Midlatitude Cirrus Clouds at the SACOL Site: Macrophysical Properties and Large‐Scale Atmospheric States

Abstract Two‐year observations of a Ka‐band Zenith Radar at the Semi‐Arid Climate and Environment Observatory of Lanzhou University (SACOL) are used to document the midlatitude cirrus cloud macroproperties. Generally, cirrus occurs 41.6% of the observation time and most frequently appear at about 7.2 km above ground level. The cirrus macroproperties are strongly coupled with large‐scale atmospheric states; thus, its occurrence and location over the SACOL have significant seasonal variations. A k ‐mean clustering method is used to classify cirrus into four distinct regimes without a prior knowledge about the meteorological process. Contrasting to the different cirrus physical properties in each regime, the cirrus event of each regime has a distinct seasonal distribution and the synoptic conditions from the ERA‐Interim reanalysis responsible for each cirrus regime are also quite different. Since global climate models typically overestimate cirrus cloud thickness due to inadequate parameterization or coarse grid resolution, we examined the probability density functions of large‐scale vertical velocity associated with each cirrus regime and the relationship between cirrus thickness and vertical velocity. It is found that the differences of the vertical velocity probability density functions among the cirrus regimes are as distinct as their macroproperties and a significant correlation exists between cirrus thickness and the vertical velocity, although the large‐scale vertical motion is nearly as likely to be descending as ascending when cirrus clouds are observed. This may imply that large‐scale vertical velocity can be used to constrain the variations of cirrus thickness simulated by global climate models.