Forecasting of Air and Water Temperatures for Fishery Purposes with Selected Examples from Northwest Atlantic

M. Stein and J. Lloret

Institut für Seefischerei, Palmaille 9
D-22767 Hamburg, Germany

M. Stein and J. Lloret. 2001. Forecasting of Air and Water Temperatures for Fishery Purposes with Selected Examples from Northwest Atlantic. J. Northw. Atl. Fish. Sci. 29: 23-30. https://doi.org/10.2960/J.v29.a2

Statistical methods were used to describe and forecast monthly mean air and bottom water temperatures from 3 sites in the Northwest Atlantic region, up to one year in advance. ARIMA (Auto-Regressive-Integrated-Moving- Average) models were developed that accounted for 92% of the total variability in the long-term time series of monthly means of air temperature and 80% for bottom water temperatures. These models were then used to forecast conditions in 1999, with results showing good agreement between the predicted and observed values of both air and bottom water temperatures. Intervention analysis that models events as step-like features was also carried out. While this provided a better model fit to the observed data series, such events can not be predicted. Since nearly all fitted interventions appeared during winter (December-March), prediction of temperatures during these months must be viewed with caution. Results showed that the use of ARIMA models yields better forecasts for highly variable time series than simple models based upon averages of previous monthly averages alone.

PDF

Language - English
Publisher - Northwest Atlantic Fisheries Oragnization (NAFO), Dartmouth, N.S., Canada
Publication Date - 2001
Publication Type - Journal Article

Citation: M. Stein and J. Lloret. 2001. Forecasting of Air and Water Temperatures for Fishery Purposes with Selected Examples from Northwest Atlantic. J. Northw. Atl. Fish. Sci. 29: 23-30. https://doi.org/10.2960/J.v29.a2