Ecological investigations on the northern distribution range of Hippocrepis comosa L. in Germany

The objective of this study was to identify the causes of the limits of the geographical range of Hippocrepis comosa. Along a gradient from the northwestern distribution boundary towards the distribution centre in Germany, 46 field plots were established where growth and microclimate were monitored simultaneously. In total, 11 vegetative and generative traits and 7 microclimatic parameters were recorded over 7 time intervals during the vegetation period, together with 14 general site parameters. Regression analyses were performed between all traits and environmental parameters in a certain interval. At the beginning of the growing season the best positive correlation coefficients for vegetative growth were observed with soil temperature. From the end of May to the middle of June, vegetative growth rates decreased and showed the best positive correlation with soil water content. Despite credible relationships between vegetative growth and microclimate, their contribution toward explaining the northern distribution boundary was found to be limited, because no correlation with the distance from the distribution boundary was observed. The only growth parameter that showed both a positive correlation with distance from the distribution boundary and a significant correlation with microclimate was the percentage of seed setting, which increased towards the distribution centre and was correlated with air temperature. Further field observations on plots outside the actual range of Hippocrepis comosa revealed no microclimatic reasons as to why the species was absent from these sites. This shows that the environmental parameters are in no way deterministic for the range limit. The frost hardiness of Hippocrepis comosa was studied in additional laboratory experiments in which significant damage was not found above -18 degrees C for adult plants and above -14 degrees C for seedlings, which is remarkably low and too low to be relevant for the northwestern distribution boundary. Another field experiment revealed that seedling establishment exhibited a positive relationship to soil water content, which became more favourable towards the range boundary. It is concluded that temperature, particularly air temperature, makes the largest contribution to explaining the northern distribution boundary of Hippocrepis comosa in Germany and that mainly generative reproduction is affected.