Design and performance of a free-air exposure system to study long-term effects of ozone on grasslands

Studying long-term effects of ozone on perennial plant communities in situ is difficult because of the need for large observation plots and long exposure times, and because of confounding effects of altered microclimatic conditions in most systems. A novel free-air fumigation system was designed for multi-year exposures of semi-natural grassland plots to either ambient or elevated ozone concentrations at a Swiss sub-alpine site. The system operates by releasing ozone-enriched air from either one of two 120° sectors, placed in each of the two main wind directions. Ozone-enriched air is rapidly mixed with ambient air behind a small transparent windscreen mounted at canopy height around the 7-m diameter circular plots. Ozone generation and dispensing during daylight hours is PC controlled and depends upon wind speed, wind direction and ambient ozone. In 2001, from April to October fumigation was possible during 85% of the time. The ozone-enrichment factor (EF, [O3]enriched/[O3]control) in ozone plots relative to control plots averaged 1.48 (±0.24 SD), and EF was highest with winds of around 2 m s−1. The seasonal cumulative ozone exposure (AOT40) at canopy level was 5 ppm h in the control and 30 ppm h in the ozone-treated plots, with less than 10% variation between replicates, and 13 ppm h in ambient air at 2 m. Variation in ozone across the plot-diameter was <5% for 24-h mean concentrations, but differed by a maximum of 20% in terms of AOT40. There was no difference in microclimatic conditions between experimental plots and ambient conditions. It is concluded that the system is suitable for long-term ozone experiments at relatively low running costs, but that the variation in ozone distribution across the plots may lead to systematic differences in cumulative exposure over longer periods of time which need to be taken into account when relating exposure to biological effects.