Plant-Mediated Controls On Nutrient Cycling In Temperate Fens And Bogs

INTRODUCTION

Mires (peat-forming ecosystems) are usually classified into fens and bogs. These wetland types are characterized by high water tables, accumulation of organic matter (peat), and low nutrient availability. The nutrient supply to fens originates from nutrients contained in precipitation, surface water, and groundwater, whereas bogs are fed by the nutrients contained in precipitation only. In general, precipitation has a lower pH and a lower alkalinity than other water sources. As a result, the interstitial water in bogs has a low pH and low alkalinity in contrast to fens. Most fens and bogs are found in the temperate and (sub)arctic region of the Northern Hemisphere (Gore 1983). For purposes of this study, the temperate region includes the Cf (warm temperate) and Df (cool temperate) climates as defined by Kopppen (1931).

Fertilization experiments have shown that plant growth in temperate fens and bogs is nutrient limited (e.g., Vermeer 1986, Aerts et al. 1992, Verhoeven et al. 1996a). This raises the question of how plants from fens and bogs cope with low nutrient availability. Adaptations of perennials to different levels of nutrient availability are often evaluated by considering their nutrient-use efficiency (NUE). "NUE" is generally defined as the ratio between aboveground biomass production and nutrient loss in litterfall (Vitousek 1982). Thus, NUE covers a variety of processes, including tissue life-spans, tissue chemistry, and nutrient resorption from senescing tissues (Aerts and De Caluwe 1994, Aerts 1995). A widely used index of NUE is the inverse of litter nutrient concentrations (Vitousek 1982, Birk and Vitousek 1986). This index of NUE is strongly determined by the partitioning of nutrients between resorption and decomposition pathways (Aerts 1997a). High nutrient resorption contributes to high NUE, but leads to lower litter nutrient concentrations, and vice versa. This may have direct implications for litter decomposition rates and nutrient release, because decomposition and nutrient release from litter are often positively related to nutrient concentrations in litter and negatively related to carbon-to-nutrient ratios (Aber and Melillo 1982, Aerts 1997b). Thus, there may be a trade-off between leaf-level NUE and leaf-litter decomposition rate (Aerts 1997a).

The amount of nutrients available for plant uptake is determined by the balance between external inputs and outputs and by internal nutrient cycling. Nutrientbudget studies for fens and bogs in the temperate zone have shown that nutrient fluxes through internal cycling exceed external inputs by far (Urban and Eisenreich 1982, Hemond 1983, Morris 1991, Koerselman and Verhoeven 1992). This implies that plants exert a strong influence on nutrient availability through litter production, litter decomposition, and mineralization of nutrients contained in the litter. These processes show clear differences among various growth forms that occur in fens and bogs. In general, temperate bogs are dominated by Sphagnum mosses and to a lesser extent by slow-growing evergreen and deciduous shrubs and trees. Temperate fens are very often dominated by graminoids (mainly Carex and Cladium species) and by deciduous shrubs and trees and have, especially when water pH exceeds 5.5, a much lower Sphagnum cover (Malmer 1993, Bridgham et al. 1996).

The aim of this paper is to summarize patterns in plant-mediated processes, which determine the rate of nutrient cycling in temperate fens and bogs. We are aware of the fact that nutrient cycling is determined by the interaction of both plant-mediated effects and abiotic effects. However, given the fact that in many wetlands the plant-mediated effects overrule the abiotic effects (Urban and Eisenreich 1982, Hemond 1983, Morris 1991, Koerselman and Verhoeven 1992) we focus in this paper on the former. We collected literature data (Appendix) on NUE and other selected components, on litter decomposition (using the negative exponential decomposition model of Olson [1963]), and on nutrient mineralization. For the fen-bog classification we used the classification provided in the original papers. We organized the paper into four parts. First we review and summarize literature on NUE by comparing growth forms (evergreen species, deciduous species, graminoids, and Sphagnum mosses) and mire types (fens and bogs). We then investigate the link between NUE and litter decomposition rate by comparing published data on rates of leaf-litter decomposition using the same growth-form and mire-type classification as for NUE. Next, we determine if patterns of NUE and decomposition are expressed at the ecosystem level through patterns of nutrient cycling in fens and bogs. Unfortunately, there have been very few comparative studies on soil nutrient-cycling processes in fens and bogs. Therefore, most of this section deals with a synthesis of an ongoing comparative study of soil nutrient cycling in wetlands in the temperate zone using a standardized measuring protocol (Verhoeven et al. 1994, 1996b). Finally, we investigate if the patterns observed in our summary for fens and bogs deviate from those in terrestrial ecosystems.