We aimed to develop a simple method to predict regional variation in nitrogen loss potentials from forested area in the forested Yahagi River watershed (1,340 km
2) in central Japan.
Firstly, we estimated within the watershed variation in nitrogen loss potentials through the following three processes: (1) We investigated the net nitrification rate of surface mineral soils using aerobic laboratory incubation at 30 °C for 4 weeks (mg N⋅kg soil
-1⋅28 d
-1); (2) we developed a multiple regression model to predict the net nitrification potential on a per-area basis for surface (0-5 cm) mineral soils (Mg N⋅km
-2) from only geographical factors; (3) we predicted the regional variation in the net nitrification potential using available GIS data as explanatory variables.
Secondly, we estimated annual outflow using a combination of a GIS model and a simplified evapotranspiration model. We examined accuracy of the model for estimation of annual outflow by comparing values predicted by the model with values measured at two outflow survey points.
Finally, we estimated nitrate concentration potentials in stream water defined as the net nitrification potential divided by annual discharge (mg N⋅L
-1) as an index of regional river water quality. We verified that a series of GIS prediction technique could produce the regional trend in measured nitrate concentrations.
View full abstract