Examination of the Merged Sea Surface Temperature Using Wavelet Analysis

In the previous study, merged sea surface temperature (SST) dataset called “New Generation SST” has been produced from several infrared and microwave satellite SSTs through an objective mapping. Here we examine the merged SST by comparison with moored buoy SST at 1 m depth, which is treated as true...

Full description

Saved in:
Bibliographic Details
Published in:Journal of oceanography 2004-10, Vol.60 (5), p.843-852
Main Authors: Hosoda Kohtaro, Kawamura, Hiroshi
Format: Article
Language:English
Subjects:
Buoys
Components
Marine
Meteorology
Oceanography
Rain
Rainfall
Response time
Sea surface
Sea surface temperature
Surface temperature
Temperature
TRMM satellite
Tropical climate
Wavelet analysis
Winter
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In the previous study, merged sea surface temperature (SST) dataset called “New Generation SST” has been produced from several infrared and microwave satellite SSTs through an objective mapping. Here we examine the merged SST by comparison with moored buoy SST at 1 m depth, which is treated as true sea surface temperature. Comparison between wavelet spectra of merged and buoy SSTs shows that the former have larger amplitudes than those of the latter, which is partly explained as an aliasing effect due to TRMM Microwave Imager (TMI) aboard Tropical Rainfall Measuring Mission (TRMM) sampling on merged products. Coherency between wavelet-decomposed merged and buoy SSTs has high values in autumn and low ones in winter to spring. In winter, phase differences between them are positive, meaning that wavelet components of merged SST lag those of buoy SST. Reasons for delay and low coherency are: (1) seasonal components of merged SSTs are strongly affected by a lack of infrared SSTs due to clouds in winter, and (2) small-scale oceanic features, undetectable by coarse-resolution microwave SSTs, are blurred by the merging process. Improvements of merging methodology are discussed with regard to present study results.
ISSN:0916-8370
1573-868X
DOI:10.1007/s10872-004-5777-7