Reliable weather and climate information for Oregon extend back about 100 years. During that time, there have been about an equal number of wet years and dry years. We find, however, that during some periods there is a preponderance of wetter than average years while in others there are mostly dry years. These wet and dry "cycles" generally span 20-25 years. The dry years tend to be warm (due to reduced cloudiness, most likely) and the wet years cool.
Figure 1 shows Water Year precipitation (Oct. - Sept.) for the Oregon Coast Division (Zone 1) since 1896. Notice the dry (and warm) periods from about 1920-1945 and 1975-1994, and the wet periods before and after. The last four years have been quite wet. Figure 2 shows Portland precipitation since 1920. The wet/dry periods are similar, coinciding almost exactly with those on the coast. Are the recent wet years a harbinger of things to come? Have we reentered a wet cycle?
The Global "Conveyor Belt"
Several years ago I heard a talk by Dr. William Gray (see Gray and Landsea, 1993) that described an exciting new finding based on the work of an oceanographer, Wallace Broeker. Broeker (1991) suggested that there is a global-scale current, operating on a time scale of several decades, which affects worldwide weather patterns. Broeker used data from ice cores to postulate that changes in this current may have been a primary triggering mechanism for the last ice age. Gray and Landsea, however, were more interested in our era. Gray has received acclaim for his work in predicting Atlantic hurricanes, with significant success.
The "conveyor belt" involves transport of warm ocean water from the Pacific through the Indian Ocean and into the Atlantic. In the north Atlantic, this warm water (now very saline due to evaporation during the journey), encounters cold water coming down from the north. The warm water cools quickly, and sinks (due to greater density). This sets up a sub-surface countercurrent which transports the cool water back to the Indian and Pacific oceans. Figure 3 illustrates the warm and cold currents associated with an active conveyor cycle.
In looking back over the last 100 years, Gray and Landsea identified four distinct periods, two when the conveyor belt was very active, two when it was quite inactive. They also found several important atmospheric phenomena which correlate quite well with the strength of the conveyor; we have added precipitation in the Northwest to the table (Figure 4). Note also that global temperatures seem to correspond to the active-inactive phases as well. This should not surprise us, since the tropical Pacific is the largest terrestrial heat source to the atmosphere; when the Pacific warms, so does the atmosphere. And during El Niño events, the Pacific temperatures (taken overall) warm significantly. Thus the warmup during El Niño-dominated inactive periods.
Gray and Landsea suggest that we may be entering a period of strong conveyor belt activity again. Consider how the four major indicators have changed over the last several years:
1. Atlantic Hurricanes. After the quietest 4-year period of the last 50 years during 1991-94, 1995 had the largest number of hurricanes since storms were first named in the early 1950s. 1996 and 1998 were active years, and 1997 was about average. The latest four-year period is the busiest on record.
2. Sahel precipitation. The last four years have seen near-average precipitation after many years of sever drought.
3. El Niños. Following a 20-year period with only one La Niña, three of the last four years have seen La Niña conditions in the Pacific.
4. Precipitation in the Northwest. Following the very dry 1975-94 period, which saw two significant statewide droughts and 10 consecutive dry years, we are now completing the fifth consecutive above-average year.
Predictions
The signs are there. All indications are that the Conveyor Belt has switched back to "active" again, portending a mostly wet regime for the next 20 years or so. If history repeats itself, we can expect:
- frequent floods
- no droughts
- about 75% of all years wetter than average
- relatively cool
Note that the above predictions apply to the cool season only. Summers do not necessarily reflect those trends. Our usual warm, dry, beautiful Northwest summers will still be here.
Technical note: Atlantic Hurricanes and Portland Precipitation
Since the number of Atlantic hurricanes seems to correspond
to precipitation in the Northwest, we decided to plot both types
of data together to see how well they correlate. As can be seen
in Figure 5, there is a very strong
correlation indeed. Active hurricane years are almost always followed
by wetter than average winter conditions in Portland, and inactive
years by dry winters. This graph not only validates the Conveyor
Belt concept, it suggest that hurricane frequency can be used
effectively to predict Portland precipitation the following winter.
References
Broeker, W.S., 1991. The Great Ocean Conveyor. Oceanography, 4, 79-89.
Gray, W.M. and C.W. Landsea, 1993. West African rainfall and
Atlantic basin hurricane activity as proxy signals for Atlantic
conveyor belt circulation strength. Conference on Hydrology,
American Meteorological Society, Anaheim, California, January.
