A number of models have been developed to simulate nitrogen cycling dynamics and movement at a watershed scale, such as SWAT, AnnAGNPS, MAGIC, and INCA. However, most of these models employ the simulated soil temperature by air temperature which is consistently higher than soil temperature. It might underestimate rates of temperature-sensitive nitrogen cycling processes in topsoil which has higher temperature than air during datetime according to field measurements. Meanwhile, slope of land use patches is taken into account for only few models although it is well-recognized to impact hydrological regime. This study considers nitrogen cycling in a watershed as a life cycle and employs life cycle model to develop a watershed inorganic nitrogen dynamic (WIND) model to simulate nitrogen cycling under various land uses in a watershed. The WIND model consists of energy-balanced surface soil temperature module and slope-introduced surface soil moisture module. Meanwhile, inorganic nitrogen cycling processes in surface soil under various land use types have been consecutively measured using a robot incubation system, i.e. mineralization, nitrification, and denitrification. A land use patch based simulated inorganic nitrogen input/output pathways are assembled into a watershed considering slope and its exposure.
Results/Conclusions
The WIND model has been testified in a small watershed in subtropical China with sum of determination coefficient (R2) and Nash-Sutcliffe efficiency (NS) above 0.7, which is calibrated and validated with surface soil sampling events in 2013 and 2016-17, respectively. The results evidently illustrate differences in inorganic nitrogen dynamics in surface soil among land use types as well as among inorganic nitrogen cycling processes. The spatial mapping of surface soil inorganic nitrogen in the watershed using the land use patch based scaling up approach matches up well with the two field surveys and their Krigging interpolations. Surface soil moisture and then inorganic nitrogen cycling processes are not sensitive to slope of land use patch, which only become significant above 15o steepness (p < 0.05). But the increased surface soil temperature by the energy-balanced module shifts inorganic nitrogen cycling processes significantly but the extents are dependent on land use types. In short, the WIND model might be an alternative tool for watershed-scale nitrogen management.