To better understand the accuracy of cloud top heights (CTHs) derived from passive satellite data, ground-based Ka-band radar measurements from 2016 and 2017 in Beijing are compared with CTH data inferred from the Moderate Resolution Imaging Spectroradiometer (MODIS) and the Advanced Himawari Imager (AHI). Relative to the radar CTHs, the MODIS CTHs are found to be underestimated by−1.10 ± 2.53 km on average and 49 % of CTH differences are within 1.0 km. The AHI CTHs are underestimated by −1.10 ± 2.27 km and 42 % are within 1.0 km. Both the MODIS and AHI CTH retrieval accuracy depends strongly on the cloud depth (CD). Large differences are mainly due to the retrieval of thin clouds of CD <1 km, especially when the cloud base height is higher than 4 km. For clouds with CD >1 km, the mean CTH difference decreases to $-\mathrm{0.48}\pm \mathrm{1.70}$ km for MODIS and to $-\mathrm{0.76}\pm \mathrm{1.63}$ km for AHI. It is found that MODIS CTHs with higher values (i.e. >6 km) show smaller discrepancy with radar CTH than those MODIS CTHs with lower values (i.e. <4 km). Statistical analysis illustrate that the CTH difference between the two satellite instruments is lower than the difference between the satellite instrument and the ground-based Ka-band radar. The monthly accuracy of both CTH retrieval algorithms is investigated and it is found that summer has the smallest retrieval difference.

The Precipitation Measurement Radar (PMR) onboard FengYun‐3G consists of a Ku‐/Ka‐band radar, which is characterized by similar configurations with the Dual‐frequency Precipitation Radar (DPR) carried by Global Precipitation Measurement mission Core Observatory. However, directly comparing observations from two radars is challenging due to a scarcity of their coincidences. In this study, sea surface echoes in their track intersections were employed to cross‐calibrate PMR. Then, we show that the dual‐frequency ratio (DFR) in snow stably increases with Ku‐band reflectivity in statistics, allowing for an assessment of the consistency between PMR and DPR observations. Surprisingly, our results reveal a underestimation of DFR in DPR inner swath, while observations from PMR are in good agreement with those from DPR outer swath. This study demonstrates the novel use of natural targets for spaceborne dual‐frequency radar calibration, and presents a unique view into the connection between the two spaceborne precipitation radar missions in operation.

The physical properties and radiative role of ice clouds remain one of the uncertainties in the Earth–atmosphere system. In this study, we present a detailed analysis of ice cloud properties based on 4 years of surface millimeter-wavelength radar measurements in Beijing, China, where the summer monsoon from the ocean and the winter monsoon from the continent prevail alternately, resulting in various ice clouds. More than 6300 ice cloud clusters were studied to quantify the properties of ice clouds, such as the height, optical depth and horizontal extent, which can serve as a reference for parameterization and characterization in global climate models. In addition, comparison between ice cloud clusters formed under the summer monsoon and the winter monsoon indicates the different formation and evolution mechanisms of cirrus clouds. Statistically, temperatures of more than 95 % of ice radar bins are below −15 ^∘C and more than 80 % of ice clouds are above 7 km. The dependence of the radar reflectivity of ice particles on height and temperature was also observed in this study, indicating that the reflectivity of ice bins increases (decreases) as the temperature (height) increases. In addition, it is found that there is a strong linear relationship between the mean reflectivity and the ice cloud depth. Due to various synoptic circumstances, the ice clouds in summer are warmer, higher and thicker, with larger reflectivity than that in winter; in particular, the mean cloud-top height of ice clouds in summer is 2.2 km higher than that in winter. Our analysis indicates that in spring, in situ-origin cirrus clouds are more common than liquid-origin cirrus clouds, while in summer liquid-origin cirrus clouds are more frequent; in autumn and winter, most cirrus clouds are of in situ origin.

A new dual-frequency Doppler polarimetric cloud radar (DDCR), working at 35-GHz (Ka-band radar, wavelength: 8.6 mm) and 94-GHz (W-band radar, wavelength: 3.2 mm) frequencies, has been in operation at Yangbajing Observatory on the Tibetan Plateau (China) for more than a year at the time of writing. Calculations and field observations show that the DDCR has a high detection sensitivity of −39.2 dBZ at 10 km and −33 dBZ at 10 km for the 94-GHz radar and 35-GHz radar, respectively. The radar reflectivity measured by the two radars illustrates different characteristics for different types of cloud: for precipitation, the attenuation caused by liquid cloud droplets is obviously more serious for the 94-GHz radar than the 35-GHz radar (the difference reaches 40 dB in some cases), and the 94-GHz radar lost signals due to serious attenuation by heavy rainfall; while for clouds dominated by ice crystals where the attenuation significantly weakens, the 94-GHz radar shows better detection ability than the 35-GHz radar. Observations in the Tibetan region show that the 35-GHz radar is prone to missing cloud near the edge, such as the cloud-top portion, resulting in underestimation of the cloud-top height (CTH). Statistical analysis based on one year of observations shows that the mean CTH measured by the 94-GHz radar in the Tibetan region is approximately 600 m higher than that measured by the 35-GHz radar. The analysis in this paper shows that the DDCR, with its dual-frequency design, provides more valuable information than simpler configurations, and will therefore play an important role in improving our understanding of clouds and precipitation in the Tibetan region.