Identification of non-linear influences on the seasonal ozone dose-response of sensitive and resistant clover clones using artificial neural networks

Ozone is a commonly occurring pollutant that has a large impact on the yiel d of agricultural crops. The dose-response of crops in the field is complex , with influences from numerous biotic and abiotic factors, including micro climatic variables. This paper presents results of a number of analysis met hods of artificial neural network (ANN) models, developed on biomonitoring data from 12 countries, to identify the importance of interacting influence s on the biomass response of sensitive (NC-S) and resistant (NC-R) clones o f white clover (Trifolium repens L. cv. Regal). These methods of analysis w ere also used to identify the importance of influences on a subset of the d ata. Empirical equations were extracted from the ANN model with the best pe rformance and these were analysed to determine their performance and to ind icate the nature of microclimatic influences. Analysis indicated that combi nations of VPD and the number of raindays were strong influences on the ozo ne dose-response and that temperature and the number of raindays had a seco ndary influence on the NC-S/NC-R biomass ratio irrespective of the ozone do se. Analysis of derived empirical equations indicated they compared well wi th the ANN model and that only a small loss in accuracy occurred. (C) 2000 Elsevier Science B.V. All rights reserved.