Patterns in nutrient availability and plant diversity of temperate North American wetlands

Few wetland studies from temperate North America have related either species richness or plant community composition to any direct measure of nutrient availability, or examined changes in species composition following experimental nutrient additions. Studies of wetlands in western Europe and of other terrestrial ecosystems in North America frequently show that nutrient enrichment leads to changes in species composition, declines in overall plant species diversity, and loss of rare and uncommon species. We therefore used an extensive literature survey and analysis of data on plant species composition, species richness, productivity or standing crop, and C:N:P stoichiometry in plant tissues and surface soils to draw conclusions about the nature of nutrient limitation in temperate North American bogs, fens, marshes, and swamps, and to infer their potential response to nutrient enrichment. We searched all major bibliographic data bases for studies containing such data (through March 1998) and added relevant data from our own ongoing research. We analyzed plant and soil data sets by wetland type and by wetland soil type, and the plant data set also by growth form. Existing studies appear to confirm common generalizations: (1) plant community type changes across broad nutrient gradients; (2) species richness declines as various indicators of nutrient availability increase beyond some threshold; and (3) rare and uncommon species are almost always associated with species-rich communities. However, (1) these generalizations do not always hold within community types; (2) for many community types, the threshold beyond which richness declines has not been established, and high or low diversity may occur below that threshold; and (3) the failure of many studies to include bryophytes precludes drawing strong conclusions about nutrient availability and diversity in peatlands. Marshes had significantly lower mean nitrogen:phosphorus (N:P) ratios in live tissue than other wetland types (bogs, fens, and swamps), which did not differ significantly from each other. Mean N:P ratios in live tissues were significantly higher in peatlands than in mineral-soil wetlands. Nitrogen:phosporus ratios in litter did not differ significantly between peatlands and mineral-soil wetlands but were higher than in live tissues. Among growth forms, the highest mean N:P ratios in live tissues occurred in bryophytes, and the lowest in vascular herbaceous species. Only bryophyte live tissues differed significantly from other growth forms; litter N:P ratios were not significantly different among growth forms. Average N:P ratios in surface soils were lower in marshes and swamps than in bogs and fens. Wetlands on mineral soils had lower average N:P ratios than wetlands on peat soils. Average surface soil N:P ratios rose sharply at high soil organic-matter contents (greater than or equal to 90%) and were generally greater than 16 at organic-matter concentrations above 20%. In combination, plant tissue and surface soil N:P ratios suggest that a large proportion of North American wetlands are either P limited or co-limited by N and P, especially those occurring on organic soils. Only marshes have N:P ratios in both live tissues and surface soils that consistently indicate N limitation, although soils data suggest that the majority of swamps are also N limited. Vascular herbaceous species appear to be N limited, but no pattern is apparent among other growth forms. Inferences drawn from N:P stoichiometry need to be verified by examining ecosystem response to experimental fertilization. High variances in plant and soil N:P ratios suggest that understanding nutrient limitation at both the community and individual-species level may be necessary to predict changes in species composition and richness with nutrient enrichment.