Agroclimatic requirements and phenological responses to climate change of local apple cultivars in northwestern Spain

In a global warming context, analyses of historic temperature records are essential to understand the potential impacts of climate change on spring phenology. To estimate flowering trends over recent decades, we analyzed long-term temperature and phenology records of eleven local apple cultivars in Asturias (northwestern Spain) in a temperate oceanic climate. Our results show that, over a period of 30 years, bloom dates of the local cultivars have experienced relatively minor changes, considering that temperatures increased strongly since 1978, by 0.30 degrees C per decade. An explanation for this weak phenological response to warming may be that these temperature changes only had a small effect on overall chill accumulation, but possibly delayed the onset date of endo-dormancy, which may have counteracted phenology-advancing effects of warming in spring. At present, chill accumulation in this area is high, at an average of 96 Chill Portions from November to March, which indicates that chill is not currently a limiting factor for the quality of flowering and fruiting in the study area. We used Partial Least Squares (PLS) regression to delineate an effective chilling period between November 12th and February 9th and effective heat accumulation between March 15th and May 4th. While these periods appear plausible, we noticed that this approach was unable to identify well-known differences in chilling requirements among many of the cultivars, with similar chill needs determined for many of them. This observation may be explained by inaccurate expectations about cultivars' climatic needs, by inaccuracy of the chill (and possibly heat) model or, most concerning, by inability of the PLS approach to correctly identify the chilling periods of apple cultivars in this region. Bloom dates were similarly responsive to mean temperature during the chill and the heat accumulation phases, indicating that both processes need to be considered when predicting future phenology.