Environmental Change Network link Module 10 final report

by
M.D. Morecroft, W.A. Scott, M.E. Taylor, T.W. Parr, S.M. Smart & R.G.H. Bunce
Contact: Mike Morecroft (mdm@ceh.ac.uk). CEH Merlewood, c/o University Field Laboratory, Wytham, Oxford, OX2 8QJ, UK

Contents

• Executive Summary - HTML
• Report - PDF, 294 Kb
  

Note: you will need Adobe® Acrobat Reader®, if you don't have this installed on your computer, click here. 

For viewers with visual difficulties, information on the services provided to improve the accessibility of Acrobat documents is available from the following web site: http://access.adobe.com

• Go to the Module 10 (Environmental Change Network link) information page

Executive Summary

• Module 10 of the Countryside Survey 2000 research programme addressed the issue of year-to-year variability in vegetation: whether it was likely to influence Countryside Survey results and how it might relate to weather patterns.

• The vegetation of plots at Environmental Change Network (ECN) sites throughout the UK was recorded in 1998 and 1999. ECN is a collaborative, long-term monitoring programme, with the aim of detecting change in a wide range of environmental variables, using a series of intensively studied sites. For many of these plots, data were also available from 1997, 1996 and in some cases 1994. The data were analysed by testing for year-to-year differences in numbers of species and the ecological characteristics of those species (using the systems of Grime and Ellenberg); these variables have also been used in the analysis of the main CS2000 vegetation results. The Countryside Vegetation System (CVS) was used for classifying the vegetation and stratifying the sampling and analysis.

• Year-to-year changes in CVS aggregate vegetation classes were found: 23% of the studied plots changed classification at some point. Between 1998 and 1999, the two years with the most data, 12% of plots changed. This compares with a change of 30% between 1990 and 1998 in CS2000 data.

• Arable Crop/ Weed communities showed the largest year-to-year variability in species number and ecological characteristics; fertile grasslands and woodlands were also relatively variable.

• The number of species per plot decreased significantly between 1997 and 1999, taking all vegetation classes together. The difference was also statistically significant in fertile grassland, lowland woodland and heath / bog classes when analysed separately. In the fertile grasslands, which showed the largest significant decrease, the decline in numbers of species was largely due to decreases in ruderal species ('weeds' able to grow quickly in temporary gaps in the grass sward). This may relate to the differences in weather between the dry conditions of 1995-7 and the substantially wetter period between 1997 and 1999.

• In infertile grassland there were significant differences in some vegetation characteristics, but not species richness, between 1999 and earlier years; the tendency was away from stress tolerant plants and towards more fast-growing competitive ones. An increase in the mean Ellenberg fertility score in this vegetation class, indicated a higher proportion of species adapted to high nutrient conditions. This may reflect an ongoing change in response to nutrient enrichment (e.g. from atmospheric pollution and spray drift), detected in both CS1990 and CS2000 data.

• Although there were significant differences between years, and climate may well have been an important factor causing these, very few significant correlations between vegetation and weather variables were found. This is probably because of the short length of the time series. Many climatic effects may be subject to a time-lag and interactions between variables.

• An understanding of year-to-year changes in vegetation can help to inform the results of Countryside Surveys. Year to year variability can be large enough to obscure or distort long-term changes and should be accounted for in the interpretation of CS2000 and similar monitoring exercises. In the case of the CS2000 survey it is likely that changes of a similar nature would have been detected if the main field survey were carried out in 1997 or 1999 rather than 1998; the size of the changes could however have been quite different.

• We recommend that annual vegetation monitoring be continued at ECN sites with further developments to improve the coverage of vegetation types and sites. More detailed analysis, using other data from ECN sites should be carried out to improve understanding of the underlying mechanisms. Ultimately it should be possible to develop models of vegetation response to climate to help interpret results of wider, intermittent monitoring programmes.