Premium
Dynamics of nitrate concentration‐discharge patterns in an urban watershed
Author(s) -
Duncan Jonathan M.,
Welty Claire,
Kemper John T.,
Groffman Peter M.,
Band Lawrence E.
Publication year - 2017
Publication title -
water resources research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.863
H-Index - 217
eISSN - 1944-7973
pISSN - 0043-1397
DOI - 10.1002/2017wr020500
Subject(s) - environmental science , watershed , storm , streamflow , hydrology (agriculture) , atmospheric sciences , nitrate , scale (ratio) , meteorology , geography , drainage basin , ecology , geology , cartography , geotechnical engineering , machine learning , computer science , biology
Concentration‐discharge (c‐Q) relations have been used to infer watershed‐scale processes governing solute fluxes. Prior studies have documented inconsistent concentration‐discharge patterns at the storm‐event scale driven by changes in end‐member concentrations. Other studies have evaluated c‐Q data from all periods in a composite fashion to quantify chemostasis (relatively invariant changes in concentration over several orders of magnitude variation in streamflow). Here we examine 3 years of high‐frequency nitrate and discharge data (49,861 data points) to complement 14 years of weekly data (699 data points) for an urban stream in Baltimore, MD, U.S. to quantify c‐Q relationships. We show that these relationships are variable through time and depend on the temporal scale at which they are investigated. On a storm‐event scale, the sensor data exhibit a watershed‐specific dQ/Q threshold when storms switch from counter‐clockwise to clockwise c‐Q behavior. On a seasonal scale, we show the influence of hydrologic variability and in‐stream metabolism as controls on stream nitrate concentrations and fluxes. On a composite scale, we evaluate the c‐Q data for chemostasis using analysis of both c‐Q slopes and CV c /CV Q , as a function of time. The slopes of c‐Q data for both long‐term weekly and high‐frequency data sets are in close agreement on an annual basis and vary between dry and wet years; the CV c /CV Q analysis is less sensitive to hydroclimate variability. This work highlights the value of both long‐term and high‐frequency c‐Q data collection for calculating and analyzing solute fluxes.