ALBERT

Description: A new integrated hydrological and nitrogen model, called TNT2 (topography-based nitrogen transfer and transformation), has been developed to study nitrogen fluxes in small catchments. This model, process-based and spatially distributed in order to take spatial interactions into account, has been kept as simple as possible. Here, only the hydrological module is discussed. The two main hypotheses of the hydrological model are taken from the TOPMODEL concept (constant hydraulic gradient equal to slope and hydraulic conductivity decreasing exponentially with depth). The model is based on a daily water balance for each cell of a regular square grid and computes an explicit cell-to-cell routing. Transfer through the vadose zone is simulated using a conceptual, layer-based algorithm analogous to the Burns model, except that a drainage water reservoir has been added to simulate mobile/immobile water processes and variations of the water table within the soil. The crop growth and nitrogen transformations are simulated using the equations of a generic plant-soil model, STICS. As an example, a preliminary study of the effect of the catchment geomorphology on denitrification is presented. The study was performed on theoretical catchments with contrasted slope shapes and pathway patterns. Results show that the whole-catchment denitrification depends on catchment geomorphology, although not directly through the extent of saturated areas. It is concluded that TNT2 seems to be a powerful tool to explore catchment processes, both by application to actual cases and by exploration on simple scenarios. © 2002 John Wiley and Sons, Ltd.

Description: In catchments with impervious bedrock, the nitrate concentrations in streamwater often show marked seasonal and small inter-annual variations. The inter-annual trends are usually attributed to changes in nitrogen inputs, due to changes in land use or in nitrogen deposition whereas seasonal patterns are explained in terms of availability of soil nitrate for leaching and of seasonality of nitrogen biotransformations. The companion paper showed that inter-annual variations of nitrogen in streamwater are not directly related to the variations of land use. The aim of this study is to describe nitrate concentration variations in a set of very small adjacent catchments, and to discuss the origin of the inter-annual and seasonal trends. Data from four catchments at the Kerbernez site (South Western Brittany, France) were used in this study. Nitrate concentrations in streamwater were monitored for eight years (1992 to 1999) at the outlet of the catchments. They exhibit contrasting inter-annual and seasonal patterns. An extensive survey of agricultural practices during this period allowed assessment of the amount of nitrogen available for leaching. The discharges measured since 1997 show similar specific fluxes but very different seasonal dynamics between the catchments. A simple, lumped linear store model is proposed as an initial explanation of the differences in discharge and nitrate concentration patterns between the catchments. The base flow at the outlet of each catchment is considered as a mixture of water from two linear reservoirs with different time constants. Each reservoir comprises two water stores, one mobile contributing to discharge, the other, immobile, where nitrate moves only by diffusion. The storm flow, which accounts for less than 10% of the annual flux, is not considered here. Six parameters were adjusted for each catchment to fit the observed data: the proportion of deep losses of water, the proportion of the two reservoirs and the size and initial concentration of the two immobile stores. The model simulates the discharge and nitrate concentration dynamics well. It suggests that the groundwater store plays a very important role in the control of nitrate concentration in streamwater, and that the pattern of the seasonal variation of nitrate concentration may result from the long term evolution of nitrogen losses by leaching. Keywords: nitrate, diffuse pollution, groundwater, seasonal variations, agricultural catchment, simulation model

Description: A new version of the Integrated Nitrogen in Catchments model (INCA) was developed and tested using flow and streamwater nitrate concentration data collected from the River Kennet during 1998. INCA is a process-based model of the nitrogen cycle in the plant/soil and in-stream systems. The model simulates the nitrogen export from different land-use types within a river system, and the in-stream nitrate and ammonium concentrations at a daily time-step. The structure of the new version differs from the original, in that soil-water retention volumes have been added and the interface adapted to permit multiple crop and vegetation growth periods and fertiliser applications. The process equations are now written in terms of loads rather than concentrations allowing a more robust tracking of mass conservation when using numerical integration. The new version is able to reproduce the seasonal dynamics observed in the streamwater nitrogen concentration data, and the loads associated with plant/soil system nitrogen processes reported in the literature. As such, the model results suggest that the new structure is appropriate for the simulation of nitrogen in the River Kennet and an improvement on the original model. The utility of the INCA model is discussed in terms of improving scientific understanding and catchment management. Keywords: modelling, water quality, nitrogen, nitrate, River Kennet, River Thames

Description: The hydrological and biogeochemical monitoring of catchments has become a common approach for studying the effect of the evolution of agricultural practices on water resources. In numerous studies, the catchment is used as a "mega-lysimeter" to calculate annual input-output budgets. However, the literature reflects two opposite interpretations of the trends of nitrate concentration in streamwater. For some authors, essentially in applied studies, the mean residence time of leached nitrate in shallow groundwater systems is much less than one year and river loads reflect annual land use while for others, nitrate is essentially transport limited, independent of soil nitrate supply in the short term and annual variations reflect changes in climatic conditions. This study tests the effect of agricultural land-use changes on inter-annual nitrate trends on stream water of six small adjacent catchments from 0.10 to 0.57 km2 in area, on granite bedrock, at Kerbernez, in Western Brittany (France). Nitrate concentrations and loads in streamwater have been monitored for nine years (1992 to 2000) at the outlet of the catchments. An extensive survey of agricultural practices from 1993 to 1999 allowed assessment of the nitrogen available for leaching through nitrogen budgets. For such small catchments, year-to-year variations of nitrate leaching can be very important, even when considering the 'memory effect' of soil, while nitrate concentrations in streamwater appear relatively steady. No correlation was found between the calculated mean nitrate concentration of drainage water and the mean annual concentration in streams, which can even exhibit opposite trends in inter-annual variations. The climatic conditions do not affect the mean concentration in streamwater significantly. These results suggest that groundwater plays an important role in the control of streamwater nitrate concentration. Keywords: nitrate, diffuse pollution, agricultural catchment, nitrogen budget, leaching, Kerbernez catchments