Premium
Changes in vegetation types and Ellenberg indicator values after 65 years of fertilizer application in the Rengen Grassland Experiment, Germany
Author(s) -
Chytrý Milan,
Hejcman Michal,
Hennekens Stephan M.,
Schellberg Jürgen
Publication year - 2009
Publication title -
applied vegetation science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.096
H-Index - 64
eISSN - 1654-109X
pISSN - 1402-2001
DOI - 10.1111/j.1654-109x.2009.01011.x
Subject(s) - grassland , fertilizer , vegetation (pathology) , indicator value , vegetation classification , long term experiment , agronomy , nutrient , environmental science , ecology , biology , medicine , pathology
Question: How does semi‐natural grassland diversify after 65 years of differential application of Ca, N, P, and K fertilizers? Is fertilizer application adequately reflected by the Ellenberg indicator values (EIVs)? Location: Eifel Mountains, West Germany. Methods: The Rengen Grassland Experiment (RGE) was established in an oligotrophic grassland in 1941. Six fertilizer treatments (Ca, CaN, CaNP, CaNP‐KCl, CaNP‐K 2 SO 4 , and unfertilized control) were applied annually in five complete randomized blocks. Species composition of experimental plots was sampled in 2006 and compared with constancy tables representing grassland types in a phytosociological monograph of a wider area. Each plot was matched to the most similar community type using the Associa method. Mean EIVs were calculated for each treatment. Results: The control plots supported oligotrophic Nardus grassland of the Polygalo‐Nardetum association ( Violion caninae alliance). Vegetation in the Ca and CaN treatments mostly resembled montane meadow of Geranio‐Trisetetum ( Polygono‐Trisetion ). Transitional types between Poo‐Trisetetum and Arrhenatheretum (both from the Arrhenatherion alliance) developed in the CaNP treatment. In the CaNP‐KCl and CaNP‐K 2 SO 4 treatments, vegetation corresponded to the mesotrophic Arrhenatheretum meadow. Major discontinuity in species composition was found between control, Ca, and CaN treatments, and all treatments with P application. EIVs for both nutrients and soil reaction were considerably higher in P treatments than in Ca and CaN treatments. Surprisingly, the control plots had the lowest EIVs for continentality and moisture, although these factors had not been manipulated in the experiment. Conclusions: Long‐term fertilizer application can create different plant communities belonging to different phytosociological alliances and classes, even within a distance of a few meters. Due to their correlated nature, EIVs can erroneously indicate changes in factors that actually did not change, but co‐varied with factors that did change. In P‐limited ecosystems, EIVs for nutrients may indicate availability of P rather than N.