Ground-based lidars show daytime low cloud cover falling at mid- to high-latitude land sites

Why it matters

Reports of falling low cloud cover come mostly from satellites and reanalyses, which are not fully independent of each other. Lidar data from the ground give a separate check on the direction of the change. The authors say this supports a role for low cloud loss in the rise in Earth’s energy imbalance, but they cannot measure that role directly.

What they did

The authors gathered and reprocessed 10–25-year records from ground-based lidars and ceilometers at 16 sites, mostly in the Northern Hemisphere. They compared these with CERES satellite data and ERA5 reanalysis at the same places and times. For ERA5, they computed low cloud cover in two ways, one based on cloud-base height and one on cloud-top pressure. Trends came from a rank-based method (Theil-Sen with Mann-Kendall tests), and the site trends were combined into one joint trend per dataset. They did this for the whole year and for each season.

Key findings

• Daytime joint trends are negative and significant at the 5% level in all four datasets: −1.08±0.58 %-pt dec−1 for lidar, −1.64±0.46 for ERA5/CBH, −1.41±0.42 for ERA5/CTP and −1.65±0.38 for CERES.
• Daytime trends are negative at 13 of 16 sites for lidar, 14 for ERA5/CBH, 13 for ERA5/CTP and 12 for CERES.
• Lidar shows a weaker decline than ERA5 and CERES. The authors could not explain why.
• At night the picture is less clear. The joint lidar trend, −0.51±0.56 %-pt dec−1, is not significant at the 5% level, and the datasets disagree more.
• At single sites, few trends are significant (0 of 16 for lidar in daytime). Seasonal trends often differ in sign within a year, so annual trends can hide them.

Limitations

• Only 16 sites, mostly in the Northern Hemisphere, so the results may not hold globally. Point lidar data also differ from grid-cell averages in ERA5 and CERES.
• Individual-site trends are hard to separate from natural variability in these short records. Seasonal lidar trends rest on few data and are less robust. Lidar is less sensitive to ice clouds, and unknown instrument problems cannot be ruled out.
• The study cannot estimate how much low cloud change contributes to Earth’s energy imbalance. ERA5 and satellite data are not fully independent. This is a preprint that has not been peer reviewed.

Glossary

• Low cloud cover (LCC): The fraction of time or sky covered by low clouds, defined differently for lidar, ERA5 and satellite data.
• Automatic low-power lidar and ceilometer (ALC): A ground instrument that sends light upward and measures the reflected signal to find clouds and their base height.
• Theil-Sen estimator: A trend method that takes the median of slopes between pairs of data points, so outliers have little effect.
• CERES: A satellite programme that measures Earth’s radiation budget and cloud properties.