Nutrient cycling in forests.

Studies of nutrient cycling in forests span more than 100 yr. In earlier years, most attention was given to the measurement of the pools of nutrients in plants and soil and of the return of nutrients from plant to soil in litterfall. The past 20 yr or so have seen a major concentration on the processes of nutrient cycling, with particular emphasis on those processes by which the supply of nutrients to the growing forest is sustained. In the more highly productive forests, up to 10 tonnes of litter of low nutritional quality is deposited annually on the forest floor. The decomposition of this litter, the mineralization of the nutrients it holds, and the uptake of nutrients by tree roots in the carbon-rich environment which results are the themes of this review. Studies of decomposition of litter in forests have been dominated by the role of nitrogen as a limiting factor, a domination which reflects the preponderance of studies of temperate forests in the Northern Hemisphere. For many forests of the world growing on soils of considerable age, it seems more probable that growth and nutrient cycling are limited by phosphorus (or some other element). There is increasing evidence for a number of forests that phosphorus is immobilized in the first stages of decomposition to a significantly greater extent than is nitrogen. Advances in research will depend, as with studies of soil organic matter, in denning and developing analytical techniques for studying biologically active forms of potentially limiting nutrients, rather than total elemental concentrations. The availability of phosphorus in forests is sustained by phosphorus cycling. More than 50% of the total phosphorus in the surface soils is in organic forms and much of the more labile phosphorus is in the form of diesters. Phosphorus availability is determined by competition between biological and geochemical sinks, and it is clear that the sinks in the rhizosphere (plant roots, microorganisms, soil mineral and organic components) are extensively modified by active processes (e.g. production of exudates, nutrient storage in a variety of organic or polymeric forms and nutrient transport away from sites of uptake). There is abundant evidence that roots of many species exude compounds which have the ability to solubilize sources of phosphorus of otherwise low availability. The significance of root exudates (for example, phosphatases, organic acids) in the functioning of perennial ecosystems has yet to be quantified and there are conflicting reports as to the effects of simple organic acids on phosphorus availability. The distribution of phosphorus sinks and their relative competitiveness and their modification are topics of fundamental importance for future research. In contrast to the mineralization of phosphorus, our knowledge of transformations and availability of nitrogen in forest soils is well-developed. Net nitrogen mineralization rates approximate rates of nitrogen return in litterfall but the contribution of nitrification is variable. Nitrification is not inhibited by the low pH of many forest soils and there is increasing evidence of nitrate immobilization by microorganisms and of increased diversity and better competitiveness for NH of nitrifying microorganisms than has previously been accepted. Variability in rates of nitrification is often interpreted as being due to allelopathy. Hypotheses invoking allelopathy are more or less untestable, and it seems likely that new techniques using N in situ will lead to a more fundamental understanding of nitrogen transformations in forest soils. Recent studies in coniferous forest soils have highlighted the short (< 1 d) turnover time of NH . Finally, it seems that forest soils are resistant to major changes in patterns of nitrogen mineralization (and certainly, because of the large number of sinks, in patterns of phosphorus mineralization) following disturbance by natural events such as wind-throw and fire, and by man-made events such as logging and fertilizing. The long-term disturbance by acid rain is a more complex matter since forest ecosystems are not adequate buffers for nitrate. Contents Summary 561 I. Introduction 562 II. Linking nutrient cycling to nutrient availability - Setting the themes 563 III. The nature of soil organic matter 566 IV. Tree roots and the availability of nutrients 566 V. The decomposition of forest litter 569 VI. Mineralization of organically-bound nutrients 571 Acknowledgements 576 References 576.