9. Estimation of Common Trends for Trophic Index Series

  1. Richard E. Chandler4 and
  2. E. Marian Scott5
  1. Alain F. Zuur1,
  2. Elena N. Ieno2,
  3. Cristina Mazziotti3,
  4. Giuseppe Montanari3,
  5. Attilio Rinaldi3 and
  6. Carla Rita Ferrari3

Published Online: 18 MAR 2011

DOI: 10.1002/9781119991571.ch9

Statistical Methods for Trend Detection and Analysis in the Environmental Sciences

Statistical Methods for Trend Detection and Analysis in the Environmental Sciences

How to Cite

Zuur, A. F., Ieno, E. N., Mazziotti, C., Montanari, G., Rinaldi, A. and Ferrari, C. R. (2011) Estimation of Common Trends for Trophic Index Series, in Statistical Methods for Trend Detection and Analysis in the Environmental Sciences (eds R. E. Chandler and E. M. Scott), John Wiley & Sons, Ltd, Chichester, UK. doi: 10.1002/9781119991571.ch9

Editor Information

  1. 4

    Department of Statistical Science, UCL, Gower Street, London WC1E 6BT, UK

  2. 5

    School of Mathematics and Statistics, University of Glasgow, Glasgow G12 8QW, UK

Author Information

  1. 1

    Highland Statistics Ltd, 6 Laverock Road, Newburgh AB41 6FN, UK

  2. 2

    Highland Statistics, Suite N 226, Av Finlandia 21, CC Gran Alacant Local 9, 03130 Santa Pola, Spain

  3. 3

    ARPA Emilia-Romagna, Struttura Oceanografica Daphne, V le Vespucci 2, 47042 Cesenatico (FC), Italy

Publication History

  1. Published Online: 18 MAR 2011
  2. Published Print: 18 MAR 2011

ISBN Information

Print ISBN: 9780470015438

Online ISBN: 9781119991571

SEARCH

Keywords:

  • additive modelling;
  • autocorrelation;
  • common trends;
  • dynamic factor analysis (DFA);
  • trophic index series

Summary

This chapter discusses the multimetric TRophic IndeX (TRIX). It describes the effects of pollution over time, in order to inform management strategies such as cleanup operations and to set new regulations. The chapter presents two different statistical approaches. First it uses additive models incorporating temporal autocorrelation along with a spatial residual correlation structures. In the second approach, dynamic factor analysis (DFA) is used. The chapter can be regarded as a guide to the model-building process for practitioners. It also discusses a univariate smoothing method to estimate underlying patterns in multivariate time series. The chapter uses additive modelling, taking a step-by-step approach to the development of an appropriate model for the TRIX series. It considers two rather different modelling approaches to investigate trends in eutrophication, as measured using TRIX, along the coast of the Emilia-Romagna region.

Controlled Vocabulary Terms

autocorrelation; factor analysis; multivariate time series analysis; trend