Workshop on Convection, Water Vapor, and Climate

March 27 - 29, 2012



Dorian Abbot, University of Chicago



Title: Cloud effect on Snowball Earth deglaciation

Author(s): Dorian Abbot

Abstract: Influential past results from the FOAM global climate model showed a Snowball Earth climate that remained cold even at very high CO2, indicating that it might be very hard to escape from a Snowball Earth by increasing CO2. I present results from the FOAM, CAM, SP-CAM, LMDz, ECHAM, and GENESIS global climate models run in a standardized Snowball Earth configuration with CO2=100 ppm and CO2=1e5 ppm (0.1 bar). FOAM is dramatically cooler than the other models. I show that this is because FOAM produces very little cloud cover compared to the other models, and clouds are generally warming in a Snowball Earth. More generally, cloud behavior is the main driver of intermodel temperature differences. The super-parameterized CAM (SP-CAM) simulations were meant as a first attempt to establish which model's cloud behavior is most realistic. I find that cloud behavior in SP-CAM and CAM is very similar. This may be due either to the realism of the CAM cloud scheme or to the similarity of non-cloud aspects between the two models.

Full Presentation

Larissa Back, University of Wisconsin - Madison


Title: Global hydrological cycle response to rapid and slow global warming

Author(s): Larissa Back

Abstract: Held and Soden (2006, hereafter HS06) proposed some robust hydrological responses to global warming.  One of these is that global water vapor $q$ increases roughly 7.5$\%$ per degree K of global mean surface warming, $\Delta T_s$ and 2). This rate has since been accepted in some circles as one of the most robust features of global climate response, because this $\approx 7\%$ K$^{-1}$ is robust across all the state-of-art climate models and has been thought to be based on fundamental thermodynamic law. Here, based on a simulation of the last 21,000 years with naturally occurring slow global warming, we find that the global increase of 7.5$\%$ per degree K rate is robust for rapid, anthropogenic-warming-like climate changes, but not for slow, paleoclimate-like changes, because of the different patterns of surface temperature response.  Despite these variations in the response of water vapor to warming, we find that the rate of change of global precipitation is robust to whether warming is rapid or slow. We explore the reasons for these findings.

Full Presentation

Michela Biasutti, Lamont-Doherty Earth Observatory -Columbia University


Title: Intense precipitation as seen by instantaneous and aggregated measurements

Author(s): Michela Biasutti and Sandra E. Yuter

Abstract: Seasonal and spatial variations in rainfall frequency and intensity are derived from TRMM precipitation radar instantaneous reflectivity measures and compared to similar quantities obtained from daily rainfall. The instantaneous data correctly identi es regions of very intense storms, as well as regions experiencing more frequent and milder systems; in particular, it captures the tendency for land convection to be especially strong in the Congo, in the plains of subtropical North and South America, and in semi-arid regions. This climatology is combined with the precipitation feature dataset of the University of Utah Precipitation Measuring Mission to show that, in a variety of continental environments, intensity is highest in the months preceding the core of the rainy season|not because convective showers themselves are more intense, but because storms have comparatively little stratiform precipitation at that time. Over oceanic regions, the relative importance of convective to stratiform rain is quite constant throughout the seasons, but the intensity of convection itself tends to vary in sync with frequency. Patterns of daily intensity are different from patterns of storm intensity. Daily intensity is determined to a large extent by the duration and frequency of rain events, so that it tends to be maximum where and when rainfall frequency is maximum; it is not closely linked to the intensity of the rain showers themselves and does not capture the distribution (in space or time) of intense rainfall events. Even the convective-to-stratiform ratio, when calculated from aggregated rainfall data, fails to highlight regions of very intense storms: thus, daily data misses severe events that are a potential cause for floods. Given the importance of understanding how extreme rainfall precipitation will change over land regions|certainly including at the timescale of individual storms|we suggest that instantaneous conditional rainfall intensity should also be saved as output by climate models.

Full Presentation

Sandrine Bony-Lena, Laboratoire de Meteorologie Dynamique (LMD/IPSL)


Title: Observational evidence for relationships between convective aggregation, water vapor and radiation over tropical oceans

Author(s): Sandrine Bony-Lena and Isabelle Tobin

Abstract: Tropical deep convection exhibits a wide diversity of spatial organizations. Idealized studies using cloud resolving models suggest that the mean state of the tropical atmosphere might depend on the degree of aggregation of deep convection. In this presentation, we will present some observational evidence for relationships between convective aggregation, water vapor and top-of-the-atmosphere radiation derived from the analysis of long time series of satellite observations. We will compare these relationships with those derived from cloud resolving models, and we will discuss the implications of the observational results for climate studies.

William Boos, Yale University


Title: Thermodynamic bias in the multi-model mean boreal summer monsoon

Author(s): William Boos

Abstract: We show that almost all climate models exhibit a common bias in the thermodynamic structure of boreal summer monsoons.  The strongest bias lies over South Asia, where the thermal maximum is too weak and is located over coastal oceans rather than in its observed continental position.  The simulated Asian upper-tropospheric temperature maximum also does not penetrate as far west over Africa as it does in observations, which suggests a remote influence of South Asian monsoon bias on simulated climate in the Sahel.  We present evidence that this model bias is caused by an overly smoothed representation of the Hindu Kush mountain range west of the Tibetan Plateau, which allows the thermal maximum to be ventilated by dry extratropical air.  We also show that this bias strongly affects predictions of next-century regional precipitation change in the entire northern hemisphere.  These results emphasize the importance of shallow moisture advection and relatively low topography west of the Tibetan Plateau for boreal summer climate.

Full Presentation

Chris Bretherton, University of Washington


Title: An LES perspective on subtropical low cloud feedback

Author(s): Chris Bretherton

Abstract: The CGILS intercomparison project has been comparing the response of an international set of single-column and large-eddy simulation models to an idealized +2K tropical warming, imposed by perturbing boundary conditions and advective forcings on representative atmospheric columns in the NE Pacific subtropical marine stratocumulus and trade-cumulus regimes.  At UW, we have also explored the cloud response to a number of other climate perturbations to isolate physical mechanisms of warming.

Here we focus on the LES results, which show reasonable consistency between models.  In stratocumulus regimes, more free tropospheric CO2 and water vapor reduce cloud-top radiative cooling and thin the cloud layer.  Decreased subsidence deepens the boundary layer, thickening stratocumulus cloud in well-mixed situations.  Changes in inversion strength also correlate with cloud thickness.  In the cumulus regime, feedbacks between precipitation and cumulus-layer depth act as a governor to damp the boundary layer response to the simulated climate perturbations.  Some of these feedbacks can also be seen in superparameterized climate model simulations.

Full Presentation

Mark Cane, Lamont-Doherty Earth Observatory - Columbia University


Title: Global Precipitation Change Shaped by Natural and Anthropogenic Forcing

Author(s): Mark Cane, Jian Liu, Bin Wang, So-Young Yim, and June-Yi Lee

Abstract: How much global precipitation would increase for a given temperature increase due to global warming has been the subject of intense debate. There is also disagreement about whether the equatorial Pacific, which strongly influences global climate, responds to increased heating by enhancing the east-west gradient or reducing it. Whether precipitation changes due to greenhouse warming will resemble those in the past millennium due to increases in solar and volcanic heating is not known. Here we show in a climate model's simulations of the last millennium that although the late 20th century is warmer than the Medieval Warm Period (MWP), precipitation is less. This is consistent with the global energy budget, but the dynamical reason for the difference is that the solar-volcanic (SV) radiative forcing and greenhouse gas (GHG) forcing induce different coupled precipitation-sea surface temperature (SST) patterns in the equatorial Pacific. The enhanced zonal SST gradient in the equatorial Pacific (a "La Niña-like" pattern) during the MWP, a time of increased SV forcing, is accompanied by increased global precipitation. On the other hand, increasing GHG forcing reduces the zonal SST gradient in the equatorial Pacific (an "El Niño-like" pattern), an SST pattern that is less effective in increasing global precipitation. The GHG forcing favors an El Niño-like weakening of the atmospheric Walker circulation, because it tends to stabilize the tropical atmosphere, whereas the SV forcing activates the oceanic thermostat mechanism, leading to a strengthened SST gradient and enhanced Walker circulation.

Full Presentation

Andrew Dessler, Texas A&M University


Title: On the control of stratospheric water vapor and the implications for climate change

Author(s): Andrew Dessler

Abstract: Results from a trajectory model of stratospheric water vapor are analyzed and compared to a 25-year record of measurements from three satellite measurements. The model and measurements show no long-term increase in water vapor in the stratosphere despite a warming surface over this period. Thus, we see no evidence of any stratospheric water vapor feedback. The model is also analyzed to understand the source of interannual variability in stratospheric water vapor. We find that the QBO and aerosol variations control stratospheric water vapor variability by controlling tropopause-layer temperatures.

Full Presentation

Kerry Emanuel, MIT


Title: Rectification of the Diurnal Cycle of Moist Convection: Implications for Tropical Circulation

Author(s): Timothy Cronin and Kerry Emanuel

Abstract: The diurnal cycle of moist convection has been studied for some decades, mostly for the purposes of understanding and predicting the timing of convective rainfall over land. It is well known that convective rainfall over land generally has a late afternoon peak, and almost as well known that parameterized convection usually fails to produce the correct phasing of the diurnal cycle. One interesting question that arises is whether failure to produce a correct diurnal cycle as implications for weather and climate on time scales much longer than a day. To explore this question, we performed simulations using a three-dimensional cloud-permitting model run on a large domain with an island consisting of land with no heat capacity but with an albedo identical to that of the sea. The model solves a full radiation code, including cloud-radiation interaction, and a diurnal cycle of insolation. We find a net radiative surplus in the column of atmosphere over the land surface, accompanied by time-averaged column-integrated ascent and excess precipitation over the island. Preliminary analysis of the results suggest that the late afternoon peak in convection over the island produces anomalous high cloud (and possibly water vapor) over land during night, reducing nighttime (and therefore diurnal average) outgoing longwave radiation and thereby providing a net radiative surplus over the land.

Full Presentation

Stephan Fueglistaler, Princeton University


Title: Atmospheric humidity and clouds - what can we learn from stratospheric water?

Author(s): Stephan Fueglistaler

Abstract: The Stratosphere may seem a rather exotic place to search for insights on the mechanisms that control atmospheric humidity and clouds. However, I will argue that stratospheric water vapor (in the stratospheric "overworld") is in many ways an ideal natural "laboratory case" to study the problem. First, stratospheric humidity is largely controlled in a thin layer around the tropopause in the inner tropics. Second, the flux of moisture into the stratosphere can be understood to good approximation as a single timeseries of "entry mixing ratios". Both of these properties considerably simplify the problem compared to the troposphere, where the full 4-dimensional fields have to be considered. Third, the resulting moisture field (in the stratosphere) feeds back only weakly on the structure of the controlling region (i.e. the vicinity of the tropical tropopause). Fourth, there exists large variability on seasonal and interannual timescales that can be measured fairly accurately. Finally, three processes regulating much of this variability have almost ideal characteristics: both the strength of the stratospheric residual circulation and the stratospheric Quasi-Biennial Oscillation induce zonally symmetric temperature perturbations, while ENSO induces strong zonally asymmetric perturbations. I will present calculations that show that "advection-condensation"-type model calculations capture the observed variability well. However, even in this case our understanding is still only diagnostic rather than prognostic. That is, for a given circulation and temperature structure we can calculate the resulting moisture field using a numerical model, but we do not have a theory that would allow prediction of changes in the moisture field in response to changes in the circulation and/or temperature structure.

Full Presentation

Dennis Hartmann, University of Washington


Title: Radiative control of deep tropical convection

Author(s): Dennis Hartmann

Abstract: The radiative controls on tropical deep convection will be investigated in the context of observations, limited-area cloud resolving models and global climate models.

Full Presentation

Isaac Held, NOAA


Title: Lessons learned from the simulation of tropical cyclone genesis in global mesoscale models

Author(s): Isaac Held

Abstract: Experience with global "mesoscale" models, with resolutions in the range of 15-75 kms, is growing rapidly, and is proving to be of particular relevance to the problem of tropical cyclone (TC) genesis. I will summarize some of the results obtained at GFDL with regard to simulations of the climatology, interannual variability, predictability, and sensiitivty to global warming of TC statistics -- and some oif the suggestive implications of these results for our understanding of genesis and convective parameterization.

Full Presentation

George Kiladis, NOAA


Title: Relative contributions of synoptic and low frequency eddies to timemean atmospheric moisture transport

Author(s): George Kiladis

Abstract: The relative contributions to mean global atmospheric moisture transport by both the time-mean circulation and by synoptic and low-frequency (periods greater than 10 days) anomalies are evaluated from the mean vertically-integrated atmospheric moisture budget based on 40 years of NCEP-NCAR Reanalysis data. In the extratropics, while the time mean circulation moves moisture primarily zonally from one part of the ocean to another, low-frequency and synoptic anomalies drive much of the moisture transport both meridionally and from ocean to land. While some low-frequency transport originates in low latitudes, much of it is of extratropical origin, due to large-scale atmospheric anomalies that extract moisture from the Northeast Pacific and Atlantic oceans. Synoptic variability drives about half of the poleward midlatitude moisture transport in both hemispheres, consistent with simple "lateral mixing" arguments. Atmospheric transport in the extratropics is also particularly focused within "atmospheric rivers", relatively narrow poleward-moving plumes of moisture associated with frontal dynamics. These results suggest that understanding potential anthropogenic changes in the Earth's hydrological cycle may require understanding not only thermodynamic changes but also corresponding changes in atmospheric circulation variability on low frequency time scales.

Full Presentation

Steven Krueger, University of Utah


Title: Mass flux, vertical velocity, and entrainment in the Giga-LES

Author(s): Steven Krueger and Ian Glenn

Abstract: We are applying the approach developed by Kuang and Bretherton (2006) to investigate various aspects of the ensemble characteristics of cumulus convection in a model dataset from a large-doman LES of deep convection. Our results agree with those of Kuang and Bretherton for the cumulus updraft properties. We are currently examining the relative merits of different entrainment and cloud-top-height assumptions in spectral plume models of cumulus updrafts, the characteristics of downdrafts, and the nature of a rapid transition from shallow to deep convection.

Full Presentation

Zhiming Kuang, Harvard University


Title: Weakly forced mock-Walker cells

Author(s): Zhiming Kuang

Abstract: Mock-Walker cells driven by weak sea surface temperature (SST) forcing are studied using planetary-scale cloud-system-resolving simulations and a simplified framework that represents convection with its linear response functions and parameterizes the large-scale flow based on the gravity wave equation. The goal is to gain insight into steady tropical circulations through this simple example. The Walker cells vary substantially with horizontal domain size. Vertical velocity trends toward broader vertical scales as the horizontal scale increases. This is explained by the fact that temperature anomalies required to sustain a vertical velocity profile of given amplitude are stronger in cases of larger horizontal and smaller vertical scales. Different treatments of convective momentum transport (CMT) significantly affect the Walker cells. The downward advection component of the CMT is important in capturing a number of features of the mock-Walker cells, such as the lower tropospheric cold anomalies over the warmer SST that are stronger than the upper tropospheric warm anomalies, and hence the high surface pressure there. It also better captures the shape of the vertical velocity profile. The extent of convective organization also affects the Walker cell through its effects on the sensitivities of convective heating and moistening to temperature and moisture anomalies. For strongly organized convection with deep inflows, these sensitivities are consistent with a layer mode of convective overturning, instead of the parcel mode as in unorganized convection, resulting in a weaker second baroclinic component in the Walker cells.

Full Presentation

Brian Mapes, University of Miami


Title: Toward an understanding of form-function relationship for mesoscale convection

Author(s): Brian Mapes

Abstract: I will show some pretty loops of mesoscale form (radar observations, cloud model outputs) before turning to the question of how we measure the "function" of deep convection, and how it might relate to form. I see a small handful of ways to measure functionality: by the responses of CRMs in a couple different types of test harnesses, or by sensitivity matrices as estimated by various methods. If we map function while form is controlled by shear, model domain geometry, boundary conditions, or conditional sampling of inputs to empirical estimation methods, we might have the beginnings of an understanding - or at least a well defined object for discourse.

Full Presentation

David Neelin, University of California - Los Angeles


Title: Model precipitation uncertainties and constraints on entrainment from convective onset

Author(s): David Neelin


Full Presentation

David Nolan, University of Miami


Title: Tropical Cyclone Genesis Parameters, Genesis Thresholds, and Mid-Level Humidity

Author(s): Michael G. McGauley and David S. Nolan

Abstract: "Genesis parameters" (GPs) attempt to predict seasonal or annual tropical cyclone activity by combining representative values of relevant environmental parameters into a single formula. GPs developed by Gray (1975), Emanuel and Nolan (2004), Emanuel (2010) and Tippet et al. (2011) used environmental measures of thermodynamic instability, wind shear, relative humidity, and environmental vorticity, with mean values of each being normalized and then multiplied together to make their respective GPs. We present an alternative to the GP approach, the "genesis frequency index", or GFI. This method uses the fraction of time (or frequency) that each relevant parameter is above or below some threshold value that permits TC genesis, given that each of the other three parameters has a highly favorable (but realistic) value. Threshold values for MPI, wind shear, and environmental vorticity are computed from idealized numerical simulations. For humidity, we found that 1) it was necessary to use an empirical approach to find the threshold value, and 2) the measure of humidity that produces the best results is the mid-level saturation deficit divided by the surface layer saturation difference, or "normalized saturation deficit." GFI results are presented for each month of the year for each ocean basin. The GFI correlates better with observed TC genesis distributions in most months in most basins, although there are some exceptions. The GFI also indicates appropriate changes in TC genesis rates for the Atlantic in El Nino and La Nina years, and illustrates the reasons for these changes.

Full Presentation

Joel Norris, University of California -San Diego


Title: Evidence for Climate Change in the Satellite Cloud Record

Author(s): Joel Norris

Abstract: Clouds have been a key uncertainty in our understanding of climate change due to disagreement among global climate models and observational datasets over what cloud changes have occurred during recent decades or will occur in response to anthropogenic global warming. We show that, after removal of spurious artifacts, several independent satellite records exhibit similar regional patterns of cloud change between the 1980s and 2000s. The observed cloud changes moreover resemble those produced by model simulations of the late 20th century climate and for the doubling of CO2. Observed and simulated cloud change patterns are consistent with poleward shifts of midlatitude storm track cloudiness, expansion of subtropical dry zones, increasing cloud optical depth at high latitudes, and increasing height of the highest cloud tops at all latitudes. These results indicate that cloud changes robustly predicted by global climate models are currently occurring in nature.

Paul O'Gorman, MIT


Title: Upward shift of the atmospheric general circulation under global warming: theory and simulations

Author(s): Paul A. O'Gorman in collaboration with Martin S. Singh

Abstract: Many features of the general circulation of the atmosphere shift upwards in response to warming in simulations of climate change with both general circulation models (GCMs) and cloud system resolving models. The importance of the upward shift is well known, but its physical basis and the extent to which it occurs coherently across all variables are not well understood. A transformation is derived here that shows how an upward shift of a solution to the moist primitive equations gives a new approximate solution with higher tropospheric temperatures. According to the transformation, all variables shift upwards with warming, but with an additional modifi cation to the temperature and a general weakening of the pressure velocity. The applicability of the vertical-shift transformation is explored using a hierarchy of models, from adiabatic parcel ascents to comprehensive GCMs. The results allow for a physical interpretation of the upward shift in terms of the governing equations and suggest that it may be thought of as a coherent response of the general circulation as a whole.

Full Presentation

Michael Pritchard, University of Washington


Title: A moist static energy budget analysis of the MJO in the Super-Parameterized Community Atmosphere Model (SPCAM)

Author(s): Michael Pritchard and Chris Bretherton

Abstract: Global climate models that enforce tight column humidity - rainfall coupling can produce MJO-like variability with signatures of moisture mode dynamics. In such models, the simulated column moist static energy budget is a useful diagnostic for understanding MJO destabilization and propagation mechanisms. However, different conclusions can be been drawn from different models about the dominant destabilization pathways, including the relative roles of longwave heating vs. latent heat fluxes in amplifying the mode. To add to this debate, we examine the representation of the MJO in the Super-Parameterized Community Atmosphere Model, extending the recent aquaplanet-mode MSE budget analysis of Andersen& Kuang (2011) to a case with a realistic basic state.

Full Presentation

David Randall, Colorado State University


Title: Future pathways for parameterization of turbulence and shallow convection

Author(s): David Randall

Abstract: Lower tropospheric turbulence and shallow moist convection are parameterized in both large-scale models, with grid spacings as large as hundreds of kilometers, and cloud-resolving models (CRMs), with grid spacings of a few kilometers. In recent years we have seen the emergence of global CRMs, and also super-parameterized global models with embedded CRMs. The "jobs" of a parameterization of turbulence and shallow convection are to determine fluxes of various quantities, the subgrid cloud fraction, and even subgrid microphysical and chemical processes. Despite the importance of these processes, CRMs usually represent them in highly simplified ways that are barely defensible on physical grounds. Over the past several years, we have seen the emergence of parameterizations that are more realistic and also more complicated. The success of these efforts is still being assessed, but it is already clear that the current versions have important shortcomings. I will discuss these issues and approaches to overcoming them.

Full Presentation

David J. Raymond, New Mexico Tech


Title: Vorticity, Moisture, and Precipitation in the Tropical Atmosphere

Author(s): David J. Raymond

Abstract: This talk explores the hypothesis that mid-level vorticity is the main factor controlling the saturation fraction and mean precipitation rate in the tropical atmosphere. The chain of causality is as follows: (1) The balanced thermodynamic response to mid-level vorticity is warming at upper levels and cooling at lower levels. Warming and cooling in tropical disturbances outside of fully developed tropical cyclones is of order 1 K. (2) Virtual temperature anomalies of 1 K are sufficient to cause significant changes in the vertical mass flux profiles of deep convection. In particular, the stabilization due to such a temperature dipole results in a dramatic lowering of the elevation of maximum vertical mass flux. (3) Such "bottom-heavy" convection exhibits smaller gross moist stability, resulting in a higher precipitation rate as well as a moister convective environment. (4) This process can be frustrated by the ingestion of dry air and the dismemberment of the mid-level vortex resulting from vertical or horizontal shear. An important consequence of a shallow convective mass flux maximum is the concentration of the convective inflow into a shallow layer near the surface. This can cause a low-level vortex to spin up beneath the mid-level vortex, a process that may be central to tropical cyclogenesis. Low-level vorticity by itself, such as occurs in weak west Pacific tropical waves, results in top-heavy convection that has maximum mass convergence at middle levels. Under favorable circumstances this convection may result in the formation of the mid-level vortex needed for further intensification of the wave. Dynamics and thermodynamics thus appear to be entwined in the tropics very differently than in middle latitudes.

Full Presentation

David Romps, University of California - Berkeley


Title: Parameterizing large-scale dynamics: a comparison of WTG and WPG

Author(s): David Romps

Abstract: The weak temperature gradient (WTG) approximation and an alternative, the weak pressure gradient (WPG) approximation, are evaluated using analytical solutions and cloud-resolving simulations. Analytically, WPG reproduces key features of the 3D solutions that WTG is unable to generate, including ascending regions of negative buoyancy and adiabatic lifting below a transient buoyancy anomaly. Numerically, the assumptions of WPG are approximately satisfied by large 3D cloud-resolving simulations, while the assumptions of WTG are violated. When implemented in a cloud-resolving model, it is found that WPG correctly reproduces the behavior of the 3D simulations, including steady-state vertical velocity and transient convective triggering by buoyancy anomalies aloft. WTG, on the other hand, is unable to replicate these features.

Full Presentation

Tapio Schneider, California Institute of Technology


Title: Understanding low cloud feedbacks through hierarchical physical modeling

Author(s): Tapio Schneider

Abstract: Low clouds cover a substantial fraction of the world's oceans and are key players in regulating Earth's energy balance. How they react to climate change is largely unknown: theories and models suggest they may represent positive or negative feedbacks that amplify or damp perturbations such as those associated with increasing concentrations of greenhouse gases. Here we present a physical framework for understanding low-cloud feedbacks. We use a hierarchy of models, ranging from high-resolution large-eddy simulations to global climate models, to study how low clouds react to climate change and to reveal physical mechanisms constraining this reaction. Simulation results and theoretical arguments support a robustly positive low cloud feedback, in agreement with recent observations.

Steven Sherwood, University of New South Wales


Title: The vertical momentum budget of a cumulus cloud and its implications for climate

Author(s): Steven Sherwood

Abstract: Efforts to tune cumulus parameterisations, notably by adjusting entrainment rates, have yielded conflicting results with no single entrainment representation producing satisfactory representation of both the sensitivity of convection to ambient humidity (which requires a high rate) and accurate representation of the mean state (which may require convection to reach the tropopause more easily). This situation calls for a rethinking of the basic assumptions upon which convection schemes are built. We argue that a crucial target is the assumption that convective plumes or thermals experience heavy drag upon ascent, which is either explicit or implicit in basic models underlying convective schemes. We present analyses of the momentum budget of numerically-simulated cumulus thermals to reassess the role of damping due to friction-like processes, and find in particular that exchanges of fluid between the thermal and its environment do not lead to a loss of momentum as widely assumed. A fundamental issue that arises in this computation is the ambiguity in defining what exactly constitutes the "parcel" that is represented in the idealised ascending-parcel calculation, in a realistic situation of a strongly deforming and turbulent fluid with rapid phase changes. We accordingly present a new method for tracking and defining thermal entities in CRMs; this calculation shows that during shallow cumulus growth, thermals rise at nearly constant rates and diameters until dissipating. This contrasts with what is implied in most current models, but we argue that it is what allows the thermal to exchange mass vigorously with surroundings without losing momentum. Finally we show that the presence or absence of frictional damping has significant implications for the macroscopic behavior of convection.

Full Presentation

Pier Siebesma, KNMI


Title: The role of tropospheric humidity and stability on the detrainment in deep convection

Author(s): A. Pier Siebesma, Steef Boing and Dirk-Jan Korpershoek

Abstract: The influence of tropospheric humidity on the strength of deep convection has been a topic of interest for the last decade. Cloud Resolving Model studies have shown that increasing relative humidity leads to intensified deep convection. This feature is absent in most parameterized convection schemes in state-of-the-art weather and climate models. A notably exception is the ECMWF model that has a convection scheme in which lateral entrainment rate is assumed to be dependent on the environmental relative humidity. In order to explore to relation between tropospheric relative humidity and deep convection in more detail we have made a large number of Large Eddy Simulations of deep convection in which both the atmospheric stability and the relative humidity is varied in a systematic way. Analyses of the results show that it is the detrainment rather than the entrainment that has a strong dependence to the relative humidity and stability. Moreover we show that the detrainment has a perfect functional dependence with the fraction of environmental air that is necessary to make moist convection just neutral buoyant. This provides an easy and direct parameterization for detrainment and we will provide simple physical arguments for this behaviour.

Full Presentation

Adam Sobel, Lamont-Doherty Earth Observatory - Columbia University


Title: CRM simulations with parameterized large-scale dynamics using time-dependent forcings from observations

Author(s): Shuguang Wang, Adam Sobel, Zhiming Kuang

Abstract: We present simulations of deep convection with a cloud-resolving model on a modest-sized, doubly periodic box, with large-scale dynamics parameterized but key forcings derived from time-dependent observations - in this case, from TOGA COARE. Two different dynamical parameterizations are used, the weak temperature gradient (WTG) approximation and the gravity wave scheme due to Kuang. Key forcings are the horizontal mean temperature profile and the surface wind speed; large-scale vertical motion is not prescribed, but is part of the solution. In these formulations, the variations of deep convective activity are not direct budgetary consequences of the forcing, and can differ widely from those observed. Yet, with appropriate parameter settings, the model is able to simulate at least the low-frequency component of the variability with some fidelity to observations. WTG and the gravity wave scheme give comparably successful simulations of precipitation, but the gravity wave scheme gives much better agreement in the large-scale vertical velocity profile. The profile is much too top-heavy under WTG, a result that can be explained in terms of WTG's ignorance of momentum.

Full Presentation

Brian Soden, University of Miami


Title: Why has the Walker circulation strengthened over the past 30 years?

Author(s): Brian Soden

Abstract: Based upon simple thermodynamic arguments one expects the atmospheric circulation to weaken as the climate warms. In models, this weakening primarily manifests itself as a reduction in the zonally-assymetric overturning circulation in the tropics. Observations of sea-level pressure (SLP) show a centennial scale weakening of the SLP gradient along the equatorial Pacific, suggesting a long-term weakening of the Walker circulation has occurred since the late 19th Century. However this trend has reversed during the last 30 years – a period in which anthropogenic warming is most pronounced. Results from CMIP5 model simulations will be presented which examine the link between a warming climate and a weakening circulation, and evaluate the time-scales on which this relationship holds.

Full Presentation

Bjorn Stevens, Max Planck Institute for Meteorology


Title: Slow Drivers and the Climatology of Precipitation

Author(s): Bjorn Stevens (selections from work in progress with Traute Crueger, Cathy Hohenegger, Benjamin Möbis, Dagmar Popke and Aiko Voigt)


Full Presentation

Minghua Zhang, Stony Brook University


Title: Cloud feedback and the interaction between parameterized convection and boundary-layer turbulence in climate models

Author(s): Minghua Zhang and CGILS Participants

Abstract: This presentation will describe the CGILS project (CFMIP-GCSS Intercomparison of Large-Eddy and Single-Column Models) to investigate the physical mechanism of low cloud feedbacks in climate models. The targeted clouds are shallow convection and stratocumulus. We will first discuss the experimental design and up-to-date results of the project. We then show how different parameterizations of the convective mass fluxes and cloud-top entrainment affect the behavior of the cloud feedback in different models.

Full Presentation



| EPS Home  |  SEAS Home  |  Harvard University
© 2012 President and Fellows of Harvard College