is it HOT ENOUGH yet?, page-88

The cloud coverage analysis shows a clear ENSO mode followed by a trend mode in terms of variance decomposition. The trend mode shows growth in cloud coverage with time over the tropical and subtropical oceans, while a shrinking in cloud coverage is revealed over most non-desert continents.

From the link I provided

The reported trends in cloud coverage are consistent with several previous estimations that were based on long-term observations and historical simulations. A few examples are the reported decreasing trend over land as revealed by surface observations (Warren et al., 2007) and the general increasing trend detected over the tropics and eastern subtropics by means of the analysis of satellite observations and historical simulations (Norris and Evan, 2015; Zhou et al., 2016; Norris et al., 2016). Another example is the analysis of the observed liquid water path from the Multi-sensor Advanced Climatology of Liquid Water Path (MAC-LWP) dataset, which showed an increasing trend over most of the oceans (Manaster et al., 2017). However, there are some contradictions between our findings and previously reported satellite observations, which show a decreasing trend over most of the Congo Basin and an increasing trend over most of the northeast tropical Atlantic over the last decades (1983–2009) (Norris and Evan, 2015