Spectral fluorescent signature (SFS) is a rapid, reagent free and inexpensive technique, which has great potential for environmental monitoring of aqueous systems, especially for predicting dissolved organic carbon (DOC) along natural waters. This technical note aimed to examine the possibility to use SFS associated with partial least squares regression (PLS) to assess the organic loading in natural water. A model was built using samples of water collected between October 1999 and February 2002 on the Passaic River at Little Falls, NJ, USA. A correlation was established between measured DOC, SFS, and the corresponding daily registered flow from United States Geological Survey (USGS) New Jersey's streamflow database. The methodology presented herein looks promising in making use of the significant organic characteristics information contained in a SFS for application and use in spatial and temporal water quality management and treatment. (C) 2004 Elsevier B.V. All rights reserved.