Skip to main content
ARS Home » Research » Publications at this Location » Publication #105338
Title: DEVELOPMENT OF A MULTI-FACTOR MODEL FOR PREDICTING THE CRITICAL LEVEL OF OZONE FOR WHITE CLOVER

Author
item MILLS, G - INST. OF TERR. ECO-BANGOR
item BALL, G - NOTTINGHAM TRENT UNIV.
item HAYES, F - INST. OF TERR. ECO-BANGOR
item FUHRER, J - IUL
item SKARBY, L - IVL
item GIMENO, B - CIEMAT-DIAE
item TEMMERMAN, L - VET. & AGROCHEM. RES. CEN
item Heagle, Allen

Submitted to: Environmental Pollution
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 1/5/2000
Publication Date: N/A
Citation: N/A

Interpretive Summary: Ground level ozone (O3) injures plant leaves, and decreases growth and yield in many areas of the world. The amount of plant stress caused by a given O3 concentration is influenced by climate factors such as temperature and humidity, and may be influenced by other pollutants such as nitrogen oxides (NOx). Estimates of how climate affects the amount of O3 stress under field conditions have not been satisfactory. To address this question, a bio-indicator system consisting of an O3-resistant (NC-R) and O3-sensitive (NC-S) clone of white clover was tested at 14 sites in 8 European countries in 1996, 1997 and 1998. Plants were exposed to ambient O3 episodes, the forage was harvested every 28 days for 2 to 5 harvests per year, and the forage dry-weight ratio (NC-S/NC-R) was calculated for each harvest. A forage ratio near 1.0 indicates no O3 stress, whereas decreasing ratios below 1.0 indicate increasing O3 stress. The forage ratio at each harvest was related to the climatic and NOx conditions at each site using multiple linear regression analysis (MLR) and a relatively new method called artificial neural network (ANN) analysis. Twenty-one input parameters describing the O3 exposure, climate, and NOx levels were considered individually and in combination, with the aim of developing a model with high accuracy and simple structure that could be used to estimate effects of O3 on clover. The non-linear ANN models generally performed better than the linear MLR models, and can provide improved estimates of O3 effects under specific conditions. The best ANN model showed a relatively minor influence of relative humidity but indicated that plants are relatively sensitive to O3 when temperatures are moderate and the air is relatively free of NOx pollution.

Technical Abstract: Ozone (O3)-resistant (NC-R) and O3-sensitive (NC-S) clones of white clover (Trifolium repens L.) were exposed to ambient O3 episodes at 14 sites in 8 European countries in 1996, 1997 and 1998. The plants were grown according to a standard protocol, and the forage was harvested every 28d for 2 to 5 months per year by excision 7cm above the soil surface. Forage dry-weight ratio (NC-S/NC-R) was related to the climatic and pollutant conditions at each site using multiple linear regression (MLR) and artificial neural networks (ANNs). Twenty-one input parameters [e.g. different O3 concentration metrics, vapor pressure deficit (VPD), daily maximum temperature] were considered individually and in combination with the aim of developing a model with high r2 and simple structure that could be used to estimate effects of O3 on clover. MLR models were generally more complex, and performed less well for unseen data than non-linear ANN models. The ANN model with the best performance had five inputs with an r2 value of 0.84 for the training data, and 0.71 for unseen data. Two inputs to the model described the O3 conditions (AOT40 and 24h mean for O3), two described temperature (daylight mean and 24h mean temperature), and the fifth input (NO concentration at 1700h) appeared to be differentiating between semi-urban and rural sites. Neither VPD nor harvest interval was an important component of the model. The model can be used to estimate effects of O3 in specific conditions. For example, a 5% reduction in the forage ratio was predicted for AOT40s in the range 0.9 - 1.7 ppm h accumulated over 28d, with plants being most sensitive in conditions of low NOx and medium-range temperature.