NON-POINT-SOURCE NUTRIENT LOSSES TO THE AQUATIC ENVIRONMENT IN DENMARK - IMPACT OF AGRICULTURE

Citation
B. Kronvang et al., NON-POINT-SOURCE NUTRIENT LOSSES TO THE AQUATIC ENVIRONMENT IN DENMARK - IMPACT OF AGRICULTURE, Marine and freshwater research, 46(1), 1995, pp. 167-177
Citations number
27
Categorie Soggetti
Oceanografhy,"Marine & Freshwater Biology",Limnology,Fisheries
ISSN journal
13231650
Volume
46
Issue
1
Year of publication
1995
Pages
167 - 177
Database
ISI
SICI code
1323-1650(1995)46:1<167:NNLTTA>2.0.ZU;2-Y
Abstract
As part of a coordinated programme for nationwide monitoring of the Da nish aquatic environment, initiated in 1988, nutrient fluxes are measu red in 270 rivers, and nutrient cycling is studied in six agricultural catchments. This enables assessment of the degree of non-point-source nutrient loading of Danish surface water and groundwater as well as o f the impact of agriculture practised under different climatic and phy siographic conditions and with different farming practices. Comparison of annual median losses of total nitrogen and total phosphorus in agr icultural catchments and undisturbed catchments revealed an average lo ss ratio of 14:1 for total nitrogen (23.4 and 1.7 kg N ha(-1), respect ively) and 4:1 for total phosphorus (0.29 and 0.070 kg P ha(-1), respe ctively). Assessment of nitrogen cycling in six small agricultural cat chments with either sandy soil or loamy soil revealed marked differenc es in annual net input to the soil (142 v. 77 kg N ha(-1)), annual lea ching from the root zone (154 v. 78 kg N ha(-1)), and annual riverine loss of nitrogen (13 v. 25 kg N ha(-1)). The differences in the rates of leaching and riverine loss in the two soil types reflect the differ ent potential for surface-water and groundwater pollution with nitroge n and also the fact that the denitrification potential is higher in bo th the soil and the riparian areas of sandy ecosystems. Modelling of d ata from 77 small representative river basins revealed significant emp irical relationships (P < 0.001) between annual loss of both nitrogen and phosphorus and various predictor variables (e.g. runoff, proportio n of agricultural land, soil type). Runoff and proportion of agricultu ral land explained, respectively, 41% and 26% of the variance in the n itrogen model and 45% and 8%, of the variance in the phosphorus model. Such loading models are valuable tools for systems analysis and manag ement at the catchment level, such as when assessing measures implemen ted to reduce non-point-source nutrient pollution.