The Geophysical Fluid Dynamics Laboratory (GFDL) has developed a coupled general circulation model (CM3) for the 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, subgrid 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 ecosystem dynamics and hydrology. Its horizontal resolution is approximately 200 km, and its vertical resolution ranges approximately from 70 m near the earth’s surface to 1 to 1.5 km near the tropopause and 3 to 4 km in much of the stratosphere. Most basic circulation features in AM3 are simulated as realistically, or more so, as 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 remains problematic, as in AM2. The most intense 0.2% of precipitation rates occur less frequently in AM3 than observed. The last two decades of the twentieth century warm in CM3 by 0.32°C relative to 1881–1920. The Climate Research Unit (CRU) and Goddard Institute for Space Studies analyses of observations show warming of 0.56° and 0.52°C, respectively, over this period. CM3 includes anthropogenic cooling by aerosol–cloud interactions, and its warming by the late twentieth century is somewhat less realistic than in CM2.1, which warmed 0.66°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.

The Madden–Julian oscillation (MJO) produces alternating periods of increased and reduced precipitation and African easterly wave (AEW) activity in West Africa. This study documents the influence of the MJO on the West African monsoon system during boreal summer using reanalysis and brightness temperature fields. MJO-related West African convective anomalies are likely induced by equatorial Kelvin and Rossby waves generated in the Indian Ocean and West Pacific by the MJO, which is consistent with previous studies. The initial modulation of tropical African convection occurs upstream of West Africa, near the entrance of the African easterly jet (AEJ). Previous studies have hypothesized that an area to the east of Lake Chad is an initiation region for AEWs. Called the “trigger region” in this study, this area exhibits significant intraseasonal convection and wave activity anomalies prior to the wet and dry MJO phases in the West African monsoon region. In the trigger region, cold tropospheric temperature anomalies and high precipitable water, as well as an eastward extension of the African easterly jet, appear to precede and contribute to the wet MJO phase in West Africa. An anomalous stratiform heating profile is observed in advance of the wet MJO phase with anomalous PV generation maximized at the jet level. The opposite behavior occurs in advance of the dry MJO phase. The moisture budget is examined to provide further insight as to how the MJO modulates and initiates precipitation and AEW variability in this region. In particular, meridional moisture advection anomalies foster moistening in the trigger region in advance of the wet MJO phase across West Africa.

This study demonstrates a link between coastal downwelling and tropical cyclone (TC) intensification. We show that coastal downwelling increases air‐sea enthalpy (heat, moisture) fluxes ahead of TCs as they approach land, creating conditions conducive to intensification even in the presence of typically inhibiting factors like strong vertical wind shear. The study uses a coupled TC model (HWRF‐B) and buoy observations to demonstrate that coastal downwelling developed as three TCs in 2020 approached land. Results show downwelling maintained warmer sea‐surface temperatures over the ocean shelf, enhancing air‐sea temperature/humidity contrasts. We found that in such cases resulting air‐sea enthalpy fluxes can replenish the boundary‐layer even when cool, dry air intrudes, as in sheared storms and storms approaching continental land‐masses. The resulting warm, moist air is advected into the TC inner core, enhancing convective development, thus providing energy for TC intensification. These results indicate coastal downwelling can be important in forecasting TC intensity change before landfall.

Future changes to the year-to-year swings between active and inactive North Atlantic tropical cyclone (TC) seasons have received little attention, yet may have great societal implications in areas prone to hurricane landfalls. This work investigates past and future changes in North Atlantic TC activity, focusing on interannual variability and evaluating the contributions from anthropogenic forcing. We show that interannual variability of Atlantic TC activity has already increased, evidenced by an increase in the occurrence of both extremely active and inactive TC seasons. TC-resolving general circulation models project a 36% increase in the variance of North Atlantic TC activity, measured by accumulated cyclone energy, by the middle of the 21st century. These changes are the result of increased variability in vertical wind shear and atmospheric stability, in response to enhanced Pacific-to-Atlantic interbasin sea surface temperature variations. Robust anthropogenic-forced intensification in the variability of Atlantic TC activity will continue in the future, with important implications for emergency planning and societal preparedness.

In this study, the design of movable multilevel nesting (MMLN) in the Hurricane Weather Research and Forecasting (HWRF) modeling system is documented. The configuration of a new experimental HWRF system with a much larger horizontal outer domain and multiple sets of MMLN, referred to as the “basin scale” HWRF, is also described. The performance of this new system is applied for various difficult forecast scenarios such as 1) simulating multiple storms [i.e., Hurricanes Earl (2010), Danielle (2010), and Frank (2010)] and 2) forecasting tropical cyclone (TC) to extratropical cyclone transitions, specifically Hurricane Sandy (2012). Verification of track forecasts for the 2011–14 Atlantic and eastern Pacific hurricane seasons demonstrates that the basin-scale HWRF produces similar overall results to the 2014 operational HWRF, the best operational HWRF at the same resolution. In the Atlantic, intensity forecasts for the basin-scale HWRF were notably worse than for the 2014 operational HWRF, but this deficiency was shown to be from poor intensity forecasts for Hurricane Leslie (2012) associated with the lack of ocean coupling in the basin-scale HWRF. With Leslie removed, the intensity forecast errors were equivalent. The basin-scale HWRF is capable of predicting multiple TCs simultaneously, allowing more realistic storm-to-storm interactions. Even though the basin-scale HWRF produced results only comparable to the regular operational HWRF at this stage, this configuration paves a promising pathway toward operations.

The Hurricane Weather Research and Forecasting (HWRF) Model is a dynamical model that has shown annual improvements in its tropical cyclone (TC) track forecasts as a result of various modifications. This study focuses on an experimental version of HWRF, called the basin-scale HWRF (HWRF-B), configured with 1) a large, static outer domain to cover multiple TC basins and 2) multiple sets of high-resolution movable nests to produce forecasts for several TCs simultaneously. Although HWRF-B and the operational HWRF produced comparable average track errors for the 2011–14 Atlantic hurricane seasons, strengths of HWRF-B are identified and linked to its configuration differences. HWRF-B track forecasts were generally more accurate compared with the operational HWRF when at least one additional TC was simultaneously active in the Atlantic or east Pacific basins and, in particular, when additional TCs were greater than 3500 km away. In addition, at long lead times, HWRF-B average track errors were lower than for the operational HWRF for TCs initialized north of 25°N or west of 60°W, highlighting the sensitivity of TC track forecasts to the location of the operational HWRF’s outermost domain. A case study, performed on Hurricane Michael, corroborated these HWRF-B strengths. HWRF-B shows the potential to serve as an effective bridge between regional modeling systems and next-generational global efforts.