The variability of internal tides during their generation and long‐range propagation in the South China Sea (SCS) is investigated by driving a high‐resolution numerical model. The present study clarifies the notably different processes of generation, propagation, and dissipation between diurnal and semidiurnal internal tides. Internal tides in the SCS originate from multiple source sites, among which the Luzon Strait is dominant, and contributes approximately 90% and 74% of the baroclinic energy for M2 and K1, respectively. To the west of the Luzon Strait, local generation of K1 internal tides inside the SCS is more energetic than the M2 tides. Diurnal and semidiurnal internal tides from the Luzon Strait radiate into the SCS in a north‐south asymmetry but with different patterns because of the complex two‐ridge system. The tidal beams can travel across the deep basin and finally arrive at the Vietnam coast and Nansha Island more than 1000–1500 km away. During propagation, M2 internal tides maintain a southwestward direction, whereas K1 exhibit complicated wave fields because of the superposition of waves from local sources and island scattering effects. After significant dissipation within the Luzon Strait, the remaining energy travels into the SCS and reduces by more than 90% over a distance of ∼1000 km. Inside the SCS, the K1 internal tides with long crests and flat beam angles are more influenced by seafloor topographical features and thus undergo apparent dissipation along the entire path, whereas the prominent dissipation of M2 internal tides only occurs after their arrival at Zhongsha Island.

Extratropical cyclones are the main generators of the strong winds that cause large ocean waves in temperate regions of the world. The severity of the winds associated with these storms is poorly represented by the coarse resolution of current global climate models (GCMs), making it challenging to produce projections of the future climate of large waves. Wind data from GCMs can be downscaled in resolution using dynamical methods, resulting in a successful reproduction of the mean wave climate, but a suboptimal reproduction of the storm wave climate 1 . Projections of large wave occurrence can also be produced using statistical downscaling methods, although such methods have previously been applied only to three or less GCMs 2,3 , preventing a robust assessment of confidence in projections based on variation between models. Consequently, considerable uncertainty remains in projections of the future storm wave climate. Here we apply a statistical diagnostic of large wave occurrence in eastern Australia to 18 di erent GCMs, allowing model variations to be examined in greater detail than previously possible. Results are remarkably consistent between di erent GCMs, allowing anthropogenic influences to be clearly demonstrated, with fewer days with large waves expected to occur in eastern Australia due to increasing greenhouse gas concentrations.

No abstract

The bulk microphysical properties and number distribution functions (N(D)) of supercooled liquid water (SLW) and ice inside and between ubiquitous generating cells (GCs) observed over the Southern Ocean (SO) during the Southern Ocean Clouds Radiation Aerosol Transport Experimental Study (SOCRATES) measured by in situ cloud probes onboard the NCAR/NSF G‐V aircraft are compared. SLW was detected inside all GCs with an average liquid water content of 0.31 ± 0.19 g m−3, 11% larger than values between GCs. The N(D) of droplets (maximum dimension D < 50 μm) inside and between GCs had only slight differences. For ice particles, on the other hand, the mean concentration (median mass diameter) with D > 200 μm inside GCs was 2.0 ± 3.3 L−1 (323 ± 263 μm), 65% (37%) larger than values outside GCs. As D increases, the percentage differences became larger (up to ~500%). The more and larger ice particles inside GCs suggest the GC updrafts provide a favorable environment for particle growth by deposition and riming and that mixing processes are less efficient at redistributing larger particles. The horizontal scale of observed GCs ranged from 200 to 600 m with a mean of 395 ± 162 m, smaller than GC widths observed in previous studies. This study expands knowledge of the microphysical properties and processes acting in GCs over a wider range of conditions than previously available.

No abstract

Though cloud fraction (CF) from Moderate Resolution Imaging Spectroradiometer (MODIS) has been widely used, it remains unclear whether it can fully represent the diurnal variations. This study evaluates the time representation (i.e., satellite passes' mean value per day to represent daily average value) error in MODIS CF by using daytime‐only total sky cover and continuous day‐and‐night radar/lidar CF (Active Remote Sensing of Clouds product, ARSCL) from 2000 to 2010 for two Atmospheric Radiation Measurement (ARM) program climate regime sites of Southern Great Plains (SGP) and Manus. By comparing the daily averaged CFs from ARSCL between using all hourly and using the MODIS‐passing‐time observations, it shows a correlation coefficient of 0.93 (0.88) and root mean square deviation (RMSD) of 12.68% (13.27%) over SGP (Manus) site for daily averaged CFs. Differently, it shows a better correlation coefficient of 0.97 (0.97) and smaller RMSD of 2.98% (3.97%) over SGP (Manus) site for monthly averaged CFs. These suggest that considerable errors could be introduced while using the MODIS CF observed at several fixed time points a day to represent average CF at different time scales. Monthly time representation errors have also been evaluated for daytime only and nighttime only, which show even larger values. A further analysis shows that uncertainties caused by the time representation account for about 23% (21%) of the total differences between surface and MODIS CFs over SGP (Manus) site at monthly time scale.

Internal tides, generated by barotropic tidal currents flowing over abrupt topographic features (such as sills), are internal gravity waves that oscillate at tidal frequencies. The dissipating energy carried by propagating internal tides contribute a significant amount of mechanical energy to abyssal mixing and, hence, the general circulation (Polzin et al., 1997). For example, internal tides with spatial inhomogeneity and temporal variability largely induce a globally averaged diapycnal diffusivity of 10 −4 m 2 /s (Koch-Larrouy et al., 2010), which maintains global meridional overturning circulation (Garrett & Kunze, 2007). Therefore, to fully understand tidal mixing, it is crucial to dynamically resolve how fast-manifold internal tides radiate and how the tidal energy influence a balanced general circulation (

These low-level clouds, such as shallow boundary layer clouds (e.g., subtropical stratocumulus and trade-wind cumulus) have significant impacts on the radiation budget and climate change (Bony & Dufresne, 2005). While most past studies have focused on r e and LWC, the relative dispersion (ε), defined as the ratio of the standard deviation of the droplet size distribution (σ) to the number mean radius (r m), represents another DSD parameter which is of great importance to estimate the indirect aerosol effect on the climate system (Peng & Lohmann, 2003).