A Cloudy Question: was 'the pause' caused by a change in global cloud cover?

Guest essay by Mike Jonas

I have been looking at some cloud data (from ISCCP: isccp.giss.nasa.gov all available monthly “EQ” data (equal-area grid)) and an interesting question arises. I haven’t seen the answer on WUWT or anywhere else.

Was there a near-global sea-change (no pun intended) in cloud cover around the turn of the millenium, and if so, what caused it?

Figure 1. Global ClearSky anomaly, 1-yr smoothing (centred).

The point is that eye-balling the above graph, it looks like ClearSky was increasing in the late 20thC and then stopped. ie, cloud cover was decreasing, then stopped.

The answer to that question might go a long way towards explaining the “hiatus” and resolving the entire climate science controversy.

If any WUWT readers can supply the answers, I would be most grateful.

– – – – –

A bit of background:

[Except where stated otherwise, graphs in this document are all of temperature anomaly, cloud anomaly over ocean only. The Cloud or ClearSky axis is on the left, Temperature axis is on the right. ClearSky% = (100 – Cloud%), so ClearSky anomaly = (- Cloud anomaly). Cloud anomaly is based on calendar month averages over all full years of cloud data. Temperature data is from UAH Lower Troposphere Ocean-only, provided by UAH in anomaly form but likely to have a different base period.]

There are short term and longer term correlations between cloud and temperature.

Short term (a month or two), cloud increases with temperature. Well, after about 1998 it does. This is only to be expected, because it is generally agreed that the water cycle increases with temperature, and the water cycle necessarily involves clouds. The relationship, as would be expected, is strongest in the tropics.

Figure 2. Monthly Temperature and Cloud anomalies in the Tropics.

Temperature moves first, which suggests that it’s the driver (in the short term). This says nothing about the rate at which the water cycle increases with temperature.

Long term, though, ClearSky increases with temperature. This is the Global picture with 11-yr smoothing:

Figure 3. Global Temperature and ClearSky anomalies, with 11-yr smoothing (centred).

There is no clear indication from the Global picture, as to which comes first, temperature or ClearSky. Temperature appears to trail ClearSky with a lag of a few years in the NH …

Figure 4. NH Temperature and ClearSky anomalies, with 11-yr smoothing (centred).

… and in the Antarctic …

Figure 5. Antarctic (to 60S) Temperature and ClearSky anomalies, with 11-yr smoothing (centred).

… but the pattern is much less clear in other regions.

That temperature increases as ClearSky increases, but with a lag, is to be expected because visible light [and some UV] penetrates many metres into the ocean thus warming it; reflective clouds affect the amount.

The IPCC claim a large positive cloud feedback (ie, that a rising temperature causes more warming by clouds [presumably they mean that higher temperature leads to less clouds, but I can’t see anywhere that they say it explicitly]. A long time ago I explained on WUWT how the way in which they derive the positive cloud feedback was invalid (https://wattsupwiththat.com/2015/09/17/how-reliable-are-the-climate-models/). The fact that the initial effect of temperature on clouds is in the opposite direction (Figure 2 above) suggests that the IPCC finding is mistaken, and that ClearSky is simply a significant driver of temperature over decadal+ periods.

It’s perhaps a bit odd that the ClearSky effect on Temperature is most visible in the Antarctic and the NH. I speculate as follows:

The period covered by the cloud data is simply not long enough to get a clear picture of the longer-term mechanisms. There are also a lot of other things going on which confuse the picture. For example, there are winds and ocean currents that flow from region to region, so regions are affected by what is going on in other regions. Over periods of a year to multiple decades, temperatures everywhere are affected by ENSO and other ocean oscillations. Clouds are presumably affected too. And then there is the short term effect of temperature on clouds, which is in the opposite direction to the longer term cloud-temperature relationship, and hence may confuse the picture further. And, of course, we always have to bear in mind that climate is a non-linear system.

ENSO in particular is strongest in the Tropics and south of the Tropics, and maybe this would make the cloud-temperature link more difficult to see there, particularly given the short period over which we have cloud data. Maybe that is why the ClearSky-Temperature lag is most visible in the Northern Extra-Tropics and the Southern Ocean.

The Southern Ocean is more isolated than other regions, if I have understood it correctly. It has virtually no incoming winds or surface currents from other ocean areas. The principal incoming ocean current is in the form of upwellings from the deeper ocean, and is therefore unaffected by weather/climate conditions in other ocean areas. Similarly, the principal wind direction is from the Antarctic continent (the katabatic wind) not from other ocean areas.

All other ocean areas, by contrast, have incoming surface ocean flows and winds from other ocean areas, and are therefore influenced by weather/climate conditions in those other ocean areas.

One implication of this is that the cloud-temperature relationship is more likely to reflect locally-generated conditions over the Southern Ocean than it is over other ocean regions. The fact that solar radiation is weakest there per unit area would suggest that a smaller temperature effect should be expected, but the effect appears to be just as strong.

The cloud-temperature relationship in the other ocean regions, as covered by UAH, is less clear than in the Southern Ocean. See worksheet Graphs in spreadsheet UAH_ClearSky.xlsx (Excel .xlsx 1.2 mb)

Mike Jonas (MA Maths Oxford UK) retired some years ago after nearly 40 years in I.T.