Primary tabs

Does prior pulse disturbance increase future sensitivity of primary production to climatic change?

Project Abstract: 
Compounding disturbances from insects and extreme weather are becoming more widespread in North American forests and the consequences for individual species as well as the temporal stability of net primary production are poorly understood. Before understanding how temporal stability of net primary production is affected by disturbance and extreme weather, the terms in the context of this study must be defined. Disturbance from insects is characterized by its severity, which is defined by the intensity and duration of the disturbance (Goetz et al., 2012). Net primary production will be studied and evaluated as above-ground wood net primary production taken from stem diameter interannual increments (ANPPw henceforth). It is also important to understand the effects of each individual factor on the net primary production of forests in order to comprehend the concept of temporal stability of forests. The effect of insect disturbance on productivity has been widely studied in Eastern and Western forests of North America. However, there are many variables that determine the increase, decrease or neutrality of productivity in a forest. Some of these are the severity and duration of disturbance, climate, vegetation and the carbon pools that are already available (Goetz et al., 2012). Essentially, a holistic view of the forest must be attained in order to predict the overall changes in carbon sequestration. At the University of Michigan Biological Station, a decadal study from the Ameriflux project has been conducted with regards to moderate defoliation of canopy trees, simulating insect invasion. The disturbance severity had an intensity of 40% canopy tree and its duration was short-term. The forest is early successional (mainly Populus and Betula), transitioning into mid-successional (mainly Quercus, Acer, and Pinus) and has been a consistent carbon pool over the past century (Gough et al., 2008). This body of literature indicates an overall positive trend for productivity, in that, this type of moderate defoliation by girdling increased productivity of the disturbed stand over a period of 13 years (Gough et al., 2013, Gough et al., 2016, Gough et al. 2021). Climate change has been found to increase productivity in Eastern North American forests, compared to Western North American forests. It is interesting to see where this phenomenon is mild compared to Western North American forests. One study utilizing forest inventory plots from across the United States, suggests such a finding (Hogan et al., 2024). The opposite effect on productivity has been found in Western North American forests, where climate change has been much harsher. Additionally, because of the intensity of climate change in the West, it is evident that the order of these two factors has been reversed, especially in the Western Mountains of Oregon (Goetz et al., 2012). Having defined the basic concepts, we arrive at the focus of this study: temporal stability of ANPPw, as opposed to ANPPw. Temporal stability in ecology is defined as the variability in an ecosystem process over time. It is calculated as the coefficient of variance (CV henceforth) of a single parameter, which is the ratio between standard deviation and mean. ANPPw has been found to have high temporal stability when disturbance is severe in two chronosequences (i.e. low CV), and low temporal stability in the reference late successional stands (i.e. high CV). (Wales et al., 2020). Objective 1: Evaluate whether a moderate severity disturbance killing 40% of all mature canopy trees increases sensitivity to anomalous annual mean photosynthetically active radiation (PAR) measured in umol/m^2*s^1; average annual air temperature measured in degrees Celsius; yearly precipitation in cm; soil temperature at 7.5 cm in degrees Celsius and soil moisture content (vol/vol) , thereby increasing the temporal stability of stand ANPPw (i.e. lower CV). Utilizing 17-year datasets from control and disturbed sites, I will analyze environmental data and biological data (as diameter increments in cm to convert into mass in kg to convert into ANPPw in kg/ha/yr) to evaluate stand-scale climate-disturbance interactions. Hypothesis 1: I hypothesize that a phloem-disrupting disturbance eliminating 40% of all canopy trees compounded with annual climate anomalies will stabilize stand-level temporal stability of ANPPw because each tree species complements the other (Gough et al. 2021). Objective 2: Evaluate whether a moderate severity disturbance killing 40% of all mature canopy trees increases sensitivity to anomalous annual mean photosynthetically active radiation (PAR) measured in umol/m^2*s^1; average annual air temperature measured in degrees Celsius; yearly precipitation in cm; soil temperature at 7.5 cm in degrees Celsius and soil moisture content (vol/vol) , thereby reducing the temporal stability of species ANPPw (i.e. higher CV). Utilizing 17-year datasets from control and disturbed sites, I will analyze environmental data and biological data (as diameter increments in cm to convert into mass in kg to convert into ANPPw in kg/ha/yr) to evaluate species-scale climate-disturbance interactions. Hypothesis 2: Disturbance will decrease species’ temporal stability of ANPPw in reference to treatment and control plots because of the nature of girdling. The early-successional species will suffer, thus having low temporal stability (i.e. high CV) and the mid-successional species will flourish and still have a low temporal stability (i.e. high CV) because they will be compensating for the dying trees’ lack of growth.
Investigators: 
Status of Research Project: 
Years Active: 
2024
Methods: 
Study site and design The study will be conducted at the University of Michigan Biological Station in northern Michigan in Emmet County (​​45°35′N, 84°43′W). The average environmental conditions in Emmet County are 5.5 degrees Celsius average annual temperature and mean annual precipitation of 817mm with 295mm of rainfall. The study site has a history of extensive logging of pre-existing Pinus and Tsugae species up until the early 1900s, which have been replaced by Populus and Betula expanses due to extensive cutting and burning of those conifers. These early-successional forests are now becoming mid-successional, varying in species composition: Quercus, Pinus (once again) and Acer. It will be based off of the Forest Accelerated Succession Experiment (FASET henceforth), initiated in 2008. FASET, as in its name, accelerated the succession of these early-successional forest through stem-girdling (simulating phloem-disrupting insects and diseases) making it the source for numerous ecological studies. The experimental design consists of paired plots between US with an average canopy defoliation of 40% among all plots (depending on tree density in the forest). Stem diameter measurements Stem diameter measurements will be achieved by the monitoring and reinstallation of dendrometer bands on FASET/Ameriflux canopy trees. This monitoring and reinstallation will occur two times in 2024, during the summer and in the fall in the paired stands and subsequent plots to assess interannual variability in growth. Utilizing aluminum tape, dendrometers with 1/10 of an inch increments will be constructed and fastened to trees at DBH using coil springs with a modulus of elasticity of 2 to 4 inches per pound (Liming, 1957). Incremental growth is measured by observing the lining up of the bottom sticker increments with the top sticker increments. Diameter increments measured in inches will be converted to cm. Diameter measurements will be converted to mass by site-specific allometric equations and wood C density data. ANPPw estimates per stand will be achieved by taking mass estimations and scaling them to the plot level and then stand level ANPPw, measured in kg/ha/yr. Same will be done for each individual species. Estimation of wood net primary production ANPPw will be extrapolated to stand-scale using species and site-specific allometric equations from the yearly cm increments (Gough et al. 2008). Species ANPPw will be calculated per species using site-specific allometric equations from the yearly cm increments. Time series of ANPPw will be calculated for each stand to include all years (2007-2024), where 2007 is a baseline year before FASET was initiated. Environmental data Ameriflux tower data will be downloaded for 2024 from the server to add to the 17-year dataset for FASET/Ameriflux. Type of climate data: Photosynthetically active radiation (PAR) measured in umol/m^2*s^1; Average annual air temperature measured in degrees Celsius; Yearly precipitation in cm; Soil temperature at 7.5 cm in degrees Celsius and soil moisture content (vol/vol) as measured in previous study by Wales et al., 2018. These variables provide light, water, air temperature and soil variability, a complete measurement of the abiotic components of the ecosystem. Time series of each variable as in the Appendix, will be recalculated for each stand to include all years (2007-2024), where 2007 is a baseline year before FASET was initiated. The yearly average of each variable will also be calculated. Data analysis The first part of the data analysis is to provide published and updated time series on environmental variables, stand ANPPw and species ANPPw for FASET/Ameriflux, as in Figure 1 in the appendix. The second part entails addressing the first and second objectives and their respective hypotheses. The formula utilized for the coefficient of variance of ANPPw will be Where the standard deviation per year will be divided by the mean per year. Yearly averages of each individual variable will be correlated with their respective ANPPw values for each paired plot to observe direct climatic effects. Linear regression will be used to ascertain statistical significance. Correlation and the corresponding linear regression between temporal stability and %BA girdled will also be conducted to observe the relationship between these other two variables. More statistical tests will be generated as better acquaintance is made with the data.