different satellite sensors will give critically important information on the precipitation suppression in the tracks (CloudSat ), the perturbation in MBL depth (CALIPSO), together with the cloud microphysical and optical properties (MODIS). This combination will lead to greatly improved understanding of the overall system response to anthropogenic perturbations, of the mechanisms for this response, and of the time and space scales over which the responses might be expected.

References:

Coakley, J.A., and C.D. Walsh, 2002: Limits to the Aerosol Indirect Radiative Effect Derived from Observations of Ship Tracks. J. Atmos. Sci., 59, 668–680.

Durkee, P.A., K.J. Noone, R.J. Ferek, D.W. Johnson, J.P. Taylor, T.J. Garrett, P.V. Hobbs, J.G. Hudson, C.S. Bretherton, G. Innis, G.M. Frick, W.A. Hoppel, C.D. O'Dowd, L.M. Russell, R. Gasparovic, K.E. Nielsen, S.A. Tessmer, E. Öström, S.R. Osborne, R.C. Flagan, J.H. Seinfeld, and H. Rand, 2000: The Impact of Ship-Produced Aerosols on the Microstructure and Albedo of Warm Marine Stratocumulus Clouds: A Test of MAST Hypotheses 1i and 1ii. J. Atmos. Sci., 57, 2554-2569.

2.4.2 Closure studies AC2, ARM (JLB)

2.4.3 Satellite observations (does it really shows changes in albedo?) (JLB)

3. HOW AEROSOL IMPACT CLOUD DYNAMICS, COVER, LIFE CYCLE AND PRECIPITATION

It has been known from modeling studies for some time (e.g. Arnason and Greenfield 1972) that microphysical changes can impact warm cloud macrophysical properties by changing the cloud dynamics. These dynamical responses are complicated and operate in a number of distinct ways. First, the condensation rate of a cloud droplet is dependent upon its radius. The timescale for droplet growth or evaporation is therefore microphysically controlled which can have a direct impact upon the energetics of the cloud system (Kogan and Martin 1994). Second, changes to the cloud droplet size influence the sedimentation rate of the cloud droplet population. This can remove water from the cloud top interface (Bretherton et al. 2007) and influence rate of coalescence in the cloud and can modify the precipitation that a cloud of a given thickness can produce (Albrecht 1989). Both of these latter effects modify the boundary layer energetics and entrainment rates which can feedback on the cloud macrophysical structure. In colder clouds there are also likely to be microphysical impacts upon the freezing process which may alter the latent heating profiles.

While there are good theoretical bases for understanding microphysical impacts upon cloud macrophysical properties, obtained observations to test these ideas has been particularly challenging. This is primarily because the microphysical impacts are often masked by natural variability in the clouds under study. For example, based upon our current understanding of the warm rain formation process in stratocumulus, the precipitation rate change expected from a 100 m increase in cloud thickness is the same as that arising from a threefold reduction in cloud droplet concentration (Pawlowska and Brenguier 2003). Most observational field studies do not measure the cloud thickness to this accuracy. Further, the case for needing to make these accurate measurements is not often articulated when planning field campaigns. In shallow and deep cumulus clouds, which are far less homogeneous than stratocumuli, the masking of microphysical effects by natural variability is yet more difficult to address.

3.1 Stratocumulus clouds

There are a number of recent studies that have demonstrated a clear impact of cloud droplet concentration upon the formation of light precipitation in stratocumulus clouds (Pawlowska and Brenguier 2003, Comstock et al. 2004, Wood 2005, VanZanten et al. 2005), but such observations require multiple case studies and demonstrate the need to account for the masking effects of natural meteorological variability. A summary of these results is shown in Fig. XX, which shows precipitation rates at cloud base (Rcb) from three recent field programs focusing upon subtropical marine stratocumulus clouds. For each field campaign (Rcb) scales well with h^3/Nd where h is the cloud thickness and Nd is the mean cloud droplet concentration. However, each dataset appears to have offsets that, while possibly caused by differences in large-scale meteorology at the different locations, may also reflect measurement biases. While errors of 30% or so are inherent in the precipitation rate measurements (Pawlowska and Brenguier 2003, Comstock et al. 2004), it should also be noted that biases of only 40-80 m in the estimation of cloud thickness are sufficient to bring these data into considerably better alignment. The observed scaling emphasizes the importance of controlling for meteorological variability when examining microphysical impacts, and highlights just how sensitive precipitation rates can be to the meteorological state.

* _Figure [DRIZZLE]: Precipitation rates measured at cloud base height from observational case studies in subtropical marine stratocumulus, plotted against the ratio of the cube of cloud thickness to the cloud droplet concentration. For the ACE-2 and DYCOMS-2 data, each point represents observations from an entire aircraft flight with precipitation measured using in-situ probes (ACE-2, Pawlowska and Brenguier 2003) and millimeter radar (DYCOMS-2, VanZanten et al. 2005), and droplet concentration measured using in-situ probes. Each data point for EPIC represents a 3 hour long period of surface-based remote sensing measurements of the cloud thickness, precipitation and cloud droplet concentration (see Comstock et al. 2004 for further details). The lines represent linear least-distance regressions to the case studies for each field campaign._

The response of stratocumulus clouds to perturbations in precipitation is poorly understood, but observations from recent field campaigns suggest that precipitation can exert a dramatic impact upon the macrophysical cloud structure through reorganization of the mesoscale variability that exists in these clouds (Stevens et al. 2005, Comstock et al. 2005). Rapid reductions in cloud cover occur when cloud base precipitation rates exceed a few millimeters per day. Such transitions from closed to open mesoscale cellular convection tend to occur during periods of very low cloud droplet concentration (Wood et al. 2008). However, one must be cautious to interpret these transitions as being driven by microphysical processes, when in fact reductions in cloud droplet concentration are expected during periods of intense drizzle through coalescence scavenging of CCN (Wood 2006). Thus, there is a complex interplay between the microphysical and macrophysical properties in stratocumulus that must be understood before examining how anthropogenic effects may perturb these clouds.

Even in the absence of dramatic changes in the mesoscale structure, precipitation perturbations can impact the thickness of stratocumuli by changing the turbulent energetics of the boundary layer. Precipitation dampens turbulence kinetic energy (TKE) production in the boundary layer as a whole by stabilizing it, and so reductions in precipitation (microphysical or otherwise) should energize the boundary layer and cause it to entrain air from aloft at a higher rate (Stevens 1998). Enhanced entrainment can result in a thinning of the cloud if the free-troposphere is sufficiently dry (Ackerman et al. 2004) but the exact response is governed by the balance of several competing processes on different timescales (Wood 2007). Observational studies have contributed substantially to understanding the processes driving entrainment in stratocumulus clouds (see Stevens et al. 2003 and discussion/references therein), but current uncertainties in the measurements of entrainment rate from in-situ studies (Faloona et al. 2005) make it difficult to difficult between existing relationships linking bulk measures of turbulence with the entrainment rate.

The question of how to constrain the impact of precipitation on TKE using observations is also a difficult one. One suggestion is to conduct Lagrangian studies (Businger et al. 2006) on shiptracks to identify the impact of precipitation suppression upon boundary layer turbulence and entrainment. One problem with such studies is that it must be appreciated that changes in the cloud field due to point sources of aerosols may differ from those resulting from an increase in the "background" levels. Small-scale perturbations to the inversion height will quickly be communicated to the surrounding cloud by gravity waves which will affect the local response. Another suggestion is to build a large database of stratocumulus turbulence measurements with the accompanying microphysical, and most importantly precipitation measurements. The study of Nicholls and Leighton (1986) demonstrated that the turbulent structure of stratocumulus cloud layers can be understood by considering the various forcings on the boundary layer, but at the time the measurements of precipitation necessary to quantify its impact upon the turbulent structure were limited to in-situ microphysical data which have large sampling errors. With modern millimeter radars, this deficiency is now removed......

3.2 Shallow cumulus clouds

Shallow cumulus clouds

References:

Ackerman, A.S., M.P. Kirkpatrick, D.E. Stevens, and O.B. Toown, 2004: The impact of humidity above stratiform clouds on indirect aerosol climate forcing. Nature, 432, 1014-1017, doi:10.1038/nature03174.

Albrecht, B. A., 1989: Aerosols, cloud microphysics and fractional cloudiness. Science, 245, 1227-1230.

Arnason, G., and R. S. Greenfield, 1972: Micro- and macro-structures of numerically simulated convective clouds. J. Atmos. Sci., 29, 342-367.

Bretherton, C. S., P. N. Blossey, and J. Uchida, 2007: Cloud droplet sedimentation, entrainment efficiency, and subtropical stratocumulus albedo. Geophys. Res. Lett., 34, L03813, doi:10.1029/2006GL027648.

Businger, S., R. Johnson, and R. Talbot, 2006: Scientific Insights from Four Generations of Lagrangian Smart Balloons in Atmospheric Research. Bull. Amer. Meteor. Soc., 87, 1539–1554.

Comstock, K. K., Wood, R., Yuter, S. E., and Bretherton, C. S., 2004: Reflectivity and rain rate in and below drizzling stratocumulus. Quart. J. Roy. Meteorol. Soc., 130, 2891-2919, 2004.

Faloona, I., D.H. Lenschow, T. Campos, B. Stevens, M. van Zanten, B. Blomquist, D. Thornton, A. Bandy, and H. Gerber, 2005: Observations of Entrainment in Eastern Pacific Marine Stratocumulus Using Three Conserved Scalars. J. Atmos. Sci., 62, 3268–3285.

Kogan, Y. L., and W. J. Martin, 1994: Parameterization of bulk condensation in numerical cloud models. J. Atmos. Sci., 51, 1728-1739.

Nicholls, S., and J. Leighton, 1986: An observational study of the structure of stratiform cloud sheets: Part 1: Structure. Quart. J. Roy. Meteorol. Soc., 112, 431-460.

Pawlowska, H., and J. Brenguier (2003), An observational study of drizzle formation in stratocumulus clouds for general circulation model (GCM) parameterizations, J. Geophys. Res., 108(D15), 8630, doi:10.1029/2002JD002679

Stevens, B., 1998: Large eddy simulations of strongly precipitating, shallow, stratocumulus-topped boundary layers. J. Atmos. Sci., 55, 3616-3638.

Stevens, B., D.H. Lenschow, G. Vali, H. Gerber, A. Bandy, B. Blomquist, J.L. Brenguier, C.S. Bretherton, F. Burnet, T. Campos, S. Chai, I. Faloona, D. Friesen, S. Haimov, K. Laursen, D.K. Lilly, S.M. Loehrer, S.P. Malinowski, B. Morley, M.D. Petters, D.C. Rogers, L. Russell, V. Savic-Jovcic, J.R. Snider, D. Straub, M.J. Szumowski, H. Takagi, D.C. Thornton, M. Tschudi, C. Twohy, M. Wetzel, and M.C. van Zanten, 2003: Dynamics and Chemistry of Marine Stratocumulus—DYCOMS-II. Bull. Amer. Meteor. Soc., 84, 579–593.

Stevens, B., G. Vali, K. Comstock, R. Wood, M. VanZanten , P.H. Austin, C.S. Bretherton, D.H. Lenschow, 2005: Pockets of Open Cells (POCs) and drizzle in marine stratocumulus. Bull. Am. Meteorol. Soc., 86, 51-57, 2005

vanZanten, M.C., B. Stevens, G. Vali and D. Lenschow 2005: Observations of drizzle in nocturnal marine stratocumulus. J. Atmos. Sci. 62, 88-106.

Wood, R., 2006: The rate of loss of cloud droplets by coalescence in warm clouds. J.Geophys. Res., 111, D21205, doi:10.1029/2006JD007553.

Wood, R., 2007: Cancellation of aerosol indirect effects in marine stratocumulus through cloud thinning. J. Atmos. Sci., 64, 2657-2669.

3.3 deep mixed phase convection (JLB)

xxx

4. AEROSOL RADIATIVE IMPACTS (RW)

In addition to acting as CCN/IN, anthropogenic aerosols can also impact the radiative properties of clouds by direct absorption of radiation, which can alter the stability by changing the heating profile (the semi-direct effect, Hansen et al. 1997), and by altering the surface energy and/or moisture budgets. The spatial distribution of aerosols is dominated by mesoscale variability (Anderson et al. 2001) but this can feed back on the system on smaller scales (e.g. by influencing the energy available to drive convection or boundary layer turbulence). This contrasts with the role of aerosols as CCN/IN which influences cloud mesoscale variability through an upward cascade originating at the microscale. However, the two types of aerosol effects will frequently interfere with one another in complex ways.

Different observational challenges and opportunities therefore exist for the study of the semi-direct aerosol effects.

Hansen J., M. Sato, and R. Ruedy, 1997: Radiative forcing and climate response, J. Geophys. Res., 102, 6831-6864.

ORIGINAL: How it impacts clouds from the mesoscale as opposed to the CCN/IN effects that proceed from micro to meso

5. STRATEGIES FOR OBSERVATIONAL APPROACHES

Stop with correlations ! Strategies shall be adapted to the cloud type and necessarily proceed with the synergy between observations and models to understand what are the mechanisms. For each one, describe why it is so difficult to separate meteo from aerosol imapcts

Importance of multi-sensor studies using aircraft, satellites, and importance of model-driven hypothesis testing in observational studies.

5.1 BL clouds (JLB)

what we already said about model features that are observable

5.2 Shallow convection (?)

still to think

5.3 deep mixed phase convection (?)

still to think even more

5.4 Radiative effect (?)

still to think even more

References

Jiang, H.,and G. Feingold, 2006: The effect of aerosol on warm convective clouds: Aerosol-cloud-surface flux feedbacks in a new coupled large eddy model. J. Geophys. Res., 111, D01202, doi:10.1029/2005JD006138.

-- RobWood - 15 Oct 2007