The Ørsted Satellite Project

Cain, J.C. (1), D.T. Mozzoni (1), B.B. Ferguson (1), and O. Ajayi (1)

1. Florida State University, Geophysical Fluid Dynamics Institute, and School of Computational Science and Information Technology, Tallahassee, FL 32306-4360, USA, Fax: 850 644 8972. E-mail: cain@gfdi.fsu.edu

Abstract. Ørsted and observatory hourly average data were selected for magnetically quiet (K_p=0 for observatory, K_p <2+ for Ørsted), nighttime (0-3^h local time for observatory, dusk to dawn for Ørsted) intervals and corrected for D_st assuming an i/e ratio of 0.28. Magsat (m102389) coefficients were adjusted up to n=20 in spatial terms with higher coefficients kept constant, n=13 in linear secular variation terms and n=12 in parabolic secular variation terms. There was also a 30 nT n=1 external term. The Gaussian envelopes of the Ørsted-F052101 misfit data have fitted means and sigmas (nT/yr): nonpolar (X = 0.5(8), Y = -1.4(5), and Z = -0.7(5)), polar (X = 3.6(15), Y = 2.1(15), and Z = -0.3(7)), ship-towed (F = -13(108)), and observatory (dx/dt = 0.1(4), dy/dt = -0.5(3) and dz/dt = -0.6(3)). Component maps of this model truncated to n=10 were compared with the IGRF2000 and generally found to be within 20 nT at the surface except for a 125 nT difference in the northern polar region. However, there were differences in secular variation ranging from a few tens of nT/year in many ocean areas up to a maximum of 75 nT/year in the South Atlantic. Secular variation represented by the IGRF2000 SV coefficients is seriously in error in some oceanic regions.