Plant development scores from fixed-date photographs: the influence of weather variables and recorder experience

In 1944, John Willis produced a summary of his meticulous record keeping of weather and plants over the 30 years 1913-1942. This publication contains fixed-date, fixed-subject photography taken on the 1st of each month from January to May, using as subjects snowdrop Galanthus nivalis, daffodil Narcissus pseudo-narcissus, horse chestnut Aesculus hippocastanum and beech Fagus sylvatica. We asked 38 colleagues to assess rapidly the plant development in each of these photographs according to a supplied five-point score. The mean scores from this exercise were assessed in relation to mean monthly weather variables preceding the date of the photograph and the consistency of scoring was examined according to the experience of the recorders. Plant development was more strongly correlated with mean temperature than with minimum or maximum temperatures or sunshine. No significant correlations with rainfall were detected. Whilst mean scores were very similar, botanists were more consistent in their scoring of developmental stages than non-botanists. However, there was no overall pattern for senior staff to be more consistent in scoring than junior staff. These results suggest that scoring of plant development stages on fixed dates could be a viable method of assessing the progress of the season. We discuss whether such recording could be more efficient than traditional phenology, especially in those sites that are not visited regularly and hence are less amenable to frequent or continuous observation to assess when a plant reaches a particular growth stage.