PowerPoint Presentation - Presentation at Yale University

lConcerns have been expressed about the reliability of model-based estimates of the natural variability of ocean temperatures (e.g., Lyman et al., 2006)

lCasts doubt on reliability of D&A results obtained with ocean temperatures

lHow do we address these concerns?

ÆBetter quantification of uncertainties in observed variability estimates (e.g., AchutaRao et al., 2007). Involves use of both physical models (ocean data assimilation products) and statistical models

ÆIdentify and adjust for the effects of instrumental biases in different ocean observing systems (Church et al., in preparation)

ÆRevisit ocean D&A studies with improved, bias-corrected observational data

ÆUse proxy data to obtain better constraints on model estimates of natural internal variability (e.g., multi-century SST reconstructions from corals)

Future research II: Improvement of observational datasets with sparse, space- and time-varying coverage


	Detecting ÒsignificantÓ climate change means showing that a given change in climate is unusual relative to some yardstick that you have of ÒnormalÓ behavior.
	If you feel hot, and measure a body temperature of 102^oF, then youÕve ÒdetectedÓ a fever - you know that that this elevated temperature is outside the limits of ÒnormalÓ body temperature.
	But this measurement doesnÕt give you the cause of the fever.
	To deduce the cause, youÕll probably need to undergo some more sophisticated tests - perhaps urine and blood analyses, maybe even x-rays or CAT scans.
	Once you have the results from these tests, itÕs much easier to narrow down the potential causes of the fever.
	This is attribution - the process of establishing cause and effect.