Date of Award

May 2016

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Civil and Environmental Engineering


Charles T. Driscoll


Adirondacks, Biogeochemical model, Critical Loads of acidity, Great Smoky Mountains National Park, Naturally occurring organic acids, Total Maximum Daily Loads

Subject Categories



Atmospheric deposition of sulfuric and nitric acids has impacted surface waters in the eastern U.S. Controls on anthropogenic S and NOx emissions have resulted in a substantial decrease in acid deposition. Long-term measurements of surface water chemistry indicate contrasting responses of surface waters to this deposition reduction. Streams in Great Smoky Mountain National Park (GRSM), in Tennessee and North Carolina, generally show a delayed response in contrast to lakes in the Adirondacks in New York State which are responding relatively rapidly to decreases in acid deposition. In this dissertation, I used a biogeochemical model, PnET-BGC, as a tool to investigate the influence of acid deposition and watershed characteristics on the extent and rate of surface water response to changes in atmospheric deposition. I applied PnET-BGC model to surface waters in two regions of the eastern US: the Adirondacks and the GRSM. These regions are both highly impacted from acid deposition, but have different watershed characteristics, which results in contrasting responses of surface waters. The focus of my analysis was on surface waters that have been identified by their respective states as acid-impaired. In addition to these impaired waters, I applied the model to other sites where data were available through intensive monitoring programs for the purpose of calibration, confirmation, or extrapolation. In the Adirondacks, 128 lakes have been identified as impaired and 12 streams in the GRSM. Simulation results indicate that surface waters in the Adirondacks are more responsive to controls on SO42- deposition than on NO3- and NH4+, while streams in the GRSM are more responsive to NH4+ rather than SO42- and NO3-.This contrasting response is most likely due to the higher SO42- adsorption capacity of the soil and the lower N retention of the ecosystems in the GRSM, compared to the Adirondacks. I used the critical loads (CLs) and Total Maximum Daily Loads (TMDLs) to project reductions in atmospheric deposition loads that would provide healthy conditions for the fisheries of surface waters in these regions. These concepts are commonly used as tools for communication between scientists and policy makers. In this study, acid neutralizing capacity (ANC) was used as an indicator of the health of surface waters.

I performed a sensitivity analysis of the model for applications to both regions; I found that model simulation of ANC is most sensitive to the model inputs of Ca2+ and Na+ weathering rates, precipitation, maximum air temperature, and SO42- wet deposition. In the Adirondacks, the model simulating ANC was also sensitive to the parameters which were used in an algorithm depicting the acid-base behavior of naturally occurring organic acids. Model sensitivity to organic acid parameters along with the findings from recent studies, which indicate that following control on acid deposition, the acidity of surface waters is shifting from inorganic to organics acids, encouraged me to improve the parameterization of the acid-base characteristics of the naturally occurring organic acids.

To accomplish this, I used long-term data from two intensive sites in the northeastern U.S.: the Adirondacks and the Hubbard Brook Experimental Forest in New Hampshire. Combining a chemical equilibrium model and an optimization algorithm, I generated a modeling framework to parameterize organic acids in order to provide the best predictions for pH, ANC and inorganic monomeric Al, given the observations of other major solutes. Model parameterization is proposed for application in biogeochemical models such as PnET-BGC. The parameterization of organic acids showed that about 5% of the dissolved organic carbon consists of associated reactive functional groups, with both strong and weak acid characteristics. The parameterization of the organic acids, made using two temporal intervals of surface water data (i.e., 1993-2001 and 2003-2012 time intervals), indicate that the charge density of organic acids increased over time. This pattern suggests that dissolved organic matter recently (i.e., 2003-2012) draining from soil has a different quality, with more acidic characteristics and a greater relative contribution to the acidity of surface waters than dissolved organic matter from an earlier period (i.e., 1993-2001).


Open Access

Included in

Engineering Commons