Abstract

This study compares CL51 ceilometer observations made at Scott Base, Antarctica, with statistics from the ERA5, JRA55, and MERRA2 reanalyses. To enhance the comparison we use a lidar instrument simulator to derive cloud statistics from the reanalyses which account for instrumental factors. The cloud occurrence in the three reanalyses is slightly overestimated above 3 km, but displays a larger underestimation below 3 km relative to observations. Unlike previous studies, we see no relationship between relative humidity and cloud occurrence biases, suggesting that the cloud biases do not result from the representation of moisture. We also show that the seasonal variation of cloud occurrence and cloud fraction, defined as the vertically integrated cloud occurrence, are small in both the observations and the reanalyses. We also examine the quality of the cloud representation for a set of weather states derived from ERA5 surface winds. The variability associated with grouping cloud occurrence based on weather state is much larger than the seasonal variation, highlighting weather state is a strong control of cloud occurrence. All the reanalyses continue to display underestimates below 3 km and overestimates above 3 km for each weather state. But the variability in ERA5 statistics matches the changes in the observations better than the other reanalyses. We also use a machine learning scheme to estimate the quantity of supercooled liquid water cloud from the ceilometer observations. Ceilometer low‐level supercooled liquid water cloud occurrences are considerably larger than values derived from the reanalyses, further highlighting the poor representation of low‐level clouds in the reanalyses.

Plain Language Summary

This study compares cloud observations from a CL51 ceilometer at Scott Base, Antarctica, with data from three weather reanalyses: ERA5, JRA55, and MERRA2. We used a lidar simulator to better match the reanalyses data with the ceilometer's measurements. The reanalyses slightly overestimate cloud presence above 3 km, but significantly underestimate it below 3 km compared to the ceilometer data. Both the servations and reanalyses show only small seasonal changes in cloud presence. However, grouping the data by weather patterns shows that these patterns strongly influence cloud presence. The reanalyses still underestimated cloud presence below 3 km and overestimated it above 3 km for all weather patterns, but ERA5 data matched the observed changes better than the other reanalyses. We also used machine learning to estimate the amount of supercooled liquid water clouds from the ceilometer data. The ceilometer detected many more low‐level supercooled liquid water clouds than the reanalyses simulations, highlighting that issues with the representation of low‐level clouds in these models are widespread.

Key Points

• Cloud occurrence is underestimated below 3 km in ERA5, JRA55, and MERRA2 reanalyses relative to
  observations, leading to cloud fraction biases
• Observed cloud occurrence is more strongly impacted by weather state than season; ERA5 simulates this pattern better than JRA55 and
  MERRA2
• Supercooled liquid cloud derived from ceilometer data have higher occurrences than the three reanalyses, with MERRA2 having the least bias