Date of Award

6-1-2015

Document Type

Open Access Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Environmental Sciences/Environmental, Earth & Ocean Sciences

First Advisor

Robert F. Chen

Second Advisor

Ellen M. Douglas

Third Advisor

Crystal B. Schaaf

Abstract

The transport of dissolved organic carbon (DOC) and chromophoric dissolved organic matter (CDOM) from land to coastal environments strongly influences coastal ecosystems. The presence of first flush phenomena due to rainwater runoff traveling from land into waterways can greatly affect carbon fluxes to coastal areas. This research utilizes sensors, autosamplers, and standard watershed sampling in order to assess for the presence of first flush and its significance. A rainfall simulator was built in order to collect runoff on two land use types. Time series data suggest that first flush of dissolved organic carbon was present for all rainfall intensities simulated on an impervious surface. At this location, approximately 40% to 51% of DOC flux occurred within the first 20% of runoff. At the permeable sampling location, first flush was observed in surface runoff collected during 12.7 and 25.4 mm hr-1 simulated storms, with 31% and 26% of DOC flux occurring within the first 22% of runoff. Seven storm events at two locations in the Neponset River Watershed, Massachusetts, USA were monitored to study the impact of storm events on DOC export from an urban watershed. Real-time CDOM fluorescence sensor measurements were better able to capture the variability present in riverine DOC and CDOM concentrations due to runoff influxes. Using modeled flow data, estimates of total DOC export fluxes during storms were compared to estimated total annual export. Based on these calculations, the seven sampled storm events account for 7 to 10 percent of the calculated yearly flux during 4 to 5 percent of the year. Additional work is needed to collect consistent year round data using sensors at these locations. DOC was sampled throughout the Neponset River Watershed monthly for seven years. Increased concentrations were observed following storm events and snowmelt, and were an average of 28% greater than concentrations observed during dry periods. Based on daily sampling data in September 2011, monthly fluxes may be underestimated by 38% or overestimated by 35%. More frequent sampling allows for better certainty in estimations of monthly and yearly fluxes from the watershed, but must be balanced with logistical and cost constraints.

Download

Share

COinS