Observational data sets in space physics often contain instrumental and sampling errors, as well as large gaps. This is both an obstacle and an incentive for research, since continuous data sets are typically needed for model formulation and validation. For example, the latest global empirical models of Earth's magnetic field are crucial for many space weather applications, and require time continuous solar wind and interplanetary magnetic field (IMF) data; both of these data sets have large gaps before 1994. Singular spectrum analysis (SSA) reconstructs missing data by using an iteratively inferred, smooth “signal” that captures coherent modes, while “noise” is discarded. In this study, we apply SSA to fill in large gaps in solar wind and IMF data, by combining it with geomagnetic indices that are time continuous, and generalizing it to multivariate geophysical data consisting of gappy “driver” and continuous “response” records. The reconstruction error estimates provide information on the physics of co variability between particular solar wind parameters and geomagnetic indices.

# Statistical methods

Gap Filling of Solar Wind Data by Singular Spectrum Analysis. Geophysical Research Letters. 2010;37 :L15101.Abstract

.
Reduced models of atmospheric low-frequency variability: Parameter estimation and comparative performance. Physica D: Nonlinear Phenomena. 2010;239 (3) :145–166.Abstract

.
Quasi-quadrennial and quasi-biennial variability in the equatorial Pacific. Climate Dynamics. 1995;12 :101–112.Abstract

.
Cluster analysis of typhoon tracks. Part II: Large-scale circulation and ENSO. Journal of Climate. 2007;20 (14) :3654–3676.

.
Interdecadal oscillations and the warming trend in global temperature time series. Nature. 1991;350 (6316) :324–327.Abstract

.
Singular spectrum analysis in nonlinear dynamics, with applications to paleoclimatic time series. Physica D. 1989;35 (3) :395–424.Abstract

.
Low-order stochastic model and ``past-noise forecasting" of the Madden-Julian oscillation. Geophysical Research Letters. 2013;40 :5305–5310.

.
Oscillatory modes of extended Nile River records (A.D. 622–1922). Geophysical Research Letters. 2005;32 (10) :L10702.Abstract

.
Trends, interdecadal and interannual oscillations in global sea-surface temperatures. Climate Dynamics. 1998;14 (7) :545–569.Abstract

.
Probabilistic clustering of extratropical cyclones using regression mixture models. Climate Dynamics. 2007;29 (4) :423–440.

.
Impacts of natural disasters on a dynamic economy. In: Extreme Events : Observations, Modeling, and Economics. American Geophysical Union and Wiley-Blackwell ; 2015. pp. 343–360.Abstract

.
Data-adaptive wavelets and multi-scale singular-spectrum analysis. Physica D. 2000;142 (3-4) :254–290.Abstract

.
Multiple Regimes in Northern Hemisphere Height Fields via Mixture Model Clustering. Journal of the Atmospheric Sciences. 1999;56 (21) :3704–3723.

.
Reply to T. Schneider's comment on "Spatio-temporal filling of missing points in geophysical data sets". Nonlinear Processes in Geophysics. 2007;14 (1) :3–4.

.
Interannual and Interdecadal Variability in 335 Years of Central England Temperatures. Science. 1995;268 (5211) :710–713.Abstract

.