SSSAJ Grow Your Career with SSSA
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 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 (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.
Soil Science Society of America Journal 67:1206-1212 (2003)
© 2003 Soil Science Society of America

DIVISION S-5—PEDOLOGY

Sources of Uncertainty Affecting Soil Organic Carbon Estimates in Northern New York

John M. Galbraith*,a, Peter J. A. Kleinmanb and Ray B. Bryantb

a Crop & Soil Environmental Sciences Dep., Virginia Polytechnic Institute and State Univ., 239 Smyth Hall (0404), Blacksburg, VA 24061
b USDA-ARS, Pasture Systems and Watershed Management Research Unit, Building 3702, Curtin Road, University Park, PA 16802-3702

* Corresponding author (ttcf{at}vt.edu)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
Estimations of regional soil organic C (SOC) stores are sensitive to map scale effects, the geography of soil resources, and uncertainty stemming from SOC values assigned to soil series. This study assessed uncertainty associated with regional SOC estimation for the Tughill Plateau (Major Land Resource Area [MLRA 141]) of northern New York. Soil samples from 103 sites, representing 30 soil series mapped within the Tughill Plateau, were collected from the upper meter and analyzed to develop a series-specific SOC database. Four 24 km2 subregions of the Plateau were identified and soil survey maps of four different scales ranging from 1:15840 to 1:750000 were compared. Soil survey maps were digitized, and their respective map unit compositions and areas were related to the series-specific SOC database to calculate SOC (weighted by land extent). Estimates using different survey map scales ranged from 16.6 to 17.6 kg C m-2. Smaller-scale soil survey maps (1:250000 and 1:750000) produced SOC estimates that were greater significantly than the 16.7 kg C m-2 estimated using the large-scale reference survey maps (1:15840). Differences were caused by the process of map generalization, as soil series with lower mean SOC that occurred on the 1:15840 survey maps were replaced by series that had higher SOC on the 1:250000 and 1:750000 survey maps. Although significantly different, area-weighted mean SOC estimates varied by <5%. The greatest uncertainty in regional SOC estimation originated from the variation in SOC values assigned to soil series, with coefficients of variation (CV) ranging from 3 to 87%.

Abbreviations: CV, coefficients of variation • MLRA, Major Land Resource Area • SOC, soil organic C


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
ACCURATELY QUANTIFYING SOC stores in soils is fundamental to global climate change modeling, because the soil C reservoir is over three times as large as atmospheric CO2 and over four times as large as biomass C reservoirs (Brady and Weil, 2002). According to Arnold (1995), uncertainty in estimation of regional SOC can result from the choice of the scale of the soil survey map used to extrapolate soil data to larger areas. Regional SOC estimation is affected by map scale, because map delineations and map unit composition change with scale. The overall effect of scale on SOC estimation remains unclear.

Soil survey map scale effects on SOC estimates are influenced by the geography of soil resources. Soils that occur in large, extensive delineations on detailed soil maps are likely to be major components of general soil map units. However, soils that only occur in small delineations on detailed maps are in danger of being eliminated or becoming map unit inclusions during the map generalization process (Rapalee et al., 1998). For instance, Histosols, key soils in terms of SOC storage in regions such as northern New York, most often occur in small delineations. Franzmeier et al. (1985), Davidson and Lefebvre (1993), and Alexeyev et al. (1996) found that Histosols contribute a greater proportion of the regional C than their proportion of the land area. Thus, the potential omission of high C soils, such as Histosols, from small-scale map unit composition would lower regional SOC estimations.

In addition to the interactive effects of soil survey map scale and the geography of soil resources, the uncertainty associated with an SOC value assigned to a soil series also affects estimates of regional SOC stores. Most soil databases (e.g., USDA-NRCS's Soil Interpretation Record Database, NRCS-SIR) do not include data from surface O horizons of mineral soils. Huntington et al. (1988) found that surface litter averaged 3.0 kg C m-2 and accounted for almost 19% of volumetric SOC in the solum of mostly moderately deep, frigid, glaciated soils in New Hampshire. Thus, surface litter C is an important component of soils in northern regions. In addition, SOC measurements are often missing from these databases for subsurface horizons. Stone et al. (1993) found about 17% of the SOC in Florida Spodosols (mainly Aquods) stored in the Bh horizon. Franzmeier et al. (1985) reported that up to half of SOC in Histosols is found in the subsoil.

Previous studies comparing regional SOC estimates reported uncertainties related to map scale and series-specific SOC values. Kern (1994) calculated regional SOC using three maps ranging in scales from 1:5000000 to a resolution of 0.5° latitude/longitude. He found that the mean SOC increased from a range of 13.6 to 15.0 kg C m-2 to a range of 24.1 to 28.0 kg C m-2 as map scales decreased. However, methods of SOC determination within the pedon database varied, and methods of aggregating pedon data to represent map units varied with map scale. Elsewhere, Homann et al. (1998) and Davidson and Lefebvre (1993) reported conflicting trends in regional SOC approximation associated with map scale. Both studies applied NRCS-SIR data (National Soil Survey Center, 1997) to maps of varying scales to estimate regional SOC. Homann et al. (1998) compared STATSGO (1:250 000 scale) and NATSGO (1:7500000 scale), while Davidson and Lefebvre (1993) compared STATSGO and the FAO/UNESCO Soil Map of the World (1:5000000 scale; Food and Agriculture Organization of the United Nations, 1988). Although Homann et al. (1998) found regional SOC estimates to decrease at smaller map scale, Davidson and Lefebvre (1993) found it to increase. These conflicting trends indicate the need for further study.

This study assesses sources of uncertainty in regional SOC estimation for the Tughill Plateau. Regional SOC is calculated by projecting a single SOC database of dominant soil series occurring in the Tughill Plateau with four soil survey maps of differing scale (from 1:15840 to 1:750000). The Tughill Plateau was selected as a focus area for the study as it is a complete MLRA in north central New York (Soil Conservation Service, 1981; National Soil Survey Center, 2001) with complex soil geography that is representative of cold, humid, glaciated parts of the northern USA. Soil organic C in such areas is likely to be underestimated when projected with small-scale maps because of changes in map unit composition during generalization or loss of small delineations of Histosols at small scales.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Study Area
The study area (Fig. 1) is MLRA 141, the Tughill Plateau, centered at 43° 30' 00'' N lat., 75° 30'0'' W long. in north central New York. The Tughill Plateau is part of the cool, temperate, wet forested life zone (Holdridge, 1947). Mean annual temperature is 8°C, and annual precipitation is between 1000 and 1500 mm. The Tughill Plateau has an area of 3080 km2 and a mean elevation of 575 m above sea level (Soil Conservation Service, 1981). Vegetation is mixed northern hardwoods and conifers, with scattered conifer plantations. Land use is dominated by forestry for pulp and timber production, and includes <10% agriculture, primarily crop and hay production for dairy farms. Tughill bedrock comprises Ordovician sandstone and shale formations. The soil moisture regime is udic and soil temperature regime is frigid (Soil Survey Staff, 1999). The area was deglaciated approximately 11 000 BP (Isachsen et al., 1991). Most mineral soils are formed in deposits of glaciofluvial material or dense glacial till derived mainly from sandstone and shale. The uplands are mainly covered by Orthods and Udepts with Aquods and Aquepts in low-lying or level areas and Histosols in the numerous glacial potholes.



View larger version (58K):
[in this window]
[in a new window]
  		Fig. 1. Map of north central New York state (in the northeastern USA) showing the Tughill Plateau (MLRA 141) in gray, centered at 43° 30' 00'' N long. and 75° 30' 00'' W lat. Digitized map areas shown by rectangles.

 
Field Sampling and Laboratory Analysis
Because the NRCS-SIR database did not include litter layer SOC data and many series lacked SOC data for some or all layers, direct sampling of soils was required to develop a database of SOC values for series in the Tughill Plateau. Soil series that were dominant components of map units in the soil surveys used for making scale comparisons in this study (Table 1) were selected for sampling. Whereas the objective was to represent individual series as they occur within the Tughill Plateau, replicate sampling sites were distributed throughout the geographic extent of where that series is mapped in the Tughill Plateau. Map units and land uses as depicted in Jefferson County (McDowell, 1989), Lewis County (Pearson and Cline, 1960), and Oswego County (Rapparlie, 1981) soil surveys were used as a guide for selecting sampling sites, but final site selection was predicated on field verification of the mapped series. Soils were not sampled if they did not represent the dominant soil in the map unit description or were physically disturbed by tree-fall, animal, or human activity. Oneida County does not have a published county soil survey, but field sheets from unpublished survey maps were available for use in site selection.


View this table:
[in this window]
[in a new window]
  		Table 1. Extent, sources, dates, and scales of soil survey maps used to project regional soil organic C.

 
A total of 103 sites were sampled representing the 30 most extensive soil series in the Tughill Plateau. Where they occurred, litter layers were sampled separately from mineral soils. Specifically, litter layers were sampled within a 20 by 20 cm frame. After removal of the litter sample, underlying mineral horizons were sampled by 8-cm open-sided auger to the shallower of 1-m below the surface of the litter layer or bedrock. All mineral soil from each auger hole was collected, air-dried, ground, sieved to <2 mm, and homogenized. Subsamples were analyzed for organic C by dry combustion using a LECO induction furnace (LECO Corporation, St. Joseph, MI). All soils were tested with 10% HCl to ensure that they did not contain carbonates.

Following organic C determination for individual samples, SOC (kg m-2) was calculated as follows:

[1]
where Sample Weight is the oven dry (<2 mm) sampled soil weight (kg), OC is the sample organic C concentration (kg kg-1), and Area is the cross-sectional area sampled (0.04 m-2 for litter layers; 0.005 m-2 for auger samples). Thus, SOC (kg m-2) for each sampled soil was calculated as the sum of the SOC of the litter layer (if present) and SOC of the underlying mineral soil to a depth of 1 m or bedrock. Mean SOC values were calculated for each of the 30 series included in the direct sampling database.

A substitute SOC value was assigned to all series that were listed as map unit inclusions but that were not among the 30 dominant series sampled. The substituted SOC was chosen by correlation between the unsampled soil and the best fit from the list of 30 dominant soils. For example, the 1:250000 scale maps contained a few small areas of Berkshire soils (well drained, coarse-loamy, isotic, frigid Typic Haplorthods formed in dense till). Since this soil was not sampled in our study, the SOC value for the Worth series (well drained, coarse-loamy, mixed, frigid Typic Fragiorthods formed in dense till) was used as a substitute. This substitution of data occurred in one to three map units at each scale, comprising no more than 3% of the area of any map. Inclusions of rock outcrop were assigned 0 kg C m-2.

Extrapolation of Soil Organic Carbon
Four subregions, consisting of atlas sheets from the Oswego County Soil Survey (1:15840) and ranging in size from 23.7 to 24.1 km2, were selected to represent a south to north transect of the Tughill soil resources (Fig. 1). Coordinates of the corners of the atlas sheets were digitized using Arc/Info, Version 7.4 (Environmental Systems Research Institute, Inc., Redlands, CA) to define the same geographic area for comparison on soil survey maps of different scales. Map units within these four geographic areas, as depicted on each of the four soil survey maps used in this comparison (Table 1), were digitized.

Mean SOC (kg m-2) for each atlas sheet was calculated as an area-weighted average of map unit SOC on a land area basis (areas depicted as water were subtracted from the total area of the map sheet). Specifically, SOC values for map units were calculated using the mean SOC value for the series, weighted by percentage of composition of the series in the map unit. Map unit compositions were either reported in the map unit descriptions (or accompanying database) of each soil survey or were taken from the MUIR Database web site (National Soil Survey, 2003). Reported percentages of areas of map unit inclusions were divided equally among the named inclusions in the map unit descriptions. For example, in a map unit of "Adams-Windsor complex, moderately steep" from the Oswego County Soil Survey, the dominant series Adams and Windsor are described as making up 50 and 40% of the unit. The remaining 10% of the unit was divided equally among the specified inclusions (i.e., Colton–5% and Hinckley–5%).

Statistical Analyses
Differences in SOC for individual series and taxonomic groups, as well as differences in SOC related to map scale, were assessed by ANOVA, pair-wise comparisons conducted by Tukey's analysis ({alpha} = 0.05; Snedecor and Cochran, 1989). Descriptive statistics were used to compare variability in SOC related to soil series (and higher taxa) and the geography of soil resources. All analyses were conducted using Minitab statistical software, version 13 (Minitab, Inc., State College, PA).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Soil Organic Carbon Database
As illustrated in Table 2, SOC contents measured in this study ranged from 9.0 to 25.8 kg C m-2. Series and higher taxa presented in Table 2 are ranked on the basis of SOC content and arbitrarily separated into categories (>20, 16–20, 12–16, and <12 kg C m-2) to facilitate discussion. For those series and higher taxa represented by five pedon samples or more, CV ranged from 18 to 50%. Notably, the Worth series, which is represented by 11 samples, has a CV value of 37%, indicating that number of observations alone is not responsible for the high variability in observed SOC contents. Comparable variation in SOC content has been documented in other studies, such as Eswaran et al. (1993) and Kern (1994), who reported CV values of 42 and 46%, respectively, for various Spodosols.


View this table:
[in this window]
[in a new window]
  		Table 2. Classification and descriptive statistics of soil organic C (SOC) of series and higher taxa sampled in the Tughill Plateau ranked in order of decreasing SOC. Soil series separated arbitrarily at levels of 20, 16, and 12 kg C m-2.

 
Series or higher taxa with the highest mean SOC contents are poorly or very poorly drained and have saturation by apparent water tables at or near the surface for most of the growing season. Similar findings were reported by Rapalee et al. (1998), who found that poorly drained soils possessed the highest SOC content in a boreal forest landscape. However, this trend is not consistent in the Tughill Plateau, as series that have aquic moisture regimes or are members of aquic subgroups also possess some of the lowest SOC contents (Table 2).

As shown in Table 3, Histosols contain the highest mean SOC by soil order, which, although not significantly different from Spodosols, is significantly greater than the mean SOC content of Inceptisols and Entisols. Notably, the mean SOC of Histosols is lower than that reported elsewhere in the literature (e.g., Franzmeier et al., 1985), as Histosols in the Tughill Plateau tend to be shallow or moderately deep organic deposits.


View this table:
[in this window]
[in a new window]
  		Table 3. Comparison of means and descriptive statistics of soil organic C (SOC) grouped by soil order.

 
Because many of the pedons sampled in the Tughill are under forest cover, litter layers contribute substantially (up to 25%) to pedon SOC contents of mineral soil series (Table 2). In those mineral soil series that possess litter layers, a mean of 8% of the total SOC derives from the litter. The importance of collecting and accounting for C in the litter layer of mineral soils is supported by other studies, such as Huntington et al. (1988), where litter accounted for 19% of pedon SOC of mostly moderately deep soils. As approximately 90% of the Tughill Plateau is forested and more than 60% of the mineral soil series sampled in this study possess litter layers, SOC databases that do not include data on litter layers would underestimate SOC for this region.

Estimation of Regional Soil Organic Carbon Reserves
Map unit delineations in the four subregions are presented in Fig. 2 for all four map scales. As the map scale becomes more generalized, smaller delineations disappear. Notably, water bodies (presented as solid dark map units) also disappear on small-scale maps. Estimates of area-weighted mean SOC from small-scale maps that have few delineations are highly sensitive to descriptions of map unit composition, and loss of information about small delineations may affect regional SOC estimates derived from general maps. The geography of soil resources in this study area does present a challenge for making small-scale maps that accurately represent SOC stores.



View larger version (66K):
[in this window]
[in a new window]
  		Fig. 2. Comparison of the four map sheet areas at different scales, with notable loss of small delineations at more general (smaller) scales. Water bodies are solid dark colors. Map sheet numbers from the Oswego County Soil Survey are shown above each column.

 
Table 4 presents area-weighted SOC estimates for each of the four subregions, as determined from the maps of different scales. For individual soil survey map scales, variability in the geography of soil resources, as represented by the four subregions illustrated in Fig. 2, is very low, with a maximum CV value of 2%. This provides some confidence that the size of the four subregions is large enough to encompass the variability in the geography of soil resources and thereby eliminate this factor as a major source of uncertainty in SOC estimates.


View this table:
[in this window]
[in a new window]
  		Table 4. Comparison of mean area-weighted soil organic C (SOC) within each map scale and descriptive statistics of SOC across four scales. Changes of percentages relative to 1:15 840 SOC are noted.

 
The Oswego County Soil Survey (1:15840 scale) provides the greatest detail of soil distribution within the Tughill Plateau and serves as a reference against which small-scale soil maps are compared. Area-weighted mean SOC contents, calculated for each of the four atlas sheets from the 1:15840 survey, ranged from 16.2 to 16.9 kg C m-2, with an overall mean of 16.7 kg C m-2 (Table 4). These SOC contents are not significantly different from area-weighted mean SOC values derived from the Oswego County General Soil Survey (1:62500 scale). However, the smaller-scale STATSGO Soil Map (1:250000 scale) and New York State General Soil Map (1:750000 scale) did produce area-weighted mean SOC estimates that are significantly greater than those derived from the 1:15840 map (4 and 5% higher, respectively). To understand why these maps produce different estimates of regional SOC, the proportional representation of individual series, as affected by map unit composition and spatial extent, must be considered.

The extent to which individual series and higher taxa are represented in maps of differing scale is shown in Table 5. In spite of the fact that the series and higher taxa with >20 kg C m-2 are substantially underrepresented in the 1:250000 relative to the 1:15840 maps, area-weighted mean SOC estimated by the 1:250000 maps was 3.9% higher than the 1:15840 map (Table 4). In the case of the 1:250000 maps, the underrepresentation of series and higher taxa >20 kg C m-2 was outweighed by the substantial overrepresentation of series and higher taxa in the 16 to 20 kg C m-2 range (Table 5). For instance, the Bice series (Coarse-loamy, mixed, active, frigid Typic Dystrudepts)(18.8 kg C m-2) that made up 23.3% of land area of the 1:250000 maps did not occur in the 1:15840 maps. Underrepresentation of the extent of soils with <12 kg C m-2 also contributes to an over-estimation of area-weighted mean SOC by the 1:250000 maps. Elsewhere, Davidson and Lefebvre (1993) found the STATSGO map (1:250000) to overestimate SOC by 13% relative to a detailed county level soil survey (1:20000). They attributed these differences in SOC to variation in the horizon depths between the databases, a source of uncertainty that is not germane to this study.


View this table:
[in this window]
[in a new window]
  		Table 5. Soil organic C (SOC) and percentage of aerial extent (land area basis) at each scale, ranked in order of decreasing SOC. Subtotals placed arbitrarily at 20, 16, and 12 kg C m-2.

 
In the case of the 1:750000 maps, soils with >20 kg C m-2 are overrepresented relative to the 1:15840 maps (Table 5). Additionally, the 1:750000 maps underrepresents the extent of soils having <16 kg C m-2. Consequently, area-weighted mean SOC estimated from the 1:750000 maps is 5.4% higher than that derived from the 1:15840 map (Table 4).

The area-weighted means estimated at all soil survey map scales evaluated by this study (16.2–18.0 kg C m-2) are comparable with those reported in the literature for similar soils. Kern (1994) estimated a range of 16.6 to 18.0 kg C m-2 area-weighted means using the 1:7500000 NATSGO map and the NRCS National Soil Survey Laboratory Pedon Database. Franzmeier et al. (1985), relating soil laboratory data to a 1:2500000 NCR-76 map for projection, reported area-weighted means of 10.9 to 19.6 kg C m-2 for soils similar to those of the Tughill Plateau (i.e., soils associated with the cool, temperate, moist forested life zone defined by Post et al., 1982) in the north central USA. Huntington et al. (1988) calculated area-weighted mean of 16.0 kg C m-2 for mostly moderately deep, frigid soils in a 23-ha glaciated upland watershed in New Hampshire based on direct sampling of representative soils and a detailed (1:10000) soil map.

In spite of the statistically significant differences observed in area-weighted mean SOC at various map scales, differences across map scales are actually quite small (maximum of 5%). In comparison, variability in mean SOC concentration of individual soil series and higher taxa is high, with CV values ranging from 3 to 87% (Table 2). These results show that, by far, the greatest source of uncertainty in estimating area-weighted SOC for a region is the database used to estimate series-specific SOC content.

Implications for Soil Survey and Regional Soil Organic Carbon Estimation
Contrary to the expectation that SOC would be underestimated by small-scale soil survey maps, because of the loss of small delineations of Histosols during the generalization process, small-scale soil survey maps significantly overestimated SOC in the Tughill Plateau. Notably, area-weighted mean SOC in this study was calculated on a land area basis. In projecting results regionally, land area is influenced by the extent of water bodies, which is usually determined from the same map used to assess soil resources. Although it did not affect our land-based estimations of SOC, we observed that small-scale maps portrayed substantially reduced areas of water bodies relative to the 1:15840 map; water bodies account for 2.1, 1.9, 0.4, and 0.6% of total area in the 1:15840, 1:62500, 1:250000, and 1:750000 maps, respectively. As a result, the extent of land area is overestimated in the small-scale maps. In making regional projections of total SOC stores, the overestimation of land area would combine with the overestimation of area-weighted mean SOC derived from the small-scale maps in this study to further inflate estimates of regional SOC stores by as much as 2%. Even with this added degree of uncertainty, variability related to soil survey map scale is relatively small. Advances in GIS tools should facilitate easier and more accurate generalization of detailed soil maps in the future, addressing uncertainties related to both map unit composition and percentage of water bodies.

As observed in this study, the largest source of uncertainty affecting regional SOC estimates derives from the SOC database for series and higher taxa. Given such variability within series, additional factors affecting mean SOC content may need to be included in predicting area-weighted mean SOC. For instance, impacts of land use on SOC may override soil-related differences, as evidenced by the contribution of surface O horizons to forested mineral soils (up to 25% of pedon SOC). Using GIS techniques, land use could be used to further refine SOC estimates within map units.


   ACKNOWLEDGMENTS
 
Funding provided by the USDA-Natural Resources Conservation Service (NRCS) National Soil Survey Center, 100 Centennial Mall North, Lincoln, NE 68508-3866.

Received for publication September 10, 2001.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 




This article has been cited by other articles:


Home page
 Soil Sci.Home page
Y. Zhao, X. Shi, D. C. Weindorf, D. Yu, W. Sun, and H. Wang
Map Scale Effects on Soil Organic Carbon Stock Estimation in North China
Soil Sci. Soc. Am. J., June 21, 2006; 70(4): 1377 - 1386.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
Y. Guo, R. Amundson, P. Gong, and Q. Yu
Quantity and Spatial Variability of Soil Carbon in the Conterminous United States
Soil Sci. Soc. Am. J., February 27, 2006; 70(2): 590 - 600.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
C. Rasmussen
Distribution of Soil Organic and Inorganic Carbon Pools by Biome and Soil Taxa in Arizona
Soil Sci. Soc. Am. J., January 6, 2006; 70(1): 256 - 265.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
J. A. Thompson and R. K. Kolka
Soil Carbon Storage Estimation in a Forested Watershed using Quantitative Soil-Landscape Modeling
Soil Sci. Soc. Am. J., June 2, 2005; 69(4): 1086 - 1093.
[Abstract] [Full Text] [PDF]

Home page
 Soil Sci.Home page
J. O. Miller, J. M. Galbraith, and W. L. Daniels
Soil Organic Carbon Content in Frigid Southern Appalachian Mountain Soils
Soil Sci. Soc. Am. J., January 1, 2004; 68(1): 194 - 203.
[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 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 (7)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Galbraith, J. M.
Right arrow Articles by Bryant, R. B.

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
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome