Soil Nitrogen Dynamics
227
(around 10 mg N kg-I) were found in deeper layers (Fig. 16.2). Ammonium
was always the dominant form of mineral N at both sites, accounting on average for 87-97% of mineral N. Mean nitrate concentrations were very low at
both sites, but they were higher in Prades than in Montseny, especially in the
upper layer (Fig. 16.2). This probably reflects the lower leaching under the
drier Prades climate. The amount of mineral N averaged 3.2 ± 0.2 kg N ha- I
at 0-5 cm and 2.3 ± 0.2 at 5-10 cm depth at Prades. At Montseny it was much
higher, especially in the lower mineral soil, with 4.4 ± 1.0 kg N ha- I at 0-5 cm
and 11.9 ± 3.5 at 5-20 cm depth. Casals et al. (1995) found similar amounts of
mineral N in the surface (0-10 cm) soil of a Scots pine forest (3-8 kg N ha- I ).
In spite of the high temporal variability in mineral N concentrations, seasonal trends were similar for all soil horizons in each site, but differed
slightly between sites (Fig. 16.2). The highest content of mineral N occurred
in spring and autumn at Prades, and in summer and early autumn at Montseny. Nitrate content reached a maximum in spring and autumn (and occasionally in summer) at Prades, and in summer at Montseny. Both ammonium
and nitrate were low in winter at both sites (Fig. 16.2). This site-specific seasonal variability could be due to the different climatic conditions, the drier
site (Prades) being more strongly influenced by the summer drought than
the wetter site (Montseny). These temporal patterns in soil mineral N content
reflect in part the changing rates of microbial N transformations, which are
the subject of the next section.
16.4 Nitrogen Mineralization
Net N mineralization and nitrification rates determined by the in situ plastic
bag incubation method (Pastor et al. 1984) were highly variable within and
between incubation periods, especially in the H horizon and the 0-5 cm mineral soil (Fig. 16.3). The forest floor is heterogeneous in both amount and
composition (Joffre et al. 1996; Fons et al. 1997) and together with the upper
mineral soil is strongly influenced by changes in temperature and moisture,
which often results in short pulses of mineralization and/or nitrification that
are heterogeneously distributed in space and time. In general, mineralization
rates per unit weight of soil were higher in the H horizon and 0-5 cm soil
than in 5-20 cm soil in both sites, although the H horizon in Prades also
showed occasionally the highest N immobilization (Le. negative net N mineralization; Fig. 16.3).
At both sites N mineralization and nitrification rates had a seasonal pattern, especially in the H horizon and the upper mineral soil. In these nearsurface layers, the highest rates occurred when soil temperature and moisture were high. In the lower mineral soil, N mineralization and nitrification
rates at Prades were less variable throughout the year than in the upper layers (Fig. 16.3), whereas at Montseny high rates of net N mineralization in July
227
(around 10 mg N kg-I) were found in deeper layers (Fig. 16.2). Ammonium
was always the dominant form of mineral N at both sites, accounting on average for 87-97% of mineral N. Mean nitrate concentrations were very low at
both sites, but they were higher in Prades than in Montseny, especially in the
upper layer (Fig. 16.2). This probably reflects the lower leaching under the
drier Prades climate. The amount of mineral N averaged 3.2 ± 0.2 kg N ha- I
at 0-5 cm and 2.3 ± 0.2 at 5-10 cm depth at Prades. At Montseny it was much
higher, especially in the lower mineral soil, with 4.4 ± 1.0 kg N ha- I at 0-5 cm
and 11.9 ± 3.5 at 5-20 cm depth. Casals et al. (1995) found similar amounts of
mineral N in the surface (0-10 cm) soil of a Scots pine forest (3-8 kg N ha- I ).
In spite of the high temporal variability in mineral N concentrations, seasonal trends were similar for all soil horizons in each site, but differed
slightly between sites (Fig. 16.2). The highest content of mineral N occurred
in spring and autumn at Prades, and in summer and early autumn at Montseny. Nitrate content reached a maximum in spring and autumn (and occasionally in summer) at Prades, and in summer at Montseny. Both ammonium
and nitrate were low in winter at both sites (Fig. 16.2). This site-specific seasonal variability could be due to the different climatic conditions, the drier
site (Prades) being more strongly influenced by the summer drought than
the wetter site (Montseny). These temporal patterns in soil mineral N content
reflect in part the changing rates of microbial N transformations, which are
the subject of the next section.
16.4 Nitrogen Mineralization
Net N mineralization and nitrification rates determined by the in situ plastic
bag incubation method (Pastor et al. 1984) were highly variable within and
between incubation periods, especially in the H horizon and the 0-5 cm mineral soil (Fig. 16.3). The forest floor is heterogeneous in both amount and
composition (Joffre et al. 1996; Fons et al. 1997) and together with the upper
mineral soil is strongly influenced by changes in temperature and moisture,
which often results in short pulses of mineralization and/or nitrification that
are heterogeneously distributed in space and time. In general, mineralization
rates per unit weight of soil were higher in the H horizon and 0-5 cm soil
than in 5-20 cm soil in both sites, although the H horizon in Prades also
showed occasionally the highest N immobilization (Le. negative net N mineralization; Fig. 16.3).
At both sites N mineralization and nitrification rates had a seasonal pattern, especially in the H horizon and the upper mineral soil. In these nearsurface layers, the highest rates occurred when soil temperature and moisture were high. In the lower mineral soil, N mineralization and nitrification
rates at Prades were less variable throughout the year than in the upper layers (Fig. 16.3), whereas at Montseny high rates of net N mineralization in July