The long-term study, initiated by Professor Arthur Willis in 1958, of the vegetation of permanent plots in road verges at Akeman Street, near Bibury, Gloucestershire, is a unique and valuable record of annual variations in shoot biomass of individual species, and of the vegetation as a whole, over a 38-year period. The study was originally devised to monitor the effects of herbicides and a growth regulator on roadside vegetation. However, more recently, in the context of the possible effects of climate change on indigenous vegetation, interest has centred on data from the untreated control plots and their potential in detecting long-term plant/weather relationships.
In Part 1 of this thesis, annual fluctuations in the de-trended or flat-trended aboveground biomass of over 40 plant species at Bibury are compared with three orders of climatic variables: individual weather variables such as temperature, rainfall and sunshine; the frequency of 'Lamb weather types' (anticyclonic, cyclonic and westerly weather), which in large part determines rainfall, sunshine and temperature; and with the changing position of the Gulf Stream in the North Atlantic, which may, in part, determine the frequency of weather types over the UK. In all comparisons, markedly non-random climate/plant relationships were detected. The responses of Bibury taxa to individual weather variables matched closely their responses to weather types. The analyses presented here agree with other studies which link a northerly Gulf Stream with increased frequencies of anticyclonic weather in spring and autumn and reduced frequencies in summer. It is suggested that plant species promoted by a more northerly Gulf Stream tend to be robust perennials with early phenologies or deep-rooted species which may exhibit a degree of drought avoidance, but which may also respond rapidly to favourable conditions in spring and summer. Overall, settled, hot, dry summers appear at Bibury to decrease the total productivity of the vegetation and promote the amount of bare ground in the ecosystem.
Part 2 of the thesis describes experiments designed to explore the mechanistic basis of the plant/weather relationships at Bibury and to extend the relevance of the data-set beyond the site itself. Five species were selected which exhibit a range of responses to individual weather variables. The plants were grown in deep containers (rigid plastic bins) in monoculture and in five-species mixture. Treatments were applied out of doors to establish whether the responses observed in the field could be repeated under semi-controlled conditions. The species were: Achillea millefolium (promoted by warm dry springs and summers), Cirsium arvense (retarded by warm dry springs and summers and promoted by mild winters), Dactylis glomerata (retarded by mild winters), Festuca rubra (retarded by mild winters) and Poa pratensis (retarded by warm dry springs and summers). The treatments were: heating with electric cables at the soil surface in late winter and spring; application of a severe frost in late spring; and the imposition of a severe drought in summer. The treatments affected both the vegetative and reproductive vigour of the species involved. In the great majority of cases predicted responses were fulfilled under the experimental conditions (for example. Dactylis glomerata was shown to be dependant on winter chilling to achieve maximum reproductive potential, reflecting its reduced performance in the field at Bibury following mild winters).
The five species were used to test current theories on genome size and plant response to unusually high temperatures in late winter and spring and to severe late frosts. As predicted, species with low nuclear DNA content tended to gain competitive advantage in mixture following the warming treatment, while those species with high nuclear DNA content showed evidence of reduced competitive ability. Also, as predicted small genome species tended to exhibit greater sensitivity to the late frost treatment. The experimental results provide valuable insights into the potential effects of weather on plant community dynamics and the relative competitive abilities of potential vegetation dominants and subordinates.