Lecture 27The long-term temperature record.This lecture: How do we know that the1990s was thewarmest decade in the last 1000 years?An important scientific issue is being able to distinguishnatural variability (solar, volcanic forcing, ENSO) fromanthropogenic influence. Coming next.(Ch. 15: p.292, then p.296-315, skip p.293-295)Holocene climateHistory of the Holocene(KKC 15:p.293-295 - not required) - come out of latest glaciation starting 14.7 ka - Younger-Dryas is relapse to cold, low CO2 period 12.9-11.7 ka - roughly, then, glaciation ended about 12 ka - warm period 5000-6000 yr b.p. - cool period ~3000 yr b.p. - Medieval warm period ~1000 yr b.p. (~1000-1300 A.D.) - Little Ice Age ~300 yr b.p. (~1400-1850 A.D.)The Little Ice AgeLate 1500s - cooling that continued through to about 1850Evidence:- re-advance of glaciers-wintertime freezing of rivers (Thames in London, Tagus in Spain/Portugal,canals of Amsterdam) and seas (Baltic Sea).Been proposed that the low number of sunspotsduring the period 1400-1650 AD was somehowrelated to the cold climate. However, nosatisfactory hypothesis has yet been developed toexplain how solar sunspots would significantlyaffect climate.Our best data (satellites since 1978) show thatchanges in solar flux are not significant over the11-year solar sunspot cycle. Only very shortwaveUV changes significantly but this only affects thetemperature of the atmosphere above 85 kmaltitude.Pieter Bruegel the Elder (1525–1569).Bitter winter, Feb 1565, HollandProxy data used to reconstruct climate1) Pollen2) Tree rings3) Corals4) Ice cores (gases, isotopes)5) Ocean sediments(seen these before for glacial cycle record)Pollen: (e.g. from lake sediments) tell us the past vegetation that lived.For example, a pine forest might be followed by an oak forest,a pattern typical of a shift to a drier climate.(Palynology - study of pollen)Trees: Annual growth rings provide the age of a tree. The width ofthe growth ring can be related to the availability of waterand the temperature in the growing season.(Leonardo da Vinci: wet/dry years evident in tree rings.1837: Charles Babbage mentions tree rings for dating (dendrochronology)1928: Andrew Douglass (US astronomer) develops the technique.Corals: Seasonal growth band and stable isotope variations in CaCO3 skeleton.d18O decreases as temperature increases because of (a) more evaporation of 18O,and (b) more direct incorporation of 16O into the CaCO3 (a kinetic effect).Yields a reliable temperature record to ±0.2°C accuracy.(Summary: Low d18O = WARM , High d18O = COLD)Pollen correlation: glacial-to-Holocenesedimentcoreice coreTREECORINGAnnual coral bands(from coral collected in 1994)[Source:Linsley et al., (1999) Coral Reefs18, 13-27.]Carbon datingMethod by William Libby in 1949 (Nobel Prize in Chemistry, 1960)In the upper atmosphere, cosmic ray neutronsinteract with nitrogen and make carbon-14:† 714N + 01n Æ 614C + 11H mass numberatomic (proton)numberRadioactive 14C gets oxidized to 14CO2. This CO2 is mixed byconvection and winds to a near constant fraction in theatmosphere. CO2 is cycled through vegetation. All living thingscontain some radioactive 14C. Can calibrate past variations in solarcosmic rays, and hence atmospheric 14C, using tree rings.When organisms die, 14C decays with a half-life of 5730 yearsto 14N. This provides an age of shells, bones, dead wood, corals, etc.“The 1990s are likely to have been the warmest decade of themillennium in the Northern Hemisphere” (IPCC 2001)T-variations 1000-2000What is the the largest, 50-year change in the pre-industrial era? What is themagnitude of this change (best-guess and upper bound)?blue = datablack = smoothed averagegray = 90% confidence boundRecent climate changes - and natural variabilityMotive: context for assessing the question of future climate change• Natural variability produces "noise" (ups and downs) in the temperaturerecord• Volcanoes provide a test of the theory that changes in energy balance cause changes in Global Annual Average Surface Temperature(GAAST)• Solar changes are frequently invoked and emphasized by the so-called "skeptics” as an alternative explanation for the observed, global warming. Does it have any scientific credibility? Answer: No.• Volcanoes and Climate (Kump 15:299-302)• Sun and Climate (Kump 15:303-306)Natural Variability, Signal, and NoiseAny measurement in science contains signal, what we are trying to detect and quantify, and noise, anything that gets in the way of detecting and quantifying the signal.When we examine the temperature record, we are trying to determine if it containsthe signal of human-induced global warming. Two types of "noise" get in theway: - errors in the measurements (discussed in Kump, Chap 1) - natural variabilityIn essence, we need to be able to answer this question: How large are natural changes in GAAST?Given that information, we can assess the likelihood that the industrial-erachanges in GAAST represent natural variation and not a human-induced effect.(In a statistical analysis, natural variability would be the "null hypothesis".)proxyinstrumentaltheory/predictionDoes the "signal" of predicted global warming rise above the"noise" of natural variability in T-observations so far?Natural climate variability due to Volcanoes and the SunTwo types of natural variability of climate: 1) unforced change (internal variability) 2) change that is forced by natural causes:Volcanoes and Solar changes are the two main mechanisms of natural, forced change.Basis for historical research - proxy records volcanoes: ash and sulfate in snow and ice-cores (many 1000s of years) sun: sunspot observations (centuries to a few thousand years) Forcing mechanism (physical basis of the forcing): - well understood for volcanoesi. sulfur dioxide (SO2) injected to the stratosphere forms sulfuric acid dropletsii. particles in the stratosphere reflect sunlight, increasing Earth's albedoiii. Pinatubo (1991): forcing and corresponding surface cooling were measured - highly speculative for the suni. changes in solar energy have been measured but are too smallii. therefore, exotic theories involving cosmic rays and clouds are invoked as "possible mechanisms". (To date, these appear