Nitrification in Dutch heathland soils

Abstract

A survey was conducted over a range of 17 Dutch heathland locations, subdivided into 41 sites dominated by either dwarf-shrubs (Calluna vulgaris or Erica tetralix) or grass species (Deschampsia flexuosa or Molinia caerulea). Among the habitats of the dominant plant species relatively little differences in general soil properties were observed. The P status of Deschampsia sites was relatively high as well as the NO3 −-N concentrations in the 0–10 cm layer (FH included) at the grass-dominated sites. At sites with a dead or degenerating dwarf-shrub vegetation, NH4 +-N concentrations reached very high levels. Net production of nitrate was observed during incubation of intact 0–10 cm soil cores (FH-layer included) in the laboratory for all sites, even though in some instances, particularly at Calluna and Erica sites, no nitrate was initially measured. Generally, a higher nitrification rate was found for the grass-dominated sites, and for Deschampsia in particular. The net production of nitrate was highly significantly correlated with net N mineralization, being a reasonable predictor of nitrification in a simple regression model (R2=0.47; P<0.001). Net nitrification was also significantly correlated with the NO3 −-N initially present at the start of the growing season (R=0.65; P<0.001) and with the labile organic P content of the soil (R=0.65; P<0.001). By including initial NO3 −-N and labile organic P, together with net N mineralization and pH, in a multiple regression model, net nitrate production could be predicted with a much higher precision (R2=0.75; P<0.001). Although apparent nitrification was not significantly correlated with pH, the latter contributed significantly to the multiple regression equation for the prediction of the former. The influence of the labile organic P pool may act via its positive correlation with microbial biomass, thus more or less reflecting the potential mineralization/nitrifying capacity of a particular site.