Spectro-fluorometric characterization of dissolved organic matter (DOM) in a Mid-Atlantic watershed

Date
2013
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Spatial and seasonal pattern of dissolved organic matter (DOM) was characterized using a unique combination of hydrologic flow path analyses and spectro-fluorometric tools. The study was conducted over a four year (2008-11) period in a headwater forested watershed located in the Piedmont region of the mid-Atlantic USA. Sampling was conducted for stream water and variety of watershed sources which include throughfall, litter, soil waters, riparian, shallow, and deep ground waters during base flow and storm flow conditions. Ultraviolet and fluorescence measurements were performed to generate excitation-emission matrices (EEMs). A site specific parallel factor analyses (PARAFAC) EEMs model for DOM was developed and compared against a generic model. Discriminant analyses indicated that the site-specific model was more sensitive to subtle differences in DOM and was able to provide a greater level of differentiation in DOM among the watershed sources. This allowed better insights into how DOM evolved as it moved through the watershed. Spatial analyses indicated that DOM derived from catchment with increased wetland coverage was more humic and aromatic. No influence of drainage area was present in base flow DOM response. Attenuated DOM response during storm flow conditions with drainage area was observed. Water flow paths appeared to be key drivers in shaping DOM response in storm conditions. Seasonal patterns of DOM were evaluated for stream water during base flow and storm flow conditions and for multiple watershed sources. Overall, seasonal pattern in stream water DOM was more pronounced for storm flow versus base flow. Similarly among watershed sources, seasonality in DOM sources was more apparent in surficial versus groundwater DOM sources. This pattern confirmed our primary hypothesis that sorption of DOM (to mineral soil surfaces) along deeper hydrologic flow paths damped the seasonal DOM signal. Seasonal “hot” moments such as autumn leaf fall appeared to have a strong influence on seasonal DOM patterns that was apparent in storm runoff occurring immediately following autumn leaf inputs. In conclusion, this study clearly demonstrated the value and benefits of using a suite of optical indices to characterize DOM.
Description
Keywords
Citation