The statistical power of two butterfly monitoring schemes to detect trends

1. Monitoring schemes for butterflies in the United Kingdom and the Netherlands are aimed at the detection of long-term trends. It is useful to examine the power of these schemes to detect trends in a given period of time. 2. The approach was based on an ANOVA-model, and the trends were compared with the random population changes from one year to another. Several assumptions were made for simplicity's sake: autocorrelation in the data was ignored and only linear trends in log 10 (N + 1) transformed data were examined. 3. The relevant variance components to examine were the year-to-year variances and year-by-site variances. These were estimated from the time series of the British Butterfly Monitoring Scheme. Year-to-year variances appeared to be higher in northern Britain than in other regions. In addition, variance components were related to the voltinism of species. 4. Power assessment was based on the estimates of variance components and on the number of sampling sites. In the British scheme, for 37 out of 51 species studied a decrease of 50% or less is detectable with a power of 80% within a 20-year period. In the Dutch scheme such a decrease is detectable for 29 out of 47 species. 5. Because the schemes lack power for a number of species, several strategies are discussed to enhance power. For species present at less than 25 sites, it is most effective to increase the number of sampling sites where they are present, if that is possible in practice. But for species that are present at more than 50 sites, a further increase hardly improves the power. For these species, it is more efficient to adjust the data for weather conditions than to increase the number of sites. 6. The assumptions we made hardly affect the results for common species. But for rare species the results are more or less questionable. To get better estimates of the power, methods to assess power for monitoring schemes need to be developed that treat count data as discrete random variables.

[1]  M. Hill,et al.  Data analysis in community and landscape ecology , 1987 .

[2]  J. Sauer,et al.  Topics in route-regression analysis , 1990 .

[3]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[4]  H. Scheffé.  The Analysis of Variance , 1960 .

[5]  Tim Gerrodette,et al.  A POWER ANALYSIS FOR DETECTING TRENDS , 1987 .

[6]  E. Pollard.  [Effects of temperature on fecundity of nine lepidopteran species in Tiantong National Forest Park, Zhejiang Province, China]. , 1988 .

[7]  P. McCullagh,et al.  Generalized Linear Models , 1984 .

[8]  C.J.F. ter Braak,et al.  Analysis of monitoring data with many missing values: which method? , 1994 .

[9]  D. Moss,et al.  Increased fluctuations of butterfly populations towards the northern edges of species' ranges , 1994 .

[10]  C. Swaay.  An assessment of the changes in butterfly abundance in The Netherlands during the 20th Century , 1990 .

[11]  M. Morris,et al.  Monitoring butterflies for ecology and conservation , 1993 .

[12]  D. Moss,et al.  Calculation of collated indices of abundance of butterflies based on monitored sites , 1993 .

[13]  E. Pollard.  Monitoring butterfly numbers , 1991 .

[14]  J. Hodgson.  Commonness and rarity in British butterflies , 1993 .

[15]  M. Warren.  A review of butterfly conservation in central southern Britain: I. Protection, evaluation and extinction on prime sites , 1993 .

[16]  D. Moss,et al.  Population trends of common British butterflies at monitored sites , 1995 .

[17]  W. Hinds,et al.  Towards Monitoring of Long-term Trends in Terrestrial Ecosystems , 1984, Environmental Conservation.