TIME-SERIES ANALYSIS OF CHANGES IN THE SOIL SOLUTION - EVIDENCE FOR APPROACH TO NITROGEN SATURATION IN DUTCH FOREST SOILS

Citation
A. Stein et N. Vanbreemen, TIME-SERIES ANALYSIS OF CHANGES IN THE SOIL SOLUTION - EVIDENCE FOR APPROACH TO NITROGEN SATURATION IN DUTCH FOREST SOILS, Agriculture, ecosystems & environment, 47(2), 1993, pp. 147-158
Citations number
21
Categorie Soggetti
Agriculture,"Environmental Sciences
ISSN journal
01678809
Volume
47
Issue
2
Year of publication
1993
Pages
147 - 158
Database
ISI
SICI code
0167-8809(1993)47:2<147:TAOCIT>2.0.ZU;2-G
Abstract
In the Netherlands, high atmospheric inputs of ammonium N followed by microbial oxidation of ammonium N to nitric acid and low N uptake from poor tree growth, has resulted in high levels of dissolved nitrate an d strong soil acidification in forest soils. So-called nitrogen satura ted soils where the supply of atmospheric N plus mineralised soil orga nic N exceeds the capacity of the ecosystem (vegetation plus soil biot a) to assimilate N have been observed occasionally. At high levels of N availability, N saturation leads to enhanced soil acidification and groundwater pollution. To test if nitrate saturation is being approach ed elsewhere, temporal trends in concentrations of dissolved nitrate a nd some other elements in a number of acidic forest soils were analyse d statistically. A procedure was developed to account simultaneously f or short-term, mainly seasonal variability plus any longer-term trend on a time scale of 4-7 years. Although results should be interpreted w ith care, because of the limited period of data collection, strong evi dence for a long-term increase in nitrate concentrations has been foun d in all plots considered. These trends could not be accounted for by trends in atmospheric deposition or soil hydrology in general, suggest ing the approach to nitrogen saturation. The procedures used in this s tudy to analyse the monitored data focus on the estimation of trends a nd calculation of its significance. They can be applied to similar stu dies dealing with temporally varying data, sampled at irregular interv als and exhibiting missing values.