Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering


Charles Driscoll


Carbon nitrogen and water dynamics;Climate change;PnET-CN-daily model;Urbanization

Subject Categories

Civil and Environmental Engineering | Engineering | Environmental Engineering


Human activities have significantly influenced the structure and function of forest ecosystems in the northeastern US. Urbanization directly reduces forest area and alters forest functions. Forest growth, health, and function can also be affected by climate change, atmospheric carbon dioxide (CO2) concentrations, and air pollution including ground-level ozone (O3) and atmospheric nitrogen (N) deposition. Critical questions for research and environmental management are: How will urbanization, climate change, and air pollution interact to influence the function of forest ecosystems, and how will changes in these environmental drivers alter the function of forests in the future? In this dissertation, I simulate the historical and possible future interactive effects of urbanization, climate change, air pollution, and CO2 fertilization on the carbon, nitrogen, and water dynamics in northeastern forest ecosystems. The dissertation is divided into four phases. In Phase 1, I calibrated and tested the forest biogeochemical model PnET-CN-daily using data from an intensive study site at Harvard Forest in Petersham MA. Following model testing I simulated historical patterns of carbon, nitrogen, and water cycling in the forest ecosystem and evaluated the contribution of individual and collective disturbance factors (past land disturbance, climate, CO2 and ozone concentrations, atmospheric N deposition) to these dynamics. In Phase 2, the calibrated PnET-CN-daily model is employed to forecast changes in carbon, nitrogen, and water processing at Harvard Forest under different future scenarios. The development of these scenarios encompasses both climate and air chemistry drivers. Specifically, the climate and CO2 scenarios utilized in this study include simulations from a suite of global circulation models (GCMs (CCSM4, GFDL-ESM2G, and GFDL-ESM2M)) for future climate conditions. Projections of air chemistry data, including N deposition and O3 concentration, were obtained from the output of the Community Multiscale Air Quality (CMAQ) model, considering two energy usage policies: a business as usual (BAU) energy policy and an aggressive policy to decarbonize the power sector. In Phase 3, I use regional data sets of air temperature, atmospheric concentrations of CO2 and O3, and N deposition to establish empirical relationships between urbanization as quantified by impervious surface area (ISA) with these environmental drivers. Then I conducted a suite of hypothetical simulations using the parameterized PnET-CN-daily model for Harvard Forest applying the empirical relations with arbitrary values of ISA to evaluate the effects of increasing urbanization on the processing of carbon, nitrogen, and water in a forest ecosystem in New England and the relative contribution of different drivers of urbanization in these processes. I also tested these simulations against experimental observations from plots established along an urbanization gradient in Massachusetts. Finally, in Phase 4 I conducted regional scale simulations of future scenarios of land cover, climate, and air quality on the landscape processing of carbon, nitrogen, and water for New England using PnET-CN-daily over the period from 2020 to 2050. This analysis builds on future land cover scenarios that were developed by the New England Landscape Future (NELF) project by scientists at Harvard Forest. I considered three of these land cover scenarios, including Yankee Cosmopolitan, Growing Global, and Recent Trends scenarios. I also considered two climate scenarios (BAU and RCP 8.5), two air quality scenarios including a BAU, and an aggressive decarbonization scenario for these land cover conditions. I tested the results of these spatial simulations against regional measurements including aboveground biomass data from the USDA Forest Service Inventory and Analysis (FIA) program and soil organic carbon data from Gridded National Soil Survey Geographic (gNATSGO). Cross-site analysis of weather records, CO2 concentrations, O3 concentrations, and N deposition data reveals significant variations in the microclimate of urban areas, characterized by elevated air temperature, CO2 concentrations, N deposition, and lower O3 concentrations along an increasing urbanization gradient. Total and plant C storage are projected to continue to increase from the beginning of the 20th century to the end of the 21st century at the regional scale. Meanwhile, ecosystem N storage is projected to increase throughout the 20th century but is expected to decline in the 21st century due to the combined impacts of urbanization and climate change. Despite an overall increase in evapotranspiration (ET) due to increasing photosynthesis and temperatures since 2000, soil water stress will not be an important limitation because of the increase in precipitation and improved water use efficiency (WUE) due to elevated CO2 concentrations. The simulation suggests that urbanization is the dominant factor for C and N storage in New England, which directly reduces C and N storage in the urbanized area, especially in Connecticut, Massachusetts, and Rhode Island. Fortunately, C storage in New England is concentrated in Maine, which has a relatively low level of urbanization. As a result, total ecosystem C storage is projected to continue to increase across all land cover scenarios, mitigating the adverse effects of urbanization. Unlike carbon, the majority of N storage is associated with soil, and atmospheric deposition represents an important external source of N to ecosystems in New England. Therefore, while urbanization directly decreases ecosystem N storage, N deposition is insufficient to compensate for this loss throughout model simulations. CO2 fertilization is the dominant environmental factor for forest growth by enhancing photosynthesis and improving WUE. However, the CO2 fertilization effect generally diminishes over time due to nutrient limitations. Continuous N uptake by plants leads to N oligotrophication in the soil, which ultimately limits forest growth. Furthermore, enhanced forest growth increases the soil C: N ratio, decreasing N mobilization and availability. Elevated air temperatures are likely to increase soil decomposition, which can offset the N immobilization resulting from the increasing soil C: N ratio. However, this phenomenon is only observed under extreme climate change scenarios (RCP8.5) not under moderate climate change (RCP4.5) in the late 21st century. Atmospheric N deposition serves as an external N resource that mitigates N oligotrophication, however, predictions indicate that N deposition will decrease due to emissions regulations.


Open Access