Search Results

You are looking at 1 - 2 of 2 items for

  • Author or Editor: Alfredo R. Navato x
  • Refine by Access: All Content x
Clear All Modify Search
Bryan C. Weare, Alfredo R. Navato, and Reginald E. Newell

Abstract

An empirical orthogonal function analysis has been performed on monthly mean sea surface temperatures for the greater part of the Pacific Ocean between 55°N and 20°S. The analysis identifies the most important modes of seasonal and non-seasonal variability during the period 1949–73. A mode is defined spatially in terms of an empirical orthogonal function which describes the degree of coherence of variation. The function's corresponding coefficient portray the evolution of the mode in time. The seasonal variation is dominated by a mode having a 12-month periodicity and greatest coherence in the higher latitudes. A second important seasonal mode has a period of approximately 6 months and is dominated by deviations in the North Pacific. The most important non-seasonal variation is identified with the, long-recognized El Niño. The spatial pattern of this mode demonstrates the large-scale nature of the El Niño phenomenon. Other important non-seasonal modes are discussed.

Full access
Alfredo R. Navato, Reginald E. Newell, Jane C. Hsiung, Clare B. Billing Jr., and Bryan C. Weare

Abstract

Multiple-regression analyses of changes in tropospheric mean temperature as predictands and Pacific, Atlantic and Indian Ocean sea surface temperatures and atmospheric aerosol concentrations as predictors show that large fractions of the variances of the tropical, Northern Hemispheric and Southern Hemispheric extratropical tropospheric temperatures may be explained by fluctuations in ocean surface temperatures and atmospheric aerosols. The sensitivity of the tropical, Northern Hemisphere and Southern Hemisphere extratropical tropospheric temperatures to the various predictors are estimated.

To improve the precision of the estimates in the presence of serial correlations in the variables we used a generalized least-squares procedure to obtain the regression models.

Full access