SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow An erratum has been published
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (20)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Grant, R.F.
Right arrow Articles by McGill, W.B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Grant, R.F.
Right arrow Articles by McGill, W.B.
Agricola
Right arrow Articles by Grant, R.F.
Right arrow Articles by McGill, W.B.
Soil Science Society of America Journal 65:205-214 (2001)
© 2001 Soil Science Society of America

DIVISION S-6-SOIL & WATER MANAGEMENT & CONSERVATION

Long-Term Changes in Soil Carbon under Different Fertilizer, Manure, and Rotation

Testing the Mathematical Model ecosys with Data from the Breton Plots

R.F. Granta, N.G. Jumaa, J.A. Robertsona, R.C. Izaurraldeb and W.B. McGilla

a Dep. of Renewable Resources, Univ. of Alberta, Edmonton, AB, Canada T6G 2E3
b Pacific Northwest National Lab., 901 D St. S.W., Suite 900, Washington, DC 20024-2115 USA

Corresponding author (robert.grant{at}ualberta.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 THEORY
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil C contents can be raised by land use practices in which rates of C input exceed those of C oxidation. Rates of C inputs to soil can be raised by continuous cropping, especially with perennial legumes, and by soil amendments, especially manure. We have summarized our understanding of the processes by which changes in soil C content are determined by rates of soil C input in the mathematical model ecosys. We compared model output for changes in soil C with those measured in a Gray Luvisol (Typic Cryoboralf) at Breton, Alberta, during 70 yr of a 2-yr wheat (Triticum aestivum L.)–fallow rotation vs. a 5-yr wheat–oat (Avena sativa L.)–barley (Hordeum vulgare L.)–forage–forage rotation with unamended, fertilized, and manured treatments. Model results indicated that rates of C input in the 2-yr rotation were inadequate to maintain soil C in the upper 0.15 m of the soil profile unless manure was added, but that those in the 5-yr rotation were more than adequate. Consequent changes of soil C in the model were corroborated by declines of 14 and 7 g C m-2 yr-1 measured in the control and fertilized treatments of the 2-yr rotation; by gains of 7 g C m-2 yr-1 measured in the manured treatment of the 2-yr rotation; and by gains of 4, 14, and 28 g C m-2 yr-1 measured in the control, fertilized, and manured treatments of the 5-yr rotation. Model results indicated that soil C below 0.15 m declined in all treatments of both rotations, but more so in the 2-yr than in the 5-yr rotation. These declines were corroborated by lower soil C contents measured between 0.15 and 0.40 m after 70 yr in the 2- vs. 5-yr rotation. Land use practices that favor C storage appear to interact positively with each other, so that gains in soil C under one such practice are greater when it is combined with other such practices.

Abbreviations: Ca, atmospheric CO2 concentration • NPP, net primary productivity • Rh, heterotrophic respiration • SOC, soil organic C


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
THEORY
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
LAND USE PRACTICES that involve soil disturbance and removal of vegetation have been widely observed to cause losses of soil C to the atmosphere. These losses account for about one-third of the atmospheric CO2 that has accumulated since preindustrial times, and they indicate widespread soil degradation. There is currently much interest in reducing these losses through land use practices that increase the sequestration of C in soils. These practices include greater cropping frequency (Bremer et al., 1994; Campbell et al., 1995), reduced tillage (Reicosky et al., 1995), fertilizer application (Gregorich et al., 1996; Nyborg et al., 1995), application of organic amendments such as manure (Sommerfeldt et al., 1988, although Janzen et al., 1998 regard this as a C transfer rather than a gain), and cropping to perennial legumes and grasses (Bremer et al., 1994; Campbell et al., 1991).

Accurate estimates of changes in soil organic C (SOC) with land use practices under site-specific conditions of soil and climate are needed. In many studies of SOC, these estimates have been made using mathematical models of C and N transformations in terrestrial ecosystems. Such models have been used to simulate changes in SOC with grazing intensity (Parton et al., 1994), fertilizer and organic amendments (Jenkinson et al., 1987; Li et al., 1994; Parton and Rasmussen, 1994), and tillage and rotations (Grant, 1997; Grant et al., 1998; Paustian et al., 1998). Because changes in SOC represent the difference between net primary productivity (NPP) and heterotrophic respiration (Rh), ecosystem models used to simulate C sequestration must be capable of simulating land use effects on both NPP and Rh.

Net primary productivity is simulated at varying levels of complexity in ecosystem models, ranging from prescribed C inputs, typically at monthly time steps (Coleman and Jenkinson, 1996), through simple associations with soil water and precipitation adjusted for mineral nutrients, typically at monthly time steps (Parton and Rasmussen, 1994), to radiation use efficiency modified by temperature, humidity, and atmospheric CO2 concentration (Ca), usually at daily time steps (e.g., Kiniry and Bockholt, 1998). The parameters in these simulations are model-specific and require site-specific calibration against seasonal data for plant growth, so that use of parameter values is limited to areas within which calibration has been conducted. Net primary productivity is also simulated from the biochemistry of CO2 fixation (Farquhar et al., 1980) and respiration (Penning de Vries, 1983). The basic nature of this simulation confers a generality of parameterization and robustness of model performance that is necessary for the simulation of diverse ecosystems. Consequently, this approach to the modeling of NPP is expanding in use.

Many ecosystem models share some common approaches to the simulation of Rh (Paustian, 1994), including first-order kinetics based on the assumption that metabolic demand of the soil biomass exceeds substrate supply, discrete soil C fractions with differing rates of decomposition, usually based on age and chemical composition, and coupled transformations of C and N based on observed C/N stoichiometry. In a review of current ecosystem models, McGill (1996a) observed that their future development would benefit from a more mechanistic treatment of soil organisms. Such a treatment is necessary because C and N transformation rates may depart from the first-order kinetics commonly assumed in many of these models. An alternative model to first-order kinetics is one in which microbial activity is explicitly represented as the agent of C and N transformations (e.g., McGill et al., 1981; Smith, 1982). The higher-order kinetics of this model can be parameterized from the known energetics of individual C and N transformations (McGill, 1996b), thereby avoiding the use of temporally and spatially aggregated, site-specific data for changes in soil C and N frequently used in the parameterization of first-order models.

In earlier work, we developed and tested a higher-order model ecosys in which NPP was driven from the energetics of CO2 fixation and respiration, and Rh was driven from the energetics of heterotrophic and autotrophic oxidation–reduction reactions conducted under aerobic and anaerobic conditions (see references listed in Appendix). This model has been used to simulate changes in plant growth under different Ca and irrigation (Grant et al., 1995, 1995c, 1999b), N and P fertilization (Grant, 1991; Grant and Robertson, 1997), salinity (Grant, 1995b), and weed competition (Grant, 1994b), and to simulate soil C under different rotation and tillage practices (Grant, 1997; Grant et al., 1998). We used this model to examine our understanding of how changes in NPP and Rh, and hence changes in SOC, are determined by the use of perennial legumes and grasses vs. annual crops, and by the use of chemical fertilizers vs. organic soil amendments. This understanding is tested with data for NPP and SOC collected from a long-term (70 yr) cropping experiment on Gray Luvisol (Typic Cryoboralf) at Breton, AB. We then used this model to examine how NPP and SOC might change with these land use practices in the future.


   THEORY
TOP
 ABSTRACT
 INTRODUCTION
 THEORY
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Net Primary Productivity
Carbon Dioxide Fixation
Carbon dioxide fixation is calculated in ecosys from coupled algorithms for carboxylation and diffusion. Carboxylation rates are calculated for each leaf surface of multispecific plant canopies as the lesser of dark and light reaction rates (Grant et al., 1999b) according to Farquhar et al. (1980). These rates are driven by irradiance, temperature, and CO2 concentration. Maximum dark or light reaction rates used in these functions are determined by specific activities and surficial concentrations of rubisco or chlorophyll, respectively. These activities and concentrations are determined by environmental conditions during leaf growth (CO2 fixation, water, N and P uptake) as described in the Plant Growth section below.

Carbon dioxide diffusion is calculated for each leaf surface from the CO2 concentration difference between the canopy atmosphere and the mesophyll multiplied by leaf stomatal conductance (Grant et al., 1999b) required to maintain a set Ci/Ca ratio at the leaf carboxylation rate. Stomatal conductance is also an exponential function of canopy turgor (Grant et al., 1999b) generated from a convergence solution for canopy water potential at which the difference between transpiration and root water uptake (Grant et al., 1999b) equals the difference between canopy water contents at previous and current water potentials. Canopy transpiration arises from a first-order solution to the canopy energy balance (Grant et al., 1999b).

Autotrophic Respiration and Senescence
The product of CO2 fixation is added to a C storage pool for each branch of each plant species from which C is oxidized to meet maintenance respiration requirements using a first-order function of storage C (Grant et al., 1999b). If the C storage pool is depleted, the C oxidation rate may be less than the maintenance respiration requirement, in which case the difference is made up through respiration of remobilizable C in leaves, sheaths, or petioles. Upon exhaustion of the remobilizable C in each leaf, sheath, or petiole, the remaining C is dropped from the branch and added to residue at the soil surface from which it undergoes decomposition as described in the Heterotrophic Respiration section below. Environmental constraints such as nutrient, heat, or water stress that reduce C fixation, and hence C storage, will therefore accelerate the loss of leaf, sheath, or petiole C. When storage C oxidation exceeds maintenance respiration, the excess is used for growth respiration to drive the formation of new biomass (Grant et al., 1999b) as described in the Plant Growth section below.

Nutrient Uptake
Nutrient (N and P) uptake is calculated for each plant species by iteratively converging towards values for aqueous concentrations of NH+4, NO-3, and H2PO-4 at its root and mycorrhizal surfaces in each soil layer at which radial transport by mass flow and diffusion from the soil solution to the surfaces equals active uptake by the surfaces (Grant and Robertson, 1997; Grant, 1998b). This convergence dynamically links rates of soil nutrient transformations with those of root and mycorrhizal nutrient uptake. Nutrient transformations control the aqueous concentrations of NH4, NO3, and H2PO4 in each soil layer through thermodynamically driven precipitation, adsorption, and ion pairing reactions (Grant and Heaney, 1997), convective-dispersive solute transport (Grant and Heaney, 1997), and microbial mineralization–immobilization (Grant et al., 1993a). Active uptake is calculated from length densities and surface areas (Itoh and Barber, 1983) given by a root and mycorrhizal growth submodel (Grant, 1993; Grant 1998b; Grant and Robertson, 1997). Active nutrient uptake is constrained by O2 uptake (Grant, 1993), by solution NH4, NO3 and H2PO4 concentrations, and by root and mycorrhizal C, N, and P storage (Grant, 1998b). The products of N and P uptake are added to root and mycorrhizal storage pools from which they are combined with storage C when driven by growth respiration to form new plant biomass as described in the Plant Growth section below. Plant species designated as legumes in the model also grow root nodules in which aqueous N2 is reduced to storage N through oxidation of storage C according to the energetics in Schubert (1982). This reduction generates concentration gradients of storage C, N, and P between nodule and root that drive nutrient exchange.

Plant Growth
Growth respiration from the Autotrophic Respiration and Senescence section above drives expansive growth of vegetative and reproductive organs through mobilization of storage C, N, and P in each branch of each plant species according to phenology-dependent partitioning coefficients and biochemically based growth yields. This growth is used to simulate the lengths, areas, and volumes of individual internodes, sheaths, or petioles, and leaves (Grant, 1994b; Grant and Hesketh, 1992) from which heights and areas of leaf and stem surfaces are calculated for irradiance interception and aerodynamic conductance algorithms used in energy balance calculations. Growth respiration also drives extension of primary and secondary root axes and of mycorrhizal axes of each plant species in each soil layer through mobilization of storage C, N, and P in each root zone of each plant species (Grant, 1993; Grant, 1998b). This growth is used to calculate lengths and areas of root and mycorrhizal axes from which root uptake of water (Grant et al., 1999b) and nutrients (Grant, 1991; Grant and Robertson, 1997) is calculated.

The growth of different branch organs and root axes in the model depends on transfers of storage C, N, and P among branches, roots, and mycorrhizae. These transfers are driven from concentration gradients within the plant that develop from different rates of C, N, or P acquisition and consumption by its branches, roots, or mycorrhizae (Grant, 1998b). When root N or P uptake rates described in the Nutrient Uptake section above are low, storage N or P concentrations in roots and branches become low with respect to those of storage C. Such low ratios in branches reduce the specific activities and surficial concentrations of leaf rubisco and chlorophyll, which in turn reduce leaf CO2 fixation rates. These low ratios also cause smaller root-to-shoot transfers of N and P and larger shoot-to-root transfers of C (Grant, 1998b), thereby allowing more plant resources to be directed towards root growth. The consequent increase in root/shoot ratios, and thus in N and P uptake, coupled with the decrease in C fixation rate redresses to some extent the storage C/N/P imbalance when N or P uptake is limiting. The model thus implements the functional equilibrium between roots and shoots proposed by Thornley (1995).

For perennial plant species, soluble C, N, and P are withdrawn from storage pools in branches into a long-term storage pool in the crown during autumn, causing leaf senescence. Soluble C, N, and P are remobilized from this pool to drive leaf and petiole or sheath growth the following spring. The timing of withdrawal and remobilization is determined by duration of exposure to low temperatures (between 3 and 8°C) under shortening and lengthening photoperiods, respectively.

Heterotrophic Respiration
Decomposition
Soil organic matter in ecosys is resolved into four substrate–microbe complexes (plant residue, animal manure, particulate organic matter, and nonparticulate organic matter) within each of which C, N, and P may move among five organic states: solid substrate, sorbed substrate, soluble hydrolysis products including acetate, microbial communities, and microbial residues (Table 1 in Grant, 1999). Each organic state in each complex is resolved into structural components of differing vulnerability to hydrolysis and into elemental fractions C, N, and P within each structural component. Microbial communities are also resolved into functional type, including obligate aerobes, facultative anaerobes (denitrifiers), obligate anaerobes (fermenters), methanogens, and diazotrophs.


View this table:
[in this window]
[in a new window]
  		Table 1. Properties of the Gray Luvisol (Typic Cryoboralf) at Breton during the 1930s used to initialize ecosys. Properties are taken from 1936 and 1938 samples of cultivated plots and 1979 samples of an adjacent undisturbed site under original forest

 
Litterfall from the Autotrophic Respiration and Senescence section above is added to the plant residue complex and partitioned into carbohydrate (hot-water extractable), protein, cellulose, and lignin structural components according to Broder and Wagner (1988) (Table 1 in Grant and Rochette, 1994). Manure amendments are added to the animal manure complex and similarly partitioned. Rates of component hydrolysis are the product of the active biomass and specific activity of each microbial functional type within each complex (Grant et al., 1993a; Grant and Rochette, 1994). Specific activity is constrained by substrate-microbe density relationships (Grant et al., 1993a; Grant and Rochette, 1994), and by the temperatures and water contents of surface residue and a spatially resolved soil profile (Grant, 1997; Grant and Rochette, 1994; Grant et al., 1998). A fraction of the hydrolysis products of lignin are coupled with those of protein and carbohydrate according to the stoichiometry proposed by Shulten and Schnitzer (1997), and the resulting compound is transferred to the solid substrate of the particulate organic matter complex. Rates of particulate organic matter formation are thus determined by substrate lignin content and heterotrophic microbial activity.

Microbial Growth
The concentration of the soluble hydrolysis products in the Decomposition section above determines rates of C oxidation by each heterotrophic population, the total of which drives CO2 emission from the soil surface. This oxidation is coupled to the reduction of O2 by all aerobic populations (Grant et al., 1993a, 1993b; Grant and Rochette, 1994), to the sequential reduction of NO-3, NO-2, and N2O by heterotrophic denitrifiers (Grant et al., 1993c, 1993d; Grant and Pattey, 1999) and to the reduction of organic C by fermenters and of acetate by heterotrophic methanogens (Grant, 1998a). The energetics of these oxidation–reduction reactions determine growth yields and hence the active biomass of each heterotrophic functional type from which its decomposer activity is calculated as described in the Decomposition section above. In addition, autotrophic nitrifiers conduct NH+4 and NO-2 oxidation (Grant, 1994a) and N2O evolution (Grant, 1995a), and autotrophic methanotrophs conduct CH4 oxidation (Grant, 1999), the energetics of which determine autotrophic growth yields and hence biomass and activity. Microbial populations in the model seek to maintain steady-state ratios of biomass C/N/P by mineralizing or immobilizing NH+4, NO-3, and H2PO-4, thereby regulating solution concentrations that drive N and P uptake by roots and mycorrhizae as described in the Nutrient Uptake section above. Microbial populations undergo first-order decomposition, products of which are partitioned between microbial residues within the same substrate–microbe complex, and the solid substrate of the nonparticulate organic matter complex according to soil clay content (Grant et al., 1993a, 1993b). Rates of nonparticulate organic matter formation are thus determined by rates of microbial decay and by soil clay content.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 THEORY
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Field Experiment
Plot Management
The Breton Plots were established in 1930 on a Gray Luvisolic soil (Typic Cryoboralf; Table 1) under a boreal climate near Breton, AB (53°07'N, 114°28'W, annual precipitation 547 mm, annual mean temperature 2.1°C). The Breton Plots site had been converted from a mixed aspen (Populus spp.)–spruce (Picea spp.) forest to cultivated agriculture {approx}10 yr earlier. The 269-m2 plots were designed to compare two rotations: a 2-yr wheat–fallow rotation and a 5-yr wheat–oat–barley–forage–forage rotation that included legumes [1930–1966: mixed legumes including red clover (Trifolium pratense L.), 1967–present: alfalfa (Medicago sativa L.)] and grasses [(1930–1966: mixed grasses including creeping red fescue (Festuca rubra L.) and timothy (Phleum pratense L.), 1967–present: bromegrass (Bromus inermis Leyss)]. Each component of each rotation was present during each year of the experiment, but rotations were otherwise unreplicated. Each rotation was amended with several combinations of fertilizer (N, P, K, and S) and with manure. Three of these amendments were selected for detailed study in each rotation: (i) check, (ii) complete N-P-K-S fertilizer, and (iii) manure. Amendment rates remained constant from 1930 to 1979, but were adjusted in 1980 to more contemporary levels, at which they have remained until the present (Table 2). All fertilizers were broadcast until 1964, after which P was drilled at seeding. Grain crops were planted with a press drill in early to mid May at 82 kg ha-1 for wheat and 89 kg ha-1 for oat and barley. All growth above 0.05 to 0.10 m in height was harvested during late August or early September. Forages were seeded at 10 kg ha-1 for alfalfa and 15 kg ha-1 for bromegrass under barley during the third year of the 5-yr rotation. All forage growth above 0.10 m was harvested in mid July and mid September of the fourth year and in mid July of the fifth year. The remaining forage was plowed under in early August. No harvested cereal or forage phytomass was returned to the plots. All plots were tilled once each fall after harvest and once each spring before planting except when forages were present. Broadleaf weeds and wild oat (A. fatua L.) were controlled by tillage until 1960, and afterwards by herbicides and tillage.


View this table:
[in this window]
[in a new window]
  		Table 2. Soil amendment rates for the fertilizer and manure treatments at the Breton Plots

 
Soil Organic Carbon Measurements
Total C and N concentrations were measured with a Carlo-Erba analyzer (model NA-1500, Carlo-Erba Inc., Milan, Italy) in 1995 on archived samples taken from the Ap and Ae horizons (0–0.15 m) of the Breton Plots in 1936 or 1938, 1972, 1979, and 1990, and in 1999 on samples taken from the Ap (0.0–0.075 m), Ae (0.075–0.15 m), and Bt horizons (0.15–0.40 m) in 1998. Total N concentrations were also measured on other samples taken in 1957 (Bentley et al., 1971) and 1968 (Khan, 1969) but not archived. These concentrations were used with C/N ratios measured from the 1936 and 1972 samples to estimate total C. An undisturbed area under original aspen–spruce forest immediately north of the Breton plots was sampled to a depth of 1 m in 1979. No carbonates were present in any of the samples. Manure used in soil amendments has been analyzed since 1980 for water content, for total C by combustion on a LECO Carbon Determinator (model CR12, LECO Corp., St. Joseph, MI), and for total N by colorimetry from an H2SO4–H2O2 digest (Technicon industrial method 334-74; Technicon Industrial Systems, Tarrytown, NY). Soil C concentrations were used to calculate soil C contents per unit area on an equivalent mass basis. Further details about plot management and soil analyses may be found in Izaurralde et al. (1996)(2001) and Juma et al. (1997).

Model Experiment
The ecosystem model ecosys was initialized with the properties of the Gray Luvisol at Breton during the 1930s estimated from samples of undisturbed soil taken from an adjacent forested area in 1979 and from cultivated field plots in 1936 and 1938 (Table 1). The model was then run with each rotation–amendment treatment for a simulated period from 1930 to 1999 with meteorological data recorded at the field site. These data included solar radiation, air temperature (maximum and minimum), precipitation, humidity, and wind speed recorded on a daily basis from 1968 to 1995, and on an hourly basis from 1996 to 1998. Data recorded between 1968 and 1998 were used in the model run between 1930 and 1967 and in 1999. Atmospheric CO2 concentration in the model was initialized in 1930 at 305 µmol mol-1 and incremented daily at an annual rate of 0.0024, thereby reaching 360 µmol mol-1 in 1998. Atmospheric N deposition occurred during model runs through N in precipitation (concentration 1 g NO-3–N m-3) and through adsorption (or volatilization) of NH3 (atmospheric concentration 0.012 µmol mol-1). All model parameters for ecosystem C transformations remained the same as those used in earlier studies cited in the Appendix.

For each year that wheat was simulated in the 2-yr rotation, the model was provided with the biological properties of wheat (Grant et al., 1995), which was initialized as seed planted on 10 May at 200 m-2 0.02 m below the soil surface. A wheat maturity group (defined as the minimum number of vegetative nodes initiated before floral induction) was selected to give an anthesis date that corresponded with that observed in the field. Tillage in the model included chisel plowing on 1 May and 1 October, and harrowing on 2 May. Fertilizer was applied at the rates indicated for the N-P-K-S treatment in Table 2 as broadcast NH4NO3 on 1 May and as broadcast (before 1964) or drilled (1964 and after) H2PO-4 on 10 May. Manure was applied on 1 October at the rates indicated for the manure treatment in Table 2, with N and P contents taken from manure analyses. The modeled wheat was harvested on 21 September at a height of 0.05 m and removed from the simulation. During fallow years, tillage included harrowing on 1 May, 1 June, 1 July, and 1 August, and chisel plowing on 1 October. Dates of management events in the model represented means of those in the field during the 70 yr of the experiment.

The 5-yr rotation was simulated as three consecutive years of cereal with the biological properties of wheat for which tillage, seeding, fertilizing, and harvesting were the same as those in the 2-yr rotation described above. The seeding of the third cereal crop was accompanied by the seeding of alfalfa and bromegrass each at 250 m-2 0.01 m below the soil surface. The alfalfa and grass thus competed with the cereal crop during the third year of the rotation and were harvested on 15 July and 15 September of the fourth year; they were harvested again on 15 July of the fifth year. All forage above 0.10 m in height was removed from the simulation at each harvest. Any remaining stubble was fully incorporated into the upper 0.10 m of the soil profile on 15 August of the fifth year of the rotation so that no further plant growth occurred. Results for plant C harvest and soil C content from the model were compared with those reported from the Breton plots for each rotation–amendment during the past 70 yr. The model runs were then continued for a further 70 yr to predict future trends in plant and soil C under current climate and current rates of increase in atmospheric CO2 concentrations. These predictions were made for two management scenarios: (i) no change from current land use practices (1980–1999), in which all cereal straw was removed, and each forage crop was harvested twice in the first year and once in the second, and (ii) reduced harvesting, in which all cereal straw was returned to the soil surface, and the forage crop was not harvested in the second year. These changes were designed to achieve an approximate 50% reduction in the fraction of aboveground biomass removed at harvest.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 THEORY
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Modeled vs. Measured Changes in Soil Carbon
Changes of soil C derived from linear regression of all soil C contents measured between 1936 and 1998 in the upper 0.15 m of the soil profile in the 2- and 5-yr rotations are given in Table 3 and Fig. 1a . These are compared with changes simulated between 1930 and 1979 and between 1980 and 1999 following changes in soil amendments (Table 2). Losses of soil C measured and modeled in the 2-yr rotation were consistent with losses of 20 g C m-2 yr-1 reported by Rasmussen and Parton (1994) from the upper 0.3 m of a 55-yr wheat–fallow rotation in which crop residues were burned. Only the addition of manure at rates greater than those at which it could be produced from crop harvests enabled C to be accumulated in the upper 0.15 m of the wheat–fallow rotation at Breton. Removal of almost all aboveground biomass from the 5-yr rotation caused moderate gains of C to be measured and modeled in the upper 0.15 m of the control and fertilizer treatments at Breton and larger gains under the manure treatment. The modeled and measured gains in the upper 0.15 m of the 5- vs. 2-yr rotations ({approx}15 and 20 g C m-2 yr-1, respectively, from Table 3) were similar to gains of 14 g m-2 yr-1 recorded in a 6-yr fallow–wheat–wheat–forage–forage–forage rotation vs. a 2-yr wheat–fallow rotation over 41 yr on a dark Brown Chernozemic (Typic Haploboroll) soil (0–0.3 m) at Lethbridge, AB (Bremer et al., 1994), and to gains of 11 g C m-2 yr-1 recorded in the same two rotations over 30 yr on a thin Black Chernozemic (Typic Cryoboroll) soil (0–0.15 m) at Indian Head, SK (Campbell et al., 1997).


View this table:
[in this window]
[in a new window]
  		Table 3. Rates of change in soil C from linear regression of all soil C contents measured in the upper 0.15 m of the soil profile between 1936 and 1938, and rates of change in soil C simulated in the upper 0.15 m of the soil profile between 1930 and 1979, and between 1980 and 1999 for control, fertilizer, and manure treatments in the 2- and 5-yr rotations

 


View larger version (30K):
[in this window]
[in a new window]
  		Fig. 1. Measured (symbols) and simulated (lines) soil C content in the upper 0.15 m of the profile during the past 70 yr of the control, fertilizer, and manure treatments in the 2- and 5-yr rotations

 
Increases in soil C measured in the fertilized vs. control treatments of both rotations before 1980 were not reproduced by the model because the low rates of fertilizer application during this period (Table 2) caused only small increases in modeled C inputs (Table 3). Some of these increases in soil C measured before 1980 in the 5-yr rotation may be attributed to a legume response to S fertilizer that was not included in the model. Soil C simulated in the upper 0.15 m of the fertilizer treatment rose above that in the control treatments of the 2- and 5-yr rotations by 3 and 5 g C m-2 yr-1, respectively, after fertilizer rates were increased in 1980. Given that 1.2 g of C is released as CO2 during the manufacture of 1 g of N as NH3 (Schlesinger, 1999), these gains in soil C have to offset fertilizer costs of 5.4 g C m-2 yr-1 in the 2-yr rotation and 3.0 g C m-2 yr-1 in the 5-yr rotation (Table 2).

Soil C modeled below 0.15 m changed between 1930 and 1999 at rates of -17, -16, and -12 g C m-2 yr-1 in the control, fertilizer, and manure treatments, respectively, of the 2-yr rotation, and at rates of -4, -3, and +6 C m-2 yr-1 in the same treatments of the 5-yr rotation (Fig. 2) . The declines modeled in the 2-yr rotation were similar to ones of 18 g C m-2 yr-1 reported over 55 yr by Rasmussen and Parton (1994) at depths between 0.3 and 0.6 m in a wheat–fallow rotation. The slower declines modeled below 0.15 m in the 5- vs. 2-yr rotation at Breton were attributed in the model to deeper, denser root systems that input more C under the perennial forages. These slower declines were corroborated by gains in soil C of 200 to 400 g m-2 measured between 0.15 and 0.40 m under the 5- vs. 2-yr rotation in 1998 (Table 4). Gains in deeper soil C modeled under the manure vs. control treatment were corroborated by gains of {approx}200 g C m-2 measured between 0.15 and 0.40 m under the manure vs. control treatment of the 5-yr rotation in 1998 (Table 4), but not under that of the 2-yr rotation. Declines in soil C below 0.15 m add to losses or offset gains above (Table 5) and indicate the need to consider more than the A horizon when evaluating the effects of land use practices on soil C storage.



View larger version (26K):
[in this window]
[in a new window]
  		Fig. 2. Soil C content simulated in the entire profile during the past 70 yr of the control, fertilizer, and manure treatments in the 2- and 5-yr rotations

 

View this table:
[in this window]
[in a new window]
  		Table 4. Soil C contents measured and simulated between 0.15 and 0.40 m (Bt horizon) during 1998 in the control, fertilizer, and manure treatments of the 2- and 5-yr rotations. Soil C content between 0.15 and 0.40 m was initialized at 1.92 kg C m-2 in 1930

 

View this table:
[in this window]
[in a new window]
  		Table 5. Annual C balances simulated between 1980 and 1999 for control, fertilizer, and manure treatments in the 2- and 5-yr rotations. Also shown is an annual C balance simulated for a mixed aspen–hazelnut forest on a gray Luvisol (Typic Cryoboralf) given in Grant et al. (1999a){dagger}{dagger}

 
The rate at which soil C changed in each modeled treatment was determined by the extent to which C, N, and P removals through harvesting and losses (runoff, leaching, denitrification) exceeded C, N, and P additions through biological fixation, atmospheric deposition, and soil amendments. Gains in soil C modeled under the 5- vs. 2-yr rotation were attributed to:

1. . Higher Net Primary Productivity Caused by Yearly vs. Biyearly Cereal Crops and by Perennial Forage Crops
Greater cropping frequency and N2–fixation by leguminous forage ({approx}35 g N m-2 rotation-1) caused average NPP modeled between 1980 and 1999 to rise from 117 to 299 g C m-2 yr-1 in the control treatments, from 212 to 408 g C m-2 yr-1 in the fertilizer treatments, and from 252 to 438 g C m-2 yr-1 in the manure treatments of the 2- vs. 5-yr rotation (Table 5).

2. . Relatively less Removal of Net Primary Productivity as Harvest
In the 2-yr rotation, 0.55, 0.67, and 0.69 of NPP was removed from the control, fertilizer, and manure treatments, respectively, while in the 5-yr rotation only 0.37, 0.49, and 0.48 of the larger NPP was removed from the same treatments (Table 5). The lower relative removals in the 5- vs. 2-yr rotation and the control vs. amended treatments were attributed in the model to higher root/shoot ratios (Table 6). These larger ratios arose in the model from the effects of a perennial growth habit on the functional equilibrium between roots and shoots described in the Plant Growth section above. Changes in root/shoot ratios are considered to be an important determinant of soil C storage under different land use practices (Parton and Rasmussen, 1994). Larger root systems caused larger soil C inputs to be modeled through root senescence and exudation, and therefore larger ratios of soil inputs to NPP. Root/shoot ratios modeled in the 2-yr treatment were consistent with ratios for wheat of 0.24, 0.11, and 0.12 measured by Izaurralde et al. (1992) in the control, fertilizer, and manure treatments, respectively. Lower root/shoot ratios in the 2- vs. 5-yr rotation were found by Izaurralde et al. (2001) to cause a doubling of the aboveground NPP required to maintain soil C.


View this table:
[in this window]
[in a new window]
  		Table 6. Root shoot ratios simulated between 1980 and 1999 for control, fertilizer, and manure treatments in the 2- and 5-yr rotations{dagger}{dagger}

 
Gains in soil C modeled under the 5- vs. 2-yr rotations were sustained by N2 fixation in the forages that raised NPP and hence soil C inputs. However, crop NPP and soil inputs modeled and measured in all rotation x amendment treatments were less than those modeled and measured in an undisturbed aspen forest on a Gray Luvisol under a similar climate (Table 5). Rises in NPP under the 5- vs. 2-yr rotations were constrained by lower soil P concentrations caused by greater removal of P in harvests and by greater immobilization of P in soil. Exchangeable P in the model (Grant and Heaney, 1997) declined from 56 µg g-1 in 1930 to 16 vs. 22 µg g-1 in the control plots of the 5- vs. 2-yr rotations by 1986. These values compared with ones for resin-extractable P of 9 vs. 23 µg g-1 measured in these plots in 1986 by McKenzie et al. (1992), who found that both control and fertilizer treatments in the 5-yr rotation were losing P more rapidly than were the corresponding treatments in the 2-yr rotation. The modeled fertilizer and manure treatments raised soil P concentrations in both rotations, although values remained lower in the 5- vs. 2-yr rotation.

The rate at which total soil C rose in the modeled manure vs. control treatment in the 2-yr rotation (+15 g C m-2 yr-1 from Table 5) was 0.17 of the rate at which manure was added (90 g C m-2 yr-1). Coleman and Jenkinson (1996) reported continuous increases of soil C during 140 yr in manured vs. unmanured barley that represented 0.15 of manure additions. The rate at which total soil C rose in the modeled manure vs. control treatment in the 5-yr rotation (+16 g C m-2 yr-1 from Table 5) was 0.23 of the rate at which manure was added (70 g C m-2 yr-1). This higher ratio in the 5-yr rotation was attributed in the model to greater N inputs from biological fixation.

Larger C inputs modeled in the 5- vs. 2-yr rotations and the manure vs. control treatments (Table 5) caused more rapid microbial growth that generated large increases in microbial C during 1981 of the model runs (Table 7). Increases of microbial C in the model were corroborated by increases of microbial C measured in 1981 by McGill et al. (1986) in the 5- vs. 2-yr rotations and in the manure vs. control treatments (Table 7). Because microbial biomass is an active agent of soil C transformation in ecosys, larger biomass caused more rapid heterotrophic respiration to be modeled in the 5- vs. 2-yr rotations and in the manure vs. control treatments (Table 5). These modeled increases in soil respiration rates were corroborated by increases measured with flux chambers over the Breton field plots by Carcamo (1997).


View this table:
[in this window]
[in a new window]
  		Table 7. Microbial biomass measured and simulated in the Ap horizon (0.0–0.05 m) during 1981 in the control, fertilizer, and manure treatments of the 2- and 5-yr rotations.{dagger}

 
Regressions of model results for total changes in soil C on soil C inputs in the 2- and the 5-yr rotations (Table 5) indicated that 0.14 of soil inputs were stabilized as soil C and that inputs of 172 g C m-2 yr-1 were required to maintain soil C between 50 and 70 yr after the start of the experiment. The C stabilization efficiency and steady-state C input in the 2-yr rotation were similar to ones of 0.18 and 189 g C m-2 yr-1 reported by Parton and Rasmussen (1994) for aboveground inputs from a wheat–fallow rotation on a soil with a higher clay content and a warmer mean annual temperature (18% and 10.2°C) than those at Breton (12% and 2.1°C).

Projected Changes in Soil Carbon under Current vs. Reduced Harvesting
If the present land use practices at the Breton Plots are maintained for another 70 yr, rates of soil C loss in the 2-yr rotation are predicted to decline from 16 to 7 g C m-2 yr-1 in the control treatment, from 13 to 5 g C m-2 yr-1 in the fertilizer treatment, and from 1 g C m-2 yr-1 to a gain of 7 g C m-2 yr-1 in the manure treatment as stable soil C contents are approached (Fig. 3) . Rates of soil C gain in the control, fertilizer, and manure treatments of the 5-yr rotation during the next 70 yr are predicted to change little from current rates. Soil C contents modeled under the control and fertilizer treatments of the 2-yr rotation are predicted to continue declining from an initial value of 7.75 kg m-2 at first cultivation to <5.5 and 5.8 kg m-2, respectively, after 140 yr. This loss of 0.33 of the C originally present in the undisturbed soil is consistent with the calculation by McGill et al. (1988) that Gray Luvisols had an average of 0.39 less C at cultivated vs. uncultivated sites, based on a comparison of A and B horizons at 72 sites in central and northern Alberta. These losses of C with cultivation persist over time. Rasmussen and Parton (1994) observed continuing losses under a wheat–fallow rotation after more than 100 yr of cultivation.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 3. Soil C content simulated in the entire profile during the next 70 yr of the control, fertilizer, and manure treatments in the 2- and 5-yr rotations under current harvest practices

 
Soil C contents modeled under the manure treatment of the 2-yr rotation are predicted to decline to {approx}7.4 kg m-2 between 80 and 100 yr after first cultivation and then to return to initial values after 140 yr (Fig. 3). Soil C contents modeled under the control, fertilizer, and manure treatments of the 5-yr rotation are predicted to continue rising above 8.0, 8.5, and 11.0 kg m-2, respectively, after 140 yr of cultivation. These gains appear to be continuous, with no evidence in the model of an upper limit to C storage within the duration of the model run. Chernozemic (Typic Cryoboroll) soils in the same agroclimatic zone as that of the Gray Luvisol (Typic Cryoboralf) at Breton typically store more than 20 kg C m-2.

If the fraction of aboveground biomass removed at harvest is reduced by {approx}50% during the next 70 yr, greater nutrient conservation is predicted by the model to raise NPP, which in combination with lower C removals, will greatly raise soil C inputs. Larger inputs will cause more rapid rates of soil C accumulation, especially in the 5-yr rotation where rates would rise by more than 30 g C m-2 yr-1 from those under current harvest practices (Fig. 4) . Soil C contents modeled under the control, fertilizer, and manure treatments of the 2-yr rotation are predicted to reach {approx}5.9, 6.1, and 8.9 kg m-2, respectively, after 70 yr under reduced harvest practices. Those modeled under the control, fertilizer, and manure treatments of the 5-yr rotation are predicted to reach {approx}10.0, 11.0, and 14.0 g m-2, respectively. Model projections indicate that reducing the harvested fraction of aboveground C by one-half would raise steady-state contents of soil C after 70 yr by 400 g m-2 in the 2-yr rotation and 2000 g m-2 in the 5-yr rotation without manure. These rises would be at least 500 g C m-2 larger in the manure treatments. Buyanovsky et al. (1996) showed that declines in soil C under continuous wheat and maize (Zea mays L.) crops from which all aboveground residues were removed could be reversed if these residues were returned. The model results suggest that the largest gains in soil C are likely to be realized when individual land use practices that favor C storage, such as forage rotations, manure application, and reduced harvest removals, are combined in conservation cropping systems.



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 4. Soil C content simulated in the entire profile during the next 70 yr of the control, fertilizer, and manure treatments in the 2- and 5-yr rotations under reduced harvest practices

 

   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 THEORY
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
Key model findings supported by long-term measurements of soil C at Breton were:
  1. Carbon is lost at rates of 5 to 10 g C m-2 yr-1 from the upper 0.15 m of the soil profile in a 2-yr wheat–fallow rotation unless manure is added.
  2. Carbon is gained at rates of 15 to 20 g C m-2 yr-1 in the upper 0.15 m of the soil profile in a 5-yr wheat–oat–barley–forage–forage rotation vs. a 2-yr wheat–fallow rotation.
  3. Further C losses and gains in the 2- and 5-yr rotation occur below 0.15 m, indicating the need for deeper soil sampling in long-term studies of soil C.
  4. Inputs of 172 g C m-2 yr-1 were required to maintain soil C in both rotations.
  5. Approximately 0.14 of crop C inputs, and 0.17 (2-yr rotation) and 0.23 (5-yr rotation) of manure C inputs were stabilized as long-term C.
  6. Land use practices such as manure, forages, and reduced harvest removal that favor C storage appear to interact positively with each other, so that gains in soil C under one such practice are greater when it is combined with other such practices.


APPENDIX 1. Background References for ecosys

Process
References
CO2 fixation Grant, 1992a, 1992b; Grant et al., 1993c, 1995, 1999a, 1999b.
Autotrophic plant respiration Grant, 1993; Grant et al., 1999a.
Nutrient transformation and uptake Grant, 1991; Grant and Robertson, 1997; Grant and Heaney, 1997; Grant et al., 1993a.
Heterotrophic soil respiration Grant, 1997, 1998a; Grant and Rochette, 1994; Grant and Pattey, 1999; Grant et al., 1993a, 1993b, 1993c, 1993d, Grant et al., 1998.
Autotrophic soil respiration
 Grant, 1994a, 1995a, 1999.


   ACKNOWLEDGMENTS
 
Current work on the Breton Plots is funded from Alberta Agriculture Research Institute (AARI) grants 980911 and 980914, and from the Breton Plots Endowment Fund. Ecosys was run on a DEC cluster made available through the Multimedia Advanced Computational Infrastructure (MACI) project at the University of Alberta. The assistance of Monica Ayala, Karen Haugen-Kozyra, and Jeff Thurston in soil analyses and data management is greatly appreciated.

Received for publication May 1, 2000.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 THEORY
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 




This article has been cited by other articles:


Home page
 Agron. J.Home page
D. K. Lee, V. N. Owens, and J. J. Doolittle
Switchgrass and Soil Carbon Sequestration Response to Ammonium Nitrate, Manure, and Harvest Frequency on Conservation Reserve Program Land
Agron. J., February 6, 2007; 99(2): 462 - 468.
[Abstract] [Full Text] [PDF]

Home page
 J. Environ. Qual.Home page
F. Fares, A. Albalkhi, J. Dec, M. A. Bruns, and J.-M. Bollag
Physicochemical Characteristics of Animal and Municipal Wastes Decomposed in Arid Soils
J. Environ. Qual., July 5, 2005; 34(4): 1392 - 1403.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
T. M. Nissen and M. M. Wander
Management and Soil-Quality Effects on Fertilizer-Use Efficiency and Leaching
Soil Sci. Soc. Am. J., September 1, 2003; 67(5): 1524 - 1532.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
R.C. Izaurralde, W.B. McGill, J.A. Robertson, N.G. Juma, and J.J. Thurston
Carbon Balance of the Breton Classical Plots over Half a Century
Soil Sci. Soc. Am. J., March 1, 2001; 65(2): 431 - 441.
[Abstract] [Full Text] [PDF]