Long-term global change effects on forest biogeochemistry in the north-central United States

Abstract

Human activities have substantially altered the composition of the atmosphere and many of these changes directly affect the biogeochemistry of forest ecosystems. Because of the geography of industrialization, these impacts are particularly acute in northern temperate forests. Unfortunately, most studies examining the effects of altered atmospheric composition on forest ecosystems may not be accurate predictors of the long-term impacts on mature forests because these studies used immature trees and were short in duration. Here, I use measurements from two large long-term collaborative experiments to examine the impacts of altered atmospheric composition on forest biogeochemistry in the north-central United States. At the Rhinelander free-air carbon dioxide (CO2) enrichment experiment in Wisconsin, I examined the independent and interactive effects of increased concentrations of atmospheric CO2 and tropospheric ozone (O3) on leaf production and soil carbon (C) storage in three forest communities. To estimate leaf production, litter traps were used to collect fallen leaves from 2002 to 2008 (years five through eleven of the experiment). In addition to leaf production (g m-2), these collections were used to assess leaf area (m2 m2 ), leaf litter nitrogen (N) concentration (mg g-1), and the leaf N content (g N m-2). On average, the factorial elevated CO2 effect (+CO2 and +CO2+O3) stimulated leaf production by 36% and the factorial elevated O3 effect (+O3 and +CO2+O3) decreased leaf production by 18%. Similar effects were observed for leaf area. However, the relative effects of the individual treatments were highly dynamic. From 2002 to 2008, the positive effect of the elevated CO2 treatment (+CO2) on leaf production relative to the ii ambient treatment decreased from +52% to +25%, while the negative effect of the elevated O3 treatment (+O3) relative to ambient changed from -5% to -18%. The CO2 and O3 treatments did not have significant overall effects on litter N concentrations. Consequently, the leaf litter N content (g m-2) was increased 30% by the elevated CO2 treatments and decreased 16% by the elevated O3 treatments. To estimate changes in soil C pools, the top 20 cm of the mineral soil was sampled seven times between 1998 and 2008. Despite an increase in the input of leaf and root litter by elevated CO2 and a decrease in litter inputs by elevated O3, there were no significant effects of CO2 and O3 on soil C storage for the overall experiment. However, within the forest community containing only aspen (Populus tremuloides), there was significantly less soil C (-17.4 Mg ha-1) beneath forests receiving the elevated CO2 treatments (+CO2 and +CO2+O3) in the 2008 samples. In addition, I was able to use the unique 13C signature of fumigation CO2 to trace the input of new C into the soil in the elevated CO2 treatments (+CO2 and +CO2+O3). Initially, soils from the +CO2+O3 treatment had less new C than soils from the +CO2 treatment, but this difference gradually disappeared. This gradual disappearance matched trends in fine root production. Combining the leaf production study with the soil C study, these results suggest that the rate of soil C cycling accelerated under elevated CO2 and declined under elevated O3 because changes in soil C accumulation did not match changes in litter production. The other long-term experiment tests the influence of atmospheric deposition on four mature northern hardwood forests spread across 500 km in northern Michigan. These four forests sit along a north to south gradient, with warmer temperatures and higher iii inputs of both acid deposition and N deposition at the southern end of the gradient. These sites were established in 1987 to examine the impacts of atmospheric deposition along this gradient, but a parallel experiment was established at the same four sites to simulate potential increases in N deposition. I utilized both aspects of this experimental design, using the existing deposition gradient to examine the ongoing effects of atmospheric deposition and using the N addition experiment to test the long-term influence of added N on leaf-level photosynthesis. Since these sites were established in 1987, there have been major changes in federal emissions regulations. These new regulations greatly restricted emissions of acid deposition precursors, but did not attempt to control overall N deposition. In the time since this policy was enacted, there have been remarkable changes in the impacts of acid deposition and N deposition on the biogeochemistry of these four forests. Using data only from the plots receiving ambient deposition, I found that there have been decreases in leaf sulfur, calcium, and aluminum concentrations over the past two decades. Acid deposition usually increases concentrations of these elements in soil solution, so the observed changes in leaf chemistry signal a waning influence of this pollutant. In comparison, leaf δ 15N and soil lysimeter data show that persistent ambient N deposition has caused widespread increases in both the availability of inorganic nitrogen and soil nitrate leaching. The declining influence of acid deposition shows that environmental policy can quickly and broadly influence forest biogeochemistry. Although there are large amounts of nitrate being leached from these forests as a result of ambient N deposition, the parallel N addition experiment at these same sites resulted in increased aboveground growth. Because of the key role of N in photosynthesis, conceptual models often attribute growth increases from increased N availability to iv higher photosynthesis. However, increases in leaf-level photosynthesis have not often been observed in long-term N addition experiments. We tested the effects of 14 years of N additions on photosynthesis in two ways: by making instantaneous measurement from both canopy towers and excised branches, and by analyzing leaf tissue for δ 13C and δ 18O, isotopes integrate changes in photosynthesis through time. Trees receiving N additions had higher foliar N concentrations, but there were no differences in instantaneous measurements of photosynthesis from canopy towers or excised branches. Further, there were no significant changes in δ 13C and δ 18O in either current foliage or leaf litter collected annually throughout the N addition experiment (1994-2007). Together, these data suggest that increases in photosynthesis are not responsible for the higher rates of aboveground growth. Together, these experiments show that changes in atmospheric composition expected to occur in the next century will alter the functioning of forest ecosystems in the northcentral United States. However, predictions from short-term experiments did not often match the results observed in these long-term projects. Alternately, the recovery of forests in the north-central United States from acid deposition suggests that forest biogeochemistry can respond positively if pollution reductions are prioritized by policy makers.