Premium
Application of partial least squares regression to the diagnosis of year‐to‐year variations in Pacific Northwest snowpack and Atlantic hurricanes
Author(s) -
Smoliak Brian V.,
Wallace John M.,
Stoelinga Mark T.,
Mitchell Todd P.
Publication year - 2010
Publication title -
geophysical research letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.007
H-Index - 273
eISSN - 1944-8007
pISSN - 0094-8276
DOI - 10.1029/2009gl041478
Subject(s) - snowpack , climatology , partial least squares regression , atlantic hurricane , snow , regression , series (stratigraphy) , geology , environmental science , oceanography , tropical cyclone , meteorology , geography , statistics , mathematics , paleontology
Application of the method of partial least squares (PLS) regression to geophysical data is illustrated with two cases: (1) finding sea level pressure patterns over the North Pacific associated with dynamically‐induced winter‐to‐winter variations in snowpack in the Cascade mountains of western Washington state and (2) finding patterns of sea surface temperature over the tropical oceans that modulate Atlantic hurricane activity on a year‐to‐year basis. In both examples two robust patterns in the “predictor field” are identified that, in combination, account for over half the variance in the target time series.