JEQ 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 Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
Related Collections
Right arrow Soil Organic Matter
Right arrow Carbon Sequestration
Right arrow Ecological Risk Assessment
Journal of Environmental Quality 32:760-766 (2003)
© 2003 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America

TECHNICAL REPORTS
Ecological Risk Assessment

Three Approaches to Define Desired Soil Organic Matter Contents

G. Sparling*,a, R. L. Parfitta, A. E. Hewittb and L. A. Schippera

a Landcare Research, Private Bag 11052, Palmerston North, New Zealand
b Landcare Research, P.O. Box 69, Lincoln, New Zealand

* Corresponding author (SparlingG{at}LandcareResearch.co.nz)

Received for publication February 13, 2002.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil organic C is often suggested as an indicator of soil quality, but desirable targets are rarely specified. We tested three approaches to define maximum and lowest desirable soil C contents for four New Zealand soil orders. Approach 1 used the New Zealand National Soils Database (NSD). The maximum C content was defined as the median value of long-term pastures, and the lower quartile defined the lowest desirable soil C content. Approach 2 used the CENTURY model to predict maximum C contents of long-term pasture. Lowest desirable content was defined by the level that still allowed recovery to 80% of the maximum C content over 25 yr. Approach 3 used an expert panel to define desirable C contents based on production and environmental criteria. Median C contents (0–20 cm) for the Recent, Granular, Melanic, and Allophanic orders were 72, 88, 98, 132 Mg ha-1, and similar to contents predicted by the CENTURY model (78, 93, 102, and 134 Mg ha-1, respectively). Lower quartile values (54, 78, 73, and 103 Mg ha-1, respectively) were similar to the lowest desirable C contents calculated by CENTURY (55, 54, 67, and 104 Mg ha-1, respectively). Expert opinion was that C contents could be depleted below these values with tolerable effects on production but less so for the environment. The CENTURY model is our preferred approach for setting soil organic C targets, but the model needs calibrating for other soils and land uses. The statistical and expert opinion approaches are less defensible in setting lower limits for desirable C contents.

Abbreviations: NSD, National Soils Database


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE MAINTENANCE of organic matter in soils is advocated by virtually all soil scientists and land managers. The advocacy is based on the desirable attributes organic matter imparts to soils, such as increased cation exchange capacity, increased moisture storage, a source of mineralizable nutrients, a food source and habitat for soil organisms, and an important contributor to soil structure and aggregate stability (Allison, 1973; Soane, 1990; Reeves, 1997; Karlen et al., 2001). Additional motivation to maintain or increase soil organic matter arises from its importance in sequestering C to decrease atmospheric CO2 concentrations (Eswaran et al., 1993; Paustian et al., 1997; Lal, 2001).

Equilibrium organic C contents in soils vary depending on the climate, soil type, mineralogy, land use, and management (Spain et al., 1983; Eswaran et al., 1993; Batjes, 1996; Percival et al., 2000). For a given soil, the greatest amount of organic matter generally accumulates in the topsoil under long-term undisturbed vegetation, typically grassland or forest. Loss of organic C is generally regarded as undesirable, although Sojka and Upchurch (1999) point out that low soil C contents can also have some unexpected benefits for agricultural use, such as the potential to reduce soil pesticide application rates due to lower sorption. It is well established that many forms of soil management can lead to changes in organic C concentrations, and the C contents of cropped and tilled soils are usually (but not always) lower than the equivalent soils under long-term grassland or forest (Wander and Bollero, 1999; Brejda et al., 2000a, b). The declines under more intensive forms of agriculture occur because there is increased loss of topsoil through erosion, decreased organic C returns from plant residues, and enhanced breakdown of previously stabilized soil organic matter (Carter and Gregorich, 1996; Reeves, 1997).

In many cases when organic matter is lost from soil, it declines in a curvilinear way (e.g., Haynes and Tregurtha, 1999; Sparling et al., 2000a). There is no clear "critical" C content, and related soil properties change along a continuum (Karlen et al., 2001). The land manager may continue to work the land profitably by "farming smarter" with improved tillage, more timely cultivation, tolerant cultivars, and strategic use of agrochemicals and irrigation (Sojka and Upchurch, 1999). Under such conditions it is very difficult to define a desirable C content, but this needs to be done before low profitability forces a change to an alternative land use (Francis and Knight, 1993; Scrimgeour and Shepherd, 1998).

For soil organic matter to be a useful soil quality indicator, as advocated by many authors (e.g., Doran et al., 1994; Wander and Bollero, 1999; Brejda et al., 2000a,b), some target value for organic C needs to be specified, even when soil C contents change along a continuum. It can be argued that "more is better" in terms of C sequestration, a stance justified by the contribution to decreased C emissions required by the Kyoto protocol (Lal, 2001). However, the "more is better" argument is weaker when applied to agricultural productivity, where the benefits of higher organic matter contents on intensively managed arable soils are sometimes obscure (Sojka and Upchurch, 1999). Organic matter equilibrium levels obtained from long-term grass pasture or forestry may not be attainable targets for intensively managed arable soils even under the best management practices. Rather than defining an maximum value, it would be much more informative for agricultural systems to define a justifiable minimum soil C below which there would be loss of desirable soil characteristics, productive capacity, and ecological functions that were not readily restored within an acceptable timeframe.

In the absence of a clear critical point and demonstrable ecological consequence, the setting of soil quality targets within a continuum requires human value judgments. The intergenerational equity argument—that future generations have a right to benefit from the earth's resources, and that the current generation does not have the right to deplete a resource irreversibly—may also be relevant (Williams, 1997). Such judgments are controversial and remain under discussion (Karlen et al., 1997, 2001; Okrent, 1999; Sojka and Upchurch, 1999).

In this paper we examine three approaches to derive minimum desirable C contents in soils. The three approaches are independent, based on (i) statistical data from national soil surveys, (ii) the rates of organic C loss and recovery predicted by the CENTURY model within time scales relevant for intergenerational equity, and (iii) the opinion of an expert panel of New Zealand soil scientists. We applied these three approaches to four major soil orders in New Zealand (Hewitt, 1998) and compared the similarities or divergence between the values obtained. We emphasize that, at this stage, it is the approaches we are examining, and not the precise numerical values they produced. For the basis of our calculations we made some arbitrary choices regarding the C targets and timeframe and we accept that other targets and time scales could be applied. We seek feedback on the acceptability and interpretation of the three approaches, and invite suggestions for defensible alternative methods.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Criteria to Select Soils, Carbon Targets, and Time Scales
Long-term pasture soils were used as the baseline to define "maximum" equilibrium C contents. Pastures are the dominant land use in New Zealand, widely distributed across all soil orders, and topsoil organic C contents are similar to indigenous forest on equivalent soils (Sparling and Schipper, 2002). We selected 80% of the pasture C content as our target value because long-term field experiments and modeling show that C recovery is asymptotic, and recovery to 100% could take an infinite amount of time (Parshotam and Hewitt, 1995). Although arbitrary, the 80% target appears attainable under New Zealand temperate climate cropping systems with appropriate farming practices (Francis and Knight, 1993; Shepherd et al., 2001). Our choice for the allowable time for recovery was 25 yr, based on the common parent–child age gap used for human intergenerational comparisons (Williams, 1997; Okrent, 1999). Only pasture soils were selected for this study, and only those soil orders where we had total C and bulk density measurements for each horizon. In addition, the orders were restricted to those where we could obtain additional C data to calibrate the CENTURY model, and orders that had been considered by the expert panel (explained below). To make our outputs comparable, we have calculated all C contents on a mass basis (Mg ha-1) to a soil depth of 20 cm.

Statistical Approach
Pasture soils (some 77% of the dataset) were selected from the New Zealand National Soils Database (NSD) (Hewitt, 1998). The samples (from soil pits) were originally described in soil surveys because they were representative of particular soil profiles. All soil orders are represented in the database, with descriptive information for each soil horizon and a varying amount of analytical data.

Four contrasting orders meeting the selection criteria were the Recent (nearest equivalent in USDA Soil Taxonomy: Endoaquept), Granular (nearest equivalent: Haplohumult), Melanic (nearest equivalent: Haplustoll), and Allophanic (nearest equivalent: Hapludand). Some topsoil characteristics of the orders are shown in Table 1. Median and lower quartile values were calculated for the C content of each order. The profile information for C content, bulk density, and horizon depth was grouped by soil order and spline curves fitted to the horizon data to enable C mass from the 0- to 20-cm depth to be derived (Bishop et al., 1999). Granular (n = 5) and Melanic soils (n = 7) were poorly represented in the database, and statistical values for these orders are less robust than for Recent (n = 31) and Allophanic soils (n = 31).


View this table:
[in this window]
[in a new window]
  		Table 1. Topsoil characteristics of the Recent, Melanic, Granular, and Allophanic soil orders under long-term pasture in New Zealand.{dagger}

 
Simulation Using the CENTURY Model
The CENTURY model (Kelly et al., 1997) was used to calculate the rate of change in total organic C contents (Mg ha-1) in the 0- to 20-cm depth of soil. Input parameters were selected for each of the four soils, for grass pasture or maize (Zea mays L.) crops, and the simulations validated against existing data for representative paired sites under long-term pasture and long-term cropping (Sparling et al., 2000a; G. Shepherd, personal communication, 2001). For each soil, input parameters were rainfall, temperature, sand, silt, clay, bulk density, field capacity, wilting point, P sorption capacity, and the P slope function. The initial sizes of the passive-C pool and slow-C pool were also changed. For the Recent soil, after a 50-yr pasture simulation, the values 30 and 45 Mg ha-1 were obtained for the passive- and slow-C pools, respectively; 40 and 50 Mg ha-1 for the Granular soil; 50 and 45 Mg ha-1 for the Melanic soil; and 75 and 50 Mg ha-1 for the Allophanic soil.

After simulating the C content under continuous (50 yr) legume–grass pasture to establish the maximum equilibrium level, the model was then changed to the scenario of a cropping regime of 25 yr of continuous maize, and the change in C pools recorded (Fig. 1) . Where the initial rate of C decline under the cropping phase was slow, the cropping phase was extended as required until the C level had declined below the 80% value. The rates of recovery after cropping were calculated by simulating a return to continuous pasture. Simulations were run for each soil order, with minimal changes to the input parameters. The equilibrium C contents under long-term pasture were used as the "maximum" C contents of each order. The target C content was defined as 80% of the maximum pasture C content. Lower C limits were estimated by modeling the minimum C content that still permitted recovery to the 80% target value within 25 yr.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 1. Example of soil C contents (Mg ha-1 to a 20-cm depth) during successive 25-yr periods under continuous pasture, continuous cropping, and back to continuous pasture in a Melanic soil in New Zealand as predicted by the CENTURY model.

 
For each soil, the annual loss or gain under pasture or cropping was calculated and the amounts under each scenario were accrued. Rates of decline and accumulation were curvilinear (Fig. 1), and a simple accounting approach was used to calculate the C balances. The maximum number of years under cropping, and hence the minimum C content, were established by the point at which the accrued annual losses could be matched by the accrued gains under the pasture recovery phase, to meet the target C content of 80% of the long-term pasture C content within 25 yr.

Expert Panel Approach
Our approach generally followed the methodology of Smith (1990), where the opinion of an expert panel was used to derive response curves and to define critical limits for environmental monitoring. Differences were that Smith (1990) used the Delphi method with anonymous participants communicating by mail, whereas we convened an intensive 2-d workshop and the group of 24 scientists was free to interact through a professional facilitator (Sparling and Tarbotton, 2000). The individuals in the group were asked to estimate soil quality score against the organic C contents for different soil orders using separate criteria for production and environmental risk. The definition of soil quality was that used by the Soil Science Society of America (1995): "the capacity of a specific soil to function, within natural or managed ecosystem boundaries, to sustain plant and animal productivity, maintain or enhance water and air quality, and support human health and habitation." This definition was used to derive production and environmental criteria for the expert panel to consider. Production criteria used in the workshops were agricultural productivity (e.g., plant dry matter, milk solids, saw logs), maximum economic yield, farm profitability, and impacts on the rural economy, all occurring on-site and within a short (<5 yr) time frame. No specific minima or maxima were defined. Environmental criteria, occurring both on- and off-site, were risks to air quality (including C sequestration), risk to water quality (surface and ground), loss of habitat, amenity, loss of diversity of indigenous species, invasions by weeds and pests, and contaminant accumulation occurring over a 25 yr (or longer) time frame. Again, no specific values for these criteria were defined.

After display and group discussion, individuals had the opportunity to redraw their response curves (nominal group technique). Some group members declined to draw curves, or more commonly, only drew those parts of curves where they felt confident. The individual curves or part-curves were converted to numeric values, and means and standard deviation calculated. An example of the response curve applicable to a Recent soil is shown in Fig. 2 . From the response curve, a soil quality value of 90% was taken as the cut-off point to calculate the organic C content below which soil quality was considered to decline. Values were converted to Mg ha-1 to a depth of 20 cm using the factors previously derived for the NSD data.



View larger version (23K):
[in this window]
[in a new window]
  		Fig. 2. Soil quality rating (arbitrary scale; 0 = minimum, 100 = maximum) determined by an expert panel of soil scientists in relation to soil organic C content for New Zealand Recent soils (USDA Soil Taxonomy nearest equivalent: Endoaquept), as assessed against criteria for production (closed circles) or environmental risk (open circles). Bars show standard deviation.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Statistical Approach
The four soil orders under long-term pasture had different median and lower quartile C contents, reflecting their different mineralogy and texture. The median C contents of long-term pastures on the Recent, Granular, Melanic, and Allophanic soils in the NSD were 72, 88, 98, and 132 Mg ha-1, respectively (Table 2). Lower quartile values were 54, 78, 73, and 103 Mg ha-1, respectively.


View this table:
[in this window]
[in a new window]
  		Table 2. Carbon contents (standardized to the 0- to 20-cm depth) of Recent, Melanic, Granular, and Allophanic soil orders as calculated from the median value of long-term pastures from the New Zealand National Soils Database (NSD); maximum C content of long-term pastures predicted by the CENTURY model, with the target value set at 80% of the maximum; predicted rates of accumulation in 25 years; the minimum C contents that still allow recovery to the target C content within 25 years; the lower quartile value from the NSD; and opinion from an expert panel considering production or environmental risk criteria.

 
Simulation Modeling Using the CENTURY Model
The CENTURY model predicted equilibrium long-term pasture C contents of 78, 93, 102, and 134 Mg ha-1 for the Recent, Granular, Melanic, and Allophanic soils, respectively (Table 2). These C contents were similar to the NSD median values for the same soil orders. Target C contents (80% of the long-term pasture value predicted by the CENTURY model) were 62, 74, 82, and 107 Mg ha-1 for the Recent, Granular, Melanic, and Allophanic soils, respectively (Table 2). The times taken for these soil orders to decline from the initial maximum to the target C content were 27, 10, 26, and 69 yr, respectively.

The lower C limits were calculated by matching the rates of depletion under a continuous cropping phase with the subsequent recovery under permanent pasture. An example of the changes in C contents of the Recent soil under the different regimes is given in Fig. 1. The minimum C contents that still allowed recovery to the 80% target over a further 25 yr were 55, 54, 67, and 104 Mg ha-1 for the Recent, Granular, Melanic, and Allophanic soils, respectively. With the exception of the Granular soil, these values were very similar to the statistical lower quartile values from the NSD (Table 2).

There were also marked differences in the predicted rates of accumulation of C for the four orders (Table 2). Accumulation over the 25-yr period was most rapid in the Granular and Melanic soils (21 and 15 Mg ha-1, respectively) and much less in the Recent and Allophanic soils (9 and 4 Mg ha-1, respectively).

Expert Panel Opinion
The panel opinion was that the C content of the various soils could be allowed to go much lower than lower limits suggested by the CENTURY model or the NSD lower quartile values, without seriously compromising production or causing adverse environmental consequences (Table 2). When expressing these opinions the panel members were not formally aware of the NSD lower quartile or the CENTURY model outputs. There was consistent opinion (small standard deviation) about a high soil quality ranking at high organic matter contents, but much greater variation in opinion (larger standard deviation) about the relationship between soil quality and organic matter content at lower organic C contents. Figure 2 shows the data for the Recent soils; similar patterns were obtained for the three other orders. The panel considered C contents as low as 20 Mg ha-1 acceptable for production purposes. When environmental rather than production criteria were included, the recommended lower C limits were approximately double those of the production criteria (Table 2).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
Target Carbon Contents and Lower Limits
The C contents derived here are presented as examples of three different approaches that could be used to derive desirable maximum and lower C contents; the numerical values themselves are not suggested as definitive or as a basis for setting environmental standards. We used various arbitrary criteria in their derivation, and using different criteria would affect the lower limits and time allowable for any period of loss and recovery. We accept that targets other than 80% of the long-term pasture C contents may be valid, but note that organic C contents of more than 80% of those under pasture are attainable with appropriate management under cropping systems in New Zealand (Francis and Knight, 1993; Shepherd et al., 2001). We also accept that statistical limits other than the lower quartile could be selected. Although the quartile value has no ecological or agronomic justification, Halvorson et al. (1996) also used the 25th percentile as a soil indicator threshold and lower quartiles are frequently used for educational and economic groupings. The use of the 25-yr timeframe is perhaps less contentious, as this is commonly adopted for intergenerational comparisons as a typical parent–child age gap. We welcome suggestions on alternative rationale to define target C contents.

As anticipated from their different textures and mineralogy, the four soils showed marked differences in C storage and dynamics. The greater amount of C storage in Allophanic soils is well-known (Goh, 1980; Eswaran et al., 1993; Batjes, 1996) and is generally attributed to the presence of allophane clay and ferrihydrite, which stabilize organic matter. Ross et al. (1982) noted comparatively little decrease in C contents of an Allophanic soil under long-term maize cropping, and Saggar et al. (1996) attributed the slower rate of organic matter decomposition in Allophanic soils to the high reactive surface area of allophane clays. Percival et al. (2000) found that a combination of pyrophosphate extractable–Al and allophane content explained a large proportion of the variability in organic C contents in New Zealand pasture soils, with a smaller additional contribution from extractable-Fe and clay contents.

The equilibrium C contents predicted by the CENTURY model for long-term pastures were very similar to the median for that soil order calculated from the New Zealand NSD. The good agreement was encouraging given that the CENTURY model was not initially developed for New Zealand conditions. An unexpected similarity was the agreement between the lower quartile value from the NSD and the lower C limit predicted by the CENTURY model (Granular soil excepted). The similarity may be fortuitous and coincidental. Other statistical limits (e.g., decile 10%) or a different intergenerational span could also be used, in which case the agreement would not be so good. The basis for using any statistical value to define outliers or target values depends on the representativeness of the sample population. The NSD data were collected by targeted sampling at sites of particular interest for soil survey, and are not claimed to be a spatially random subset or a regular grid. However, when Tate et al. (2001) compared NSD soil C contents of volcanic soils against a new set selected using random spatial sampling, the two sets did not differ significantly. The NSD does therefore appear to be representative of C contents of New Zealand soils, at least for the volcanic order (which includes the Allophanic soils).

A lower C limit can be justified in terms of maintaining soil C contents for future generations of land users. The argument would be that the soils could potentially be restored to the target value within a 25-yr period, and hence would not be open to the accusation of depleting the resource for the next human generation, and so maintaining intergenerational equity. This reflects an ethical stance, rather than an environmental or production argument. As well as failing the equity argument, low-C soils have a greater risk of structural degradation (Soane, 1990; Reeves, 1997; Shepherd et al., 2001) and lower profits (Scrimgeour and Shepherd, 1998). Lower organic C contents on cropping farms required greater inputs for tillage, more critical timing, and greater use of fertilizers (Scrimgeour and Shepherd, 1998). Despite this, the expert panel clearly felt that low C contents were tolerable, and that farm profitability would be adequate to meet any higher production and management costs resulting from lower soil C contents. That view is substantiated by the numerous examples of Granular (Haynes and Tregurtha, 1999), Recent (Shepherd et al., 2001), and Melanic soils (Sparling et al., 2000a, b) with C contents below the so-called lower limit, that are still used for arable cropping in New Zealand. A similar point about low-C soils still being productive under appropriate management is made by Sojka and Upchurch (1999) for cropland soils in the USA.

We are currently using provisional C target figures to interpret data on soil C contents under different land uses (Lilburne et al., 2002). In the absence of good data for modeling, the target figures were derived using the statistical and expert opinion approaches only. However, even these basic targets have been useful to quantify soil and land use combinations of concern and to assist with land management policy and state of the environment reporting (Sparling and Schipper, 2002). The targets for desirable C contents will continue to be refined as better data or other methods become available.

Resistance and Resilience
The C contents and the time taken for the decline and recovery in organic C provide a measure of the resistance and resilience characteristics of soils (Greenland and Szabolcs, 1994; Lal, 1994, Seybold et al., 1999). The NSD data for long-term pastures showed that Allophanic soils have the greatest C contents, suggesting the soils have a resistance ranking (to C loss) of Allophanic > Melanic > Granular > Recent. On the basis of the time taken to be depleted to the 80% target, as modeled by CENTURY, the soils had the ranking Allophanic >> Recent = Melanic > Granular. Based on the minimum C content, the order was Allophanic > Melanic > Granular = Recent. This general pattern, with Allophanic soils consistently having high resistance to C loss, is similar to the structural resistance ranking of Allophanic > Granular > Melanic > Recent found by Shepherd et al. (2001). Both C loss and structural degradation methods agree that Allophanic soils are the most resistant, but differed particularly in the ranking of the Granular order. Granular soils are known to maintain good structural stability even when C is lost, because they contain large amounts of Al and Fe oxides and hydroxides (Parfitt et al., 1997). These minerals contribute to stable soil aggregates (Shepherd et al., 2001), and this characteristic could explain the different position of the Granular soil in the resistance rankings.

Carbon Sequestration
Increased storage of C as soil organic matter is now advocated as a means whereby nations could decrease gaseous C emissions and lower greenhouse gas concentrations.

In countries such as the USA, with large areas of croplands with low soil organic C, there is the potential for large amounts of storage (Paustian et al., 1997; Lal, 2001). Projects such as Conservation Reserves, no-till systems, residue retention, and rotations provide management options to enhance and maintain soil organic matter contents (Karlen et al., 1998). However, even with these measures, recovery of C contents in depleted soils can be slow (Karlen et al., 1998; Baer et al., 2000). Approaches such as those suggested here may be useful to identify soils most suitable for rapid C incorporation and storage.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Soil C content is an important attribute for productive soils, ecological function, and C storage. We suggest that a maximum C content and lowest desirable C content, specific to soil order, need to be defined. Of the three approaches examined here, the modeling approach has appeal, in that it can be used to define maximum and lower limits and recognizes the dynamic nature of soil organic matter, and the lower limit can be justified by the ability of a soil to recover to a target value within specified timeframes. A limitation is that we have very little data to calibrate the models and extend them to a wider range of soils and land management options.

The statistical approach was useful to identify maximum desirable C contents, provided a representative set of soils is available. Our preference is not to use a statistical approach to set lower limits. Lower quartiles as used in this study were an arbitrary choice and did not have any ecological basis, despite the reasonable agreement with the modeling predictions. Further, setting lower C contents at the lower quartile means, by definition, that 25% of the sample will fall below that limit, even if the actual C content is not depleted.

To supplement modeling and statistical data, the judgements from expert panels can be used. Panel opinions have the advantage being able to integrating existing scientific knowledge, personal experience, anecdotal evidence, and informed guesses. However, in the absence of hard evidence, panel decisions may be difficult to justify and open to challenge from opposing opinion.

We have used statistical and expert opinion approaches to define provisional target C contents for a limited number of soils and land uses in New Zealand. This has enabled us to identify soil and land use combinations outside the target ranges that can be used for land management advice and state of the environment reporting.


   ACKNOWLEDGMENTS
 
Funding for part of this project was provided through the New Zealand Foundation for Science Research and Technology, Contract C09X0016 and the Ministry for the Environment Sustainable Management Fund.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 


Related articles in JEQ:

This Issue in Journal of Environmental Quality

JEQ 2003 32: 745-750. [Full Text]  



This article has been cited by other articles:


Home page
 J. Environ. Qual.Home page
G. P. Sparling, D. Wheeler, E.-T. Vesely, and L. A. Schipper
What is Soil Organic Matter Worth?
J. Environ. Qual., March 1, 2006; 35(2): 548 - 557.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Related articles in JEQ
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (14)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Sparling, G.
Right arrow Articles by Schipper, L. A.
Related Collections
Right arrow Soil Organic Matter
Right arrow Carbon Sequestration
Right arrow Ecological Risk Assessment

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome