EMPIRICAL-EVIDENCE OF GROWTH DECLINE RELATED TO VISIBLE OZONE INJURY

Differences in radial growth at breast height of yellow-poplar (Liriod endron tulipifera L.) and black cherry (Prunus serotina Ehrh.) were te sted between individual trees with a history of visible foliar ozone i njury and those not expressing foliar injury to ozone at three sites i n Great Smoky Mountains National Park, USA. No significant differences in growth for black cherry were found although there was a 12% reduct ion in radial growth over 5 yrs (1990-1994) (p-level 0.4) and 8% over 10 yrs (1985-1994) (p-level 0.6). There was a significant difference i n radial growth For yellow-poplar of 43% over 5 yrs (p-level 0.001) an d 30% over 10 yrs (p-level 0.005). Even though the trees of both speci es were selected to balance the diameter distribution of each species between the two groups at each site, it was still possible that the di fferences were due to some other factors than sensitivity to ozone exp osure. Therefore, a series of multiple linear regressions were used to identify the most explanatory model based on principal components der ived from the following independent variables: Diameter at breast heig hts total height, height to the live crown, percent slope, and a numbe r of competition indices based on the diameter and distance to competi tors. These regressions were then tested for different intercepts and slopes between the sensitive and nonsensitive trees. Once again, no si gnificant differences occurred for black cherry (p-levels of 0.4 and 0 .7 for five-year and ten-year radial growth. respectively) and some di fferences for yellow-poplar (p-levels of 0.04 and 0.1 for five-year an d ten-year radial growth, respectively). Although the conclusions did not change, the importance of proper balancing of the diameter distrib ution and accounting for the effects of uncontrollable independent var iables are discussed. (C) 1998 Elsevier Science B.V.