Projecting the Risk of Future Climate Regime Shifts

Example Applications

The nomograph in Figure 4 provides quantitative estimates of the probability for a future shift in the AMO climate regime. Thus, it is generally thought that the AMO switched from cool to warm during the 1994-95 time frame, 10 years prior to this writing. If we enter Figure 5 with t1 = 10 years , we find a rather low probability (< 30%) that the AMO will switch back to its cool phase in less than t2 = 5 years from now. For t2 = 10 and 15 years, the risk increases to ~51% and ~70%, respectively, and a regime shift within 20 years is highly likely (~86%). Such a shift, when it occurs, would imply a return to more frequent droughts in Florida, fewer droughts in the Colorado River basin, and less frequent severe hurricanes in the tropical Atlantic [10,12,16]. As expected, (Figure 4) shows that the risk for any of these t2 values increases as time advances and the last regime shift (1994-95) recedes further into the past (t1 increases).

Consider the situation in 1990, more than 20 years into a period of cool North Atlantic sea surface temperatures (AMO) associated with dry conditions in Florida, wet conditions in the southwestern region and less frequent hurricanes. It is not difficult to imagine management decisions that could have been made then as an AMO reversal became imminent within operational time horizons. Where water was expected to become more plentiful, mosquito-breeding areas could have been drained in anticipation of more persistent water accumulations. Where more persistent and/or frequent droughts were expected, more water could have been shunted to aquifer storage, water access leases shortened, reservoir withdrawals reduced, conservation measures implemented and agricultural practices modified. Underwriting associations could have increased the funding of windstorm contingency pools in anticipation of more frequent, destructive hurricanes.