ECOSYSTEM DISTURBANCE IN A WARMER AND WETTER NORTHEASTERN US: HOW WILL SOIL C AND N LOSSES AND MICROBIAL COMMUNITIES RESPOND TO A FOREST HARVEST IN THE FUTURE?

Open Access
- Author:
- McDaniel, Marshall Douglas
- Graduate Program:
- Soil Science
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- April 25, 2011
- Committee Members:
- Jason Philip Kaye, Dissertation Advisor/Co-Advisor
Jason Philip Kaye, Committee Chair/Co-Chair
Maryann Victoria Bruns, Committee Member
Margot Wilkinson Kaye, Committee Member
Roger Tai Koide, Committee Member
Hangsheng Lin, Committee Member - Keywords:
- soil microbial communities
gas flux
nitrogen
carbon
forest
soil
clear cutting
climate change
forest harvest - Abstract:
- Following whole-tree forest harvest, soils are susceptible to carbon (C) and nitrogen (N) losses owing to increased mineralization and decreased inputs from vegetation. While soil responses to harvesting have been extensively studied over several decades, a major gap in our knowledge is predicting how forest harvesting will interact with a changing climate. Climate models predict increases in temperature, as well as changes in precipitation in the northeastern United States. Increases in soil temperature and moisture alone, separate from harvesting effects, have been shown to increase C and N mineralization. Therefore, predicted regional climate change may exacerbate post-harvest C and N losses in some areas such as northeastern forests. The major goal of this dissertation was to test whether or not simulated climate change may indeed increase soil C and N fluxes and losses from mineral soil and examine the mechanisms driving this. A post-harvest climate manipulation experiment was established in the ridge and valley region of central Pennsylvania. The experimental design was a 2 x 2 factorial which included four treatments: warmed, wetted, warmed+wetted, and ambient. Warmed treatments had a targeted increase in surface temperatures of ~2 °C. Wetted plots received a target of +20% of long-term average precipitation. Soil biogeochemical responses to the climate manipulations were measured for 2.5 years as the forest regenerates from a whole-tree harvest. My overarching objectives were to 1) quantify changes in soil C and N pools and fluxes due to increased temperature and precipitation treatments in a post-harvest forest and 2) understand how changes in C and N cycling are linked to changes in soil microbial community composition and enzyme activity. Relative to controls, warmed treatments increased mean daily temperature of surfaces and soil (5 cm depth) by 1.8 and 2.5 °C, respectively, over the duration of the experiment. Warming also decreased the number of soil freeze-thaw events, and increased growing degree days, frost-free days, and the amount of time leaf surface temperatures were in optimal photosynthetic range. Wetted treatments increased mean monthly precipitation by 23%, but did not change the amount of time the soil water potential at 5 cm depth was below the permanent wilting point (-1.5 MPa). There was a significant interaction of warming and wetting on surface and soil temperature, emphasizing the importance of multivariate climate change experiments. Soil C and N fluxes, labile pools, and treatment effects on those pools and fluxes were all greatest the first year immediately after harvest compared to subsequent years, which is consistent with results from prior harvesting studies. The warmed and warmed+wetted treatments significantly increased NH4-N (p = 0.0275) and NO3-N (p = 0.0016) accumulating on ion exchange membranes in soil during the first year compared to the non-warmed treatments. The warmed treatment increased soil-atmosphere CO2 efflux (p = 0.026) over 2.5 years relative to ambient. The wetted treatment significantly increased cumulative soil N2O efflux over 2.5 years (p = 0.0270). Soil temperature and moisture were strong predictors of soil respiration (R2 = 0.75, p = 0.0001), but treatments significantly affected this relationship. The two-factor treatment (warmed+wetted) had a negative interaction decreasing CO2 effluxes compared to ambient during the growing season, but showing no overall effect on cumulative soil CO¬2 efflux. Despite significant treatment differences in C and N fluxes, there were no treatment effects on labile or bulk soil C and N contents. Total N in all treatments declined on average 34% from immediately after harvest to the end of the experiment. Post-harvest soils under climate change showed significantly different trace gas emissions and N availability, meaning disturbances under future climates may have a greater positive feedback with climate change (through increased CO2 and N2O fluxes) and release more N to eutrophication-sensitive bodies of water. I used ecologically meaningful extracellular enzyme (EE) ratios to explain treatment effects observed in soil C and N fluxes and, in an exploratory study, determine what soil C and N metrics may best be predicted by soil EEs and their ratios. Post-harvest soils were measured for β-1,4-glucosidase (BG), cellobiohydrolase (CBH), leucine aminopeptidase (LA), N-acetyl glucosaminidase (NAG), peroxidase (PER), and polyphenol oxidase (PPO) activities over 2.5 years. Soil hydrolase enzymes (BG, CBH, LA, and NAG) all showed greater mean activities than those reported from other studies including study sites with similar climatic and edaphic characteristics. This suggested high overall soil microbial activity after forest harvest which is supported by the high soil C and N fluxes that year. Only BG and NAG showed significant treatment effects from ambient with a two-factor interaction with BG (p = 0.029) and warmed effect on NAG (p = 0.007). Warming decreased BG relative to ambient and NAG relative to wetted plots. There was substantial variability in the season and time since harvest, but water-extractable organic C (WEOC), C:N, and total N best explained the variability of EEAs over the experiment (p’s < 0.05). Two exoenzyme ratios, BG:(LA+NAG) and BG:PPO, did not vary among treatments. However, the ratio of nitrogen acquiring enzymes (LA+NAG) to that of the other enzyme activity did show a significant warmed by wetted effect and increased over the time of the experiment indicating possible N limitation in N-acquiring enzymes over time and greater within warmed only treatments. Extracellular enzyme activity in a post-harvest forest soil was affected more by increased temperature but not by increased precipitation, and surprisingly the warming treatment had the opposite effect on soil C and N mineralization (i.e. warming increased the soil C and N fluxes, but decreased the BG and NAG). Soil microbial biomass, fungal to bacterial biomass (F:B), community-level physiological profiles (CLPP), and a reciprocal transplant incubation (RTI) experiment were used on soils after receiving the simulated climate treatments for 2.5 years. This suite of experiments was used to determine if there were changes in the soil microbial community due to the climate manipulations. There was no noticeable treatment effects on soil microbial biomass or F:B. All three climate change treatments altered CLPPs as measured with 15 substrates, with 2.5 years of warming having the strongest effect on substrate usage. Warmed and warmed+wetted soils responded more to low C:N substrates. The RTI experiment showed that there was a significant soil x inoculum effect (p = 0.0005), but that inoculum was the only significant simple effect (p = 0.001). This suggests that the soil microbial communities were not limited by the substrates available in their soil, but that they were structurally or physiologically different from one another. Furthermore, it was shown that the two-factor treatment had a significant home-field disadvantage effect, reflecting a soil microbial community that is maladapted to utilizing the substrates in their own soil. This I hypothesized was due to combined increase in stress events from wetting-drying events and the infrared lamps decreasing soil moisture. To summarize, the climate treatments showed significantly different fluxes in soil C and N but no observable treatment effects on soil pools. This could be due to either inputs balancing losses or a lack of precision and accuracy in the methods used in this study to measure soil C and N pools. Soil microbial physiology, community structure, or both were altered due to the simulated climate scenarios. Warming had a greater effect on the soil microbial EEA and CLPPs and indicated possible increase in simple N substrate limitation. I show evidence that whole-tree forest harvest in a future warmer and wetter northeastern US could cause even greater positive feedback to climate change relative to forest harvested in recent decades. Furthermore, it is possible that harvesting could increase N leaching rates to aquatic ecosystems. However, an interesting finding which contradicts many climate manipulation experiments in mature ecosystems was that when post-harvest soils were warmed and wetted there was a negative interaction on soil C and N fluxes, and that this was brought about mostly through a shift in microbial community physiology or composition. This research provides evidence that soil C and N biogeochemistry and the microbial communities that drive the cycles in post-harvest soils will be altered under different climate change scenarios. This alteration could also lead to distinct trajectories in ecosystem succession due to C and N loss.