GLERL 1998 Milestone Reports

Milestone 1998 Home

GOAL: IMPLEMENT SEASONAL TO INTERANNUAL CLIMATE FORECASTS

OBJECTIVE: CONDUCT RESEARCH FOR IMPROVED CLIMATE PREDICTIONS

PM: Determine the seasonally varying character of the North American climate response to El Niño.

Milestone: Investigation of Implications of Strong Warm ENSO Events for North American Laurentian Great Lakes Winter Severity and Ice Cover

Scientist: Ray.Assel@noaa.gov 734-741-2268

Improved long-range models / forecasts of early winter, mid-winter, late winter ice formation and ice loss have applications for a better understanding of under ice ecology; improved models of whitefish year class strength; improved understanding of spring lake ecosystem processes; improved lake levels forecasting; and cost savings for Great Lakes winter operational activities, such as US Coast Guard support of shipping, US Army Corps of Engineers and St Lawrence Seaway Development Corporation deployment of ice control structures on the connecting channels of the Great Lakes to prevent ice jams & to protect hydro-electric-generating plants, and operation and maintenance of navigation locks.

The study compares Great Lakes winter severity and annual maximum ice cover for the six strongest warm El Nino/Southern Oscillation (ENSO) events relative to other winters over the past half century with the objective of providing information potentially useful for developing long-range models and predictions of Great Lakes ice cover.

Figure 1: Relative strength of the six strongest warm ENSO events since 1950. (Click on the image to download it as a postscript file).

The graph above is from http://www.cdc.noaa.gov/ENSO/enso.mei_index.html. The 1997 ENSO event is similar to the record 1982 ENSO. The 1982-83 winter following the 1982 event had much-above-average temperatures and much-below-average ice cover (Assel et al. 1985). Thus, the implication made last summer when it appeared that the 1997 ENSO might be even stronger than the 1982 ENSO if it continued to intensify during the fall of 1997 and winter of 1998 was that the Great Lakes would have a extremely mild winter and much-below-average ice cover. (Winter 1997-98 was among the mildest and ice cover much-below-average).

Comparing Great Lakes winter (temperature) severity and annual maximum ice cover for the six strongest warm ENSO events relative to other winters since 1950 we found that a regional winter severity index averaged 1.2°C higher and modeled annual regional maximum ice cover averaged 15% lower for the winters following the onset year of a strong warm ENSO event relative to the average of other winters in the 1950-1994 base period. These results are significant at the 5% level (Students t-test).

Figure 2: USCG Cutter Mackinaw assisting a commercial bulk carrier in ice covered waters. The USCG typically spends between 1000 and 2000 hours a year keeping Great Lakes shipping channels open. During the extremely strong warm 1983 ENSO the USCG spent less than 100 hours because of the much-below-average ice cover. The cost for operating buoy tender class vessel is over $1000/hour(over $3000/hour for the USCGC Mackinaw). Thus, improved long-range forecast of extremely warm ENSO events would likely lead to substantial cost saving. (Click on the image to download it as a postscript file).

This analysis provides evidence of a statistically significant teleconnection between Great Lakes winter severity (and ice cover) and strong warm ENSO events. Given the caveat that the sample size is small, these results suggest that the strong warm ENSO events may be useful (perhaps as part of a suite of variables) in developing a long-range model of Great Lakes winter severity and ice cover.

This preliminary study focused on one aspect of one teleconnection (strong warm ENSO events) and one aspect of ice cover (annual maximum ice cover). There are other ice cover teleconnections and other aspects of ice cover that should be investigated for long-range ice models and forecasts.

Milestone Home Climate Change and Variability Program Homepage