INFLUENCE OF CATCHMENT CHARACTERISTICS, FORESTRY ACTIVITIES AND DEPOSITION ON NITROGEN EXPORT FROM SMALL FORESTED CATCHMENTS

Citation
A. Lepisto et al., INFLUENCE OF CATCHMENT CHARACTERISTICS, FORESTRY ACTIVITIES AND DEPOSITION ON NITROGEN EXPORT FROM SMALL FORESTED CATCHMENTS, Water, air and soil pollution, 84(1-2), 1995, pp. 81-102
Citations number
51
Categorie Soggetti
Environmental Sciences","Water Resources
ISSN journal
00496979
Volume
84
Issue
1-2
Year of publication
1995
Pages
81 - 102
Database
ISI
SICI code
0049-6979(1995)84:1-2<81:IOCCFA>2.0.ZU;2-S
Abstract
The ability to predict nitrogen export from forested catchments is ess ential in order to evaluate the effects of anthropogenic activities on the trophic status of lakes and sea areas, and to extrapolate the res ults to catchments from which no measurements are available. Data from 20 forested catchments (0.3-42 km(2)) in Finland and Sweden during th e 10-year period 1979-88 were used to develop empirical multivariate r egression models of average NO3-N, NH4-N and organic N export fluxes a s a function of geomorphological, meteorological, hydrological, deposi tion and forest management variables. A combination of high inorganic N deposition or air temperature and a low extent of organic soils was related to (R(2) = 0.64) high losses of NO3-N. A strong correlation be tween N deposition and air temperature makes it difficult to distingui sh the effects of one variable from the other. Retention of deposited nitrogen is still high in most of the catchments. High losses of NH4-N had the strongest correlation with forestry activities and stream den sity. A combination of drainage percentage and temperature was related to (R(2) = 0.53) losses of NH4-N. The most important factors explaini ng spatial variability of organic N losses were clearly forestry activ ities. A combination of high percentage of drainage and clear-cutting was related to (R(2) = 0.81) high organic N losses. However, within th e catchments, large-scale forest management practices were needed befo re any clear effect on spatial variability was detected. All the equat ions obtained were influenced by the choice of a limited number of cat chments. No causal relationships between losses and the explaining var iables can be inferred from this type of study.