SCOPE 56 - Global Change: Effects on Coniferous Forests and Grasslands

10.1 INTRODUCTION 

10.1.1 Soil organic matter as a C reservoir

Detrital inputs to world soils are estimated at about 60 x 10¹⁵ g yr^-1 and decomposition at 50-60 x 10¹⁵ g yr^-1. There is no a priori reason to suspect that litter and soil organic matter (SOM) are in a steady-state condition at present, given the large changes in land use patterns that continue to occur at a global scale and the continuous input of C to the global cycle from fossil fuel burning. Schlesinger(1990) analysed chronosequence studies for soil C accumulation rate, and concluded that, on a world-wide basis, only approximately 0.4 x 10¹⁵ g C yr^{- 1}, or less than 1% of primary production, escapes decomposition and is added to a soil C store of 1500 x 10¹⁵ g. However, erosion balances this influx of soil C. Given the uncertainties associated with C balances on a global scale, let alone on a plot scale, an input-output imbalance of 5-10% could easily escape detection, and imbalances of this magnitude would make soils either a sizeable source or sink for C. It is therefore important to determine first of all if soils are a net source or a net sink of C, and, if possible, to make some estimate as to the magnitude of the imbalance between inputs and outputs of C to the soil.

10.1.2 Soil organic matter as a nutrient reservoir

The major part of the N in biological circulation is bound to living or dead organic matter. Soderlund and Svensson (1976) estimate that the soil contains 300 x 10¹⁵ g of N in organic form versus 16 x 10¹⁵ g as insoluble inorganic N. They do not provide any estimates of soluble inorganic N but it should be negligible; Burns and Hardy (1975) assume it to be 1% of the insoluble. Post et al. (1985), based on direct soil samples, give a lower value, 95 x 10¹⁵ g, for the total soil N store. However, this change in total N should not affect the relative distribution between N forms. There is also a considerable variability in N storage between life zones with a range from l00 to 2800 g m^-2 to a depth of 1 m. However, much of this variability is associated with differences in SOM storage such that the range of C/N values is from 10 to 30.

Dalal (1977) has reviewed the occurrence of soil organic P and concludes that the organic P constitutes 20-80% of the total P in the surface layer of the soil, although both higher and lower values have been observed. The importance of the organic P to the turnover of P can be greater than indicated by just its relative contribution to the total soil P because in the soil water solution around 80% of the P is organic but with some variation depending on soil type (Pierre and Parker 1927).

As with N and P, S in soils is mainly in organic form. Biederbeck (1978) states that 'well over 90% of the total S in most noncalcareous surface soils is present in organic form'. Scott (1985) goes even further: 'It is well established that 95% or more of the total S in most soils from humid and semi-humid areas is organic. The C:S (as the C:N:S) ratio is also bounded to rather narrow limits, between 70 and 300 (Biederbeck 1978).

Soil organic matter also has a major influence upon site productivity because of its effects upon physical (bulk density, water holding capacity), biological (microbial populations) and chemical (cations exchange capacity) properties of soils (Chen and Aviad 1990). The cation exchange capacity (CEC) of organic matter arises primarily from H+ dissociation from those carboxyl and phenolic hydroxyl groups, which are acidic enough to dissociate at normal soil pH (Coleman and Thomas 1967). The CEC of organic matter is therefore strongly pH-dependent, in contrast to the CEC of clay minerals, which is derived from cationic substitution (e.g. Al for Si) within the lattice structure of the clays. Thus, organic matter represents a potential cation exchanger for the whole soil, especially in view of the fact that the CEC of organic matter (100-900 meq g^-l soil) is generally much greater than that of clays (5-150 meq g^-l soil).

10.2 ENVIRONMENTAL CONTROLS ON SOM

The environment controls the levels of SOM by regulating both the inputs of organic matter to the soil and the rate of losses. An analysis of observed levels of SOM is therefore, in contrast to a theoretical analysis (see section 10.5) difficult to do in isolation from the plant and animal communities that provide the source of the organic matter.

10.2.1 Temperature and soil moisture controls

There are strong correlations between soil C pools and climate. An extensive analysis of the global variability was made by Post et al. (1982), who concluded that 'Soil carbon density generally increases with increasing precipitation, and there is an increase in soil carbon with decreasing temperature for any particular level of precipitation.' Temperature increases over most normal ranges could increase both inputs to and outputs from the soil C. The analysis by Post et al. (1982) shows that increasing temperatures increase the rate of soil C output more than the input. This implies that the temperature response function for decomposition must be steeper than for production. As the response of SOM to increasing temperature is one of the most important likely changes with climate change, it is important to attempt to understand how SOM is likely to respond.

The effects of climate on decomposition are often expressed as a temperature factor and a moisture factor. The temperature response is generally expressed as a Q₁₀ function where

where k₂ and k₁ are the rate constants at two observed temperatures T₂ and T₁. While this relationship is easy to use and provides a ready indication of the temperature sensitivity of any system, it lacks a theoretical justification. Other relationships, such as the Arrhenius function, have been used (e.g. Ellert and Betany 1992) to provide a better theoretical framework. It is unclear, however, how this should be applied to a system such as a population of soil organisms, where the total activity is determined by the combined activity of a whole range of different organisms with probably quite different individual responses to temperature.

In the absence of a useful theoretical model, the use of Q₁₀ values is a convenient tool to summarize observed responses. It must be recognized, however, that Q₁₀ values are not constant for any system, but change with temperature ( Kirschbaum 1994). This is illustrated in Figure 10.1, which summarises Q₁₀ values for SOM or N mineralization for a range of soils as reported by different workers. Figure 10.1 indicates that Q₁₀'S are generally around 2 for temperatures above 20 °C, but increase at lower temperatures to some very high values. There is a large amount of scatter in the data, part of which may be due to experimental error. Researchers who have compared the temperature responses in different systems, such as different soil types, different soil horizons or different litter types, also generally report differences in the temperature sensitivities of different systems. The figure also shows a widely used temperature dependence of net primary productivity (Lieth 1973). The Q₁₀ for net primary production is generally lower than the Q₁₀ for decomposition. A comparison of the temperature dependencies of decomposition and net primary production therefore suggests that with increasing temperature, decomposition should be stimulated much more than primary production leading to a considerable loss in SOM (Kirschbaum 1994).

Figure 10.1 The Q_l0 of soil or litter respiration or net mineralization as a function of temperature (after Kirschbaum 1994). Shown are experimental data, indicated by different letters, and a least squares fit (solid line). The figure also shows the Q_l0 of net primary productivity according to Lieth (1973) (broken line). Letters refer to: (A) Bunt and Rovira (1955); (B) Drobnik (1962); (C) Flanagan and Veum (1974); (D) Moureaux (1967); (E) Nyhan (1976); (F) O'Connell (1990); (G) Ross and Cairns (1978); (H) Schleser (1982); (I) Waksman and Gerretsen (1931); (J) Wiant (1967)

The effects of soil moisture are also expressed in diverse ways. Various modifications of the precipitation to potential evapotranspiration ratio (P/ PET) used in the Holdridgelife zone classification system provide easily accessible variables. In the CENTURY model this approach is used and parameterized such that for P/PET above1, soil moisture has little effect on decomposition rates. Other alternatives are to calculate directly the soil moisture deficit (e.g.. Jenkinson et al. 1987).

If we take the data by Post et al. (1982) and plot the cumulative fraction of soil C as a function of soil C density (Figure 10.2) we find that approximately 65% of the world's soil C is found in soils with a C density above 10 kg m^-2. In the CENTURY model, the C density of 10 kg m^-2 corresponds approximately to a potential evapotranspiration ratio of I. If we accept this parameterization, then, soil moisture has only a small influence on decomposition rates for a majority of soils. We can thus conclude that, for SOM, precipitation is more important in regulating the inputs to (plant production) than in regulating outputs from (decomposition) the soil.

Figure 10.2Cumulative distribution of soil carbon over carbon densities for the world's different biomes. Based on data in Post et al. (1982)

10.2.2 Biological controls

 10.2.2.1 Productivity

The discussion in the previous section shows that the understanding of SOM is inseparable from understanding production. Lashof (1989) compiled data on production and C storage in the world's major biomes. We have plotted these data in Figure 10.3. If the tundras are excluded, net primary production explains over 60% of the variation in soil C storage, A primary determinant of soil C storage appears therefore to be the production of the system. However, it is difficult to isolate productivity from decomposition as these two factors often depend in the same fashion on underlying environmental controls. There is also a large amount of small-scale variability within any biome (cf. section 10.2.2.2).

10.2.2.2 Litter quality

Differences in decomposition rates of litters of different origins have been extensively studied (e.g. Aber et al. 1984; Berg et al. 1986, 1991a, b; Harmon et al. 1986). The studies of the long-term consequences of additions of different litters are fewer. In one of these, six different types of organic materials were added to an agricultural soil in central Sweden in equal amounts every second year from 1956 to 1991 ( Kirchmann et al. 1994). Depending on the type of material added, between 17 and 74% of the added material was estimated to remain in 1991 (Table 10.1) (Hyvonen et al. 1996).

Figure 10.3 Relation between soil carbon store and net primary production for different biomes. Solid line with all biomes included (r² = 0.30), broken line with tundra (solid symbol) excluded (r² =0.62). Based on data in Lashof (1989)

Table 10.1 Total amounts of C added and remaining with different types of added substrates in some of the treatments of the Ultuna long-term SOM experiment (from Hyvonen et al. 1996)

The same variability is found in e.g. the studies by McClaugherty et al. (1985) on litter dynamics and SOM formation in different forest types on Blackhawk Island, Wisconsin. Although aboveground litter production was approximately equal in three deciduous stands and a white pine stand, the forest floor of the latter contained twice as much organic matter. Likewise, a hemlock stand in which the litterfall was 40% of the litterfall in the deciduous stands had 60% more organic matter in the forest floor.

10.2.2.3 Decomposer community

It is generally recognized that the composition of the decomposer community influences decomposition rates. In particular, soil animals can accelerate the decomposition rate either directly by modifying the physicochemical environment or indirectly by stimulating microbial activity through grazing (Anderson and Ineson 1983, 1984). Quantitatively, less is known about the effects and variation with substrate but such effects are likely. For example, Couteaux et al. (1991b) found in decomposition experiments with chestnut leaves from plants grown in elevated CO₂ (up to 700 ppm) that different combinations of soil fauna could produce a fourfold difference in evolved CO₂. With litter from plants grown in ambient CO₂ only a 50% difference was observed. However, Setala and Huhta (1990) found much smaller faunal effects with birch and spruce litters, and during the later stage of their experiment, decomposition rates were lower with fauna present than without. This might be the result of a decline in litter quality due to the more rapid initial decomposition with fauna present. The decomposition studies on Blackhawk Island by Aber et al. (1984) show that litter quality dominates over differences in soil community, (Figure 10.4) (see also McClaugherty et al. 1985). There are no striking obvious differences in decomposition rate due to litter bag type (exclusion of fauna of different sizes) or placement in the soil. It is possible that there is a difference in decomposition rate between the different ecosystems, with the sugar maple stand having a higher rate than the other ones, but, again, the differences are small compared to the differences between the litter types. One could therefore expect that the diversity of the soil microbial community in natural systems is large enough to be able to cope with different litters almost similarly in any system.

10.2.3 Fire

The effects of fire on SOM are highly complex and differ greatly for different sites, with different soils, climates, species composition, time scales and for different nutrients. Here, we can only deal with fire in a brief and most general manner. For more detailed accounts, see Viro (1974), Walker et al. (1986), and Raison et al. (1989).

Fire potentially has three effects on SOM, and we will deal with them in turn: 

(a) it causes a short-term loss of C due to direct combustion and indirectly due to changes in microclimate that enhance decomposition rate;

(b) it causes a short- to medium-term loss of N and a long-term loss of p and other nutrients;

(c) it may change species composition of a site that may restore lost N.

Figure 10.4 Fraction of different litters remaining after 732 d of decomposition in different environments. Open symbols: in sugar maple stand. Filled symbols: in the stand of the origin of the litter. (�): fine-meshed litter-bags on the forest floor; (D): fine-meshed litter-bags in the mineral soil; (Ñ): coarse-meshed litter-bags on the forest floor; (RMW) red maple wood; (WPW) white pine wood; (SML) sugar maple leaves; (AL) aspen leaves; (WOL) white oak leaves; (WPN) white pine needles; (HN) hemlock needles; (ROL) red oak leaves; (HB) hemlock bark; (RMB) red maple bark; (WPR) white pine roots; (SMR) sugar maple roots (from Aber et at. 1984)

Fire leads to the direct combustion of large amounts of C above the soil surface. The combustion may be almost complete for dry grasslands or consume only a small fraction of potentially available C, such as in a low-intensity burn in an established and fire resistant forest (Walker et al. 1986). Intense fires may also burn part of the organic matter that is already in the soil (Walker et al. 1986), although soils tend to heat only very slowly, and even very intense fires are unlikely to directly affect the SOM to a depth greater than a few centimetres (Aston and Gill 1976; Humphreys and Craig 1981).

Of great importance are also post-fire changes to the soil microclimate. After the canopy and understory have been burned off, the soil is no longer protected by shading and much more sunlight may reach the soil surface. This can raise both maximum and minimum soil temperatures by several degrees for one or more years (Armson 1979). Viro (1974) reports an increase in mean summer temperature in the year after the fire in the humus layer from 18 °C under the unburned tree canopy to 31°C in the open. With a loss of leaves from vegetation, the water transpiring capacity is also reduced; consequently, more water is retained in the soil profile. Hence, for one to several years after a fire, soils may be warmer and wetter than in the unburned state, which may greatly increase the rate of organic matter decomposition. However, the decrease in water-holding capacity due to loss of organic matter together with the increased soil temperature will increase evaporation from the soil (Pritchett 1979).

While fire clearly leads to a loss of soil C in the short term, the longer-term C dynamics are principally controlled by the soil's nutrient economy. During fires, N compounds readily volatilize so that large quantities may be lost (Raison et al. 1985). Viro (1974) estimates the average loss of N due to burning after clear-felling to be 320 kg ha^-1 with about 180 kg coming from the slash and 140 kg from the humus layer. Phosphorus compounds, on the other hand, are generally less volatile and a greater proportion of P is likely to be retained within the system (Raison et al. 1985), although ash deposited on the soil surface may readily be lost due to erosion (Raison et al. 1989). Another consequence of fires can be a substantial increase in soil pH on acid soils, up to 2-3 pH units (Viro, 1974). This increase in soil pH can enhance N fixation and mineralization (Johansson 1984) and may increase the amount of available nutrients (Nykvist 1977). If the nutrients associated with the lost C become available to plants and promote greater C gain, then, after a number of years the soil C to nutrient ratios would return to the values characteristic for the site and the same soil C content as before the occurrence of fire would be re-established. The direct loss of C due to fires would then have few long-term consequences. In the long run, fire affects the amount of SOM primarily through the loss of nutrients that cannot be replenished by the ecosystem. Hence, the loss of P, or other nutrients if they are the primary limiting element, looms large in determining the long-term soil C dynamics of the system.

If fire frequency changes due to climate change it may have some short-term consequences for soil C storage. In the medium term, soil C storage is primarily controlled by the system's N economy. If the system is given sufficient time to recover after each fire episode, its overall N content can recover to pre-fire values, but very frequent fire intervals could prevent that. In the longer term, the movement of nutrients other than N are likely to be the most rate-limiting. In regions where fire frequency and intensity may increase, more nutrients may be lost and overall productivity and soil C storage may eventually decrease. The converse may be true for regions with fewer fires.

10.2.4 Edaphic and physical factors

In addition to temperature, moisture, and litter quality, there are several important edaphic and physical factors that strongly influence the accumulation of organic matter in soils. Chief among these are clay content, base status, and Fe and Al hydrous oxide content (Oades 1988; Anderson 1992). Intercalation of organic matter in smectite clays and carbohydrate adsorption onto clay surfaces can result in substantial SOM stabilization (Oades 1988; Anderson 1992). However, the often noted positive correlation between clay and organic matter content in soils may not represent a true cause-effect relationship because clay content is usually highly correlated with other factors that influence SOM accumulation such as polyvalent cations (Oades 1988). These cations stabilize SOM by a number of mechanisms, including electrostatic cation bridging of organic colloids (Ca and Mg in alkaline and circumneutral soils; Al in acidic soils) and specific adsorption onto Fe and Al hydrous oxide surfaces (acidic soils) (Oades 1988). Because Ca is seldom limiting to terrestrial plant productivity, the often noted long-term positive effects of liming on SOM (e.g. Gilmore 1980; Jenkinson 1991; see also review by Johnson 1992a) are likely due to these reactions rather than a direct effect upon plant primary productivity.

In addition to polyvalent cations, the effects of non-biological reactions between N and organic matter on SOM stabilization deserve mention here. Non-biological condensation reactions of phenols with ammonium and amino compounds are important in the production of humus (Mortland and Wolcott 1965; Paul and Clark 1989; see also review by Johnson 1992b). These reactions are enhanced by high pH (because NH₃ is the reactive form of N) and high NH₃ and/or NH₄⁺ concentrations. The reaction can occur slowly at pH below neutrality, also (Mortland and Wolcott 1965), but the importance of non-biological reactions in forest soil or litter N retention under ambient pH conditions is unclear. Nommik (1970) found little non-biological NH₄⁺ retention in acid Norway spruce humus unless pH was raised to neutrality or higher. However, Schimel and Firestone (1989) found that non-biological reactions accounted for 20% of the N retained in an acid (pH 4.3-4.5) forest soil.

10.3 OBSERVED PATTERNS OF SOM

Long- term studies of SOM have mostly been done on agricultural soils. The most well-known of these is the 150-year-long Rothamsted experiments (e.g. Jenkinson 1991). A similar study, but of only a 35-year duration and on a smaller spatial scale, is the Ultuna long-term SOM experiment (Kirchmann et al. 1994). These studies show that the loss of C from old agricultural sites can be very slow; plots that were left bare lost annually on average 0.4% at Rothamsted over a 120-year period (Jenkinson and Rayner 1977) and 1.1% at Ultuna over a 35-year period, respectively. The slower rate at Rothamsted might be a result of the longer observation period and a decreasing rate with time. When allowed to return to deciduous woodlands, two Rothamsted areas increased their soil C content by 65 and 165%, respectively, in less than 100 years (Jenkinson 1991). Depending on type, the amounts of litter required to maintain these soils at a steady-state level vary between 1 and 3 Mg C ha^-1 (Jenkinson et al. 1992).

There seem to be no similar studies of forest soils. The effects of forest management on soil C storage was recently reviewed by Johnson (1992c) showing that C storage does not decrease due to forest harvesting unless it is followed by intense burning or cultivation.

10.4 MODELLING SOM DYNAMICS

There exists at present a number of different approaches to summarize and explain the observations of SOM dynamics discussed above. They differ mainly in the degree to which they resolve the SOM into components of different properties but also in the way they handle external factors, e.g. climate and soil texture.

10.4.1 Single pool models

The simplest approach is found in Olson's (1963) classic paper which combines a rate of litterfall, I_o with a loss rate constant, k, to calculate the change in stored C:

(2)

or in integrated form

In combination with estimates of litterfall and decomposition constants (k's) under different climatic conditions (e.g. Bray and Gorham 1964; Olson 1963), it is simple to calculate changes in the detrital pool under given climatic scenarios. This approach is extremely simple, but at the expense of lumping all information into a single parameter, which limits the possibilities of generalization. At steady state, equation (3) is formally equal to equation (11) (see section 10.4.3) but the term 1/k is replaced by a richer expression allowing more insights into the mechanisms exerting control over the decomposition rate. More complicated versions of the single pool model that take into account changes in litter quality with time have also been developed (e.g. Janssen 1984; Godshalk and Wetzel 1978).

10.4.2 Multiple pool models

As an alternative to letting all SOM be combined into one pool, other authors have divided it into several fractions characterized by different decomposition rates. The rationale for this is that there are differences in chemical structure and physical protection.

Using a matrix notation, all these models can be written in the format

where C is a vector representing the amounts of C in the different pools, T is a transition matrix describing the fluxes of C between the pools, and I a vector containing the litter inputs. To simplify the comparisons and stress the qualitative differences among the models as they appear in the topology of the models, we will only indicate by -, 0, or + in the transition matrix whether a certain matrix element causes a decrease in a pool, has no effect or increases the pool. We will contrast three different approaches of this class (many of these models undergo continuous revisions and exist therefore in several versions with more or less substantial differences), all used to describe long-term carbon accumulation under grasslands. The three are:

(a) CENTURY (Parton et al. 1987; see also Chapter 11):

 

(5)

   

 (b) Rothamsted model (Jenkinson and Rayner 1977; Jenkinson et al. 1987; see also Jenkinson 1990; Jenkinson et al. 1992, for a newer version that is close to CENTURY):

      

(6)

(c) V&P (van Veen and Paul 1981):

Although the models are structurally quite similar with entering litter divided into two classes of decomposability and the microbial biomass (active soil C in the terminology of CENTURY) as the wheel around which everything turns, there are also clear differences. On the one hand, the reason for differentiating the soil C into the pools differs. In CENTURY, there is no explicit physical protection of SOM but the division into slow and passive soil C is based on both soil texture and SOM chemical properties. The Rothamsted and the V&P model both make the corresponding division by explicitly distinguishing between physical and chemical stabilization. The differences are also apparent in the turnover rates associated with the different pools (Table 10.2).

10.4.3 Continuous quality models

As an alternative to the over-simplification in the Olson model and a way of avoiding the ambiguities arising when dividing SOM into a small number of pools, Bosatta and Ågren (1985, 1991a, b, 1994; Ågren and Bosatta 1987,1988) have suggested that one could assume that there exists an infinite number of 'pools'. Thus, rather than assigning a certain C compound to a discrete C fraction, the compound is described by a continuous variable, q, named 'quality' (the CQ-model). Quality is a measure of the degradability of the compound as measured by the microbial growth rate for microbes feeding on that particular compound. Carbon in the soil is then described by a continuous

Table 10.2 Turnover times for the different pools of soil C. The values are only approximate as in all models climatic variables regulate the decomposition rates

distribution, such that rc(q, t)dq is the amount of C in the quality interval [q, q + dq] at time t.

To formulate the interaction between the soil C and the decomposing microbial community, the latter is described with three functions: e(q), the microbial efficiency or production-to-assimilation ratio which tells how much soil C is lost in producing new microbial biomass; u(q), the microbial growth rate; and D(q, q'), the dispersion function, which determines the fraction of C assimilated as quality q' that is returned to the soil C as quality q. The dispersion function plays the role of the off-diagonal elements in the matrix T in equation (4). In addition, the C concentration in the microbial biomass, f_c, is needed. The requirement of mass balance for C then leads to the following equation:

where I(q, t) is the input of C. This equation can be analysed at different levels of approximation depending upon the detail of information used to specify the three functions e(q), u(q), and D(q,q'). Equation (8) has also been modified to explicitly describe the microbial biomass (Bosatta and Ågren 1994). Decomposition of marine sediments has also been analysed by a similar approach (Boudreau and Ruddick 1991).

10.4.4 Comparison of the modelling approaches

Using the CQ-approach, it is easy to see how the various modelling approaches differ. The introduction of discrete classes of C is then simply a way of approximating the continuous spectrum with a set of discrete variables; the extreme case is Olson's model with a single variable. The effects of physical soil structure is another feature differentiating the models. In the CQ-model, the microbial growth rate on different qualities is differentially affected. In the CENTURY model, the way the C quality spectrum is discretized depends on soil texture. The Rothamsted model and the V&P model, on the other hand, both assume the existence of two C quality spectra, one for physically protected soil C and one for soil C that is not physically unprotected. The discretization of these spectra is fixed, but soil texture shifts C between them. The latest version of CENTURY (Chapter 11) has added a spatial dimension by separating surface litter from SOM.

10.5 EXPECTED EFFECTS OF CLIMATE CHANGE

The expected changes in SOM as a result of a climate change can be direct through some of the mechanisms discussed above or indirect through the changes in litter inputs. Anderson (1992) has written an extensive review of the effects of climate change on soils. In addition to what has already been discussed, he notes that organic matter turnover becomes more superficial in the profile and divorced from SOM as soils become colder. Under cold conditions, SOM accumulation is a function of litter quality, there is little stabilization of SOM (e.g. by reactions with clays and polyvalent cations), and SOM is generally younger. Presumably the reverse is true as soils warm. Changes in soil physical structure through the suppression of freezing-thawing cycles could also alter the decomposition rates.

With respect to changes in plant properties, the most difficult to predict are those associated with changes in plant chemical properties. Some investigators have noted substantially decreased tissue N concentrations with elevated CO₂ concentrations (e.g. Strain 1985; Norby et al.1986a, b; see also review by Johnson and Ball 1991). This litter quality-decomposition feedback is also known to occur as stands age due to increasing returns of a very high C: N ratio in woody litterfall following crown closure (Switzer and Nelson 1972; Turner 1981; Miller 1981; Johnson and Ball1991)and could therefore occur at an earlier stand age if climate change were to increase production rates.

At the moment, the picture with respect to CO₂-litter quality-decomposition is somewhat confused. Based upon lignin:N and lignin:P ratios in white oak (Querqus alba) leaves, Norby et al. (1986b)conclude that rates of litter decomposition will not be greatly affected by CO₂. Couteaux et al. (1991a) found that CO₂ enrichment caused increased litter C/N ratios but no differences in lignin, cellulose + hemicellulose, or hydrosoluble compounds in chestnut leaves. During subsequent decomposition tests of the litter, the authors noted two phases of CO₂ effects. Initially, litter quality effects dominate in that microbial respiration from untreated (ambient CO₂) leaves was greater than in treated (elevated CO₂) leaves. During the second period, the pattern reversed because of differences in organism composition (increased presence of lignin decomposers in the treated litter).

We can now elucidate some possible consequences of climate change for SOM stores by using the continuous quality approach (section 10.4.3). Although the other theories and models could have been used (e.g. Schimel et al. 1990; Jenkinson et al. 1991; Kohlmaier et al. 1991; Thornley et al. 1991; Kirschbaum 1993), we believe that this theory is the most convenient in this context as it summarizes in a very compact form. a range of the features that have been discussed above. We will restrict the analysis to the soil only, neglecting feedbacks through the plant system.

For simplicity we will consider a system at steady state. Hence, there is a constant flux of C, I_o, of some organic matter fraction to the soil. The organic matter will start at some initial quality, qo but as it is used by the microbes as an energy source, it will decrease in quality on average for each cycle through the microbial community by an amount h₁ (h₁ is an approximation for D(q, q')). At each cycle a fraction, 1 -e(q), of the C is also lost as respiration, where e(q) is the microbial efficiency (production-to-assimilation ratio). We assume that this efficiency can be described as

where e₁ is a parameter. Similarly, the microbial growth rate, u(q), is assumed to depend on the substrate quality as

where Uo is a base growth rate and b a shape parameter which measures the degree of physical protection of the substrate. We can then show that we will have a steady-state C store, C_ss of (Ågren and Bosatta 1987)  

(11)

The steady-state store of C derived from one type of original organic matter can thus be expressed in terms of the seven parameters required to define equation (11), all of which have well-defined biological meanings. We can measure I_o and f_c directly, whereas the other parameters at the moment need to be estimated from decomposition curves for each particular type of litter.

We can now use equation (11) to calculate, by changing the parameters, which category of processes can have the largest impact on soil C storage. To compare the importance of the different parameters, we define a relative sensitivity, S_p to a parameter p as

For three of the parameters (I_o ,f_c , and u_o ) the outcome of the sensitivity analysis is trivial because the steady-state C store is proportional (or inversely proportional) to the parameter. The effects of the other four parameters will depend upon where in the parameter space the analysis is performed. As an example, we will choose a set of parameters for Scots pine needles decomposing in a forest stand in central Sweden, q_o = 1,  h₁= 0.893, e₁ =0.187, b= 3 (Ågren and Bosatta 1987). We then get the results shown in Table 10.3.

For this type of system, the analyses show that, for equal relative changes in the parameters, litter quality (q_o) and microbial efficiency (e₁) are most important. Changes in microbial growth rate (climatic factors operating through (u_o) rank equal to changes in litter production rate and changes in the growth rate--quality curve (b) come next. The average change in quality for each decomposition cycle (h₁) has least importance, which is fortunate as this parameter is the most difficult to estimate.

To see how representative a Scots pine forest in central Sweden is for other types of soils, we have performed a Monte Carlo simulation where we have created 1000 artificial soils by randomly assigning parameter values to q_o h₁e₁

Table 10.3 Relative sensitivity of steady-state soil C stores to the parameters defining the steady-state value

Table 10.4 Results from a Monte Carlo simulation of the relative sensitivity of the C store to the parameters defining the steady-state value. One thousand different parameter values have been used in the ranges given in the table. Mean, maximum, minimum values, and standard deviations of the relative sensitivity are shown

and b from a uniform distribution and then calculated the relative sensitivities. Ranges for the parameters were selected on the basis of the discussions in section 10.2 to be what we consider ecologically relevant. We have also taken into account restrictions on parameter combinations, e.g. e_lq_o<1. The results are given in Table 10.4.

The mean sensitivities in this Monte Carlo simulation are close to what we found for the pine forest. This might, however, only reflect the ranges chosen for the parameters. The ranges and variability in the sensitivities are, however, large. In some cases we can even have a change in the sign of the sensitivity. We have also looked for patterns in the sensitivities, but have not detected any. We do not know to what extent the parameters we are using should be correlated in real systems, e.g. a high litter quality implies efficient microbes, which implies additional constraints on parameter combinations. Some of the extreme sensitivities can therefore be the consequence of unreal parameter combinations, but we should be aware that there might be large differences, both quantitatively and qualitatively, between different ecosystems. These differences are probably more important to understand than deviations from the presumed steady state.

In conclusion, it does not seem possible at the moment to predict the balance between the forces that increase SOM accumulation under climate change and those that decrease accumulation. The comparisons of forest models (Chapter 13) show that different ways of integrating our current understanding lead to predictions that even differ in the signs of the predictions, let alone the magnitudes. At the same time, the model analyses of the different grassland sites (Chapter 13) show that there are large differences between sites with different current climates.

10.6 REFERENCES

Aber, J. D., McClaugherty, C. A. and Melillo, J. M. (1984) Litter decomposition in Wisconsin forests-mass loss, organic-chemical constituents and nitrogen. School of Natural Resources, University of Wisconsin-Madison. Report R3284.

Ågren, G. I. and Bosatta, E. (1987) Theoretical analysis of the long-term dynamics of carbon and nitrogen in soils. Ecology 68,1181-1189.

Ågren, G. I. and Bosatta, E. (1988) Nitrogen saturation of terrestrial ecosystems. Environ. Pollut. 54, 185-197.

Anderson, J. M. (1992) Responses of soils to climate change. Adv. Ecol. Res. 22, 163-210. 

Anderson, J. M. and Ineson, P. (1983) Interaction between soil arthropods and micro-organisms in carbon, nitrogen and mineral element fluxes from decomposing leaf litter. In: Lee, J.A., McNeill, S. and Robinson, I. H. (Eds) Nitrogen asan Ecological Factor, pp. 413-432. Blackwell, Oxford.

Anderson, J. M. and Ineson, P. (1984) Interaction between microorganisms and soil invertebrates in nutrient flux pathways of forest ecosystems. In: Anderson, J. M., Rayner, A. D. and Walton, D. W. H. (Eds) Invertebrate-Microbial Interactions, pp. 59-88. Cambridge University Press, Cambridge.

Armson, K. A. (1979) Forest Soils: Properties and Processes. University of Toronto Press, Toronto.

Aston, A. R. and Gill, A. M. (1976) Coupled soil moisture, heat and water vapour transfer under simulated fire conditions. Aust. J. Soil Res. 14, 55-66.

Berg, B., Booltink, H., Breymeyer, A., Ewertsson, A., Gallardo, A., Holm, B., Johansson, M.-B., Koivuoja, S., Meentemeyer, V., Nyman, P., Olofsson, J., Reurslag, A., Staaf, H., Staaf, I. and Uba, L. (1991a) Data on needle litter decomposition and soil climate as well as site characteristics for some coniferous forest sites. Part 1. Site characteristics. Dep. Ecol. Environ. Res., Swed. Univ. Agric. Sci. Rep.41. Uppsala.

Berg, B., Booltink, H., Breymeyer, A., Ewertsson, A., Gallardo, A., Holm, B., Johansson, M.-B., Koivuoja, S., Meentemeyer, V., Nyman, P., Olofsson, J., Reurslag, A., Staaf, H., Staaf, I. and Uba, L. (1991b) Data on needle litter decomposition and soil climate as well as site characteristics for some coniferous forest sites. Part 2. Decomposition data. Dep. Ecol. Environ. Res., Swed Univ. Agric. Sci. Rep. 42. Uppsala.

Berg, B., Ewertsson, A., Ferenczi, J., Holm, B., Jansson, P.-E., Kuivuoja, S., Meentemeyer, V., Nyman, P., Staaf, H. and Staaf, I. (1986) Data on needle litter decomposition and soil climate as well as site characteristics for some coniferous forest sites. Dep. Ecol. Environ. Res., Swed. Univ. Agric. Sci. Rep. 22. Uppsala.

Biederbeck, V. O. (1978) Soil organic sulfur and fertility. In: Schnitzer, M. and Khan, S. U. (Eds) Soil Organic Matter. Developments in Soil Science 8, pp. 273-310. Elsevier, Amsterdam.

Bosatta, E. and Agren, G. I. (1985) Theoretical analysis of decomposition of heterogeneous substrates, Soil Biol. Biochem. 17, 601-610.

Bosatta, E. and Agren, G. I. (1991a) Dynamics of carbon and nitrogen in the organic matter of the soil: a generic theory. Am. Nat. 138, 227-245.

Bosatta, E. and Agren, G. I. (1991b) Theoretical analysis of carbon and nutrient interactions in soils under energy-limited conditions. Soil Sci. Soc. Am. J. 55, 728- 733.

Bosatta, E. and Agren, G.I. (1994) Theoretical analysis of microbial biomass dynamics in soils. Soil Biol. Biochem. 26,143-148.

Boudreau, B. P. and Ruddick, B. R. (1991) On a reactive continuum representation of organic matter diagenesis. Am. J. Sci. 291, 507-538.

Bray, L. R. and Gorham, E. (1964) Litter production in forests of the world. Adv. Ecol. Res. 2, 101-157.

Bunt, J. S. and Rovira, A. D. (1955) The effect of temperature and heat treatment on soil metabolism. J. Soil Sci. 6, 129-136.

Burns, R. C. and Hardy, R. W. F. (1975) Nitrogen Fixation in Bacteria and Higher Plants. Springer-Verlag, Berlin, Heidelberg, New York.

Chen, Y. and Aviad, T. (1990) Effects of humic substances on plant growth. In: McCarthy, P., Clapp, C. E., Malcolm, R. L. and Bloom, P. R. (Eds) Humic Substances in Soil and Crop Sciences: Selected Readings, pp. 161-186. Am. Soc. Agron., Madison, Wis.

Coleman, N.T. and Thomas, G. W. (1967) The basic chemistry of soil acidity. In: Pearson, R. W. and Adams, F. (Eds) Soil Acidity and Liming, pp.1-41. Am. Soc. Agron., Madison, Wis.

Coûteaux, M.-M., Bottner, P., Rouhier, H. and Billes, G. (1991a) Atmospheric CO₂ increase and plant material quality: production, nitrogen allocation and litter decomposition of sweet chestnut. In: Teller, A., Mathy, P. and Jeffers, J. N. R. (Eds) Responses of Forest Ecosystems to Environmental Changes, pp. 429-436. EIsevier Applied Science, London and New York.

Coûteaux, M.-M., Mousseau, M., Célérier, M.-L. and Bottner, P. (1991b) Increased atmospheric CO₂ and litter quality: decomposition of sweet chestnut leaf litter with animal food webs of different complexities. Oikos 61, 54-64.

Dalal, R. C. (1977) Soil organic phosphorus. Adv. Agron. 29, 83-117.

Drobnik, J. (1962) The effect of temperature on soil respiration. Folia Microbiol. 7,132-140.

Ellert, B. H. and Bettany, J. R. (1992). Temperature dependence of net nitrogen and sulfur mineralization. Soil Sci. Soc. Am. J. 56, 1133-1141.

Flanagan, P. W. and Veum, A. K. (1974) Relationship between respiration, weight loss temperature and moisture in organic residues in tundra. In: Holding, A. J ., Heal, 0. W ., MacLean, S. F. and Flanagan, P. W. (Eds) Soil Organisms and Decomposition in Tundra, pp. 249-277. Tundra Biome Steering Committee, Stockholm, Sweden.

Gilmore, A. R. (1980) Changes in a reforested soil associated with tree species and time. IV. Soil organic content and pH in pine plantations after 24 years. Univ. Illinois Agr. Exp. Sta. For. Res. Rep. No.80-3. Urbana-Champaign.

Godshalk, G. L. and Wetzel, R. G. (1978) Decomposition of aquatic angiosperms. II. Particulate components. Aquat. Bot. 5, 301-327.

Harmon, M. E., Franklin, J. F., Swanson, F. J., Sollins, P., Gregory, S. V., Lattin, J. D., Anderson, N. H., Cline, S. P., Aumen, N. G., Sedell, J. R., Lienkaemper, G. W., Cromack, K. Jr and Cummins, K. W. (1986) Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res. 15, 133-302.

Humphreys, F. R. and Craig, F. G. (1981) Effects of fire on soil chemical, structural and hydrological properties. In: Gill, A. M., Groves, R. H. and Noble, I. R. (Eds) Fire and the Australian Biota, pp. 177-200. Aust. Acad. Science, Canberra, Australia.

Hyvönen, R., Agren, G. I. Land Andren, O. (1996) Modelling long-term carbon and nitrogen dynamics in an arable soil receiving organic matter amendments. Ecol. Appl. (in press).

Janssen, B. H. (1984) A simple model for calculating decomposition and accumulation of 'young' soil organic carbon. Plant Soil 76, 297-304.

Jenkinson, D. S. (1990) The turnover of organic carbon and nitrogen in soil. Philos. Trans. R. Soc. Lond. B329, 361-368.

Jenkinson, D. S. (1991) The Rothamsted long-term experiments: are they still of use? Agron. J. 83,2-10.

Jenkinson, D. S., Adams, D. E. and Wild, A. (1991) Model estimates of CO₂ emissions from soil in response to global warming. Nature 351,304-306.

Jenkinson, D. S., Harkness, D. D., Vance, E. D., Adams, D. E. and Harrison, A. F. (1992) Calculating net primary production and annual input of organic matter to soil from the amount and radiocarbon content of soil organic matter. Soil Biol. Biochem. 24, 295-308.

Jenkinson, D. S., Hart, P. B. S., Rayner, J. H. and Parry, L. C. (1987) Modelling the turnover of organic matter in long-term experiments at Rothamsted. INTECOL Bull. 15, 1-8.

Jenkinson,D. S. and Rayner, J. H.(1977)Theturnover of soil organic matter in some of the Rothamsted classical experiments. Soil Sci. 123, 298-305.

Johansson, M.-B. (1984) Litter decomposition rate at burned and clear-felled areas compared to closed stands. Dep. For. Soils, Swed. Univ. Agric. Sci., Report 49. Uppsala.

Johnson, D. W. (1992a) Effects of forest management on soil carbon storage. Water Air Soil Pollut. 64, 83-120.

Johnson, D. W. (1992b) Nitrogen retention in forest soils. J. Environ. Qual. 21, 1-12. 

Johnson, D. W. (1992c) Effects of forest management on soil carbon storage. NCASI Tech. Bull. No.628. New York.

Johnson, D. W. and Ball, J. T. (1991) Environmental pollution and impacts on soils and forests of North America. Water Air Soil Pollut. 54, 3-20.

Kirchmann, H., Persson, J. and Carlgren, K. (1994) The Ultuna long-term soil organic matter experiment, 1956-1991. Dep. Soil Sci., Swed. Univ. Agric. Sci., Rep. Diss. 17, 55 pp. Uppsala.

Kirschbaum, M. U. F. (1993) A modelling study of the effects of changes in atmospheric CO₂ concentration, temperature and atmospheric nitrogen input on soil organic carbon storage. Tellus 45B, 321-334.

Kirschbaum, M. U. F. (1994) The temperature dependence of soil organic matter decomposition and the effect of global warming on soil organic carbon storage. Soil Biol Biochem. (in press).

Kohlmaier, G. H., Lüdeke, M., Janecek, A., Benderoth, G., Kindermann, J. and Klaudius, A. (1991) Land biota, source or sink of atmospheric carbon dioxide: positive and negative feedbacks within a changing climate and land use development. In: Schneider, S.H. and Boston, P.J. (Eds) Scientists on Gaia, pp. 223-239. MIT Press, Cambridge, Mass.

Lashof, D. A. (1989) The dynamic greenhouse: feedback processes that may influence future concentrations of atmospheric trace gases and climatic change. Clim. Change 14, 213-242.

Lieth, H. (1973) Primary production: terrestrial ecosystems. Hum. Ecol. 1, 303-332. 

McClaugherty, C. A., Pastor, J., Aber, J. D. and Melillo, J. M. (1985) Forest litter decomposition in relation to soil nitrogen dynamics and litter quality. Ecology 66, 266-275.

Miller, H. G. (1981) Forest fertilization: some guiding concepts. Forestry 54, 157-167.

Mortland, M. M. and Wolcott, A. R. (1965) Sorption of inorganic nitrogen compounds by soil materials. In: Bartholomew, W. V. and Clark, F. E. (Eds) Soil Nitrogen, pp. 150--197. Agron. 10. Amer. Soc. Agron., Madison, Wis.

Moureaux, C. (1967) Influence de la temperature et de l'humidite sur les activites biologiques de quelques sols ouest-africains. Cah. Pedol. Orstom 5, 393-420.

Nömmik, H. (1970) Non-exchangeable binding of ammonium and amino nitrogen by Norway spruce raw humus. Plant Soi1 33, 581-595.

Norby, R. J., O'Neill, E. G. and Luxmoore, R. J. (1986a) Effects of atmospheric CO₂ enrichment on the growth. and mineral nutrition of Querqus alba seedlings in nutrient-poor soil. Plant Physiol. 82, 83-89.

Norby, R. J., Pastor, J. and Melillo, J. M. (1986b) Carbon-nitrogen interactions in CO₂-enriched white oak: physiological and long-term perspectives. Tree Physiol. 2, 233-241.

Nyhan, J. W. (1976) Influence of soil temperature and water tension on the decomposition rate of carbon-14 labelled herbage. Soil Sci. 121,288-293.

Nykvist, N. (1977) Skoliga åtgärders inverkan på storlek och tillgänglighet av ekosystemets näringsförråd. Sver. skogsvårdsförb. tidskr. 2-3, 167-178.

Oades, J. M. (1988) The retention of organic matter in soils. Biogeochemistry 5,35-70. 

O'Connell, A. M. ( 1990) Microbial decomposition (respiration) of litter in eucalypt forests of south-western Australia: an empirical model based on laboratory incubations. Soil Biol. Biochem. 22, 153-160.

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

Parton, W. J., Schimel, D. S., Cole, C. V. and Ojima, D. S. (1987) Analysis of factors controlling soil organic matter levels in Great Plains grasslands. Soil Sci. Soc. Am. J. 51, 1173-1179.

Paul, E. A. and Clark, F. E. (1989) Soil Microbiology and Biochemistry. Academic Press, New York.

Pierre, W. H. and Parker, F. W. (1927) Soil phosphorus studies: II. The concentration of organic and inorganic phosphorus in the soil solution and soil extracts and the availability of organic phosphorus to plants. Soil Sci. 24, 119-128.

Post, W. M., Emanuel, W. R., Zinke, P. J. and Stangenberger, A. G. (1982) Soil carbon pools and world life zones. Nature 298, 156-159.

Post, W. M., Pastor, J., Zinke, P. J. and Stangenberger, A. G. (1985) Global patterns of soil nitrogen storage. Nature 317,613-616.

Pritchett, W. L. (1979) Properties and Management of Forest Soils. John Wiley & Sons, New York.

Raison, R. J., Keith, H. and Khanna, P. K. (1989) Effects of fire on the nutrient-supplying capacity of forest soils. In: Dyck, W. J. and Mees, C. A. (Eds) Impact of Intensive Harvesting on Forest Site Productivity, pp. 39-54. IEA/BE T6/A6 Report No.2, Forest Research Institute, Rotorua, New Zealand.

Raison, R. J ., Khanna, P .K. and Woods, P. V. (1985) Mechanisms of element transfer to the atmosphere during vegetation fires. Can. J. For. Res. 15, 132-140.

Ross, D. J. and Cairns, A. (1978) Influence of temperature on biochemical processes in some soils from tussock grasslands. New Zeal. J. Sci. 21, 581-589.

Schimel, D. S., Parton,W. J., Kittel, T.G.F.,Ojima, D. S. and Cole, C. V. (1990) Grassland biogeochemistry: Clim. Change 17, 13-25.

Schimel, J. P. and Firestone, M. K. (1989) Inorganic N incorporation by coniferous forest floor material. Soil Biol. Biochem. 21,41-46.

Schleser, G. H. (1982) The response of CO₂ evolution from soils to global temperature changes. Z. Naturfor. 37a, 287-291.

Schlesinger, W. H. (1990) Evidence from chronosequence studies for a low carbon-storage potential of soils. Nature 348, 232-234.

Scott, N.M. (1985) Sulphur in soils and plants. In: Vaughan, D. and Malcolm, R. E. (Eds) Soil Organic Matter and Biological Activity. Developments in Plant and Soil Sciences Vol. 16, pp. 379-401. Martinus Nijhoff/Dr W. Junk Publishers, Dordrecht.

Setälä, H. and Huhta, V. (1990) Evaluation of the soil fauna impact on decomposition in a simulated coniferous forest soil. Biol. Fertil. Soils 10, 163-169.

Söderlund, R. and Svensson, B. H. (1976) The global nitrogen cycle. In: Svensson, B. H. and Söderlund, R. (Eds) Nitrogen, Phosphorus and Sulphur-Global Cycles. Scope Report 7. Ecol. Bull. 22, 23-73.

Strain, B. R. (1985) Physiological and ecological controls on carbon sequestering in terrestrial ecosystems. Biogeochemistry 1, 219-232.

Switzer, G. L. and Nelson, L. E. (1972) Nutrient accumulation and cycling in loblolly pine (Pinus taeda L.) plantation ecosystems: the first twenty years. Soil Sci. Soc. Am. Proc. 36, 143-147.

Thornley, J. H. M., Fowler, D. and Cannell, M. G. R. (1991) Terrestrial carbon storage resulting from CO₂ and nitrogen fertilization in temperate grasslands. Plant Cell Environ. 14, 1007-1011.

Turner, J. (1981) Nutrient cycling in an age sequence of Western Washington Douglas-fir stands. Ann. Bot. 48, 159-169.

Van Veen, J. A. and Paul, E. A. (1981) Organic carbon dynamics in grassland soils. 1. Background information and computer simulation. Can. J. Soil Sci. 61, 185-201.

Viro, P. J. (1974) Effects of forest fire on soil. In: Kozlowski, T. T. and Ahlgren, C. E. (Eds) Fire and Ecosystems, pp. 7-45. Academic Press, New York.

Waksman, S. A. and Gerretsen, F. C. (1931) Influence of temperature and moisture upon the nature and extent of decomposition of plant residues by microorganisms. Ecology 12, 33-60.

Walker, J., Raison, R. J. and Khanna, P. K. (1986) Fire. In: Russell, J. S. and Isbell, R. F. (Eds) Australian Soils: The Human Impact, pp. 185-216. University of Queensland Press, St Lucia, Australia.

Wiant, H. V. (1967) Influence of temperature on the rate of soil respiration. J. For. 65, 489-490.