Numerical general circulation models of the atmosphere are generally required to conserve mass and energy for their application to climate studies. Here we draw attention to another conserved global integral, viz. the component of angular momentum (AM) along the Earth's axis of rotation, which tends to receive less consideration. We demonstrate the importance of global AM conservation in climate simulations with the example of the Community Atmosphere Model (CAM) with the finite-volume (FV) dynamical core, which produces a noticeable numerical sink of AM. We use a combination of mathematical analysis and numerical diagnostics to pinpoint the main source of AM non-conservation in CAM–FV. We then present a method to enforce global conservation of AM, and we discuss the results in a hierarchy of numerical simulations of the atmosphere of increasing complexity. In line with theoretical expectations, we show that even a crude, non-local enforcement of AM conservation in the simulations consistently results in the mitigation of certain persistent model biases.

This paper describes the implementation of a coarser-resolution physics grid into the Community Atmosphere Model (CAM), containing 5/9 fewer grid columns than the dynamics grid. The dry dynamics is represented by the spectral element dynamical core, and tracer transport is computed using the Conservative Semi-Lagrangian Finite Volume Method (CAM-SE-CSLAM). Algorithms are presented that map fields between the dynamics and physics grids while maintaining numerical properties ideal for atmospheric simulations such as mass conservation and mixing ratio shape and linear-correlation preservation. The results of experiments using the lower-resolution physics grid are compared to the conventional method in which the physics and dynamical grids coincide. The lower-resolution physics grid provides a volume mean state to the physics computed from an equal sampling of the different types of nodal solutions arising in the spectral-element method and effectively mitigates grid imprinting in regions with steep topography. The impact of the coarser-resolution physics grid on the resolved scales of motion is analyzed in an aquaplanet configuration, across a range of dynamical core grid resolutions. The results suggest that the effective resolution of the model is not degraded through the use of a coarser-resolution physics grid. Since the physics makes up about half the computational cost of the conventional CAM-SE-CSLAM configuration, the coarser physics grid may allow for significant cost savings with little to no downside.

Atmospheric modeling with element-based high-order Galerkin methods presents a unique challenge to the conventional physics–dynamics coupling paradigm, due to the highly irregular distribution of nodes within an element and the distinct numerical characteristics of the Galerkin method. The conventional coupling procedure is to evaluate the physical parameterizations ( physics) on the dynamical core grid. Evaluating the physics at the nodal points exacerbates numerical noise from the Galerkin method, enabling and amplifying local extrema at element boundaries. Grid imprinting may be substantially reduced through the introduction of an entirely separate, approximately isotropic finite-volume grid for evaluating the physics forcing. Integration of the spectral basis over the control volumes provides an area-average state to the physics, which is more representative of the state in the vicinity of the nodal points rather than the nodal point itself and is more consistent with the notion of a “large-scale state” required by conventional physics packages. This study documents the implementation of a quasi-equal-area physics grid into NCAR’s Community Atmosphere Model Spectral Element and is shown to be effective at mitigating grid imprinting in the solution. The physics grid is also appropriate for coupling to other components within the Community Earth System Model, since the coupler requires component fluxes to be defined on a finite-volume grid, and one can be certain that the fluxes on the physics grid are, indeed, volume averaged.

It is the purpose of this paper to provide a comprehensive documentation of the new NCAR (National Center for Atmospheric Research) version of the spectral element (SE) dynamical core as part of the Community Earth System Model (CESM2.0) release. This version differs from previous releases of the SE dynamical core in several ways. Most notably the hybrid sigma vertical coordinate is based on dry air mass, the condensates are dynamically active in the thermodynamic and momentum equations (also referred to as condensate loading), and the continuous equations of motion conserve a more comprehensive total energy that includes condensates. Not related to the vertical coordinate change, the hyperviscosity operators and the vertical remapping algorithms have been modified. The code base has been significantly reduced, sped up, and cleaned up as part of integrating SE as a dynamical core in the CAM (Community Atmosphere Model) repository rather than importing the SE dynamical core from High-Order Methods Modeling environment as an external code.

Spaceborne lidar observations from the Cloud‐Aerosol Lidar and Infrared Pathfinder Satellite Observation (CALIPSO) satellite are used to evaluate cloud amount and cloud phase in the Community Atmosphere Model version 5 (CAM5), the atmospheric component of a widely used state‐of‐the‐art global coupled climate model (Community Earth System Model). By embedding a lidar simulator within CAM5, the idiosyncrasies of spaceborne lidar cloud detection and phase assignment are replicated. As a result, this study makes scale‐aware and definition‐aware comparisons between model‐simulated and observed cloud amount and cloud phase. In the global mean, CAM5 has insufficient liquid cloud and excessive ice cloud when compared to CALIPSO observations. Over the ice‐covered Arctic Ocean, CAM5 has insufficient liquid cloud in all seasons. Having important implications for projections of future sea level rise, a liquid cloud deficit contributes to a cold bias of 2–3°C for summer daily maximum near‐surface air temperatures at Summit, Greenland. Over the midlatitude storm tracks, CAM5 has excessive ice cloud and insufficient liquid cloud. Storm track cloud phase biases in CAM5 maximize over the Southern Ocean, which also has larger‐than‐observed seasonal variations in cloud phase. Physical parameter modifications reduce the Southern Ocean cloud phase and shortwave radiation biases in CAM5 and illustrate the power of the CALIPSO observations as an observational constraint. The results also highlight the importance of using a regime‐based, as opposed to a geographic‐based, model evaluation approach. More generally, the results demonstrate the importance and value of simulator‐enabled comparisons of cloud phase in models used for future climate projection.

Satellite observations and their corresponding instrument simulators are used to document global cloud biases in the Community Atmosphere Model (CAM) versions 4 and 5. The model observation comparisons show that, despite having nearly identical cloud radiative forcing, CAMS has a much more realistic representation of cloud properties than CAM4. In particular, CAM5 exhibits substantial improvement in three long-standing climate model cloud biases: 1) the underestimation of total cloud, 2) the overestimation of optically thick cloud, and 3) the underestimation of midlevel cloud. While the increased total cloud and decreased optically thick cloud in CAM5 result from improved physical process representation, the increased midlevel cloud in CAM5 results from the addition of radiatively active snow. Despite these improvements, both CAM versions have cloud deficiencies. Of particular concern, both models exhibit large but differing biases in the subtropical marine boundary layer cloud regimes that are known to explain intermodel differences in cloud feedbacks and climate sensitivity. More generally, this study demonstrates that simulator-facilitated evaluation of cloud properties, such as amount by vertical level and optical depth, can robustly expose large and at times radiatively compensating climate model cloud biases.

The authors have decomposed the anthropogenic aerosol radiative forcing into direct contributions from each aerosol species to the planetary energy balance through absorption and scattering of solar radiation, indirect effects of anthropogenic aerosol on solar and infrared radiation through droplet and crystal nucleation on aerosol, and semidirect effects through the influence of solar absorption on the distribution of clouds. A three-mode representation of the aerosol in version 5.1 of the Community Atmosphere Model (CAM5.1) yields global annual mean radiative forcing estimates for each of these forcing mechanisms that are within 0.1 W m(-2) of estimates using a more complex seven-mode representation that distinguishes between fresh and aged black carbon and primary organic matter. Simulating fresh black carbon particles separately from internally mixed accumulation mode particles is found to be important only near fossil fuel sources. In addition to the usual large indirect effect on solar radiation, this study finds an unexpectedly large positive longwave indirect effect (because of enhanced cirrus produced by homogenous nucleation of ice crystals on anthropogenic sulfate), small shortwave and longwave semidirect effects, and a small direct effect (because of cancelation and interactions of direct effects of black carbon and sulfate). Differences between the three-mode and seven-mode versions are significantly larger (up to 0.2 W m(-2)) when the hygroscopicity of primary organic matter is decreased from 0.1 to 0 and transfer of the primary carbonaceous aerosol to the accumulation mode in the seven-mode version requires more hygroscopic material coating the primary particles. Radiative forcing by cloudborne anthropogenic black carbon is only -0.07 W m(-2).

The Community Atmosphere Model version 3 (CAM3) is the latest generation of a long lineage of general circulation models produced by a collaboration between the National Center for Atmospheric Research (NCAR) and the scientific research community. Many aspects of the hydrological cycle have been changed relative to earlier versions of the model. It is the goal of this paper to document some aspects of the tropical variability of clouds and the hydrologic cycle in CAM3 on time scales shorter than 30 days and to discuss the differences compared to the observed atmosphere and earlier model versions, with a focus on cloud-top brightness temperature, precipitation, and cloud liquid water path. The transient behavior of the model in response to changes in resolution to various numerical methods used to solve the equations for atmospheric dynamics and transport and to the underlying lower boundary condition of sea surface temperature and surface fluxes has been explored.The ratio of stratiform to convective rainfall is much too low in CAM3, compared to observational estimates. It is much higher in CAM3 (10%) than the Community Climate Model version 3 (CCM3; order 1%-2%) but is still a factor of 4-5 too low compared to observational estimates. Some aspects of the model transients are sensitive to resolution. Higher-resolution versions of CAM3 show too much variability (both in amplitude and spatial extent) in brightness temperature on time scales of 2-10 days compared to observational estimates. Precipitation variance is underestimated on time scales from a few hours to 10 days, compared to observations over ocean, although again the biases are reduced compared to previous generations of the model. The diurnal cycle over tropical landmasses is somewhat too large, and there is not enough precipitation during evening hours. The model tends to produce maxima in precipitation and liquid water path that are a few hours earlier than that seen in the observations over both oceans and land.

The direct radiative forcing by aerosols over the Indian Ocean region is simulated for the Indian Ocean Experiment (INDOEX) Intensive Field Phase during Spring 1999. The forcing is calculated for the top‐of‐atmosphere (TOA), surface, and atmosphere by differencing shortwave fluxes computed with and without aerosols. The calculation includes the effects of sea‐salt, sulfate, carbonaceous, and soil‐dust aerosols. The aerosol distributions are obtained from a global aerosol simulation including assimilation of satellite retrievals of aerosol optical thickness (AOT). The time‐dependent, three‐dimensional aerosol distributions are derived with a chemical transport model driven with meteorological analyses for this period. The surface albedos are obtained from a land‐surface model forced with an identical meteorological analysis and satellite‐derived rainfall and insolation. These calculations are consistent with in situ observations of the surface insolation over the central Indian Ocean and with satellite measurements of the reflected shortwave radiation. The calculations show that the surface insolation under clear skies is reduced by as much as 40 W/m2 over the Indian subcontinent by natural and anthropogenic aerosols. This reduction in insolation is accompanied by an increase in shortwave flux absorbed in the atmosphere by 25 W/m2. The inclusion of clouds in the calculations changes the direct effect by less than 2 W/m2 over the Indian subcontinent, although the reduction is much larger over China. The magnitude of the difference between all‐sky and clear‐sky forcing is quite sensitive to the three‐dimensional spatial relationship between the aerosol and cloud fields, and other estimates of the difference for the INDOEX Intensive Field Phase are as large as 5 W/m2.

A system for simulating aerosols has been developed using a chemical transport model together with an assimilation of satellite aerosol retrievals. The methodology and model components are described in this paper, and the modeled distribution of aerosols for the Indian Ocean Experiment (INDOEX) is presented by Rasch et al . [this issue]. The system generated aerosol forecasts to guide deployment of ships and aircraft during INDOEX. The system consists of the Model of Atmospheric Transport and Chemistry (MATCH) combined with an assimilation package developed for applications in atmospheric chemistry. MATCH predicts the evolution of sulfate, carbonaceous, and mineral dust aerosols, and it diagnoses the distribution of sea salt aerosols. The model includes a detailed treatment of the sources, chemical transformation, transport, and deposition of the aerosol species. The aerosol forecasts involve a two‐stage process. During the assimilation phase the total column aerosol optical depth (AOD) is estimated from the model aerosol fields. The model state is then adjusted to improve the agreement between the simulated AOD and satellite retrievals of AOD. During the subsequent integration phase the aerosol fields are evolved using meteorological fields from an external model. Comparison of the modeled AOD against estimates of the AOD from INDOEX Sun photometer data show that the differences in daily means are −0.03±0.06. Although the initial application is limited to the Indian Ocean, the methodology could be extended to derive global aerosol analyses combining in situ and remotely sensed aerosol observations.

We use the aerosol assimilation procedure described by Collins et al. [2000] to help explain INDOEX aerosol distributions. The procedure combines modeled aerosol with AVHRR satellite estimates. The result is consistent with satellite measurements, regular in space and time, and provides information where retrievals are difficult (over land, coincident with clouds, and at night). Extra information on aerosol composition, vertical distribution, and region of origin is also produced. Carbonaceous, sulfate, and sea salt aerosols agree with the in situ measurements to 10–20%. Carbonaceous aerosols were estimated to be the dominant contributor (36%) to the aerosol optical depth (AOD); dust (31%) and sulfate (26%) were also important. The residence time for sulfate and carbon is ∼7 and ∼8 days respectively, longer than globally averaged residence times of many modeling studies. Thus aerosols produced here during the winter monsoon may have a larger climate impact than the same emissions occurring where the residence time is shorter. Three points of entry are found for anthropogenic aerosol to the INDOEX region: a strong near surface southward flow near Bombay; a deeper plume flowing south and east off Calcutta and a westward flow originating from southeast Asia and entering the Bay of Bengal. All three plumes are strongly modulated by a low‐frequency change of meteorological regime associated with the Madden Julian Oscillation. The analysis suggests that India is the dominant source of aerosol in the Arabian Sea and Bay of Bengal near the surface but that Asia, Africa and the rest of world also contribute at higher altitudes. India and Asia contribute ∼40% each to the total column mass of air reaching the Maldives, the balance of air comes from other source regions. The assimilation procedure produces an analysis that is a synergy in information about aerosols, that is not easily accessible by independent estimates from remote sensing, in situ measurements, or global transport models by themselves.

A fundamental difficulty in 3‐D models is addressed which can arise due to inconsistencies between advection schemes and winds. It is shown that a model will fail to meet certain basic criteria, desired of 3‐D transport models, when the density field computed by the advection scheme from the winds differs from the implied density field based on the surface pressure and the sigma (or hybrid) coordinates. To allow a rigorous mathematical formulation, the focus is on the example of a mass flux advection scheme in a model where the winds and surface pressure are derived from different advection schemes (e.g. a spectral scheme in a climate model or a weather centre model); however, in principle the discussion applies to nearly any situation in which the pressure levels change in a model. To illustrate the potential severity of such problems, a mass conserving gridto‐grid transformation scheme is constructed which only uses the current tracer mass mixing‐ratio distribution. It is shown that only one solution exists that is comprehensively valid for any arbitrary tracer distribution, and that this type of correction introduces an additional undesired artificial vertical diffusion component into the model transport that increases with increasing tracer mass mixing‐ratio gradients and may exceed the physical vertical transport itself. It is demonstrated that the results of any supplementary fix, either mass fixer or grid‐to‐grid transformation, are generally unacceptable for global modelling applications. From this, it is concluded that the only alternative which can produce reliable results for any arbitrary tracer is to maintain a consistent grid throughout the entire model time step, where all changes in pressure levels due to modelled advection exactly match the changes implied by the surface pressure at the next time step. Although this is already done in some models, this would require significant changes in the structure of the advection scheme or its input wind fields in several other contemporary general circulation and chemistry transport models.

A model of atmospheric photochemistry and transport has been developed and applied toward investigating global tropospheric chemistry. The Model of Atmospheric Transport and Chemistry Mas-Planck-Institute for Chemistry version (MATCH-MPIC) is described and key characteristics of its global simulation are presented and compared to available observations. MATCH-MPIC is an "offline" model which reads in gridded time-dependent values for the most basic meteorological parameters (e.g., temperature, surface pressure, horizontal winds), then uses these to compute further meteorological parameters required for atmospheric chemistry simulations (convective transport, cloud microphysics, etc.). The meteorology component of MATCH-MPIC simulates transport, by advection, convection, and dry turbulent mixing, as well as the full tropospheric hydrological cycle (water vapor transport, condensation, evaporation, and precipitation). The photochemistry component of MATCH-MPIC represents the major known sources (e.g., industry, biomass burning), transformations (chemical reactions and photolysis), and sinks (e.g., wet and dry deposition) which affect the O-3-HOx-NOy-CH4-CO photochemical framework of the "background" troposphere. The results of two versions of the model are considered, focusing on the more recent version. O-3 is in relatively good agreement with observed soundings, although it is generally underestimated at low levels and overestimated at high levels, particularly for the more recent version of the model. We conclude that the simulated stratosphere-troposphere flux of O-3 is too large, despite the fact that the total flux is 1100 Tg(O-3)/yr, whereas the upper limit estimated in recent literature is over 1400 Tg(O-3)/yr. The OH distribution yields a tropospheric CH4 lifetime of 10.1 years, in contrast to the Lifetime of 7.8 years in the earlier model version, which nearly spans the range of current estimates in the literature (7.5-10.2 years). Surface CO mixing ratios are in good agreement with observations. NO is generally underestimated, a problem similar to what has also been found in several other recent model studies. HNO3 is also considerably underestimated. H2O2 and CH3OOH, on the other hand, are in relatively good agreement with available observations: though both tend to be underestimated at high concentrations and overestimated at low concentrations. Possible reasons for these differences are considered.

To what extent can we trust the results of global atmospheric models? It is well known that a substantial degree of uncertainty exists in the parameters used in these models, for instance in parameterizations of complex physical processes such as convection and in reaction rates for photochemical models. Anyone familiar with computer work is also well aware of occasional “bugs,” such as an “l” typed where a “j” was intended.However, many newcomers to the field of atmospheric modeling (or other similar Earth systems modeling endeavors) are not yet very familiar with the technical side of modeling, have little or no formal education in computer programming, and are instead expected to learn the art of programming on the job. These researchers are often unaware of some of the types of artifacts that can be generated by the software and hardware they use.

We compare chemical transport simulations performed in a model using archived meteorological data (an off‐line transport model) to those performed in a model in which the meteorological data are predicted every time step (an on‐line model). We identify the errors associated with using data sampled at timescales much longer than those operating in the atmosphere or in the on‐line model, and strategies for ameleorating those errors. The evaluation is performed in the context of a global off‐line chemical transport model called the Model of Atmospheric Transport and Chemistry (MATCH) for three test problems: (1) the passive advection of blobs initially concentrated in the lower and upper troposphere; (2) the surface emission of radon and its decay to lead; and (3) the removal of lead from the atmosphere by wet and dry deposition processes. These problems exercise the important processes of transport by resolved scale winds, rapid transport by smaller scale convection processes, and wet removal (which depends on the representation of the hydrologic cycle). We show that the errors in off‐line model simulations (compared to the on‐line simulations) can be made small when the sampling interval is order 6 hours or less. We also show that one can accurately reproduce the subgrid‐scale processes within the off‐line model, rather than needing to archive the results of those processes as input to the off‐line model. This suggests that for the spatial and temporal scales treated in global models it is possible to treat many problems nearly as accurately in an off‐line mode as one can with an on‐line treatment.

The Model of Atmospheric Transport and Chemistry (MATCH) is used to simulate the transport of 222Rn using both European Centre for Medium-Range Weather Forecasts (ECMWF) winds and National Center for Environmental Prediction/National Center for Atmospheric Research (hereafter referred to as NCEP) reanalysis winds. These winds have the advantage of being based on observed winds but have the disadvantage that the subgrid-scale transport processes are not routinely archived. MATCH derives subgrid-scale mixing rates for the boundary layer using a nonlocal scheme and for moist convective mixing using one of two parameterizations (Tiedtke (1989) or Pan and Wu (1997)). This paper describes the ability of the model to recreate mixing rates of 222Rn using the forecast center winds. Radon 222 is a species with a continental crust source and a simple sink involving radioactive decay with an e-folding timescale of 5.5 days. This atmospheric constituent is therefore a good tracer for testing the vertical transport in the chemical transport model, as well as the horizontal transport from continental regions to remote oceanic regions. The various simulations of 222Rn are compared with observations as well as with each other, allowing an estimate of the uncertainty in transport due to uncertainties in the winds and subgrid-scale processes. The calculated vertical profiles over the western United States are somewhat similar to observed, and the upper tropospheric concentrations compare reasonably well in their spatial distribution with data collected during Tropospheric Ozone TT (TROPOZ TT), although the model values tend to be higher than observed values, especially in the upper troposphere. The model successfully simulates specific observed pollution events at Cape Grim. It has more difficulty at sites farther from continental source regions, although the model captures the seasonal structure of the pollution events at these sites (Macquarie Island, Amsterdam Island, Kerguelen Island, and Crozet Island). Inclusion of a moist convective mixing scheme in 222 MATCH increases Rn concentrations in the upper troposphere by 50% compared to not having moist convective mixing, while surface concentrations do not appear to be very sensitive to moist convection. In addition, differences between the upper tropospheric concentrations of radon predicted using the ECMWF and NCEP winds can be 30% for large areas of the globe, due to either differences in the forecast center winds themselves or the moist convective mixing schemes used in conjunction with them. This has implications for model simulations of radiatively and chemically important trace species in the atmosphere.

The Model of Atmospheric Transport and Chemistry (MATCH) is used to simulate the transport of 222Rn using both European Centre for Medium‐Range Weather Forecasts (ECMWF) winds and National Center for Environmental Prediction/National Center for Atmospheric Research (hereafter referred to as NCEP) reanalysis winds. These winds have the advantage of being based on observed winds but have the disadvantage that the subgrid‐scale transport processes are not routinely archived. MATCH derives subgrid‐scale mixing rates for the boundary layer using a nonlocal scheme and for moist convective mixing using one of two parameterizations (Tiedtke [1989] or Pan and Wu [1997]). This paper describes the ability of the model to recreate mixing rates of 222Rn using the forecast center winds. Radon 222 is a species with a continental crust source and a simple sink involving radioactive decay with an e‐folding timescale of 5.5 days. This atmospheric constituent is therefore a good tracer for testing the vertical transport in the chemical transport model, as well as the horizontal transport from continental regions to remote oceanic regions. The various simulations of 222Rn are compared with observations as well as with each other, allowing an estimate of the uncertainty in transport due to uncertainties in the winds and subgrid‐scale processes. The calculated vertical profiles over the western United States are somewhat similar to observed, and the upper tropospheric concentrations compare reasonably well in their spatial distribution with data collected during Tropospheric Ozone II (TROPOZ II), although the model values tend to be higher than observed values, especially in the upper troposphere. The model successfully simulates specific observed pollution events at Cape Grim. It has more difficulty at sites farther from continental source regions, although the model captures the seasonal structure of the pollution events at these sites (Macquarie Island, Amsterdam Island, Kerguelen Island, and Crozet Island). Inclusion of a moist convective mixing scheme in MATCH increases 222Rn concentrations in the upper troposphere by 50% compared to not having moist convective mixing, while surface concentrations do not appear to be very sensitive to moist convection. In addition, differences between the upper tropospheric concentrations of radon predicted using the ECMWF and NCEP winds can be 30% for large areas of the globe, due to either differences in the forecast center winds themselves or the moist convective mixing schemes used in conjunction with them. This has implications for model simulations of radiatively and chemically important trace species in the atmosphere.

Computed geophysical fields from a 3‐D general circulation model are coupled with the Coastal Zone Color Scanner (CZCS) satellite data on chlorophyll content of surface ocean waters. The CZCS satellite data on chlorophyll content of surface ocean waters are used to estimate the ‘photochemical lability’ of dissolved organic matter in the surface ocean. Monthly estimates are made of the global ocean to atmosphere flux of a biogeochemically important gas, carbonyl sulfide (OCS), with 2.8° × 2.8° latitude‐longitude spatial resolution. This novel technique provides a conceptual and computational method for integrating data collected as part of future satellite measurement campaigns, such as the Earth Observing System (EOS) and Sea‐viewing Wide‐field‐of view Sensor (Sea WiFS), with 3‐D chemistry‐climate prediction models.