Typical liberal environmental science. It is all anecdotal with no firm statistical analysis
The Nature article (not the summary article I posted) does have statistical analysis, and cites 97 references that were consulted to write the paper. Furthermore, the early occurrence of spring has been statistically confirmed by Magnuson et al. (spring thaw and autumn freeze data for Northern Hemisphere bodies of water), and by Robeson using temperature data in Illinois (cited in the World Climate Report, Volume 7, No. 14). John Daly's Still Waiting for Greenhouse site has a feature about the Nenana Ice Classic (also written up in this month's issue of Discover magazine) that cites a similar trend. So what the Nature report describes is supported by actual data.