Sunday 4 December 2011

Frozen Planet: Fact Checking Nigel Lawson.

In the 3 - 9 December issue of the Radio Times (RT) there is a substantial piece by Sir David Attenborough about the final episode of Frozen Planet - On Thin Ice. I've fact checked that too, but here I'll fact check Lord Nigel Lawson's response, printed on page 15 of that issue of RT. The blockquotes below are Lawson's factual statements from the RT article.

Had he [Attenborough] wished to be objective, he would have pointed out that while satellite observations confirm that the extent of  Arctic sea-ice has been declining over the past 30 years, those satellite observations show that, overall, Antarctic sea-ice has been expanding over the same period.
This is a point I use as a litmus test of how climate-savvy a person is. If they raise this point it's a sure bet they don't know their arse from their elbow.
1) It implies that they don't know that whereas the Arctic is an ocean surrounded by land, the Antarctic is land surrounded by ocean.
2) It implies that even if they do get point 1, then either a) they think their audience are too stupid to figure out that this will have en effect, or b) they are too stupid to realise that this will have an effect.
3) It implies that either a) they're too stupid or lazy to do a quick search on Google, or b) they think their audience are too stupid or lazy to do a quick search on Google. Quick Search on Google, third link down.
Leaving aside the fact that any reasonable person would find no surprise at the difference. The difference is of course a reasonable question for the genuinely enquiring to ask. Why is the Antarctic sea-ice increasing while the Arctic is decreasing? In short: The answer is that while the Arctic sea-ice is decreasing, the main reason being human driven global warming, the Antarctic is increasing due to the differing rates of ice melt from ocean heat flux and ice growth (Zhang 2006).

Had he wished to be objective, he would have pointed out that the polar bear population has not been falling but rising.
This is incorrect, the only reason I do not call it a lie is that to lie one has to know that one is telling an untruth, the rest of Lawson's response implies he just doesn't know what he's talking about. The Polar Bear Specialist Group have combined research to give a picture of the status of Polar Bear populations in the Artic. By 2010, of 19 regions, 7 have too little data to say anything, 1 is increasing, 3 are stable, 8 are in decline (source). The denialist lobby have attempted to call the threat to Polar Bears into question, but scientists have felt it worth investing valuable time to rebut these specious claims. The bottom line remains - Polar Bears rely on sea-ice for hunting and that sea-ice is declining.

Had he wished to be objective, he would have mentioned that recent research findings show that the increased evaporation from the Arctic Ocean, as a result of warming, will cause there to be more cloud cover, thus counteracting the adverse effect he is so concerned about.
The changes in the Arctic are already well underway, we are not speculating about some abstract future here. In 2007 the Arctic lost 25% of it's summer minimum area, it has not recovered since then (ref), we're now 4 melt seasons on, ample time for a recovery.

So where is this negative cloud feedback? Why is it not already stopping the rapid and vast changes in the Arctic?

Yes it is true that clouds reduce surface sunlight, and it is true that clouds increase over areas of open water in the Arctic. However what Lawson fails to do is explain the caveats, that infra-red radiation is emitted from clouds which can cause warming. For example Francis et al find that for melt season: "Relationships between Maximum Ice Retreat Anomalies (MIAs) and individual forcing anomalies are predominantly linear; [Downward Longwave Flux] DLF anomalies are consistently positively correlated with ice retreat, while [Downward Shortwave Flux] DSF anomalies are consistently negatively correlated. This implies that variability in solar fluxes, owing primarily to varying cloud opacity, is overwhelmed by opposing changes in emitted longwave radiation."

Furthermore increased water vapour during the summer drives convection and increased influx of low pressure systems into the Arctic both act to mechanically break the ice and mix warmer water up from deeper layers. Finally Lawson does not inform his readers that some scientists are concerned that cloud feedback from infra red radiation may cause 'tipping point' behaviour in the future loss of winter Arctic sea-ice (i.e. Abbott et al, 2010).

The most parsimonious answer to the question "where is this negative feedback" is that it is not as strong as Lawson wants people to think, and indeed may even end up being a positive feedback due to infra-red back radiation from clouds.

Had he wished to be objective, he would have have noted that while there was indeed a modest increase in mean global temperature (of about half a degree centigrade) during the last quarter of the 20th century, so far this century both the UK Met Office and the World Meteorological Office (sic) confirm that there has been no further global warming at all.
To start with, simplistically using short periods to try to say something about global warming is ill advised: Natural variability caused by processes like the Pacific's El Nino can overwhelm the relatively small annual warming from global warming (about 0.017 degrees C per year) over short periods. This is why the standard timescale for climate is a 30 year baseline.

Furthermore if you're interested in global average temperature you'll use GISS, if you're interested in what's going on in the global grid boxes with data you'll use HAD/CRU (UK Met Office) or GHCN. This is because GISS use the findings of Hansen & Lebedeff 1987, who found that the global average temperature signal is robust at distances of over 1200km (figure 3 and text). Whereas Had/CRU don't cover regions with no data, such as most of the Arctic, using interpolation as GISS do. In a 2005 paper Vose et al found that compared to GISS, "GHCN and CRU have larger trends because the gridding techniques used in those analyses do not provide estimates for unsampled grid boxes, which are more prevalent in the Southern Hemisphere." So if you don't think the warming in the Arctic is relevant, or lower trends in the Southern Hemisphere, and you'd prefer to ignore these regions you can use HAD/CRU or GHCN. The World Meteorological Organisation don't publish a temperature dataset however in a recent press release they find: "Global temperatures in 2011 are currently the tenth highest on record and are higher than any previous year with a La NiƱa event, which has a relative cooling influence. The 13 warmest years have all occurred in the 15 years since 1997." Source: WMO, 29 November 2011.

But perhaps it'll make more sense to see some graphs. In the following two graphs I've calculated the trend from 1975 to 1999, then projected that trend onwards to the latest data for 2010. This is a good test of whether the trend has changed*, if it has then the previous 1975 to 1999 trend won't fit the 2000 to 2010 trend well. *a good test while keeping it simple.

HAD/CRU Temp3v

GISS Land Ocean Temperature Index (LOTI).

It's clear that while HAD/CRU shows a flat period in the 2000s (above then below the preceding trend), GISS, which includes the Arctic, remains more firmly on trend. This map from GISS reveals the pattern of changes from 2000 to 2010, comparison with 1980 to 1990, and 1990 to 2000, periods of warming, is informative. Perhaps more telling is what happens when one considers global changes over a longer period, like 1975 to 2010. Over short periods the pattern of global warmng is mixed, but the recent warming of the Arctic is clear. The insinuation that global warming has stopped is incorrect.

So who are we to trust? Sir David Attenborough who has a long track record of excellent scientifically accurate natural history documentories. Or Lord Nigel Lawson, former chancellor, economist, and member of the Global Warming Policy Foundation. On the basis of this performance I would suggest Lord Lawson is an unreliable guide to the science.

HAD/CRU data
GISS data

Abbot et al, 2011, "Bifurcations leading to summer Arctic sea ice loss."

Francis et al, 2006, "New Insight Into the Disappearing Arctic Sea Ice."

Hansen & Lebedeff, 1987, "Global Trends of Measured Surface Air Temperature."

Jinlun Zhang, 2006, "Increasing Antarctic Sea Ice under Warming Atmospheric and Oceanic Conditions."

Vose et al, 2005, "An intercomparison of trends in surface air temperature analyses at the global, hemispheric, and grid-box scale."

No comments: