A unified turbulence and cloud parameterization based on multi-variate probability density functions (PDFs) has been incorporated into the GFDL atmospheric general circulation model AM3. This PDF-based parameterization not only predicts sub-grid variations in vertical velocity, temperature, and total water, which bridge sub-grid scale processes (such as aerosol activation and cloud microphysics) and grid-scale dynamic and thermodynamic fields, but also unifies the treatment of planetary boundary layer (PBL), shallow convection, and cloud macrophysics. This parameterization is the “Cloud Layers Unified By Binormals” (CLUBB) parameterization. With the incorporation of CLUBB in AM3, coupled with a two-moment cloud microphysical scheme, AM3-CLUBB allows for a more physically-based and self-consistent treatment of aerosol activation, cloud micro- and macro-physics, PBL, and shallow convection.
The configuration and performance of AM3-CLUBB are described. Cloud and radiation fields, as well as most basic climate features, are modeled realistically. Relative to AM3, AM3-CLUBB improves the simulation of coastal stratocumulus, a longstanding deficiency in GFDL models, and their seasonal cycle, especially at higher horizontal resolution, but global skill scores deteriorate slightly. Through sensitivity experiments, we show that (1) the two-moment cloud microphysics helps relieve the deficiency of coastal stratocumulus; (2) using the CLUBB sub-grid cloud water variability in the cloud microphysics has a considerable positive impact on global cloudiness; and (3) the impact of adjusting CLUBB parameters is to improve the overall agreement between model and observations.
Pincus, Robert, S Platnick, S A Ackerman, Richard S Hemler, and R J P Hofmann, July 2012: Reconciling simulated and observed views of clouds: MODIS, ISCCP, and the limits of instrument simulators. Journal of Climate, 25(13), DOI:10.1175/JCLI-D-11-00267.1. Abstract
The properties of clouds that may be observed by satellite instruments, such as optical thickness and cloud top pressure, are only loosely related to the way clouds are represented in models of the atmosphere. One way to bridge this gap is through “instrument simulators,” diagnostic tools that map the model representation to synthetic observations so that differences between can be interpreted as model error. But simulators may themselves be restricted by limited information or by internal assumptions. This paper considers the extent to which instrument simulators are able to capture essential differences between MODIS and ISCCP, two similar but independent estimates of cloud properties. The authors review the measurements and algorithms underlying these two cloud climatologies, introduce a MODIS simulator, and detail data sets developed for comparison with global models using ISCCP and MODIS simulators. In nature MODIS observes less mid-level cloudiness than ISCCP, consistent with the different methods used to determine cloud top pressure; aspects of this difference are reproduced by the simulators. Differences in observed distributions of optical thickness, however, are not captured. The largest differences between can be traced to different approaches to partly-cloudy pixels, which MODIS excludes and ISCCP treats as homogeneous. These cover roughly 15% of the planet and account for most optically thinnest clouds. Instrument simulators can not reproduce these differences because there is no way to synthesize partly-cloudy pixels. Nonetheless, MODIS and ISCCP observation are consistent for all but the optically-thinnest clouds, and models can be robustly evaluated using instrument simulators by integrating over the robust subset of observations.
The Geophysical Fluid Dynamics Laboratory (GFDL) has developed a coupled general circulation model (CM3) for atmosphere, oceans, land, and sea ice. The goal of CM3 is to address emerging issues in climate change, including aerosol-cloud interactions, chemistry-climate interactions, and coupling between the troposphere and stratosphere. The model is also designed to serve as the physical-system component of earth-system models and models for decadal prediction in the near-term future, for example, through improved simulations in tropical land precipitation relative to earlier-generation GFDL models. This paper describes the dynamical core, physical parameterizations, and basic simulation characteristics of the atmospheric component (AM3) of this model.
Relative to GFDL AM2, AM3 includes new treatments of deep and shallow cumulus convection, cloud-droplet activation by aerosols, sub-grid variability of stratiform vertical velocities for droplet activation, and atmospheric chemistry driven by emissions with advective, convective, and turbulent transport. AM3 employs a cubed-sphere implementation of a finite-volume dynamical core and is coupled to LM3, a new land model with eco-system dynamics and hydrology.
Most basic circulation features in AM3 are simulated as realistically, or more so, than in AM2. In particular, dry biases have been reduced over South America. In coupled mode, the simulation of Arctic sea ice concentration has improved. AM3 aerosol optical depths, scattering properties, and surface clear-sky downward shortwave radiation are more realistic than in AM2. The simulation of marine stratocumulus decks and the intensity distributions of precipitation remain problematic, as in AM2.
The last two decades of the 20th century warm in CM3 by .32°C relative to 1881-1920. The Climate Research Unit (CRU) and Goddard Institute for Space Studies analyses of observations show warming of .56°C and .52°C, respectively, over this period. CM3 includes anthropogenic cooling by aerosol cloud interactions, and its warming by late 20th century is somewhat less realistic than in CM2.1, which warmed .66°C but did not include aerosol cloud interactions. The improved simulation of the direct aerosol effect (apparent in surface clear-sky downward radiation) in CM3 evidently acts in concert with its simulation of cloud-aerosol interactions to limit greenhouse gas warming in a way that is consistent with observed global temperature changes.
The formulation and simulation characteristics of two new global coupled climate models developed at NOAA's Geophysical Fluid Dynamics Laboratory (GFDL) are described. The models were designed to simulate atmospheric and oceanic climate and variability from the diurnal time scale through multicentury climate change, given our computational constraints. In particular, an important goal was to use the same model for both experimental seasonal to interannual forecasting and the study of multicentury global climate change, and this goal has been achieved.
Two versions of the coupled model are described, called CM2.0 and CM2.1. The versions differ primarily in the dynamical core used in the atmospheric component, along with the cloud tuning and some details of the land and ocean components. For both coupled models, the resolution of the land and atmospheric components is 2° latitude × 2.5° longitude; the atmospheric model has 24 vertical levels. The ocean resolution is 1° in latitude and longitude, with meridional resolution equatorward of 30° becoming progressively finer, such that the meridional resolution is 1/3° at the equator. There are 50 vertical levels in the ocean, with 22 evenly spaced levels within the top 220 m. The ocean component has poles over North America and Eurasia to avoid polar filtering. Neither coupled model employs flux adjustments.
The control simulations have stable, realistic climates when integrated over multiple centuries. Both models have simulations of ENSO that are substantially improved relative to previous GFDL coupled models. The CM2.0 model has been further evaluated as an ENSO forecast model and has good skill (CM2.1 has not been evaluated as an ENSO forecast model). Generally reduced temperature and salinity biases exist in CM2.1 relative to CM2.0. These reductions are associated with 1) improved simulations of surface wind stress in CM2.1 and associated changes in oceanic gyre circulations; 2) changes in cloud tuning and the land model, both of which act to increase the net surface shortwave radiation in CM2.1, thereby reducing an overall cold bias present in CM2.0; and 3) a reduction of ocean lateral viscosity in the extratropics in CM2.1, which reduces sea ice biases in the North Atlantic.
Both models have been used to conduct a suite of climate change simulations for the 2007 Intergovernmental Panel on Climate Change (IPCC) assessment report and are able to simulate the main features of the observed warming of the twentieth century. The climate sensitivities of the CM2.0 and CM2.1 models are 2.9 and 3.4 K, respectively. These sensitivities are defined by coupling the atmospheric components of CM2.0 and CM2.1 to a slab ocean model and allowing the model to come into equilibrium with a doubling of atmospheric CO2. The output from a suite of integrations conducted with these models is freely available online (see http://nomads.gfdl.noaa.gov/).
Manuscript received 8 December 2004, in final form 18 March 2005
Pincus, Robert, Richard S Hemler, and Stephen A Klein, 2006: Using Stochastically Generated Subcolumns to Represent Cloud Structure in a Large-Scale Model. Monthly Weather Review, 134(12), DOI:10.1175/MWR3257.1. Abstract
A new method for representing subgrid-scale cloud structure in which each model column is decomposed into a set of subcolumns has been introduced into the Geophysical Fluid Dynamics Laboratory’s global atmospheric model AM2. Each subcolumn in the decomposition is homogeneous, but the ensemble reproduces the initial profiles of cloud properties including cloud fraction, internal variability (if any) in cloud condensate, and arbitrary overlap assumptions that describe vertical correlations. These subcolumns are used in radiation and diagnostic calculations and have allowed the introduction of more realistic overlap assumptions. This paper describes the impact of these new methods for representing cloud structure in instantaneous calculations and long-term integrations. Shortwave radiation computed using subcolumns and the random overlap assumption differs in the global annual average by more than 4 W m−2 from the operational radiation scheme in instantaneous calculations; much of this difference is counteracted by a change in the overlap assumption to one in which overlap varies continuously with the separation distance between layers. Internal variability in cloud condensate, diagnosed from the mean condensate amount and cloud fraction, has about the same effect on radiative fluxes as does the ad hoc tuning accounting for this effect in the operational radiation scheme. Long simulations with the new model configuration show little difference from the operational model configuration, while statistical tests indicate that the model does not respond systematically to the sampling noise introduced by the approximate radiative transfer techniques introduced to work with the subcolumns
The climate response to idealized changes in the atmospheric CO2 concentration by the new GFDL climate model (CM2) is documented. This new model is very different from earlier GFDL models in its parameterizations of subgrid-scale physical processes, numerical algorithms, and resolution. The model was constructed to be useful for both seasonal-to-interannual predictions and climate change research. Unlike previous versions of the global coupled GFDL climate models, CM2 does not use flux adjustments to maintain a stable control climate. Results from two model versions, Climate Model versions 2.0 (CM2.0) and 2.1 (CM2.1), are presented.
Two atmosphere–mixed layer ocean or slab models, Slab Model versions 2.0 (SM2.0) and 2.1 (SM2.1), are constructed corresponding to CM2.0 and CM2.1. Using the SM2 models to estimate the climate sensitivity, it is found that the equilibrium globally averaged surface air temperature increases 2.9 (SM2.0) and 3.4 K (SM2.1) for a doubling of the atmospheric CO2 concentration. When forced by a 1% per year CO2 increase, the surface air temperature difference around the time of CO2 doubling [transient climate response (TCR)] is about 1.6 K for both coupled model versions (CM2.0 and CM2.1). The simulated warming is near the median of the responses documented for the climate models used in the 2001 Intergovernmental Panel on Climate Change (IPCC) Working Group I Third Assessment Report (TAR).
The thermohaline circulation (THC) weakened in response to increasing atmospheric CO2. By the time of CO2 doubling, the weakening in CM2.1 is larger than that found in CM2.0: 7 and 4 Sv (1 Sv 106 m3 s−1), respectively. However, the THC in the control integration of CM2.1 is stronger than in CM2.0, so that the percentage change in the THC between the two versions is more similar. The average THC change for the models presented in the TAR is about 3 or 4 Sv; however, the range across the model results is very large, varying from a slight increase (+2 Sv) to a large decrease (−10 Sv).
Pincus, Robert, Cecile Hannay, Stephen A Klein, K-M Xu, and Richard S Hemler, 2005: Overlap assumptions for assumed probability distribution function cloud schemes in large-scale models. Journal of Geophysical Research, 110, D15S09, DOI:10.1029/2004JD005100. Abstract
Cloud vertical structure influences the fluxes of precipitation and radiation throughout the atmosphere. This structure is not predicted in large-scale models but is instead applied in the form of overlap assumptions. In their current guise, overlap assumptions apply to the presence or absence of clouds, and new data sets have led to the development of empirical formulations described by exponential decay from maximum to random overlap over a characteristic length scale. At the same time, cloud parameterizations in many large-scale models have been moving toward assumed PDF schemes that predict the distribution of total water within each grid cell, which will require overlap assumptions that may be applied to cells with specified internal variability. This paper uses a month-long cloud-resolving model simulation of continental convection to develop overlap assumptions for use with assumed PDF cloud schemes in large-scale models. An observing system simulation experiment shows that overlap assumptions derived from millimeter-wavelength cloud radar observations can be strongly affected by the presence of precipitation and convective clouds and, to a lesser degree, by limited sampling and reliance on the frozen turbulence assumption. Current representations of overlap can be extended with good accuracy to treat the rank correlation of total water in each grid cell, which provides a natural way to treat vertical structure in assumed PDF cloud schemes. The scale length that describes an exponential fit to the rank correlation of total water depends on the state of the atmosphere: convection is associated with greater vertical coherence (longer scale lengths), while wind shear decreases vertical coherence (shorter scale lengths). The new overlap assumptions are evaluated using cloud physical properties, microphysical process rates, and top-of-atmosphere radiative fluxes. These quantities can be reproduced very well when the exact cloud structure is replaced with its statistical equivalent and somewhat less well when the time mean vertical structure is imposed. Overlap formulations that treat total water can also be used to determine the variability in clear-air relative humidity, which might be used by convection and aerosol parameterizations.
for climate research developed at the Geophysical Fluid Dynamics Laboratory (GFDL) are presented. The atmosphere model, known as AM2, includes a new gridpoint dynamical core, a prognostic cloud scheme, and a multispecies aerosol climatology, as well as components from previous models used at GFDL. The land model, known as LM2, includes soil sensible and latent heat storage, groundwater storage, and stomatal resistance. The performance of the coupled model AM2–LM2 is evaluated with a series of prescribed sea surface temperature (SST) simulations. Particular focus is given to the model's climatology and the characteristics of interannual variability related to E1 Niño– Southern Oscillation (ENSO).
One AM2–LM2 integration was performed according to the prescriptions of the second Atmospheric Model Intercomparison Project (AMIP II) and data were submitted to the Program for Climate Model Diagnosis and Intercomparison (PCMDI). Particular strengths of AM2–LM2, as judged by comparison to other models participating in AMIP II, include its circulation and distributions of precipitation. Prominent problems of AM2– LM2 include a cold bias to surface and tropospheric temperatures, weak tropical cyclone activity, and weak tropical intraseasonal activity associated with the Madden–Julian oscillation.
An ensemble of 10 AM2–LM2 integrations with observed SSTs for the second half of the twentieth century permits a statistically reliable assessment of the model's response to ENSO. In general, AM2–LM2 produces a realistic simulation of the anomalies in tropical precipitation and extratropical circulation that are associated with ENSO.
Uncertainty in cloud feedback is the leading cause of discrepancy in model predictions of climate change. The use of observed or model-simulated radiative fluxes to diagnose the effect of clouds on climate sensitivity requires an accurate understanding of the distinction between a change in cloud radiative forcing and a cloud feedback. This study compares simulations from different versions of the GFDL Atmospheric Model 2 (AM2) that have widely varying strengths of cloud feedback to illustrate the differences between the two and highlight the potential for changes in cloud radiative forcing to be misinterpreted.
We report model simulations of the effect of deep convection on aerosol under typical Intertropical Convergence Zone (ITCZ) conditions in the tropical Indian Ocean as encountered during the Indian Ocean Experiment (INDOEX). Measurements taken during various phases of INDOEX showed significant aerosol mass concentrations of nss-sulfate, carbonaceous, and mineral dust over the northern Indian Ocean. During the winter dry season aerosol species accumulate and are transported long distances to the tropical regions. In contrast, aerosol measurements south of the ITCZ exhibit significantly lower aerosol concentrations, and the convective activity, mixing, and wet removal in the ITCZ are responsible for their depletion. Our results, based on a cloud-resolving model, driven by National Centers for Environmental Prediction analysis, show that convection and precipitation can remove significant amounts of aerosol, as observed in the Indian Ocean ITCZ. The aerosol lifetime in the boundary layer (BL) is of the order of hours in intense convection with precipitation, but on average is in the range of 1-3 days for the case studied here. Since the convective events occur in a small fraction of the ITCZ area, the aerosol lifetime can vary significantly due to variability of precipitation. Our results show that the decay in concentration of various species of aerosols is comparable with in situ measurements and that the ITCZ can act to reduce the transport of polluted air masses into the Southern Hemisphere especially in cases with significant precipitation. Another finding is that aerosol loading typical to north of ITCZ tends to induce changes in cloud microphysical properties. We found that a difference between clean air masses as those encountered south of the ITCZ to aerosol polluted air masses as encountered north of the ITCZ is associated with a slight decrease of the cloud droplet effective radius (average changes of about 2 :m) and an increase in cloud droplet number concentration (average changes by about 40 to 100 cm-3 ) consistent with several in situ measurements. Thus polluted air masses from the northern Indian Ocean are associated with altered microphysics, and the extent of these effects is dependent on the efficiency of aerosol removal by ITCZ precipitation and dilution by mixing with pristine air masses from the Southern Hemisphere.
Donner, Leo J., Charles J Seman, Richard S Hemler, and Songmiao Fan, 2001: A cumulus parameterization including mass fluxes, convective vertical velocities, and mesoscale effects: thermodynamic and hydrological aspects in a general circulation model. Journal of Climate, 14(16), 3444-3463. Abstract PDF
A cumulus parameterization based on mass fluxes, convective-scale vertical velocities, and mesoscale effects has been incorporated in an atmospheric general circulation model (GCM). Most contemporary cumulus parameterizations are based on convective mass fluxes. This parameterization augments mass fluxes with convective-scale vertical velocities as a means of providing a method for incorporating cumulus microphysics using vertical velocities at physically appropriate (subgrid) scales. Convective-scale microphysics provides a key source of material for mesoscale circulations associated with deep convection, along with mesoscale in situ microphysical processes. The latter depend on simple, parameterized mesoscale dynamics. Consistent treatment of convection, microphysics, and radiation is crucial for modeling global-scale interactions involving clouds and radiation.
Thermodynamic and hydrological aspects of this parameterization in integrations of the Geophysical Fluid Dynamics Laboratory SKYHI GCM are analyzed. Mass fluxes, phase changes, and heat and moisture transport by the mesoscale components of convective systems are found to be large relative to those of convective (deep tower) components, in agreement with field studies. Partitioning between the convective and mesoscale components varies regionally with large-scale flow characteristics and agrees well with observations from the Tropical Rainfall Measuring Mission (TRMM) satellite.
The effects of the mesoscale components of convective systems include stronger Hadley and Walker circulations, warmer upper-tropospheric Tropics, and moister Tropics. The mass fluxes for convective systems including mesoscale components differ appreciably in both magnitude and structure from those for convective systems consisting of cells only. When mesoscale components exist, detrainment is concentrated in the midtroposphere instead of the upper troposphere, and the magnitudes of mass fluxes are smaller. The parameterization including mesoscale components is consistent with satellite observations of the size distribution of convective systems, while the parameterization with convective cells only is not.
The parameterization of convective vertical velocities is an important control on the intensity of the mesoscale stratiform circulations associated with deep convection. The mesoscale components are less intense than in TRMM observations if spatially and temporally invariant convective vertical velocities are used instead of parameterized, variable velocities.
Hamilton, Kevin P., R John Wilson, and Richard S Hemler, 2001: Spontaneous stratospheric QBO-like oscillations Simulated by the GFDL SKYHI general circulation model. Journal of the Atmospheric Sciences, 58(21), 3271-3292. Abstract
The tropical stratospheric mean flow behavior in a series of integrations with high vertical resolution versions of the Geophysical Fluid Dynamics Laboratory (GFDL) "SKYHI" model is examined. At sufficiently high vertical and horizontal model resolution, the simulated stratospheric zonal winds exhibit a strong equatorially centered oscillation with downward propagation of the wind reversals and with formation of strong vertical shear layers. This appears to be a spontaneous internally generated oscillation and closely resembles the observed quasi-biennial oscillation (QBO) in many respects, although the simulated oscillation has a period less than half that of the real QBO. The same basic mean flow oscillation appears in both seasonally varying and perpetual equinox versions of the model, and most of the analysis in this paper is focused on the perpetual equinox cases. The mean flow oscillation is shown to be largely driven by eddy momentum fluxes associated with a broad spectrum of vertically propagating waves generated spontaneously in the tropical troposphere of the model. Several experiments are performed with the model parameters perturbed in various ways. The period of the simulated tropical stratospheric mean flow oscillation is found to change in response to large alterations in the sea surface temperatures (SSTs) employed. This is a fairly direct demonstration of the link between the stratospheric mean flow behavior and tropical convection that is inherent in current theories of the QBO. It is also shown in another series of experiments that the oscillation is affected by the coefficients used for the subgrid-scale diffusion parameterization. These experiments demonstrate that at least one key reason why reasonably fine horizontal resolution is needed for the model to simulate a mean flow oscillation is the smaller horizontal diffusion that can be used at high resolution.
Eluszkiewicz, J, Richard S Hemler, Jerry D Mahlman, Lori Bruhwiler, and L L Takacs, 2000: Sensitivity of age-of-air calculations to the choice of advection scheme. Journal of the Atmospheric Sciences, 57(19), 3185-3201. Abstract PDF
The age of air has recently emerged as a diagnostic of atmospheric transport unaffected by chemical parameterizations, and the features in the age distributions computed in models have been interpreted in terms of the models' large-scale circulation field. This study shows, however, that in addition to the simulated large-scale circulation, three-dimensional age calculations can also be affected by the choice of advection scheme employed in solving the tracer continuity equation. Specifically, using the 3.0º latitude x 3.6º longitude and 40 vertical level version of the Geophysical Fluid Dynamics Laboratory SKYHI GCM and six online transport schemes ranging from Eulerian through semi-Lagrangian to fully Lagrangian, it will be demonstrated that the oldest ages are obtained using the nondiffusive centered-difference schemes while the youngest ages are computed with a semi-Lagrangian transport (SLT) scheme. The centered-difference schemes are capable of producing ages older than 10 years in the mesosphere, thus eliminating the "young bias" found in previous age-of-air calculations.
At this stage, only limited intuitive explanations can be advanced for this sensitivity of age-of-air calculations to the choice of advection scheme. In particular, age distributions computed online with the National Center for Atmospheric Research Community Climate Model (MACCM3) using different varieties of the SLT scheme are substantially older than the SKYHI SLT distribution. The different varieties, including a noninterpolating-in-the-vertical version (which is essentially centered-difference in the vertical), also produce a narrower range of age distributions than the suite of advection schemes employed in the SKYHI model. While additional MACCM3 experiments with a wider range of schemes would be necessary to provide more definitive insights, the older and less variable MACCM3 age distributions can plausibly be interpreted as being due to the semi-implicit semi-Lagrangian dynamics employed in the MACCM3. This type of dynamical core (employed with a 60-min time step) is likely to reduce SLT's interpolation errors that are compounded by the short-term variability characteristic of the explicit centered-difference dynamics employed in the SKYHI model (time step of 3 min). In the extreme case of a very slowly varying circulation, the choice of advection scheme has no effect on two-dimensional (latitude- height) age-of-air calculations, owing to the smooth nature of the transport circulation in 2D models.
These results suggest that nondiffusive schemes may be the preferred choice for multiyear simulations of tracers not overly sensitive to the requirement of monotonicity (this category includes many greenhouse gases). At the same time, age-of-air calculations offer a simple quantitative diagnostic of a scheme's long-term diffusive properties and may help in the evaluation of dynamical cores in multiyear integrations. On the other hand, the sensitivity of the computed ages to the model numerics calls for caution in using age of air as a diagnostic of a GCM's large-scale circulation field.
Hemler, Richard S., 2000: Key elements of the user-friendly, GFDL SKYHI general circulation model. Scientific Programming, 8(1), 39-47. Abstract PDF
Over the past seven years, the portability of the GFDL SKYHI general circulation model has greatly increased. Modifications to the source code have allowed SKYHI to be run on the GFDL Cray Research PVP machines, the TMC CM-5 machine at Los Alamos National Laboratory, and more recently on the GFDL 40-processor Cray Research T3E system. At the same time, changes have been made to the model to make it more usable and flexible. Because of the reduction of the human resources available to manage and analyze scientific experiments, it is no longer acceptable to consider only the optimization of computer resources when producing a research code; one must also consider the availability and cost of the people necessary to maintain, modify and use the model as an investigative tool, and include these factors in defining the form of the model code. The new SKYHI model attempts to strike a balance between the optimization of the use of machine resources (CPU time, memory, disc) and the optimal use of human resources (ability to understand code, ability to modify code, ability to perturb code to do experiments, ability to run code on different platforms).
Two of the key features that make the new SKYHI code more usable and flexible are the archiving package and the user variable block. The archiving package is used to manage the writing of all archive files, which contain data for later analysis. The model-supplied user variable block allows the easy inclusion of any new variables needed for particular experiments.
A high-resolution limited area nonhydrostatic model was used to simulate sulfate-cloud interactions during the convective activity in a case study from the Tropical Ocean Global Atmosphere Coupled Ocean Atmosphere Response Experiment, December 20-25, 1992. The model includes a new detailed sulfate-cloud microphysics scheme designed to estimate the effects of sulfate on cloud microphysics and radiative properties and the effects of deep convection on the transport and redistribution of aerosol. The data for SO2 and SO4(2-) species were taken from the Pacific Exploratory Mission West B observations during February-March 1994. Results show that a change in sulfate loading from the minimum to the maximum observed value scenarios (i.e., from about 0.01 to 1 µg m-3) causes a significant decrease of the effective radius of cloud droplets (changes up to 2 µm on average) and an increase of the diagnostic number concentration of cloud droplets (typical changes about 5-20 cm-3). The change in the average net shortwave (SW) radiation flux above the clouds was estimated to be on average -1.5 W m-2, with significant spatial and temporal variations. The horizontal average of the changes in the net SW radiation fluxes above clouds has a diurnal cycle, reaching typical values approximately -3 W m-2. The changes in the average net longwave radiation flux above the clouds were negligible, but they showed significant variations, typically between -10 W m-2 and 10 W m-2 near the surface. These variations were associated mainly with the changes in the distribution of cloud water, which showed typical relative changes of cloud water path of about 10-20%. Other notable changes induced by the increase of aerosol were the variations in air temperature of the order of 1°C. The case study presented here suggests that characteristics of convective clouds in tropical areas are sensitive to atmospheric sulfate loading, particularly during enhanced sulfate episodes.
Andronache, C, Leo J Donner, V Ramaswamy, Charles J Seman, and Richard S Hemler, 1999: Possible impact of atmospheric sulfur increase on tropical convective systems: A TOGA COARE Case In Proceedings of a Conference on the TOGA Coupled Ocean-Atmosphere Response Experiment (COARE) - COARE-98, WCRP-107, WMO/TD-No. 940, Geneva, Switzerland, WMO, 243-244.
Donner, Leo J., Charles J Seman, and Richard S Hemler, 1999: Ice microphysics and radiative transfer in deep convective systems In 10th Conference on Atmospheric Radiation, 28 June-2 July 1999, Madison, WI, American Meteorological Society, 611-614.
Deep convection and its associated mesoscale circulations are modeled using a three-dimensional elastic model with bulk microphysics and interactive radiation for a composite easterly wave from the Global Atmospheric Research Program Atlantic Tropical Experiment. The energy and moisture budgets, large-scale heat sources and moisture sinks, microphysics, and radiation are examined.
The modeled cloud system undergoes a life cycle dominated by deep convection in its early stages, followed by an upper-tropospheric mesoscale circulation. The large-scale heat sources and moisture sinks associated with the convective system agree broadly with diagnoses from field observations. The modeled upper-tropospheric moisture exceeds observed values. Strong radiative cooling at the top of the mesoscale circulation can produce overturning there. Qualitative features of observed changes in large-scale convective available potential energy and convective inhibition are found in the model integrations, although quantitative magnitudes can differ, especially for convection inhibition.
Radiation exerts a strong influence on the microphysical properties of the cloud system. The three-dimensional integrations exhibit considerably less sporadic temporal behavior than corresponding two-dimensional integrations. While the third dimension is less important over timescales longer than the duration of a phase of an easterly wave in the lower and middle troposphere, it enables stronger interactions between radiation and dynamics in the upper-tropospheric mesoscale circulation over a substantial fraction of the life cycle of the convective system.
Hamilton, Kevin P., R John Wilson, and Richard S Hemler, 1999: Middle atmosphere simulated with high vertical and horizontal resolution versions of a GCM: Improvements in the cold pole bias and generation of a QBO-like oscillation in the tropics. Journal of the Atmospheric Sciences, 56(22), 3829-3846. Abstract
The large-scale circulation in the Geophysical Fluid Dynamics Laboratory "SKYHI" troposphere-stratosphere-mesosphere finite-difference general circulation model is examined as a function of vertical and horizontal resolution. The experiments include one with horizontal grid spacing of ~35 km and another with ~100 km horizontal grid spacing but very high vertical resolution (160 levels between the ground and about 85 km). The simulation of the middle-atmospheric zonal-mean winds and temperatures in the extratropics is found to be very sensitive to horizontal resolution. For example, in the early Southern Hemisphere winter the South Pole near 1 mb in the model is colder than observed, but the bias is reduced with improved horizontal resolution (from ~70°C in a version with ~300 km grid spacing to less than 10°C in the ~35 km version). The extratropical simulation is found to be only slightly affected by enhancements of the vertical resolution. By contrast, the tropical middle atmospheric simulation is extremely dependent on the vertical resolution employed. With level spacing in the lower stratosphere ~1.5 km, the lower stratospheric zonal-mean zonal winds in the equatorial region are nearly constant in time. When the vertical resolution is doubled, the simulated stratospheric zonal winds exhibit a strong equatorially centered oscillation with downward propagation of the wind reversals and with formation of strong vertical shear layers. This appears to be a spontaneous internally generated oscillation and closely resembles the observed QBO in many respects, although the simulated oscillation has a period less than half that of the real QBO.
Convective clouds in tropical areas can be sensitive to the atmospheric sulfate loading, particularly during enhanced sulfate episodes. This assertion is supported by simulations with a high resolution limited area non-hydrostatic model (LAN) employing a detailed sulfate-cloud microphysics scheme, applied to estimate the effects of sulfate on convective clouds in a case study from the Tropical Ocean Global Atmosphere Coupled Ocean Atmosphere Response Experiment (TOGA COARE). Results show that a change in sulfate loading for scenarios using the minimum to the maximum observed values produces a change in the average net flux of shortwave radiation above clouds. This time-average change was estimated between -1.1 and -0.3 Wm -2 over the integration domain.
Hemler, Richard S., 1998: Key elements of the user-friendly, GFDL SKYHI general circulation model In Second International Workshop on Software Engineering and Code Design in Parallel Meteorological and Oceanographic Applications, NASA/Goddard Space Flight Center, 29-43.
Donner, Leo J., Charles J Seman, Richard S Hemler, and John P Sheldon, 1997: Radiative transfer in a three-dimensional cloud-system-resolving model In IRS '96: Current Problems in Atmospheric Radiation, Proceedings of the International Radiation Symposium, Fairbanks, Alaska, 19-24 August 1996. Hampton, Deepak Publishing, 109-112. Abstract
A three-dimensional, non-hydrostatic cloud-system-resolving model is used to study radiative transfer in convective systems. The model domain covers approximately 50,000 km2. Prognostic equations determine the evolution of liquid and ice mixing ratios. The three-dimensional distribution of liquid and ice is used in shortwave and long-wave radiative-transfer calculations.
A tropical convective system with a mesoscale anvil circulation is analyzed. The distribution of radiative forcing is examined, and its role in the evolution of the convective system is considered.
Ice clouds associated with large-scale atmospheric processes are studied using the SKYHI general circulation model (GCM) and parameterizations for their microphysical and radiative properties. The ice source is deposition from vapor, and the ice sinks are gravitational settling and sublimation. Effective particle sizes for ice distributions are related empirically to temperature. Radiative properties are evaluated as functions of ice path and effective size using approximations to detailed radiative-transfer solutions (Mie theory and geometric ray tracing). The distributions of atmospheric ice and their impact on climate and climate sensitivity are evaluated by integrating the SKYHI GCM (developed at the Geophysical Fluid Dynamics Laboratory) for six model months. Most of the major climatological cirrus regions revealed by satellite observations appear in the GCM. The radiative forcing associated with ice clouds acts to warm the Earth-atmosphere system. Relative to a SKYHI integration without these clouds, zonally averaged temperatures are warmer in the upper tropical troposphere with ice clouds. The presence of ice produced small net changes in the sensitivity of SKYHI climate to radiative perturbations, but this represents an intricate balance among changes in clear-, cloud-, solar-, and longwave-sensitivity components. Deficiencies in the representation of ice clouds are identified as results of biases in the large-scale GCM fields which drive the parameterization and neglect of subgrid variations in these fields, as well as parameterization simplifications of complex microphysical and radiative processes.
Hamilton, Kevin P., and Richard S Hemler, 1997: Appearance of a supertyphoon in a global climate model simulation. Bulletin of the American Meteorological Society, 78(12), 2874-2876.
Jones, P, Christopher Kerr, and Richard S Hemler, 1995: Practical considerations in development of a parallel SKYHI general circulation model. Parallel Computing, 21(10), 1677-1694. Abstract PDF
We have developed a parallel version of the SKYHI atmospheric general circulation model. The new parallel model has been designed for shared and distributed memory machines that support data parallel, message passing or worksharing programming paradigms. The newly developed model has a framework that makes the code easier to understand, maintain, and modify, increasing the model's flexibility and scientific productivity. Numerous model changes are described (code design, programming models, language choice, data decomposition, communications, table lookups, memory management, and i/o) that were necessary to develop the model. The performance and verification of the model is described on several systems including a shared-memory machine with high-level worksharing and a distributed-memory system with a data parallel programming paradigm.
Radiative-convective statistical equilibria are obtained using a two-dimensional model in which radiative transfer is interactive with the predicted moisture and cloud fields. The domain is periodic in x, with a width of 640 km, and extends from the ground to 26 km. The lower boundary is a fixed-temperature water-saturated surface. The model produces a temperature profile resembling the mean profile observed in the tropics. A number of integrations of several months' duration are described in this preliminary examination of the model's qualitative behavior.
The model generates a QBO-like oscillation in the x-averaged winds with an apparent period of ~60 days. This oscillation extends into the troposphere and influences the convective organization. In order to avoid the associated large vertical wind shears, calculations are also performed in which the x-averaged winds are constrained to vanish. The convection then evolves into a pattern in which rain falls only within a small part of the domain. The moisture field appears to provide the memory that localizes the convection. If the vertical shears are fixed in a modest nonzero value, this localization is avoided. Comparing calculations with surface temperatures of 25°C and 30°C, the planetary albedo is found to decrease with increasing temperature, primarily due to a reduction in low-level cloudiness.
Lipps, F B., and Richard S Hemler, 1992: On the downward transfer of tritium to the ocean by a cloud model. Journal of Geophysical Research, 97(D12), 12,889-12,900. Abstract
Observational evidence analyzed by Eriksson [1965] and Weiss and Roether [1980] suggests that over the globe the ratio of tritium deposition into the ocean by vapor diffusion to that by rainfall should be near or slightly greater than two, while Koster et al [1989] found in a general circulation model study that the diffusion to rainout ratio was closer to one. This study investigates the convective transport of tritium from the atmosphere to the ocean using a two-dimensional warm-rain cloud model. It is found that the deposition ratio is strongly dependent on the frequency and duration of rain events, with typical values for a monotonic tritium profile being about 1.0, but with values as low as 0.7 when long-duration events occur frequently and as high as 1.9 when convective events are short-lived and infrequent. On the basis of this study it appears that explicit treatment of convection in the general circulation model would not resolve the discrepancy in the deposition ratio between the model and the observations. It is also shown that the process of isotopic adjustment of tritium between the rain and the vapor phases is a key factor in determining the deposition ratio, since this process allows tritium to escape from the raindrops and ultimately diffuse to the surface. When tritium is "frozen" in the droplets, deposition ratios are reduced significantly.
Hemler, Richard S., F B Lipps, and Bruce B Ross, 1991: A simulation of a squall line using a nonhydrostatic cloud model with a 5-km horizontal grid. Monthly Weather Review, 119(12), 3012-3033. Abstract PDF
A three-dimensional nonhydrostatic cloud model is used to simulate the squall line observed in central Texas on 11 April 1979. The cloud model covers an area 400 x 400 km2 with a 5-km horizontal resolution and is supplied initial and boundary conditions by a larger hydrostatic mesoscale model.
The model produces a back-building squall line ahead of the surface cold front, as would be expected based on an analysis of the pre-squall-line environment. A well-defined gust front and cold pool develop with the squall line. At the end of the 5-h simulation, deep convection is found along a line nearly 400 km long. The simulated squall line compares favorably both with observations and with a higher-resolution model simulation in an environment of similar shear, suggesting that the 5-km horizontal resolution is adequately representing the significant features of the squall line.
The major shortcoming of this study is the failure of the cloud model to produce the observed squall line at the proper time. Without the observed small-scale forcing, which was unresolved in the Severe Environmental Storms and Mesoscale Experiment (SESAME) dataset, the model is unable to generate the squall line until a larger-scale convergence area evolves, some 2-3 h after the appearance of the observed squall line.
Lipps, F B., and Richard S Hemler, 1991: Numerical modeling of a midlatitude squall line: Features of the convection and vertical momentum flux. Journal of the Atmospheric Sciences, 48(17), 1909-1929. Abstract PDF
A 4-h simulation is carried out for the 22 May 1976 squall line that passed through the mesonetwork of the National Severe Storms Laboratory in central Oklahoma. This squall line was more than 100 km wide, oriented north-south and traveled eastward at approximately 14 m s-1. It produced rainfall of 2-h duration at surface stations.
The simulation was obtained from a three-dimensional convective cloud model with open lateral boundary conditions on the east and west, and periodic conditions on the north and south boundaries. The model domain is 96 km long (east-west) and 32 km wide (north-south) with a horizontal grid resolution of 1.0 km and a vertical resolution of 0.5 km. A squall line develops and moves eastward at 13.7 m s-1 during the last two hours of the simulation. The present mesoy-scale model, however, can only simulate the leading edge of the squall line, with rain at specific surface locations lasting only 30 min. Realistic features of the modeled flow include the surface westerlies moving faster than the line behind the gust front, the strong easterlies in the lower cloud levels, and the cold boundary layer behind the gust front.
Two-hour time means of the vertical momentum flux are calculated in a 60-km-wide domain (east-west) following the squall line. The vertical disturbance momentum flux for momentum normal to the line agrees with observations and is primarily confined to this region adjacent to the squall line. Horizontal-averaged time-mean momentum budgets are also calculated in this domain. For the normal component of momentum, this budget is in a quasi-steady state. It cannot be in a fully steady state as the gust front moves 1.2 m s-1 faster than the area of rain behind the line for the 2-h time mean.
The parameterization of Schneider and Lindzen for the vertical momentum flux associated with active clouds is compared with mean data from the simulation. Their parameterization accounts for the in-cloud vertical momentum flux reasonably well, but ignores the remaining flux associated with convective-scale downdrafts, which is significant in lower levels.
Lipps, F B., and Richard S Hemler, 1988: An investigation of the role of snow in a squall line anvil circulation In Report of the Second International Cloud Modelling Workshop, Toulose, 8-12 August 1988, WMP Report No. 11, WMO/TD No. 268, World Meteorological Organization, 163-167.
Lipps, F B., and Richard S Hemler, 1988: Numerical modeling of a line of towering cumulus on Day 226 of GATE. Journal of the Atmospheric Sciences, 45(17), 2428-2444. Abstract PDF
A three-dimensional numerical model with warm rain bulk cloud physics is used to investigate the shallow convection observed on day 226 of GATE. This convection had cloud tops at 3.0 km, cloud bases at 0.4 km and approximately 0.1 cm of rain at the surface. The simulated convection shows a strong sensitivity to the criterion for the onset of autoconversion of cloud water into rain water. The strongest convection occurs for the case in which no rain water forms. This case, however, does not conform to the observed convection, lacking the downdraft below cloud base and the observed strong surface outflow.
The primary simulation produces a "finger" of convection propagating to the northeast, perpendicular to the northwest-southeast orientation of the larger-scale line of convection. The orientation and propagation has a well-defined leading edge and strong surface outflow as observed. In poorer agreement, the cloud base was too high and the rainfall at the surface was less than observed.
Present calculations indicate that the boundary layer air is flowing through the line from southwest to northeast below cloud base. The primary moisture source for the cloud is the upper half of the subcloud layer, with nearly horizontal flow entering the cloud.
Lipps, F B., and Richard S Hemler, 1986: Numerical simulation of deep tropical convection associated with large-scale convergence. Journal of the Atmospheric Sciences, 43(17), 1796-1816. Abstract PDF
A set of four-hour simulations has been carried out to study deep moist convection characteristic of the Global Atmospheric Research Program (GARP) Atlantic Tropical Experiment (GATE). The present model includes warm rain bulk cloud physics and effects associated with a large-scale, time-invariant convergence. The convection took approximately two hours to develop from a random moisture disturbance. The cloud efficiency, in terms of the total water vapor condensed, was near 40%.
The heat and moisture budgets and the time-mean vertical fluxes of mass, heat, and moisture were calculated for the last 80 minutes of the simulations. In this study the primary emphasis was placed upon run A, the three-dimensional calculation. For this calculation, the layer centered near 4.0 km was a region of low mean cloudiness but of strong convection. The upward mass flux was strong and upward heat and moisture fluxes had maximum values in this layer. The strongest downward mass flux was due to weak downward velocities in the rainy area below cloud base.
Time-mean data were also calculated for vertical velocity cores and compared with observed data. In run A, virtually all updraft cores are in-cloud and for a deep layer between 2.5 and 8.0 km the in-cloud upward mass flux is nearly all associated with cores. In this layer the upward mass flux due to cores is approximately twice the mass flux associated with the large-scale convergence. The fractional area of updraft cores is small, varying between 2.5% and 4.0% for the vertical levels between 1 and 11 km. Calculated values of core diameter D are in relatively good agreement with the observed data. For values of mean vertical velocity w, however, the agreement is not nearly as good. For downdraft cores, values of w are significantly smaller than the observations. For updraft cores, values of w at lower levels are small, whereas values in the upper levels are in reasonable agreement with observations. The weak updraft cores at lower levels may be related to the absences of strong gust fronts in the present simulations.
Lipps, F B., and Richard S Hemler, 1985: Another look at the scale analysis for deep moist convection. Journal of the Atmospheric Sciences, 42(18), 1960-1964. Abstract PDF
In this note, a more rational approach is given to specify the parameters G and B in the scale analysis of Lipps and Hemler. The thermodynamic equation is written in a different form so that a closed expression for B can be derived. The present values of G and B are very similar to those in the previous scale analysis. A new result is that the time scale t is expressed in terms of moist convective instability rather than the inverse of the Brunt-Vaisala frequency.
The ratio of volume integrated kinetic energy to volume integrated first-order sensible heat is also discussed in more detail. It is found that for an accurate estimate of sensible heat the region of compensating downward motion between the active clouds must be taken into account. As indicated by earlier authors, the amount of sensible heat produced inside the clouds is relatively small.
Lipps, F B., and Richard S Hemler, 1985: Numerical modelling of a midlatitude squall line In 14th Conference on Severe Local Storms, Boston, MA, American Meteorological Society, 183-185.
Lipps, F B., and Richard S Hemler, 1982: A scale analysis of deep moist convection and some related numerical calculations. Journal of the Atmospheric Sciences, 39(10), 2192-2210. Abstract PDF
A scale analysis valid for deep moist convection is carried out. The approximate equations of motion are anelastic with the time scale set by the Brunt-Vaisala frequency. A new assumption is that the base state potential temperature is a slowly varying function of the vertical coordinate. It is this assumption that eliminates the energetic inconsistency discussed by Wilhelmson and Ogura (1972) for a non-isentropic base state. Another key result is that the dynamic pressure is an order of magnitude smaller than the first-order temperature and potential temperature. In agreement with observations, the kinetic energy is found to be an order of magnitude smaller than the first-order thermodynamic energy.
A set of six numerical simulations representing moderately deep moist convection is carried out. The base state is an idealized maritime tropical sounding with no vertical wind shear. The first calculation (Run A) shows the growth and dissipation of a typical shower cloud. The remaining calculations have small changes in either initial conditions or model equations from Run A. These calculations indicate the sensitivity of the present model to different approximations and give additional evidence for the validity of the scale analysis.
Lipps, F B., and Richard S Hemler, 1981: Reply. Monthly Weather Review, 109(3), 675. PDF
Lipps, F B., and Richard S Hemler, 1980: Another look at the thermodynamic equation for deep convection. Monthly Weather Review, 108(1), 78-84. Abstract PDF
The study considers deep moist convection involving only a liquid-vapor phase change. An alternative form of the classical thermodynamic equation for reversible saturated flow is derived. Four approximate forms of this equation are obtained and their relative errors compared to the full equation are evaluated by using parcel theory. The best approximation is found to be an adequate respresentation of the full equation throughout the total depth of the convection.
The two best approximations are compared with some forms of the thermodynamic equation used by other investigators.
