Experimental CCA Forecasts of Canadian Temperature and Precipitation --- Apr-May-Jun 1997

Over the past few years, forecasts of Canadian temperature and precipitation using the multivariate statistical technique of canonical correlation analysis (CCA) have been presented in this Bulletin. For Canada, predictive relationships between evolving large-scale patterns of quasi-global sea surface temperature, Northern Hemisphere 500 mb circulation, and the subsequent Canadian surface temperature and precipitation have been developed. Here, we present the forecasts for Apr-May-Jun 2002 using the predictor fields through February 2002. More details about the Canadian CCA-based seasonal climate prediction can be found in Shabbar (1996a, 1996b) and Shabbar and Barnston (1996).

Figure 1 shows the CCA-based temperature forecast for the 3-month period of Apr-May-Jun 2002 expressed as standardized anomaly. Table 1 shows the value of standard deviation in ^oC at selected stations. The mean skill over all 51 stations is given in the caption beneath each forecast map. The field significance is also shown, reflecting the probability of randomly obtaining overall map skill equal to higher than that which actually occurred. Field significance is evaluated using a Monte Carlo procedure in which the forecast versus observation correspondences are shuffled randomly 1000 times. The field of cross-validated historical skill (correlation) for the forecast time period is shown in Figure 2. The forecast has a modest expected skill: a mean national score of 0.14 and a field significance of 0.082. The skill of the temperature forecast drops off considerably in spring in Canada. Local skills are highest over the northern Canadian Prairies, and modest skill is found on the west coast of Canada. A large area of Canada from the Yukon through central Canada and into Atlantic Canada is expected to have negative temperature anomaly; positive temperature anomalies are forecast over southern British Columbia and the high Arctic Islands.

Figure 3 shows the CCA-based precipitation forecast for the 3-month period of Apr-Jun 2002 expressed as standardized anomaly. Table 1 shows the value of standard deviation (mm) at a selected few stations. Cross-validated historical skill (correlation) for this time period is shown in Figure 4. The forecast has moderate expected skill: a mean national score of 0.14 and a field significance of 0.050. Local skills are highest over sections of the Prairies, the Yukon and the Atlantic Canada. An area stretching from the lower Great Lakes through the St. Lawrence Valley into central Quebec is expected to have deficit in Apr-May-Jun precipitation. Northwestern Ontario and southern Manitoba show above normal values. Elsewhere, near-normal precipitation amounts are expected.

Both atmospheric and oceanic indices have been showing a moderate to weak strength of the cold phase of ENSO for the past two years in the tropical Pacific. Over the past three months, however, the cold phase of ENSO has ended and the warm phase of ENSO has begun to emerge. Most dynamical models are predicting a neutral to slightly warmer conditions into early summer. While some statistical models are forecasting warm conditions by fall, still other models are showing neutral conditions to persist in the tropical Pacific well into 2002. The Apr-May-Jun 2002 forecast recognizes the demise of the cold ENSO event. More importantly, it would appear that a shift from the positive phase to negative of the Pacific Decadal Oscillation (Mantua et al. 1997) over the last several months has a bigger influence on the Canadian climate from the spring to early summer season.

References:

Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, R. C. Francis, 1997: A pacific interdecadal climate oscillation with impacts on salmon production. Bull. Amer. Soc., 78, 1069-1079.

Shabbar, A., 1996a: Seasonal prediction of Canadian surface temperature and precipitation by canonical correlation analysis. Proceedings of the 20^th Annual Climate Diagnostic Workshop, Seattle, Washington, Oct. 23-27, 1995, 421-424.

Shabbar, A., 1996b: Seasonal forecast of Canadian surface temperature by canonical correlation analysis. 13^th Conference on Probability and Statistics in Atmospheric Sciences. American Meteorological Society, San Francisco, California, Feb. 21-23, 339-342.

Shabbar, A. and A. G. Barnston, 1996: Skill of seasonal climate forecasts in Canada using canonical correlation analysis. Mon. Wea. Rev., 124, 2370-2385.

Table 1. Standard deviation of temperature (Temp) and precipitation (Prcp) for the 3-month period April through June at selected Canadian stations.

Station	(mm)	(^oC)
Whitehorse	1.6	13.2
Fort Smith	2.5	19.5
Innujjuak	1.9	18.2
Eureka	2.6	3.5
Vancouver	1.3	26.6
Edmonton	1.7	26.3
Regina	2.1	30.9
Winnipeg	2.2	37.4
Churchill	2.1	24.6
Moosonee	1.9	27.6
Toronto	1.6	30.0
Quebec City	1.3	35.3
Halifax	1.2	42.7
St. John’s	1.6	46.6

Figure captions:

Fig. 1. CCA-based temperature forecast for the 3-month mean period of Apr-May-Jun 2002. Forecasts are represented as standardized anomalies.

Fig. 2. Geographical distribution of cross-validated historical skill for the forecast shown in Fig. 1, calculated as temporal correlation coefficient between forecasts and observations. Areas having forecast skill of 0.30 or higher are considered to have utility. The mean score over 51 stations is 0.14. Field significance is 0.08.

Fig. 3. CCA-based precipitation forecast for the 3-month mean period of Apr-May-Jun 2002. Forecasts are represented as standardized anomalies.

Fig. 4. Geographical distribution of cross-validated historical skill for the forecast shown in Fig. 3, calculated as temporal correlation coefficient between forecasts and observations. Areas having forecast skill of 0.30 or higher are considered to have utility. The mean score over 69 stations is 0.14. Field significance is 0.05.

contributed by Amir Shabbar

Fort Smith

2.5

19.5