mortality of midwestern overstory oaks as a bioindicator of environmental stress

Tree mortality rates are potential indicators of the impact of environmental stresses on forests. But the relationship between environmental stress and tree mortality is poorly understood, particularly in forests not experiencing epidemic mortality. To determine whether there is a sufficiently direct link between stress and tree mortality to allow mortality rate to serve as a stress bioindicator, I addressed the question: Are trees more likely to die during years when environmental stress is more intense? Subject trees were 62 overstory oaks (Quercus spp. L., ≥20 cm diameter at breast height) that died during 1968-1987.These trees were from seven oak-hickory forests in the midwestern USA that were experiencing low rates of tree mortality. A previous study had found these oaks to be growing more slowly than surviving oaks from the same forests, suggesting that these now-dead trees may have been physiologically compromised and particularly vulnerable to stress. Based on two Monte Carlo method-based tests, there was no consistent positive association between the likelihood of tree death in a year and the intensity of stress during that year or the previous year, where stress was represented by the growth rates of surviving neighbor trees (mean chronologies), growing season precipitation, or the mean Palmer drought severity index during the growing season. An evaluation of the statistical power of one of the two tests found a low probability of failing to detect a modest positive association between year of tree death and stress intensity. The results indicate the need for additional research on the link between stress and tree mortality if overstory tree mortality rates are to serve effectively as a bioindicator of environmental stresses.