Water nutrient concentrations in channels in relation to occurrence of aquatic plants: a case study in eastern Croatia |
| |
Authors: | Aleksandra Kočić Tomislav Hengl Janja Horvatić |
| |
Affiliation: | (1) Department of Biology, J. J. Strossmayer University, Trg Lj. Gaja 6, 31000 Osijek, Croatia;(2) Institute for Biodiversity and Ecosystem Dynamics, Faculty of Science, University of Amsterdam, Nieuwe Achtergracht 166, 1018 Amsterdam, WV, Netherlands |
| |
Abstract: | In this paper we analyzed nutrient concentration in the channels of eastern Croatia and investigated whether certain plant species and associations can be used as direct estimators of water quality. One hundred and twenty-two channel sites were visited and water samples taken for laboratory analysis (pH and concentrations of sulfate, chloride, ammonium, nitrate, and total phosphorus). At each site, macrophyte vegetation was recognized and its occurrence recorded. Three groups of analyses were performed: (a) principal component analysis to describe habitat characteristics of the investigated channels, (b) stepwise regression analysis to build estimation models for nutrient concentrations, and (c) geostatistical analysis including fitting of variograms and interpolation of values over the whole area of interest. High values of water nutrients in the eastern Croatian channels were reported (90% intervals): 5.3–29.4 for nitrates, 27.8–54.2 for sulfates, 0.1–0.4 for total P, and 0.18–0.34 mg l−1 for ammonium. Water nutrient concentrations can be successfully mapped over the channel network in eastern Croatia using geostatistics (regression kriging). The nutrient concentration variables required log transformation prior to regression or variogram analysis because their distributions were distinctly skewed towards lower values. Species were found to be a more successful estimator of nutrient concentrations than plant associations. In all cases, species had a higher adjusted R-square value, ranging from 0.302 (ammonium) to 0.485 (sulfates). Additional load of nutrients in water could lead to the disappearance of the more-sensitive species Lemna trisulca, Riccia fluitans, and Ricciocarpus natans and the spread of Potamogeton pectinatus, Glyceria maxima, and Glyceria fluitans. Further studies are needed to develop strategies for incorporating permanent monitoring networks to observe environmental changes and succession of vegetation. |
| |
Keywords: | Sulfates Nitrates Ammonium Stepwise regression Kriging Croatia |
本文献已被 SpringerLink 等数据库收录! |
|