Routine Estimation of Dissolved Organic Matter Sources Using Fluorescence Data and Linear Least Squares

Abstract

Dissolved organic matter (DOM) is an important component of the biogeochemistry and ecosystem function of streams and rivers. Unlike inorganic nitrogen nutrients, organic nitrogen (DON) nutrients can vary considerably depending on the contributions of various sources. The properties of DOM in natural waters have been measured using fluorescence spectroscopy, which is inexpensive compared to more sophisticated chemical analysis and is fast enough to enable its use in routine water quality monitoring. However, estimating the contributions of DOM sources to surface waters from fluorescence data in a similarly efficient manner remains an area of active research. In this study, we develop a simple linear regression method for estimating the relative amounts of DOM sources in a stream or river sample. The regression estimates are more accurate than those of previous approaches and are orders of magnitude faster to compute. Codes and data to implement the method are provided for a variety of computing environments to facilitate use for water quality and research applications.