Printer Friendly

NUTRIENT LIMITATION OF DECOMPOSITION IN HAWAIIAN FORESTS.

SARAH E. HOBBIE [1]

Abstract. In Hawaiian montane forests, we assessed whether the same nutrients limit decomposition and aboveground net primary production (ANPP) along a soil chronosequence where nutrients demonstrably and predictably limit ANPP. At three sites that vary in parent material age (300, 20000, and 4.1 X [10.sup.6] yr), we used fertilization to assess whether nitrogen (N) and/or phosphorus (P) limit decomposition. Reciprocal transplants using litter bags allowed us to distinguish limitation by externally supplied nutrients vs. limitation by nutrients within litter. Nutrient limitation of decomposition was not predictable from nutrient limitation of ANPP, in that elevated litter and soil N had only small, if any, effects on decomposition, even at the young site where N limits ANPP. At the oldest site where P limits ANPP, both elevated litter P and increased availability of soil N and P increased decomposition rates. Thus, nutrients may limit decomposition more strongly in low-P than in low-N ecosystems. Fertilizati on affected litter nutrient dynamics more strongly than it did decomposition, and we observed uptake of both N and P by decomposers that was not always accompanied by changes in decomposition rates. Such nutrient incorporation into decomposing litter may retain nutrients within ecosystems, even when nutrients do not limit decomposition rates.

Key words: decomposition; fertilization; Hawaii; litter; Metrosideros polymorpha; nitrogen; nutrient limitation; phosphorus.

INTRODUCTION

Decomposition of plant litter regulates recycling of nutrients in ecosystems, influences net ecosystem carbon storage, and is the first step in the formation of soil humus. Commonly cited limitations to decomposition include factors that directly affect the activity of decomposer organisms (e.g., temperature, moisture, and pH) and factors that affect resource availability for decomposers (e.g., soil nutrient availability and carbon and nutrient chemistry of the decomposing litter) (Swift et al. 1979, Aber and Melillo 1991).

Despite widespread acceptance of nutrient availability as a controller of decomposition (Swift et al. 1979), the nature of nutrient regulation of decomposition is far from clear. Much of the evidence for nutrient limitation of decomposition is indirect. For example, litter often immobilizes nutrients (particularly nitrogen [N] and phosphorus [P]) early during decomposition, suggesting that fresh litter contains insufficient nutrients to meet the growth and maintenance requirements of decomposers (Gosz et al. 1973, Staaf and Berg 1981). Furthermore, litter decay rates often correlate positively with absolute nutrient concentrations or with nutrient concentrations relative to that of carbon (or lignin, Melillo et al. 1982, Taylor et al. 1989, Enriquez et al. 1993).

Direct evidence for nutrient limitation to decomposition is provided only by fertilization experiments that enhance the external supply of nutrients and therefore should increase decomposition rates if nutrients limit litter decay. However, the response of decomposition to fertilization is sometimes positive (Gill and Lavender 1983, Hunt et al. 1988, Prescott et al. 1992, O'Connell 1994, Conn and Day 1996, Downs et al. 1996), but is often neutral (Staaf 1980, Pastor et al. 1987, Hunt et al. 1988, Theodorou and Bowen 1990, Van Vuuren and Van der Eerden 1992, Andren et al. 1993, Prescott 1995, Downs et al. 1996) or even negative (Gill and Lavender 1983, Titus and Malcolm 1987, O'Connell 1994, Prescott 1995; reviewed by Fog 1988).

Several possible reasons exist for this inconsistent response of decomposition to fertilization. First, nutrients may interact with aspects of litter carbon chemistry. For example, decomposers may be unable to use additional nutrients due to low availability of labile carbon substrates. Or, N may actually inhibit the decomposition of the lignin fraction of litter either by inhibiting synthesis of lignolytic enzymes or by reacting with breakdown products of lignin degradation to form other compounds that resist decay (Berg 1986, Fog 1988). Such inhibition of lignin degradation could result in a neutral or negative effect of N on overall decomposition. Second, nutrients may indeed limit decomposition, but studies showing neutral effects may not have added the limiting nutrient or sufficient nutrient to overcome limitation.

Nevertheless, indirect evidence suggests that nutrients do limit decomposition in some ecosystems. For example, along a soil chronosequence in Hawaiian montane forest, decomposition of native litter and of a common litter was fastest in sites with relatively high soil N and P availability and slowest in sites where soil nutrient availability is low and nutrients strongly limit aboveground net primary production (ANPP) (Crews et al. 1995, Vitousek and Farrington 1997). Nutrient limitation of ANPP at the infertile sites is a consequence of low soil nutrient availability resulting from differences in soil age (Vitousek and Farrington 1997). However, limitation could be exacerbated by a positive feedback between plants and decomposers; in low-nutrient sites, nutrient limitation of decomposition may slow nutrient recycling through litter, further decreasing nutrient availability to plants (Vitousek 1982, Hobbie 1992).

The objective of this study was to determine whether the same nutrients that limit ANPP also limit decomposition at a series of sites in which nutrients demonstrably limit ANPP. In addition, we wanted to determine whether nutrient limitation of decomposition (if it occurs) results primarily from direct limitation by low soil nutrient availability, or indirectly, from low concentrations of litter nutrients, or both. Other studies have demonstrated that litter and soil nutrients do not always similarly influence decomposition rates (Prescott 1995).

METHODS

Site description

We assessed nutrient limitation of decomposition at three sites on the soil chronosequence described by Crews et al. (1995). These sites have similar present climate, parent material, topography, and plant species composition, but vary substantially in parent material age (Table 1). All sites receive 2500 mm of annual precipitation, are located between 1130 and 1180 m elevation, and are on constructional surfaces. The youngest site (Thurston) was formed [sim]300 years ago by eruptions of Kilauea volcano on the island of Hawaii. The soils are derived from 0.4 m of tephra overlying lava. The intermediate-aged site (Laupahoehoe) is also located on the island of Hawaii, and was formed from [sim]20 000-year-old tephra deposits resulting from eruptions of Mauna Kea volcano. The oldest site (Kauai) is located in Kokee State Park on the island of Kauni; its parent material was formed [sim]4.1 X [10.sup.6] yr BP. The myrtaceous tree Metrosideros polymorpha dominates all sites. Surface soil horizons at all sites are h ighly organic, with [greater than]20% carbon (Torn et al. 1997).

Fertilization with N and P, as well as other essential elements, has demonstrated that nutrient limitation of ANPP differs among the three sites. Nitrogen limits ANPP in the youngest site, P limits ANPP in the oldest site, and both N and P in combination, but neither element alone, limit tree growth at the intermediate-aged site (Vitousek et al. 1993, Herbert and Fownes 1995, Vitousek and Farrington 1997).

All sites had been fertilized at a similar rate with N and P for at least three years prior to the beginning of this study. The duration and exact experimental design differ slightly among sites (Table 1). At Thurston and Kauai, 15 X 15 m plots (n = 4) received N alone, P alone, N and P together, or no fertilizer (controls). At Laupahoehoe, fertilization has been tree- rather than plot-based because of low tree density. Individual trees (n = 6) receive either N, P, N and P, or no fertilizer (controls) in a 5 m radius circle centered on the tree. Fertilizer has been semiannual at rate of 100 kg.[ha.sup.-1].[yr.sup.-1] N and/or P broadcast with commercial fertilizer on the surface of the forest floor. Nitrogen has been applied as an even mixture of urea and ammonium nitrate and P as triple superphosphate.

Decomposition

We assessed nutrient limitation of decomposition at each site in several ways. First, at each site we compared control litter decomposed in control plots with litter produced in each of the fertilizer treatments decomposed in its respective fertilizer treatment (Fig. la: In Situ Experiment). This comparison allowed us to assess the combined effects of direct and indirect fertilizer influences on decomposition rate. Second, at each site we compared decomposition of litter produced in the control plots (a common litter), but placed in the different fertilizer treatments (Fig. 1b: Soil Nutrient Experiment). This allowed us to assess nutrient limitation of decomposition by external nutrient supply; we will refer to this as control by soil nutrients. Third, we compared decomposition of litter produced in the fertilized plots and the more fertile site (Laupahoehoe) with that of litter produced in control plots when all of these substrates were placed in the control treatment (Fig 1c: Litter Nutrient Experiment). T his allowed us to assess nutrient limitation of decomposition by nutrients within the litter (or by other changes in litter structure or chemistry); we will refer to this as control by litter nutrients.

We collected leaf litter in litter traps placed under trees in control and fertilized plots during fall 1995, cleaning litter traps at least once per month. We discarded obviously green leaves, litter with psyllid galls, and litter of species other than Metrosideros polymorpha. Litter was pooled within treatments, air-dried for storage, and dried at 50[degrees]C for 48 h prior to litter bag construction.

Subsamples of initial litter were weighed, dried at 70[degrees]C and reweighed to develop 50[degrees] to 70[degrees] conversions. In addition, we analyzed subsamples of initial litter for total N and P using Kjeldahl digestion followed by colorimetric analysis on an Alpkem autoanalyzer (Alpkem, a division of OI Analytical, Wilsonville, Oregon); for cations (K, Mg, and Ca) using nitric acid digestion followed by analysis with atomic absorption (Perkin-Elmer Corporation, Norwalk, Connecticut); and for lignin using acetyl bromide digestion (Iiyama and Wallis 1990), with National Bureau of Standards pine as a standard (% lignin = 21.8; J Pastor and B. Dewey, personal communication). Samples were also analyzed for various carbon fractions using forest products techniques (Ryan et al. 1989) at the Center for Water and the Environment (Natural Resources Research Institute, University of Minnesota, Duluth, Minnesota). Fractions determined included nonpolar extractives (NPE: fats, oils, waxes), water-solubles (WS: am ino acids, simple sugars, soluble phenolics), acid-solubles (AS: cellulose, hemicellulose, starch, polypeptides, nucleic acids), and acid-insolubles (AIS: primarily lignin). Carbon fractions are presented on an ash-free dry mass basis.

We constructed 10 X 10 cm litter bags out of 1-mm mesh fiberglass windowscreen and filled litter bags with 1-2 g litter. Enough litter bags were constructed to allow harvests after 1, 3, 6, 9, 12, and 24 months. We strung together (using nylon chord) litter bags to be harvested at a particular date. Each string destined for a particular plot held litter bags containing litter from each of the four treatments at that site (control, N, P, and NP). In addition, strings destined for Thurston and Kauai held litter bags containing litter from Laupahoehoe (control trees only). Litter bags were at least 15 cm apart, and the order of the bags on the string was randomized. Insufficient litter was collected from Laupahoehoe control trees to allow for six harvests, so the 9-mo harvest was skipped for this litter type.

Six strings of litter bags (one for each harvest) were placed in each replicate of each treatment at all sites from 29 January to 2 February 1996. Strings were arranged like spokes on a wheel, radiating from a washer that was marked with a pin flag. Litter bags were placed on the forest floor in areas that were dominated by litter of Metrosideros polymorpha.

Litter bag collections were timed to be at least 3 and often 6 mo after the semiannual fertilization events, and collections at all three sites were made within 1 wk of each other. After collecting, we removed litter from litter bags and gently cleaned it of roots, invertebrates, frass, and soil. We dried (at 70[degrees]C) and weighed litter and analyzed it for N and P using Kjeldahl digestion followed by colorimetric analysis. We calculated decomposition as the percentage of the original mass remaining. Nutrient content was determined by multiplying nutrient concentration by mass and was expressed as the percentage of the original nutrient content remaining.

To determine fertilization effects on pH, we measured pH of the forest floor ([O.sub.i] horizon) in 0.01 mol/L [CaCl.sub.2] (10:1 [CaCl.sub.2]:litter by mass). We composited subsamples of litter from each quadrant of a plot and placed a 2-g subsample in 20 mL of 0.01 mol/L [CaCl.sub.2] for 30 min, with shaking every 5 min. Slurries were allowed to settle for 30-60 min before reading on a pH meter.

Statistical analyses

We analyzed decomposition (percentage mass remaining) for each site separately, using a series of two-and three-way analyses of variance (ANOVA), depending on the specific hypothesis being tested.

In situ experiment.--To determine the combined effect of direct and indirect influences of fertilization on in situ decomposition (Fig. lA), we used a three-way ANOVA with harvest date, N, and P as factors. For this particular analysis, we did not separate the effects of litter source and litter destination. (At each site, litter produced in control, N, P, and NP treatments was decomposed in control, N, P, and NP treatments, respectively.) Therefore, nutrient factors (N and P) refer to treatments in which litter was both produced and decomposed.

Soil nutrient experiment.--To determine the effect of soil nutrients alone on decomposition (Fig. 1B), we again used a three-way ANOVA with harvest date, N, and P as factors, but we restricted our analysis to litter produced in the control treatments that was decomposed in the control, N, P, and NP treatments at each site. In this analysis, nutrient factors (N and P) refer to treatments in which litter was decomposed.

Litter nutrient experiment.--We examined effects of litter nutrients alone on decomposition in two ways (Fig. 1C). First, we again used a three-way ANOVA with harvest date, N, and P as factors, but for this analysis, nutrient factors (N and P) refer to the treatments in which litter was produced. In other words, we analyzed how litter produced in the various treatments decomposed in the control treatment at each site. In addition, we determined whether litter produced in the fertilization treatments at the two infertile sites (Thurston and Kauai) decomposed similarly to litter produced at the more fertile site (Laupahoehoe) using two-way ANOVA with harvest date and litter source as main effects. Levels of litter source included native litter produced in the control, N, P, and NP treatments as well as litter produced at Laupahoehoe.

Analysis of site differences in decomposition.--We determined whether nutrient limitation of decomposition could explain site differences in decomposition using a series of two-way ANOVAs with site of decomposition and harvest as main effects. For these analyses, we compared native litter decomposing in situ (at each site); litter produced in the NP treatments and decomposed in the NP treatments at Thurston and Kauai compared to in situ decomposition at Laupahoehoe; or litter produced at Laupahoehoe and decomposed in the NP treatments at Thurston and Kauai compared with in situ decomposition at Lauphoehoe.

ANOVAs similar to those described under Soil nutrient experiment and Litter nutrient experiment were conducted for litter N and P content (percentage of initial content). Homogeneity of variances was checked using Cochran's test. When variances were heterogeneous, data were ln transformed. In no cases did ln transformations change the significance of an interaction term. Sometimes variances were still heterogeneous even after data were transformed (we attempted several different transformations). We have still presented the ANOVA results, since in a large (multifactorial) design such as this one, heterogeneous variances are less likely to increase the probability of Type I error (Underwood 1997).

RESULTS

Sites did not differ in pH of the [O.sub.i] horizon (Table 1). Three months after a fertilization event, fertilization with N slightly decreased the pH of the [O.sub.i] horizon at Thurston (pH [mean [pm] 1 SE] = 3.9 [pm] 0.1 and 3.4 [pm] 0.2 in control and N-fertilized plots, respectively, P = 0.05), but not at the other sites (data not shown). This effect disappeared by six months after fertilization. Fertilization with P did not affect pH of the [O.sub.i] horizon.

Fertilization had significant effects on initial litter nutrient chemistry, but the size of these effects differed both across sites and among fertilization treatments (Table 2). Fertilization with N significantly increased litter N concentrations at all three sites, but this increase was generally small in magnitude. At Laupahoehoe, P fertilization slightly increased litter N concentrations, and at Kauai, P slightly decreased litter N concentrations, in the absence of N (N X P interaction). However, N fertilization did not cause litter from infertile sites to resemble litter from the more fertile site: litter from Laupahoehoe had more than twice the litter N concentrations of litter from Thurston and Kauai. Fertilization with P increased litter P concentrations significantly at all three sites; the effect was relatively small at Thurston and Laupahoehoe. However, P additions quadrupled litter P concentrations at Kauai. At Thurston and Kauai, N additions decreased litter P concentrations, while at Laupahoehoe, N fertilization increased litter P concentrations. Again, litter from Laupahoehoe had double the P concentrations of Thurston litter, although litter from P-fertilized plots at Kauni exceeded litter from Laupahoehoe in P concentrations.

Concentrations of K and Mg were highest at Laupahoehoe and lowest at Thurston (Table 2). Concentrations of Ca were highest at Thurston and declined with soil age. Calcium responded most to fertilization, decreasing with addition of the limiting nutrient at Thurston and increasing with addition of the limiting nutrient(s) at Laupahoehoe and Kauai, although that may have resulted directly from additions of triple superphosphate, which contains Ca.

Initial concentrations of acetyl bromide lignin were lower at Kauni than they were at Thurston and Laupahoehoe, and fertilization with P significantly reduced acetyl bromide lignin concentrations at Thurston (Table 2). Initial concentrations of acid-insolubles were similar across sites, although a substantial proportion of the acid-insoluble fraction appears to be something other than lignin, at least as measured by acetyl bromide digestion or by the Van Soest and Wine (1968) procedure (Crews et al. 1995) (Table 2). Laupahoehoe litter had half the concentrations of nonpolar extractives than the other sites. Thurston had lower concentrations of water-solubles (total, tannin-, and glucose-equivalent), but higher concentrations of acid-soluble glucose-equivalents.

Decomposition

As expected from previous work, decomposition was fastest at the most fertile site, Laupahoehoe, and slowest at Kauai (Fig. 2A). Decay constants from an exponential decay model (Olson 1963) for in situ decomposition in the control treatments were 0.38, 0.81, and 0.26 per year for Thurston, Laupahoehoe, and Kauai, respectively. For in situ decomposition in all treatments (where litter and soil nutrients were not distinguished), the same nutrient(s) did not always limit decomposition and ANPP. At Thurston and Laupaboehoe, N and P in combination significantly stimulated decomposition, although the effect of fertilization was small at both sites (Fig. 2a, significant N X P interaction). At Kauai, N and P independently stimulated decomposition, particularly at later harvest dates, and the effect of fertilization was larger than at the other sites.

Litter and soil nutrients had opposing effects on decomposition at Thurston. Increasing soil N stimulated decomposition (Fig. 2B); however, litter produced in N and NP treatments decomposed more slowly than litter produced in control or P treatments by the final harvest date (Fig. 2C). Litter produced at Laupahoehoe decomposed significantly faster at Thurston than did native litter produced in the N and NP treatments (Fig. 2C, Tukey's hsd P [less than] 0.05).

At Laupahoehoe, neither soil nutrients nor litter nutrients significantly affected decomposition when considered alone (Fig. 2B,C). The significant N X P interaction for in situ decomposition must have resulted from small but similar effects of soil and litter N and P.

At Kauai, soil and litter nutrients both stimulated decomposition, but different nutrients were limiting, depending on whether they were increased in soil or litter. N and P together (but not alone) stimulated decomposition when added to the soil (Fig. 2B). However, litter P alone significantly increased decomposition rates, and this effect became larger later in decomposition (Fig. 2C). The significant N and P effects on in situ decomposition represent a combination of these stimulatory effects by soil and litter nutrients.

Nutrient limitation of decomposition was insufficient to explain site differences in decomposition (Fig. 3). As mentioned previously, in situ decomposition was fastest at Laupahoehoe and slowest at Kauai, although the relative differences among sites changed with time (Fig. 3, significant site and site X harvest effects). When we compared litter produced in the NP treatments and decomposed in the NP treatments at Thurston and Kauai with in situ decomposition at Laupahoehoe, decomposition was still fastest at Laupahoehoe (Fig. 3, significant site and site X harvest effects). In other words, fertilizing at Thurston and Kauai did not make these sites resemble Laupahoehoe in decomposition. Finally, when we compared decomposition of Laupahoehoe litter in the NP treatments at Thurston and Kauai with in situ decomposition at Laupahoehoe, decomposition was still faster at Laupahoehoe, although the differences among sites changed through time (Fig. 3, significant site and site X harvest effects).

Litter nutrient dynamics

Soil nutrients had larger (and more frequent) effects on litter nutrient dynamics than they did on decomposition. Litter decomposing at Thurston and Kauai immobilized N over the course of the experiment, and increasing soil N significantly stimulated this N immobilization (Fig. 4A). At Kauai, this stimulation increased through time. Litter decomposing at Laupahoehoe exhibited a net release of N over time that was unaffected by fertilization with N or P.

Effects of increased soil P on litter P dynamics were more dramatic: at all three sites, the effect of P on litter P content was much greater than the effect of P on mass loss (where one existed). At both Thurston and Kauai, litter decomposing in controls or N treatments changed little in P content through time. However, at Thurston and Kauai, litter decomposing in P and NP treatments had up to twice its initial P content by the end of the experiment (Fig. 4B). At Laupahoehoe, litter P declined over the experiment, but this decline was slightly lessened by fertilization with P.

Whether litter was produced in fertilized treatments had little effect on litter N dynamics (Fig. 5A). However, litter produced at Laupahoehoe had substantially different litter N dynamics than did native litter at Thurston and Kauai. While native litter immobilized N at both of these sites, litter from Laupahoehoe exhibited a net decline in litter N content by the end of the experiment, presumably because of its high initial N concentrations.

In contrast, whether litter was produced in fertilized treatments had large effects on litter P dynamics at Thurston and Kauai. At Thurston, litter produced in the N alone treatment (i. e., litter with elevated N concentrations, but low P concentrations) immobilized substantially more P than litter produced in the other treatments (Fig. 5B). At Kauai, litter produced in the P-fertilized treatments (P and NP) lost much of its initial P immediately, presumably due to leaching. Litter produced in the non-P treatments (control and N) immobilized P during the experiment. Litter produced at Laupahoehoe exhibited intermediate P dynamics.

DISCUSSION

Decomposition

Across a series of sites in which the nutrient that limits ANPP changes predictably, the same nutrients did not always limit ANPP and decomposition in the same way. Fertilization with N in particular had only small effects on decomposition at these sites, even at sites where N availability is low enough (relative to demand) to limit plant growth. One explanation for a small (or no) effect is that some other factor besides N availability imposes a more profound constraint on decomposition. Demand for N by decomposers may be low relative to supply, even in the low-N site, because of the poor carbon quality of the litter produced at these sites. Alternatively, fertilization with N may directly inhibit the degradation of lignin, offsetting any stimulatory effect of N on decay of more labile carbon fractions (Fog 1988). Several studies in temperate systems have demonstrated such inhibition of lignin degradation by N in situ (Berg 1986, O'Connell 1994, Conn and Day 1996).

At the oldest site (Kauai), fertilization had a larger effect on decomposition: elevated litter P and soil N and P combined to increase mass loss of litter by 60% above that in the control treatment. The response of decomposition to increased nutrients at Kauai may reflect an interaction between the low mobility of P in soil, the demands of decomposing microbes for P, and the greater response of litter nutrient concentrations to fertilization at Kauai than at other sites. Sources of immobilized nutrients during decomposition include translocation from soil or litter via fungal hyphae, throughfall, and fixation (for N) (Aber and Melillo 1991). Translocation may be more limited for P than for N because of the relatively slow diffusion of P through soil to fungal hyphae. Furthermore, the discrepancy between microbial and litter nutrient concentrations is greater for P than it is for N, resulting in relatively high microbial demand for P (Paul and Clark 1996). Thus, low supply and high demand may make low P avai lability more profoundly limiting to decomposition than low N availability.

Phosphorus additions at Kauai were presumably more available to decomposers than were N additions at Thurston, since they caused large increases in concentrations of soluble P in litter (indicated by rapid leaching of P during the first month of decomposition of litter from P treatments). Such leaching would have increased throughfall P in the P-fertilized treatments at Kauai, augmenting P availability to decomposers. A study of nutrient limitation to root decomposition across these sites also found an effect of P at Kauai, but not of N at Kauai or of N or P at Thurston (Ostertag and Hobbie 1999). The large increases in litter P concentrations with P fertilization at Kauai probably reflect population differences in growth rates and P storage. In a common garden, seedlings of Metrosideros polymorpha from Kauai had inherently slower growth and showed less growth response to added nutrients than did seedlings of M. polymorpha from Thurston or Laupahoehoe (Treseder 1999). Thus, Kauai trees appear to accumulate P rather than to dilute it with large growth responses to P additions.

The stimulation of decomposition at Kauai by soil N and P together may also represent an indirect effect of fertilization. Nitrogen and P together increase litterfall by 50-60% at Kauai (Herbert and Fownes 1995; R. Harrington, unpublished data). Litter decomposing in this treatment was relatively more incorporated into the 0 horizon than litter decomposing in other treatments, potentially increasing its water content and facilitating colonization by soil biota.

Site differences

Greater nutrient availability at the intermediate-aged site (Laupahoehoe) can only partially explain faster rates of decomposition at that site relative to the youngest and oldest sites. Even with added nutrients, litter decomposing at Thurston and Kauai lagged behind that decomposing at Laupahoehoe, especially early in decomposition. These results suggest that additional characteristics besides nutrient availability are responsible for some of the differences in decomposition rates among sites. One potentially important site difference is the abundance of soil fauna. Despite the small mesh size of the litter bags used in this study (1 mm), we frequently observed insect larvae and earthworms inside litter bags, suggesting that soil fauna has access to litter inside bags. Earthworms and lepidopteran larvae appear to be more abundant at Laupahoehoe (personal observation) and may be partially responsible for the faster decomposition there. Ultimately their higher abundance may be both a cause and a consequence of the greater site fertility at Laupahoehoe.

Litter nutrient dynamics

Fertilization had larger and more frequent effects on litter nutrient dynamics than it did on decomposition. Although P fertilization increased immobilization of P by litter at all three sites, it influenced decomposition only at Kauai. Nitrogen fertilization stimulated N immobilization at Kauai, although N alone had no effect on litter decomposition there. Thus, decomposers at these sites appear to have a large capacity to take up both N and P, even when additional N and P have no effect on rates of decomposition. Numerous studies have shown that fertilization can affect litter nutrient dynamics without simultaneous effects on decomposition (Pastor et al. 1987, Titusand Malcolm 1987, Fog 1988, Hunt et al. 1988, Theodorou and Bowen 1990, Van Vuuren and Van der Eerden 1992, Berg and Tamm 1994, O'Connell 1994, Downs et al. 1996). These results suggest that decomposers have fairly plastic carbon:nutrient ratios and that nutrient incorporation into decomposing litter is potentially an important mechanism of nutr ient retention in ecosystems, even when nutrients do not limit decomposition rates.

Conclusions

Our results suggest that the strong N limitation of ANPP at Thurston is not strengthened by positive plant--litter or soil--litter feedbacks resulting from N limitation of decomposition. In other words, low soil N availability and low litter nutrient concentrations (that presumably result from low N availability) impose little influence on litter decomposition at Thurston. At Thurston, different factors control rates of litter inputs to soil (i.e., NPP) and rates of litter decomposition. The numerous studies showing little or no effect of N fertilization on decomposition (Staaf 1980, Gill and Lavender 1983, Pastor et al. 1987, Titus and Malcolm 1987, Fog 1988, Hunt et al. 1988, Theodorou and Bowen 1990, Van Vuuren and Van der Eerden 1992, Andren et al. 1993, O'Connell 1994, Prescott 1995, Downs et al. 1996), even in temperate ecosystems where N likely limits NPP, suggest that this uncoupling of factors that regulate litterfall and decomposition may be a general phenomenon at sites where N limits NPP. At Kauai, on the other hand, both low soil nutrient availability and the resultant low litter P concentrations constrain litter decomposition rates, and probably contribute to the maintenance of nutrient limitation of ANPP at Kauai.

ACKNOWLEDGMENTS

We are grateful to Hawaii Volcanoes National Park, the Division of Forestry and Wildlife of the State of Hawaii, and the Joseph Souza Center at Kokee State Park for access to field sites. We thank Heraldo Farrington and Rebecca Ostertag for help in the field, Doug Turner for help with sample analysis, and Rebecca Ostertag and Kurt Pregitzer for comments on the manuscript. Brad Dewey and John Pastor of the University of Minnesota, Duluth, provided carbon fraction analyses. This research was supported by an NSF post-doctoral fellowship to S.E. Hobbie and by NSF grant DEB 9628803 and a USDA-NRI grant to PM. Vitousek.

(1.) Present address: Department of Ecology, Evolution, and Behavior, University of Minnesota, 100 Ecology Building, 1987 Upper Buford Circle, St. Paul, Minnesota 55108 USA. E-mail: shobbie@tc.umn.edu

LITERATURE CITED

Aber, J. D., and J. M. Melillo. 1991. Terrestrial ecosystems. Saunders College Publishing, Orlando, Florida, USA.

Andren, O., R. Kalman, and T. Katterer. 1993. Water and temperature dynamics in a clay soil under winter wheat: influence on straw decomposition and N immobilization. Biology and Fertility of Soils 15:1-8.

Berg, B. 1986. Nutrient release from litter and humus in coniferous forest soils--a mini review. Scandinavian Journal of Forest Research 1:359-370.

Berg, B., and C. O. Tamm. 1994. Decomposition and nutrient dynamics of litter in long-term optimum nutrition experiments. II. Nutrient concentrations in decomposing Picea abies needle litter. Scandinavian Journal of Forest Research 9:99-105.

Conn, C. E., and F. P. Day. 1996. Response of root and cotton strip decay to nitrogen amendment along a barrier island dune chronosequence. Canadian Journal of Botany 74:276-284.

Crews, T. E., K. Kitayama, J. H. Fownes, R. H. Riley, D. A. Herbert, D. Mueller-Dombois, and P. M. Vitousek. 1995. Changes in soil phosphorus fractions and ecosystem dynamics across a long chronosequences in Hawaii. Ecology 76:1407-1424.

Downs, M. R., K. J. Nadelhoffer, J. M. Melillo, and J. D. Aber. 1996. Immobilization of a [[blank].sup.15]N-labeled nitrate addition by decomposing forest litter. Oecologia 105:141-150.

Enriquez, S., C. M. Duarte, and K. Sand-Jensen. 1993. Patterns in decomposition rates among photosynthetic organisms: the importance of detritus C:N:P content. Oecologia 94:457-471.

Fog, K. 1988. The effect of added nitrogen on the rate of decomposition of organic matter. Biological Review 63:433-462.

Gill, R. S., and D. P. Lavender. 1983. Litter decomposition in coastal hemlock (Tsuga heterophylla) stands: impact of nitrogen fertilizers on decay rates. Canadian Journal of Forest Research 13:116-121.

Gosz, J. R., G. E. Likens, and F. H. Bormann. 1973. Nutrient release from decomposing leaf and branch litter in the Hubbard Brook forest, New Hampshire. Ecological Monographs 43:173-191.

Herbert, D. A., and J. H. Fownes. 1995. Phosphorus limitation of forest leaf area and net primary productivity on highly weathered tropical montane soils in Hawaii. Bio-geochemistry 29:223-235.

Hobbie, S. E. 1992. Effects of plant species on nutrient cycling. Trends in Ecology and Evolution 7:336-339.

Hunt, H. W., E. R. Ingham, D. C. Coleman, E. T. Elliott, and C. P. P. Reid. 1988. Nitrogen limitation of production and decomposition in prairies, mountain meadow, and pine forest. Ecology 69:1009-1016.

Iiyama, K., and A. F A. Wallis. 1990. Determination of lignin in herbaceous plants by an improved acetyl bromide procedure. Journal of the Science of Food and Agriculture 51:145-161.

Melillo, J. M., J. D. Aber, and J. F. Muratore. 1982. Nitrogen and lignin control of hardwood leaf litter decomposition dynamics. Ecology 63:621-626.

O'Connell, A. M. 1994. Decomposition and nutrient content of litter in a fertilized eucalypt forest. Biology and Fertility of Soils 17:159-166.

Olson, J. S. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44:322-331.

Ostertag, R., and S. E. Hobbie. 1999. Early stages of root and leaf decomposition in Hawaiian forests: effects of nutrient availability. Oecologia 121:564-573.

Pastor, J., M. A. Stillwell, and D. Tilman. 1987. Little bluestem litter dynamics in Minnesota old fields. Oecologia 72:327-330.

Paul, E. A., and F. E. Clark. 1996. Soil microbiology and biochemistry. Academic Press, San Diego, California, USA.

Prescott, C. E. 1995. Does nitrogen availability control rates of litter decomposition in forests? Plant and Soil 168-169:83-88.

Prescott, C. E., J. P. Corbin, and D. Parkinson. 1992. Immobilization and availability of N and P in the forest floors of fertilized Rocky Mountain coniferous forests. Plant and Soil 143:1-10.

Ryan, M. G., J. M. Melillo, and A. Ricca. 1989. A comparison of methods for determining proximate carbon fractions of forest litter. Canadian Journal of Forest Research 20:166-171.

Staaf, H. 1980. Influence of chemical composition, addition of raspberry leaves and nitrogen supply on decomposition rates and dynamics of nitrogen and phosphorus in beech (Fagus sylvatica) leaf litter. Oikos 35:55-62.

Staaf, H., and B. Berg. 1981. Accumulation and release of plant nutrients in decomposing Scots pine needle litter. Long-term decomposition in a Scots pine forest II. Canadian Journal of Botany 60:1561-1568.

Swift, M. J., O. W. Heal, and J. M. Anderson. 1979. Decomposition in terrestrial ecosystems. Blackwell Scientific, Oxford, UK.

Taylor, B. R., D. Parkinson, and W. F. J. Parsons. 1989. Nitrogen and lignin as predictors of litter decay rates: a microcosm test. Ecology 70:97-104.

Theodorou, C., and G. D. Bowen. 1990. Effects of fertilizer on litterfall and nitrogen and phosphorus release from decomposing litter in a Pinus radiata plantation. Forest Ecology and Management 32:87-102.

Titus, B. D., and D. C. Malcolm. 1987. The effect of fertilization on litter decomposition in clearfelled spruce stands. Plant and Soil 100:297-322.

Torn, M. S., S. E. Trumbore, O. A. Chadwick, P. M. Vitousek, and D. M. Hendricks. 1997. Mineral control of soil organic carbon storage and turnover. Nature 389:170-173.

Treseder, K. K. 1999. Plant-soil interactions across a fertility gradient in Hawaii: nutrient acquisition strategies and effects of genetic variation on ecosystem function. Dissertation. Stanford University, Standford, California, USA.

Underwood, A. J. 1997. Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge University Press, Cambridge, UK.

Van Soest, P. J., and R. H. Wine. 1968. Determination of lignin and cellulose in acid detergent fiber with permanganate. Journal of the Association of Official Analytical Chemists 51:780-785.

Van Vuuren, M. M. I., and L. J. Van der Eerden. 1992. Effects of three rates of atmospheric nitrogen deposition enriched with [[blank].sup.15]N on litter decomposition in a heathland. Soil Biology and Biochemistry 24:527-532.

Vitousek, P. M. 1982. Nutrient cycling and nutrient use efficiency. American Naturalist 119:553-572.

Vitousek, P. M., and H. Farrington. 1997. Nutrient limitation and soil development: experimental test of a biogeochemical theory. Biogeochemistry 37:63-75.

Vitousek, P. M., L. R. Walker, L. D. Whiteaker, and P. A. Matson. 1993. Nutrient limitation to plant growth during primary succession in Hawaii Volcanoes National Park. Biogeochemistry 23:197-215.
COPYRIGHT 2000 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:HOBBIE, SARAH E.; VITOUSEK, PETER M.
Publication:Ecology
Geographic Code:1U9HI
Date:Jul 1, 2000
Words:6115
Previous Article:MICROBIAL IMMOBILIZATION AND THE RETENTION OF ANTHROPOGENIC NITRATE IN A NORTHERN HARDWOOD FOREST.
Next Article:NITROGEN TURNOVER IN FOREST FLOORS OF COASTAL DOUGLAS-FIR AT SITES DIFFERING IN SOIL NITROGEN CAPITAL.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters