Thinning method and intensity influence long-term mortality trends in a red pine forest

Tree mortality shapes forest development, but rising mortality can represent lost production or an adverse response to changing environmental conditions. Thinning represents a strategy for reducing mortality rates, but different thinning techniques and intensities could have varying impacts depending on how they alter stand structure. We analyzed trends in stand structure, relative density, stand-scale mortality, climate, and correlations between mortality and climate over 46 years of thinning treatments in a red pine forest in Northern Minnesota. USA to examine how thinning techniques that remove trees of different crown classes interact with growing stock manipulation to impact patterns of tree mortality. Relative density in unharvested plots increased during the first 25 years of the study to around 80%, then began to plateau, but was lower (12-62%) in thinned stands. Mortality in unharvested plots claimed 2.5 times more stems yr(-1) and 8.6 times as large a proportion of annual biomass increment during the last 21 years of the study compared to the first 25 years, but showed few temporal trends in thinned stands. Mortality in thinning treatments was generally lower than in controls, particularly during the last 21 years of the study when mortality averaged about 0.1% of stems yr(-1) and 4% of biomass increment across thinning treatments, but 0.8% of stems yr(-1) and 49% of biomass increment in unharvested plots. Treatments that combined thinning from above with low growing stock levels represented an exception, where mortality exceeded biomass production after initial thinning. Mortality averaged less than 0.1% of stems yr(-1) and less than 1% of annual biomass production in stands thinned from below. These trends suggest thinning from below minimizes mortality across a wide range of growing stock levels while thinning from above to low growing stock levels can result in dramatic short-term increases in mortality. Moderate to high growing stock levels (21-34 m(2) ha(-1)) may offer greater flexibility for limiting mortality across a range of thinning methods. Mean and maximum annual and growing season temperatures rose by 0.6-1.8 degrees C during the study, and temperature variables were positively correlated with mortality in unharvested plots. Mortality increases in unharvested plots, however, were consistent with self-thinning principles and probably not driven by rising temperatures. These results suggest interactions between thinning method and intensity influence mortality reductions associated with thinning, and demonstrate the need for broader consideration of developmental processes as potential explanations for increased tree mortality rates in recent decades. (C) 2010 Elsevier B.V. All rights reserved.