Spring areas: ecology, vegetation, and comments on similarity coefficients applied to plant communities

Spring areas in Jutland, Denmark, representing gradients in physico‐chemical conditions and providing a range of stable conditions in space and time, were examined for different aspects of community biology and population biology. Special attention was paid to aspects of the dynamic equilibrium resulting from interactions between stable environments and their biological communities. Localities were chosen to represent spring areas with clear differences in ground water chemistry and where stability in hydrological conditions had existed for many years. This was ensured by confirmation of the occurrence of arctic‐boreal‐montane plant species, being late glacial relicts. The degree of environmental stability of organisms of the study areas selected was otherwise determined by examination of a number of physical factors and the chemistry of the water. Tracheophytes and bryophytes comprising the macroflora of 14 main localities were described and the microflora represented by the diatoms described from the same localities plus one reference locality, Addit. These data provided a favourable basis for ordination and grouping of the spring areas and for discussing similarity coefficients applied to biological communities in general and to ecosystem stability in particular. Factors affecting the spring areas were surveyed by studies of peat deposits, hydrostatic potentials, historical events and human impact. Brief discussions on the functional aspects of the spring areas are given in an outline of the nutrient funds within the systems and the flow of nutrients in and out of them. Glacial deposits from the quaternary Period constitute the upper 10‐150 m of land mass of Jutland, a low‐lying moraine landscape reaching a maximum altitude of only 173 m a.s.l. A general description of the hydrological situation for watercourses in the 29800 km 2 peninsula of Jutland is given, together with environmental data for each of the localities as they are located along a geological SW‐NE transect with poor sandy soils in the southwest, grading off to clayey and calcareous soils in the north‐east. The close relationship between the ecology of the spring areas and their local geography, including the human impact on their respective catchment areas, is stressed. The geological differences between the sites selected along the gradient are shown to be clearly reflected in the chemical composition of the deposits of the soil profiles. Only at one locality, Hald, are peat layers lacking, due to strong movements of water and subsequent mineralisation of organic material. Peat layers at most localities are several metres deep. The amount of SiO 2 and mineral constituents, as well as the Ca/Mg ratio, reflect a strong influence of the subsoil through the outflow of water in all localities. A discussion of the hydraulic properties of these peat layers is given, together with figures from measurements of discharge related to catchment areas and mean runoff through main water courses from six of the localities. The degree of physical and chemical uniformity was examined over a two‐year period by means of monthly observations at 54 permanent sampling plots. Attention was paid to seasonal variations in temperature, hydrostatic potential, pH, conductivity, alkalinity and concentrations of Ca 2+ , Mg 2+ , K + , Na + , Cl ‐ , phosphorus and nitrogen. The annual temperature maximum is reached about 1 August, except for the sites at Bredsgaarde and Kjellerup, where the maximum is reached two weeks later, and the only truly south‐facing site, Krogensmölle, where the maximum is reached a month earlier. The annual minimum temperature is reached about 15 February, although at Krogensmölle it is reached about two weeks earlier. The curves for the annual tempeerature cycles are asymmetrical, the rise in the spring being slower than the decline in the autumn. Diurnal cycles superimposed on seasonal changes do not occur in the small rivulets as long as they are still running within the spring areas. Shifts from short periods with high temperatures to low temperature periods and vice versa are traceable, even at the site at Hald, a site with a very high rate of discharge. The significance of irregularly occurring aquiferous layers in the uppermost profiles and their influence on the rates of water movements is described by means of a correlation between the results obtained from the profile borings and observations on the hydrostatic potentials at different depths down‐slope. The pH values, ranging from a minimum of 6.1 at Bredsgaarde to a maximum of 8.0 at Hald, were constant all the year round except for a drop each year in the early summer. This is seen as an effect of an accelerated release of exudates from the roots during the start of the growth season. The specific conductivity followed the seasonal variation in alkalinity, being highest in the summer and ranging from a minimum of 123 mS/m at Höliund Bro to 484 mS/m at Vinkel. A high degree of constancy of ionic composition was found in the water at all sites, upon which a seasonal variation corresponding to the growing season could be correlated. The amounts of soluble Ca 2+ Mg 2+ and Na + increased regularly during the summer, together with that of Cl ‐ . The concentrations of K + in solution dropped during the summer. Seasonal changes in the concentrations of soluble nitrogen and phosphorus were also found, but with big differences in maximum and minimum values. Nitrogen concentrations were at a maximum during the summer whereas phosphate were at minimum at this time, although neither of these were particularly high, a reflection of the rapid recycling of these elements within these ecosystems. The macroflora was dominated by hemicryptophytes, at 52%, followed by bryophytes and geophytes at 28% and 17%. The groups phanerophytes, chamaephytes and therophytes play a minor role, at about 1% each. In all, 129 taxa are represented, if we omit all species not scoring a frequency of occurrence of at least 5% at one of the 14 sample stands. The dicotyledons are represented by 57%; monocotyledons, mostly grasses and sedges, together with ferns and horsetails, by 43%. The average number of species per 0.1 m 2 ranged from 11 (11.35) at Tinnet to 19 (19.05) at Hellum. Above‐ground phytomass taken as a mean from 8 localities was 710 g m ‐2 for 1978, ranging from 540 to 980 g m ‐2 . For the same localities it was 560 g m ‐2 for 1979, ranging from 490 to 660 g m ‐2 . The lower values for 1979 are ascribed to the summer being extraordinarily cool and wet. The microflora, represented by the diatoms, is dominated by species with large, strongly silicified valves, consisting of epiphytic and more or less aerophilous species. A total of 196 samples comprising more than 98000 valves of diatoms were examined, being epiphytes on the most abundant and widespread moss species from the 14 main localities plus the reference locality for diatoms at Addit. In all, 364 taxa were found through these frequency counts. Far more taxa appear in this investigation than in the sparse earlier analyses from mosses not growing submerged. A number of taxa are recorded in Denmark for the first time. It is concluded that common freshwater diatoms are able to live aerophilously, and presumably often do. Centric freshwater species commonly occurring in plankton are infrequent in the material. The passive dispersal of diatoms is discussed, based on annotations from the single sampling plots and their dates of collection. The 14 sample stands are classified into one community based on similarity in botanical structure. In order to calculate the differences between the stands, a process of arranging them numerically is presented. The ordination method used is based on information on the frequency of macrophytes, with application of the usual statistical test. The geometrical grouping of the stands represents the numerical distances or coefficients of dissimilarity. The depiction of stands or groups of stands leads to a result distinctly marked by the chemical conditions along the SW‐NE gradient. The relative abundance of rare and common species of macrophytes as well as diatoms were normally distributed at almost all localities, making it possible to do simulation studies of ordination methods based on presence/absence data. The way in which such indices are influenced by the most fundamental decision from a biological point of view which should precede any mathematical processing is demonstrated: include only resistant and