Species richness, biomass production and recent vegetation changes of Estonian floodplain grasslands

We studied vegetation diversity and standing biomass in relation to site moisture (moist, wet) and different management regimes (regularly mown (A), recently abandoned (B), un-mown for at least 15 years but with mowing re-introduced recently (C), and abandoned for at least 15 years without re-introduction of mowing (D)) on floodplain grasslands in Soomaa National Park, SW Estonia. A flexible quadrate size (area inhabited by 500 ramets) was used for the estimation of life-form distribution, species richness, ramet density and weight of standing biomass. The size of the actual species pool (S[pool]) was also estimated and relative richness (S[rel500]), mean plant unit area (PUA), plant unit biomass (PUB), ramet density per square metre and standing dry biomass per square metre were calculated. Additionally, changes in Estonian floodplain grassland vegetation between the 1960s and the 1990s were analysed by comparing the life-form distribution of frequent species on floodplain grasslands in the 1960s and 1990s. Life-form distribution depended both on moisture conditions and on the management regime. Tussock-forming graminoids dominated in regularly mown moist sites and long-term un-mown wet sites, while mat-forming graminoids dominated in sites with irregular management. The proportion of herb ramets was highest in unmanaged moist sites without regular mowing. In general, the life-form composition of floodplain grassland flora has shifted from low-growing herbaceous plants to tall-growing herbaceous plants during the abandonment period (from 1960s to 1990s). The effect of site moisture on the standing biomass-species richness relation and PUA was not significant, but the effect of the management regime was notable. PUA varied tenfold (mean 2.6 cm[^2] at moist sites with the management regime B but was 25.2 cm[^2] at un-managed wet sites). Dry standing biomass varied threefold depending on the management regime (from 263 to 763 g m[^-2]). Ramet density, PUA, PUB, S[pool], and life-form distribution on plots with recently re-introduced mowing (regime C) differed significantly from those of plots with regime D but did not differ from regularly mown plots (regime A). Management cessation led to a change in life-form distribution. Dominance of certain lifeforms depended on site moisture but the most obvious change was the increase of vegetation height.