Individual study: Changes in flowering plant species in annually sheep-grazed and ungrazed chalk grassland at Aston Rowant National Nature Reserve, Oxfordshire, England
Ward L.K. & Jennings R.D. (1990) Succession of disturbed and undisturbed chalk grassland at Aston Rowant National Nature Reserve: details of changes in species. Journal of Applied Ecology, 27, 913-923
In the UK, ancient chalk grasslands support a high diversity of characteristic plant and invertebrate communities. However, it is a habitat very vulnerable to a lack of grazing, and accordant successional changes when grazing is reduced or ceases altogether. Most management on nature reserves is related to past agricultural or silvicultural practices, but these may not necessarily be the only or best methods of conservation management.
This paper considers the details of the individual field layer species in grazed plagioclimax grassland and in ungrazed grassland in the early years of succession (1969-81). An earlier paper considers the dynamics of changes in flowering plants that occurred in experimental plots (Ward & Jennings 1990).
Study site: The study was undertaken at Aston Rowant National Nature Reserve (National Grid ref. SU 730966), Oxfordshire, southern England.
Experimental design: The experiment had a randomized block design with four replicates of four treatments: annual grazing, no treatment, and ungrazed after disturbance by burning or rotavating in 1969. All plots were 10 x 12 m, arranged in a row in each block. All blocks were on shallow chalky rendzina soils.
Plant monitoring: The first 14 years of secondary succession, from 1969 to 1982, are described. During this period most plots had no more than 50% cover by woody plants. Up to this time, shading by the woody plants had not greatly affected other vegetation. Flowering plants were recorded in the last week of June every year from 1970; but on a presence and absence basis only in 1969, in two permanent 1m² quadrats/plot positioned at random in each plot, and on the additional flowering plant species occurring in the plot outside the quadrats. The fixed recording date avoided major diversity changes which occur throughout the year, but could not avoid some seasonal variation.
After 14 years, only one plot (untreated) had greater than 50% cover of woody plants. However, changes in the field layer had occurred in all plots, especially in ungrazed plots where vegetation included taller grasses.
In the ungrazed plots, species could be grouped into one of four categories:
i) those which gained in cover (especially tall coarse grasses);
ii) those which increased initially, then declined (more characteristic of disturbed treatments);
iii) those which decreased (mainly smaller and shorter-lived species);
iv) those which showed no consistent gains or losses.
The species which most consistently increased in cover were taller coarser grasses i.e. false oat-grass Arrenatherum elatius, false brome Brachypodium sylvaticum and upright brome Bromus erectus. A.elatius became dominant very quickly in some quadrats where it tended to suppress many smaller species.
In the disturbed plots (especially rotavated ones), many seed-bank species appeared, several persisting for many years.
Hazel Corylus avellana (a woody shrub) and black bryony Tamus communis (a climbing species) colonized a few plots after several years, indicating the importance of dispersal (i.e. hazel nuts and berries of black bryony may be dispersed by mammals or birds) rather than the stage of succession reached.
Conclusions: The experiment at Aston Rowant demonstrated that the original flora has not been entirely lost over the 14 years of secondary succession and could probably be restored if management were re-instated. Some species were associated with disturbed plots, e.g. the uncommon wild candytuft Iberis amara, and continued in the early succession to scrub.
Ward L.K. & Jennings R.D. (1990) Succession of disturbed and undisturbed chalk grassland at Aston Rowant National Nature Reserve: dynamics of species changes. Journal of Applied Ecology, 27, 897-912.
Note: If using or referring to this published study, please read and quote the original paper, this can be viewed at: