Fall Forecast Update

The following update provides a summary of Environment Canada's fall seasonal forecast. By definition, a seasonal forecast covers large areas and therefore this summary includes non-agricultural areas.

Highlights

• Chances of getting above average temperature are highest in BC, northwest Territories, northern Quebec and Newfoundland (>70% probability).
• Confidence levels elsewhere for both temperature and precipitation are low due to the absence of significant drivers of climate in the oceans or atmosphere.

The fall season is normally a transition period in Canada marking the end of the agricultural year and the start of a new year in terms of moisture accumulation. Early fall precipitation replenishes the depleted soil moisture before freeze-up. This, along with the winter precipitation, contributes to spring runoff and soil moisture conditions at seeding time. Pastures and winter seeded cereals require fall precipitation.

The fall temperature forecast as produced by Environment Canada on September 1 (Figure 1), shows that above average temperatures are likely across British Columbia and western Alberta, the Yukon, Northwest Territories, Nunavut and Newfoundland. Below average temperature is forecast from eastern Saskatchewan to southern Quebec. For the rest of the country, near average temperature is forecast. The precipitation forecast for the fall (Figure 2) shows below average precipitation across BC and normal to above normal precipitation from Alberta to Atlantic Canada. A large area of the Yukon is in the average precipitation category while the Northwest Territories are in the near average to above average category. The central part of Nunavut is in the below average category.

How much confidence is there in the fall forecast? In spite of the low skill in forecasting climate for the fall season (using past experience from 1969 to 1994), there is a relatively higher confidence in the fall temperature forecast in BC, Northwest Territories northern Quebec and Newfoundland where the probability of getting above average temperatures exceeds 70%.This means, there was agreement on model outcomes between runs, or that model results did not differ by a large margin when repeat experiments were made with different starting conditions. Precipitation did not show a tendency towards any of the three categories used to describe a forecast; near average, average or above average. This came about due to the fact that there was a lot of spread in the experimental results between simulations. A large spread or variance in the results does not provide confidence in the likelihood of any one event occurring. Instead, it suggests equal chance for any of below average, near average or above average conditions to occur. The confidence levels across Canada are comparable to those from the USA. Their Climate Prediction Centre's (CPC) forecast shows a tendency towards average values at most continental locations. An average forecast does not carry sufficient confidence since it assumes equal likelihood of the three events discussed above.

Further to the low skill during the fall season, the current forecast lacks significant signals from the drivers of climate. There is no strong forcing in the oceans or atmosphere following the dissipation of the mild El Nino episode of 2002. The current atmospheric and oceanic conditions (as summarized from several simulation models) indicate that neutral conditions across the tropical Pacific will characterize the remainder of 2003. Some atmospheric indices such as SOI (Southern Oscillation Index), which a number of researchers have used in the past to predict Canadian climate, have flip-flopped from month to month. This has complicated further the predictive potential of the fall climate using these teleconnection indices (teleconnection indices such as SOI explain the linkage over great distances of seemingly unrelated weather anomalies such as droughts on the Prairies and the occurrence of warm temperatures in the Tropical Pacific Ocean).

How did the summer forecast perform? This question is answered by examining the observed summer temperature and precipitation. The summer of 2003 was generally hot (Figure 3) while the forecast called for normal temperature across the country. Precipitation was below normal (Figure 4) in most of the agricultural areas of Canada while the forecast called for normal to above normal precipitation. The combination of hot temperatures and below average precipitation resulted in moisture deficits in western Canada, leading to drought conditions in northwest and Interlake regions of Manitoba, eastern Saskatchewan and parts of northern Alberta. The hot and dry conditions were also associated with a high incidence of crop insects such as grasshoppers in Alberta, Manitoba and Saskatchewan. The poor performance of the summer forecast illustrates the challenges of representing both atmospheric and oceanic drivers of the climate system in a climate prediction model successfully.

As new information becomes available, future updates and outlooks will be issued accordingly. For comments and questions regarding this report, contact Aston Chipanshi (306-780-7680) or Ted O'Brien (306-780-6000).

Acknowledgements

We acknowledge with thanks the support of Environment Canada whose products we have used in this update.

PREPARED BY THE AAFC - PFRA AGROCLIMATE UNIT, REGINA, SASKATCHEWAN

Figure 1: Temperature forecast for the fall season

Figure 2: Precipitation Forecast for the Fall season.

Figure 3: Summer 2003 temperature verification

Figure 4: Summer 2003 precipitation verification