Hydrological controls on nitrogen (ammonium versus nitrate) fluxes from river to coast in a subtropical region: Observation and modeling.

Increased anthropogenic nutrient input and losses has caused eutrophication problems in freshwater and coastal ecosystems worldwide. High-frequency observations and modeling of river fluxes in subtropical regions are required to understand nutrient cycling and predict water quality and ecological responses. In 2014, a normal hydrologic year, we carried out daily sampling of the North Jiulong River in southeast China, which drains an agricultural watershed and experiences the Asian monsoon climate. We focused on the distinct characteristics of two important inorganic nitrogen forms (ammonium and nitrate). Our results show contrasting hydrological controls on the seasonal timing and magnitude of ammonium and nitrate concentrations and loads, likely due to differing sources and transport pathways (surface runoff versus baseflow) to the river. Both nitrogen concentrations were enriched in the dry season and diluted in the wet season. Arrival of rains in the pre-wet period in March caused a "first flush" peak event with the highest concentrations of the year, during which ammonium peaked two weeks earlier than nitrate. In contrast, the majority of nitrogen transport occurred during the lower concentrations of the wet season, with seven storms inducing flood events that lasted 24% of the time, contributed 52% of the runoff, and exported 47% of the ammonium and 42% of the nitrate. We found that seasonally piecewise LOADEST models (for pre-wet, wet and post-wet periods) performed better (5-8% error) than a year-round model (12-24% error) in estimating monthly nitrogen loads. However, not all nitrogen dynamics are easily synthesized by this approach, and extreme floods might produce a greater deviation in estimating nitrogen loads. These findings represent important implications for coastal ecology and provide opportunity on improving observation and modeling.