The Role of Soil Organic Matter and Manures in Sustainable Nutrient Cycling - Sustainable Management of Soil Organic Matter - page 233

Fig. 4.2.1. Changes in mineral nitrogen in soil (0-80 cm) during 2 years at Kerlavic (treatment
0N): measured values (soil cores) are compared with simulated data (LIXIM).
Inorganic N in soils showed clear differences between the two sites: the
maximum soil mineral N (observed in late summer) was 300-380 kg N
ha −1 at Kerlavic and 120-150 kg N ha −1 at Kerbernez. Half of the mineral-
ized N occurred within 3 months after glyphosate application, and would
have been available for a growing crop in late May. The LIXIM model
was able to reproduce the evolution of mineral N in the whole soil profile
(Fig. 4.2.1). The kinetics of water drainage simulated by the model were
also in good agreement with the measured kinetics, on both sites (data
not shown), which suggested that the model was able to predict N
mineralization and leaching well.
The leaching losses, calculated and measured from the time of grass
destruction to the end of the second winter, are reported in Fig. 4.2.2 for
pure grass treatments. For a similar annual drainage (700-800 mm), the
amount of leached N during the first winter was about twice that during the
second winter. N leached at the Kerlavic site during the first year
(360-420 kg N ha −1 ) was much higher for all treatments than N leached at
the Kerbernez site (170-180 kg N ha −1 ). At the first site, a slight significant
difference was observed only between the non-fertilized treatment (KL0N)
and the three others. No statistical test can be easily used to test model
results (Mary et al ., 1999).
The LIXIM model gave slightly higher estimates of leaching losses than
direct measurements did during the first year (Table 4.2.1) but total
leached N for the 2 years was estimated accurately by the model for most
treatments (Fig. 4.2.3).