Mann (2004) Time Series Smoothing Routine


  • Provides smoothed time series and measure of misfit with option of boundary conditions useful in series with long-term trends or non-stationary behavior


  • The Smoothing Routine is Described in:

    Mann, M.E., On Smoothing Potentially Non-Stationary Climate Time Series, Geophysical Research Letters, 31, L07214, doi: 10.1029/2004GL019569, 2004.


    Matlab Smoothing Routine Code


    Updated Version yields a smoothed series based on combination of boundary constraints that minimizes MSE relative to raw time series





    Mann & Lees Multi-Taper Method (MTM)


  • Multitaper spectral analysis which provides an optimally low-variance, high-resolution spectral estimate.
  • Assumptions regarding signal (narrowband, but not strictly periodic) and noise ("red") that are most appropriate in the context of climate studies.
  • A "robust" method for accurate determination of the noise component of the spectrum.

    The MTM Method is Described in:

    Mann, M.E., Lees. J., Robust Estimation of Background Noise and Signal Detection in Climatic Time Series, Climatic Change, 33, 409-445, 1996.

    The MTM Method is Used in the SSA-MTM Toolkit


    MTM Fortran code

    Includes required subroutines and sample data for comparison wih results shown in the above paper.

    An enhanced version with "evolutive" spectral analysis and spectral coherence estimation is also now available.


    A separate package for performing complex demodulation of a time series as used in,

    Mann, M.E., Park. J., Greenhouse Warming and Changes in the Seasonal Cycle of Temperature: Model Versus Observation, Geophysical Research Letters, 23, 1111-1114, 1996.







    Mann & Park MTM-SVD Multivariate Signal Analysis


  • Detection of irregular spatiotemporal oscillatory signals immersed in spatially-correlated coloured noise with optimal signal detection properties
  • "Evolutive'' approach to detecting intermittent and/or frequency-modulated spatiotemporal oscillations.
  • Reconstruction of spatial and temporal patterns of oscillatory climate signals

    MTM-SVD Fortran codes, synthetic test dataset, and analysis results

    Codes Include:
    1. "LFV" multivariate spectrum estimation
    2. Spatiotemporal signal reconstruction
    3. Bootstrap confidence level estimation procedure, along with required subroutines, makefiles, and the synthetic test input and output.


    MTM-SVD References