The results are displayed as a two-dimensional coloured raster (the SiZer map) where the y axis indicates the smoothing bandwidth considered, x axis the point in time, and the cell colour the presence of a significant rising trend ( red ), a significant falling trend ( blue ), the lack of a significant trend ( purple ), or the lack of data for meaningful inference ( grey ).
← all excerpts
Machine-learning based reconstructions of primary and secondary climate variables from North American and European fossil pollen data.
1
—
—