Water, Air, and Soil Pollution

, Volume 85, Issue 3, pp 1491–1496

Estimation of the uncertainty in crop loss predictions through statistical modelling of the AOT40 levels for ozone

  • R. I. Smith
  • C. W. Anderson
  • I. Castro Rivadeneyra
Part III Direct Effects of Air Pollutants on Vegetation

Abstract

Ozone critical levels in Europe are defined in terms of an accumulated exposure over a threshold of 40 ppb, AOT40. For agricultural crops, for example, the critical level is an AOT40 of 5300 ppb.h during daylight in May to July in the year with the highest cumulative exposure in the last five years. In a region of the size of the UK, however, the worst case year is not the same over the whole region and maps become difficult to interpret. Prediction of crop losses on the basis of a single year out of five also wastes potentially valuable information. An alternative approach estimates the distribution of aggregate exceedances over a threshold by means of a compound Poisson model for episodes of raised ozone concentration with linear modelling techniques used to allow direct incorporation of covariate information. The use of spatial and environmental covariates, along with temporal and spatially correlated random effects, is explored using data from the UK ozone monitoring network. The model produces results similar to those from other mapping methods. By combining this model with a crop loss relationship, crop losses of 5–15% for the UK are predicted but the errors range between 2% and 6% indicating that fine detail in crop loss mapping is unlikely to be very accurate.

Keywords

ozone AOT40 linear modelling spatial correlation uncertainty 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Fuhrer, J. and Achermann, B. (editors): 1994, Critical Levels for Ozone: a UN-ECE workshop report, FAC, Liebefeld.Google Scholar
  2. Gnedenko, B. V. and Kolmogorov, A. N.: 1968, Limit Distributions for Sums of Independent Random Variables, Addison-Wesley.Google Scholar
  3. Leadbetter, M. R.: 1995, J. Statist. Planning & Inference 45, 247–260.Google Scholar
  4. Pleijel, H., Wallin, G., Karlsson, P., Skärby, L. and Selldèn, G.:1995, Water, Air and Soil Pollut. 85, (in press).Google Scholar
  5. Smith, R. I., Anderson, C. W. and Fowler, D.: 1994, J. Res. Natl. Inst. Stand. Technol. 99, 353–359.Google Scholar

Copyright information

© Kluwer Academic Publishers 1995

Authors and Affiliations

  • R. I. Smith
    • 1
  • C. W. Anderson
    • 2
  • I. Castro Rivadeneyra
    • 2
  1. 1.Edinburgh Research StationInstitute of Terrestrial EcologyPenicuik, MidlothianScotland
  2. 2.School of Mathematics and StatisticsUniversity of SheffieldSheffieldEngland

Personalised recommendations