gkernel (pyleoclim.utils.tsutils.gkernel)
- pyleoclim.utils.tsutils.gkernel(t, y, h=3.0, step=None, start=None, stop=None, step_style='max')[source]
Coarsen time resolution using a Gaussian kernel
- Parameters
t (1d array) – the original time axis
y (1d array) – values on the original time axis
step (float) – The interpolation step. Default is max spacing between consecutive points.
step_style ('string') – step style to be applied from grid_properties [default = ‘max’]
start (float) – where/when to start the interpolation. Default is min(t).
stop (float) – where/when to stop the interpolation. Default is max(t).
h (scalar; kernel e-folding scale) –
- Returns
tc (1d array) – the coarse-grained time axis
yc (1d array) – The coarse-grained time series
References
Rehfeld, K., Marwan, N., Heitzig, J., and Kurths, J.: Comparison of correlation analysis techniques for irregularly sampled time series, Nonlin. Processes Geophys., 18, 389–404, https://doi.org/10.5194/npg-18-389-2011, 2011.