The HRRR model surface thermodynamic biases are seasonally dependent, presenting a systematic warm and dry bias during the warm season

The primary locations of the summer warm and dry biases are over farmland, on days with optically thick clouds

A hydrological bias underestimating of spring snowmelt is consistent with subsequent summer warm and dry biases over farmland

H₂O vertical profiles were measured using Multi-AXis Differential Optical Absorption Spectroscopy (MAX-DOAS)

The correlation coefficient between the aerosol optical depth (AOD) and the H₂O VCD was higher during heavily polluted weather

Heavy air pollution easily occurred in Qingdao due to a combination of high H₂O concentration and low wind speeds

True gravity g(λ, φ, z) represents vertical and has latitudinal and longitudinal components

Replacement of the standard gravity -g₀k (g₀=9.81 m s^-2) by g(λ, φ, z) in the classical Ekman layer equation leads to a new solution

With data from EIGEN-6C4 and COADS, the Ekman pumping velocity is much larger due to g(λ, φ, z) than due to geostrophic vorticity

The influence of aerosol on clouds is investigated using multi-source data over the North China Plain and north of the Yangtze Plain

Both aerosol properties and atmospheric conditions affect cloud droplet , while drizzle has a greater impact on aerosol-cloud interaction

The cloud droplet and PM_2.5 concentration show a positive (negative) correlation for all clouds (clouds without drizzle)

For larger hydrometeors, hydrometeor shape plays an important role in streamer initiation

When hydrometeor radius is smaller than 1mm, the critical field depends on electric quantity carried by hydrometeor evidently

Streamer discharge can be predicted by Meek number, but stable streamer can not be ensured when ambient field is lower than 0.3Ek

A high percentage (90) of dancing sprites were produced by a large asymmetric MCS which developed in strong convective conditions over the Mediterranean Sea

The sprite parent flashes mainly initiated at the edge of the convective region and propagated 100-200 km across the stratiform region

An especially bright and wide sprite sequence was produced by three distinct positive CG strokes that occurred within 3 ms and were separated by 54 km

Rare in situ measurements during hurricane-force winds over the Terra Nova Bay revealed a jet core at ∼200 m elevation

The temperature increased by 10°C and specific humidity by a factor of 6 as air crossed the warm surface of a coastal polynya

The humidity of the air was always at saturation with respect to ice at an elevation of ∼15 m, resulting in ice crystal growth

Surface heat losses of up to 2500 Wm^-2 occurred over the Terra Nova Bay Polynya during a strong katabatic wind event

Where open ocean water was exposed, there was strong evidence that sea spray more than doubled the surface sensible heat flux

The low-level air flow was super-critical and in an approximate frictional-inertial balance for more than 100 km

Smaller cold pool radii in large-eddy simulations diminishes the time to reach convective self-aggregation

We report cold pools' generation time and suppression radius by evaluating the distance between rain events connected in time

A mathematical model captures the effect of domain size, suppression radius, and maximum cold pool radius in convective self-aggregation

Moist convection frequently reaches the tropopause in the United States and mixes air between the troposphere and stratosphere

Tropopause-overshooting convection occurring during May drastically increases lower stratospheric water vapor mixing ratios

The impact that convection has on UTLS composition appears to largely depend on the height of the tropopause