Nutrient Mineralization and Plant Nutrient Uptake

The net mineralization of nutrients is about an order of magnitude lower in arctic ecosystems than in the boreal region, largely because of constrained microbial activity (Nadelhoffer et al, 1992). There is also a pronounced seasonal variation in net mineralization. Several studies of both litter and soil organic matter mineralization have shown that winter mineralization is higher than summer mineralization (Giblin et al, 1991; Hobbie and Chapin, 1996; Shaver et al, 1998), which is surprising given the expected decrease of microbial activity rate with decreasing temperature. Furthermore, net N mineralization may even be negative during the growing season; i.e., inorganic N is immobilized instead of being released. For instance, Hobbie and Chapin (1996) found high winter release of N in litter contrasting with immobilization of N during summer in spite of mass loss (Fig. 1). In fact, the total amounts of N in the litter even increased during summer, showing that N was transported into the litter mass. Similarly, five of six ecosystem types in an Alaskan tundra showed negative summer net N mineralization in the SOM, while annual mineralization was positive because of high non-growing-season mineralization rates (Giblin et al, 1991).

A possible reason for this annual pattern of net mineralization could be that the decomposing microorganisms themselves absorbed and immobilized the nutrients they mineralized from the litter and soil organic matter during the summer. In contrast, nutrients could have been released passively from the microbes during winter when their activity was lower and part of the populations probably died, or during repeated freezing and thawing in autumn and spring (Giblin et al, 1991; Schimel et al, 1996). However, most studies have shown a discrepancy between estimated low annual net nutrient mineralization of, e.g., N and a much higher annual plant nutrient uptake (Schimel and Chapin, 1996), which needs to be explained. There are at least three possible explanations: (1) Net mineralization measured in summer only may underestimate annual mineralization, as described above. (2) Nutrient mineralization measured by the "buried bag method," which excludes plant roots, allows microbes to monopolize and immobilize the nutrients, of which a part otherwise would have been taken up by the plants if they had not been denied access. (3) Much of the nitrogen may be acquired by plants in the organic form and bypass the mineralization step. The relative importance of these issues in explaining the discrepancy between measured nutrient mineralization and nutrient uptake is currently unknown.

Strong regulation of plant nutrient availability by the population dynamics of soil microorganisms seems logical because the microbial biomass contains large amounts of nutrients, even in

FIGURE 1 Comparison of changes in mass (left) and nitrogen content (right) in Betula papyrifera leaf litter in litterbags deposited on the soil surface, in Hylocoiuium or Sphagnum moss mats or in Eriophorum tussocks during three growing seasons and two nongrowing seasons. (From Hobbie and Chapin, 1996, with kind permisson from Kluwer Academic Publishers).

FIGURE 1 Comparison of changes in mass (left) and nitrogen content (right) in Betula papyrifera leaf litter in litterbags deposited on the soil surface, in Hylocoiuium or Sphagnum moss mats or in Eriophorum tussocks during three growing seasons and two nongrowing seasons. (From Hobbie and Chapin, 1996, with kind permisson from Kluwer Academic Publishers).

comparison to the vegetation. For instance, in a subarctic heath, the plant-microbial-SOM pools (to 15 cm depth) contained C in the proportions 19:2.5:78.5, N in the proportions 10:6.5:83.5, and P in the proportions 11:30:59. The proportions of soil inorganic N and P were below 1 (Jonasson et al, 1999a). Hence, while the microbial C pool was much smaller than the plant C pool, the microbial N content approached the N content in plants and microbial P exceeded plant P content. Indeed, the microbial N pool in Alaskan tundra approximately equalled the amount in plant roots (Hobbie and Chapin, 1998).

Microbes contained a relatively constant proportion (2.5-2.7%) of total ecosystem C across seven Alaskan tundra sites, with the quantity of C being determined by two independent methods (Cheng and Virginia, 1993). The N incorporated in the soil microorganisms was about 7% of the total soil N, i.e., a proportion almost identical to that estimated in the subarctic heath (Jonasson et al., 1999a, calculated from Cheng and Virginia, 1993).

The quantities of nutrients in microbes are large compared with the annual plant nutrient uptake, suggesting that even relatively limited dieback of the microbial populations can lead to release of an appreciable proportion of the plants' annual nutrient requirement. Indeed, it is known that the annual uptake of P by wet tundra vegetation can be almost entirely accounted for by P released through nutrient flushes from the microbial biomass (Chapin et al., 1978). It is possible, therefore, that the supply rate of nutrients to the soil inorganic pool varies depending on the conditions for microbial population growth or decline and that plant nutrient availability fluctuates inversely to microbial nutrient demand.

If the annual pattern of net mineralization is regulated mainly by microbial immobilization-mobilization cycles, microbes may be more effective than plants as competitors for nutrients during periods of the growing season (Fiarte and Kinzig, 1993; Jonasson et al., 1996; Schimel and Chapin 1996; Schimel et al., 1996). Indeed, a laboratory experiment showed that stimulated microbial activity after addition of a labile C source to the soil increased microbial nutrient uptake to the extent of causing strong limitation of plant growth due to nutrient deficiency (Schmidt et al, 1997a, b). Hence, the stimulated microbial activity led to increased competition for nutrients between soil microbes and plants and not to increased decomposition and release of inorganic nutrients. These observations suggest that labile C regulates the microbial nutrient mineralization by increasing immobilization as the amount of available carbon increases, while net mineralization increases under conditions of microbial C limitation. This indeed corresponds to field observations. A higher rate of net N and P mineralization of low-quality tundra moss litter than of higher quality vascular plant litter has been reported from arctic tundra (Hobbie, 1996) and other comparable nutrient-limited ecosystems (Verhoeven et al, 1990; Updegraff et al, 1995). In contrast, the decomposition rate positively correlated with litter quality and, hence, was negatively correlated with the rate of mineralization, showing that differences in decomposability between litter types do not always result in differences in mineralization rates.

0 0

Post a comment