Decadal variations in surface solar irradiance as observed in a globally remote network

Twenty‐eight years (1977–2004) of surface solar irradiance observations at five remote climate observatories are analyzed for long‐term temporal variations and spatial representativeness. Autocorrelated temporal variations are suggested in the annual mean irradiance records at some of the sites and...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of Geophysical Research: Atmospheres 2006-10, Vol.111 (D19), p.n/a
Hauptverfasser: Dutton, Ellsworth G., Nelson, Donald W., Stone, Robert S., Longenecker, David, Carbaugh, Gloria, Harris, Joyce M., Wendell, James
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Twenty‐eight years (1977–2004) of surface solar irradiance observations at five remote climate observatories are analyzed for long‐term temporal variations and spatial representativeness. Autocorrelated temporal variations are suggested in the annual mean irradiance records at some of the sites and in normalized anomaly records for various combinations of the surface sites. These variations are illustrated by first‐, second‐, and third‐degree polynomials as well as by a sinusoidal fit, all of which are used as data smoothers not necessarily representative of any related physical processes, although some fits yield statistically significant coefficients. In general, there is an indication of a decreasing tendency in the early portion of the records and then an increasing tendency until toward the very end of the record. Variations among field sites are seen, with the one Arctic site showing the most departure from the others, with primarily a downward trend for most of the record. The spatial representativeness of each site's annual mean record is investigated using spatial cross correlation with satellite‐derived global estimates of the surface solar irradiance. These cross correlations show significant correlations with relatively large areas both contiguous and noncontiguous with the sites.
ISSN:0148-0227
2156-2202
DOI:10.1029/2005JD006901