International Journal of Scientific Research in Knowledge, 2(5), pp. 249-256, 2014 Available online at http://www.ijsrpub.

com/ijsrk ISSN: 2322-4541; 2014 IJSRPUB http://dx.doi.org/10.12983/ijsrk-2014-p0249-0256

Full Length Research Paper Soil Organic Carbon: Relating the Walkley-Black Wet Oxidation Method to Loss on Ignition and Clay Content
Henry Oppong Tuffour1,2*, Israel Boateng Yeboah2, Mensah Bonsu2, Thomas Adjei-Gyapong2, Abdul Aziz Khalid2, Awudu Abubakar2, Caleb Melenya2, Pearl Kpotor1
1

School of Agriculture and Bio-Resources Engineering, Anglican University College of Technology, Nkoranza Campus, Nkoranza, Ghana 2 Department of Crop and Soil Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana *Corresponding author: hoppongtuffour@gmail.com; (+233) 208 542 308
Received 17 March 2014; Accepted 26 April 2014

Abstract. Accurate estimation of soil organic carbon (SOC) is needed in many fields of science including ecology, soil science and global climate change. Loss on ignition method (LOI) of estimating SOC has been proposed as rapid, accurate and cheap method but it has some limitations. Walkley-Black method is also widely used. For this study, 12 soil samples from 0-15 cm depth were collected from four sites of the Crop Plantation Section of the Department of Crop and Soil Sciences, KNUST. The amount of SOC (%) estimated by LOI and W-B methods, and clay content were used to develop simple and multiple linear regression equations to describe their relationships. The highest estimate of SOC (3.28%) was produced by Loss on ignition method, whereas W-B gave lower estimates with 1.35% being the highest produced by this method. The coefficient of determination, R2, for the equations ranged from 0.35 to 0.84. Very weak relationship was observed between LOI and W-B as evidenced by R2 value of 0.35. The clay content had a strong positive correlation (r = 0.70, p < 0.001) with LOI and gave R 2 value of 0.49, depicting a moderately strong relationship. However, multiple linear regression analysis of clay content and LOI values as independent variables and W-B values as the dependent variable provided reliable estimates of SOC with R 2 value of 0.84. Results from t-test analyses showed that estimates of the constants and coefficients of all variables in the equations were significantly different. Keywords: Loss on ignition, Walkley-Black, Soil organic carbon, Soil organic matter, Regression equations

1. INTRODUCTION Soil organic carbon occurs in soils as plant and animal matter, their immediate decomposition products and humus. It is the chief element of soil organic matter and it plays an important role in the global carbon cycle and climate change (Lal and Kimble, 2000). Naturally, SOC forms are derived from the decomposition of living organisms but other sources are those that are derived as a result of contamination through anthropogenic activities (Schumacher, 2002). The determination of total SOC is an essential part of site characterization since its presence or absence can markedly influence how chemicals will react in the soil. Again, it is typically requested with contaminant analysis as part of an ecological risk assessment data package. Furthermore, most terrestrial carbon is held in soils, more than twice as much as in vegetation or the atmosphere, and changes in SOC content can have a large effect on the global carbon budget (Bellamy et al., 2005). Soil organic carbon can be used for

assessing soil quality; it indicates the level of productivity of a given soil. Thus, large amounts of SOC will enhance water retention, prevent leaching of soluble nutrients, serve as source of sink for nutrients and maintain soil fertility (Murrage et al., 2000; Fullen et al., 2007; Bationo et al., 2009). Estimates of soil organic carbon often are needed for applications in ecology, soil science, and global climate change. Many methods are available for measuring soil organic carbon, each with advantages and disadvantages in terms of accuracy, expense, and convenience (Nelson and Sommers, 1996). Loss on ignition is cheap, fast and easy method of estimating SOC and Elemental carbon analyzer is very expensive than the other methods but very accurate (Konen et al., 2002). However, it is liable to errors from ignition temperature and duration and mass of soil to be ignited. Very low ignition temperatures will underestimate organic carbon and high temperatures and longer duration for ignition will consequently overestimate organic carbon (Heiri et al., 2001). It is

249

Tuffour et al. Soil Organic Carbon: Relating the Walkley-Black Wet Oxidation Method to Loss on Ignition and Clay Content

also significantly affected by clay content due to loss of structural water held in clay minerals during ignition in the muffle furnace and also not all SOM ignites (Konen et al., 2002; Brunetto et al., 2006; Escosteguy et al., 2007; Abella and Zimmer, 2007; Yerokun et al., 2007; Salehi et al., 2011). On the other hand, W-B has significant uncertainties regarding oxidation of other constituents in the soil and the proportion of total soil organic carbon that is oxidized. It also shows variable soil organic carbon recovery and generates wastes that contain strong acid and Chromium (Page et al., 1982; Abella and Zimmer, 2007). However, W-B has been proposed to be economically suitable for determining the SOC of Chilean volcanic soils (Matus et al., 2009). According to Beaudoin (2002) LOI and W-B, show very little variance in the estimation of SOC. The former gave good estimates of SOC and SOM, but it had the worse precision (Schumacher, 2002), among other methods including Walkley-Black method and it is generally deemed unreliable for soils with low organic carbon contents. When LOI and WB is compared with the reference method (i.e. dry combustion), an excellent relation existed between them, with a mean conversion factor of 1.25, which compared well with the recommended value of 1.32 (Walkley and Black, 1934; Grewal et al., 2006). Over the years, there have been different regression equations and models developed for the estimation of SOC by various methods. Regression equations between LOI and SOC determined by elemental carbon analyzer were developed by Konen et al. (2002) and had coefficient of determination for the equations ranging from 0.94 to 0.98 which showed very strong relationships. Clay content and clay mineralogy had significant effect on these equations. Abella and Zimmer (2007) developed multiple regression equations using LOI, SOC measured with elemental carbon analyzer and clay content. Clay content explained 78% (0-15 cm) of the variation in SOC. It was the only variable that improved slightly, the predictions in the multiple regression equations. Mean conversion factor of 14.9 based on clay content, was obtained by Sutherland (1998) for improving LOI. This high mean conversion factor reflected significant water loss by dehydration of Fe, Al, and Mn oxides at a muffle furnace temperature of 450C. Again, the LOI method has been widely used for estimating SOM in a muffle furnace at 360oC for 2 h (Schulte et al., 1991; Konen et al., 2002; Brunetto et al., 2006; Escosteguy et al., 2007; Yerokun et al., 2007) and at 300, 450, and 600oC for 2 h (Abella and Zimmer, 2007). Accurate estimation of within-field SOC is currently an important priority for precision agriculture, given its importance in defining precise

fertilizer and pesticide management practices, thus optimizing field productivity and minimizing groundwater contamination risks (Sudduth and Hummel, 1996; Ingleby and Crowe, 2001). With recent interest in studying the carbon cycle on arid land (Stone, 2008; Wang et al., 2012), there is a need to quantify SOC of arid soils in order to assess the most efficient method. In view of this, a cheap, simple and easy method for its estimation in the laboratory will be of a great necessity. The main objective of the study was to propose a simple and easy method for estimating SOC in the laboratory. 2. MATERIALS AND METHODS 2.1. Study Area Soil samples for the study were collected from the Crop Plantation Section of the Department of Crop and Soil Sciences, KNUST, Kumasi from four sites including; cocoa plantation, oil palm plantation, uprooted oil palm plantation land and arable land under cultivation (plantain intercropped with cassava). The soils belong to the Bomso series and classified as Acrisol (FAO, 1990). The vegetation is semideciduous forest and has bimodal rainfall distribution. 2.2. Soil sample Collection and Preparation Three soil samples each of soil depth, 0-15 cm were collected from the four different sites of the Crop Plantation. The soil samples were air-dried for 10 consecutive days in the soil chemistry laboratory of the Department of Crop and Soil Sciences, KNUST and later sieved through a 2 mm sieve to remove debris. Part of the air-dried soil samples was ovendried to a constant mass at a temperature of 105oC for 24 hours. These were used for all the chemical and physical analysis. 2.3. Laboratory Analysis Particle size analysis was determined following the hydrometer method and the pH, also determined as the soil pH in distilled water using pH meter when 1:2.5 soil: water ratio was used. Percentage organic carbon was estimated using W-B by oxidizing soil with excess dichromate solution and titrating against a standard ferrous sulphate solution to determine the amount of unreduced dichromate solution. The organic carbon content (%) was further calculated from this. Oven-dried soil sample was ignited in a muffle furnace at a temperature of 400oC for four hours. The percentage loss in mass expressed per unit mass of oven-dried soil was divided by the Van

250

International Journal of Scientific Research in Knowledge, 2(5), pp. 249-256, 2014

Bemmelen constant (1.724) to derive the SOC by LOI. 2.4. Statistical Analysis The percentage organic carbon for the 12 soil samples were used for correlation and regression analysis. Values from LOI were semi-logarithmically transformed and were used with W-B values to develop simple regression equation. Using a clay base conversion factor of 2.04, conversions on LOI were made. The converted LOI was used together with WB to develop simple regression equation. Clay content and LOI were used to run correlation analysis and also develop regression equation. Multiple regression equation using LOI, clay and W-B was also obtained. The coefficient of determination in each case was determined. All the analysis was done by using GenStat Statistical Package (GenStat Discovery, 2008). 3. RESULTS AND DISCUSSIONS 3.1. Regression and Correlations Very low coefficient of determination between LOI and W-B (0.35) depicted weak relationship. Loss on ignition gave the higher estimates with the highest estimate of SOC as 3.28% whereas W-B gave lower ones with the highest one as 1.35%. This is in agreement with Rowel (1970) that the LOI always gives a higher estimate of soil organic carbon due to effect of clay content. Again, W-B had wider variability than LOI which is in consonance with reports by Konen et al. (2002) and Abella and Zimmer

(2007). The above results might be the cause of the weak relationship between the two methods. Transformed LOI and W-B had a significant positive correlation (r = 0.59, p = 0.05%) that showed good relationship between them. A simple linear regression equation obtained is shown below: y = 0.37x + 0.53 (1) 2 R = 0.35 Where, y = transformed LOI (lnLOI) values; x = W-B values. The clay content of soil samples had some effect on LOI. There was strong positive correlation (r = 0.70, p < 0.001) between clay and LOI. Coefficient of determination, R2, of 0.49 depicted moderately strong relationship. It implied that 49% of the variation in the SOC is explained by the variation in clay content of the soils. Effect of clay content on SOC is due to loss of structural water held in clay minerals, within temperature range of 280oC 400oC as reported by Hesse (1971). However, R2 value was very low when compared with what Abella and Zimmer (2007) had (R2 = 0.78). This could be due to the mass of soil and consequently the amount of clay that was ignited, the ignition temperature and duration (Heiri et al., 2001). Again, R2 = 0.49 clearly shows that other factors, other than the above could affect LOI; decomposition of some inorganic constituents of the soil such as carbonates (Alexander and Byers, 1992; Schulte and Hopkins, 1996). A simple linear regression equation (Eqn. 2) was obtained for clay content and LOI. y1 = 0.12x1 + 1.86 (2) 2 R = 0.49 Where, y1 = LOI ;x1 = clay content of soil

Fig. 1: A Graph of lnLOI against W-B

251

Tuffour et al. Soil Organic Carbon: Relating the Walkley-Black Wet Oxidation Method to Loss on Ignition and Clay Content

Fig. 2: Relationship between LOI and Clay content

The effect of clay content on LOI produced a conversion factor of 2.04. This situation could have arisen due to loss of small amount of structural water resulting from the ignition of small amount of clay from the small mass of soil ignited or to very low clay content in the soil samples. In contrast, Sutherland (1998) reported a mean conversion factor of 14.7 and attributed it to the significant loss of structural water in clay minerals. The conversion

factor value was used to derive converted values for LOI that were used to produce a linear regression equation (Equation 3) with W-B values. y2 = 0.61x2 + 0.40 (3) R2 of 0.40 Where, y2 = W-B values; x2 = converted values of LOI.

Fig. 3: Relationship of W-B and converted LOI

The conversion factor for LOI improved its relationship with W-B slightly (r = 0.63, p = 0.02). Equations with coefficient of determination, R2 of 0.35 and 0.40 were very low when compared with what Konen et al. (2007) had 0.94 to 0.98 when they developed regression equation between LOI and Elemental carbon analyzer. This might have arisen from the higher variability in W-B (Page et al., 1982; Abella and Zimmer, 2007) and possible errors in LOI rendering it, an approximate measure of SOC (Rowell, 1970; Heiri et al., 2001). Multiple linear regression equation (Equation 4) using clay content of soil and LOI as independent factors and W-B as

dependent factor provided reliable estimates of SOC since the coefficient of determination was improved significantly (R2 = 0.84, p < 0.01). Some researchers have also shown that the LOI estimation is improved where clay content as an independent variable is included in multiple regressions (Konen et al., 2002; Abella and Zimmer, 2007; Fullen et al., 2007; Yerokun et al., 2007). y3 = 0.65x3 - 0.08x4 + 0.05 (4) R2 of 0.84 Where, y3 = W-B, x3 = LOI, x4 = clay content of soil.

252

International Journal of Scientific Research in Knowledge, 2(5), pp. 249-256, 2014

From equation (4), the R2 value of 0.84 shows that 84% of the variation in organic carbon estimated by W-B could be accounted for by LOI and percent clay content. Further results from t-test analysis revealed that estimates of the constants and coefficients of all variables in the equations showed very significant differences. 4. CONCLUSIONS Clay content of soil samples had some effect on SOC estimated by LOI, with about 49% of the variations in LOI explained by clay content. Conversion factor for LOI based on clay content of the samples could be used for improving it. Again, the multiple regression equation provided a reliable estimate of organic carbon from W-B using the LOI and clay content. Hence, LOI could be very useful for determining SOC in soils with large amounts of SOM but low clay contents. Values obtained from the W-B method had wider variability than LOI, however, LOI values gave higher estimates of SOC, but, very weak relationship was observed between them. The results from the study suggest that further investigations on other factors other than clay content should be carried out to evaluate their effects on SOC estimation by LOI. Furthermore, the multiple regression equation should be tested under different edaphic conditions for its efficacy. ACKNOWLEDGEMENTS The authors wish to acknowledge the field and laboratory facilities provided by Department of Crop and Soil Sciences, KNUST, Ghana, to carry out this work. The intellectual support provided by the scholars whose articles are cited and included in references of this manuscript is also fully acknowledged. The authors are also grateful authors / editors / publishers of all those articles, journals and books from where the literature for this article has been received and discussed REFERENCES Abella SR, Zimmer BW (2007). Estimating Organic Carbon from Loss on ignition in Northern Arizona Forest Soils. Soil Science Society of America Journal, 71: 545-550. Alexander LT, Byers HG (1932). A critical laboratory review of methods of determining organic matter and carbonates in soil. USDA Tech. Bull. 317. U.S. Gov. Print. Office, Washington, DC. Bationo A, Kihara J, Vanlauwe B, Waswa B, Kimetu J (2005). Soil Organic Carbon Dynamics,

Functions and Management in West African Agro-Ecosystems. Science direct, 1325. Beaudoin B (2003). A comparison of two methods for estimating the organic content of sediments. Journal of Paleolimnology, 29: 387-390. Bellamy PH, Loveland PJ, Bradley RI, Lark RM, Guy JDK (2005). Carbon losses from all soils across England and Wales. Nature, 437(7056): 245248. Brunetto G, Melo GW, Kaminski J, Furlanetto V, Fialho FB (2006). Evaluation of loss-onignition method in the organic matter analysis in soils of the Serra Gaucha of the Rio Grande do Sul. Cienc. Rural. 36(6): 1936-1939. Escosteguy PAV, Galliassi K, Ceretta CA (2007). Soil organic matter determination by weight loss-onignition samples from the state of Rio Grande do Sul. Rev. Bras. Cien. Solo. 31: 247255. FAO, 1990. Soil Map of the World-Revised Legend, 4th Draft. FAO. Rome. Fullen MA, Benediktas J, Genovaite J, Colin AB, Alvyra S (2007). Inter-relationships between soil texture and soil organic matter content in eroded Eutric Albeluvisols in Lithuania. Zemes Ukio Mokslai, 14(3): 9-18 Fullen MA, Jankauskas B, Jankauskiene G, Booth CA, Slepetiene A (2007). Inter-relationships between soil texture and soil organic matter content in eroded Eutric Albeluvisols in Lithuania. Zemes Ukio Mokslai. 14(3): 9-18. GenStat Discovery (2008). 3rd Edition. Grewal KS, Buchan GD, Sherlock RR (2006). A comparison of three methods of organic carbon determination in some New Zealand soils. European Journal of Soil Science, 42(2): 251257. Heiri O, Lotter AF, Lemcke G (2001). Loss on Ignition as a method for estimating organic and carbonate content in sediment: reproducibility and comparability of results. Journal of Paleolimnology, 25: 101-102. Hesse PR (1971). A textbook of soil chemical analysis. Chemical Publishing Co. Inc. New York, pp. 204-209. Ingleby HR, Crowe TG (2001). Neural network models for predicting organic matter content in Saskatchewan soils. Canadian Biosystems Engineering, 43: 7.1-7.5. Konen ME, Jacobs PM, Burras CL, Talaga BJ, Mason JA (2002). Equations for predicting soil organic carbon using loss-on-ignition for north central U.S. soils. Soil Science Society of America Journal, 66: 1878-1881. Lal R, Kimble JM (2000). Pedogenic carbonate and the global carbon cycle. In Lal R, Kimble JM, Eswaran H, Stewart BA (eds.) Global Climate

253

Tuffour et al. Soil Organic Carbon: Relating the Walkley-Black Wet Oxidation Method to Loss on Ignition and Clay Content

Change and Pedogenic. Boca Raton, FL, USA: CRC press, 1-14. Matus FJ, Mauricio E, Forster JE, Marlen G, Chang AC (2009). Communications in Soil Science and Plant Analysis, Agronomy, 40(11): 1-18. Murrage EW, Karanja NK, Smithson PC, Woomer PL (2000). Diagnostic Indicators of Soil Quality in Productive and Non Productive Smallholders Fields of Kenyas Central Highlands. Agriculture, Ecosystems and Environment, 79: 1-8. Nelson DW, Sommers LE (1996). Total carbon, organic carbon, and organic matter. In DL Sparks (ed.) Methods of soil analysis, pp. 961980. Page AL, Miller RH, Keeny DR (1982). Methods of soil analysis, Chemical and microbiological Properties. Publisher Madison, Wisconsin USA, 167-169. Part 3. Chemical methods. SSSA Book Ser. 5. SSSA, Madison, WI. Rowell DL (1970). Soil science: methods and application. Addison Wesley Longman Limited, pp. 42-48. Salehi MH, Hashemi Beni O, Beigi Harchegani H, Esfandiarpour Borujeni I, Motaghian HR (2011). Refining soil organic matter determination by loss-on-ignition. Pedosphere, 21(4): 473-482. Schulte EE, Hopkins BG (1996). Estimation of soil organic matter by weight loss-on-ignition. Schulte EE, Kaufmann C, Peter BJ (1991). The influence of sample size and heating time on soil weight loss-on-ignition. Communications

of Soil Science and Plant Analyses, 22: 159168. Schumacher BA (2002). Methods for the determination of total organic carbon in soils and sediments. United States Environmental Protection Agency, pp. 18. Stone R (2008). Have desert researchers discovered a hidden loop in the carbon cycle? Science 320: 1409-1410. Sudduth KA, Hummel JW (1991). Evaluation of reflectance methods for soil organic matter sensing. Transactions of ASAE, 34(4): 19001909. Sutherland RA (2004). Loss-on-ignition estimates of organic matter and relationships to organic carbon in fluvial bed sediments. Hydrobiologia, 389:13 Walkley A, Black IA (1934). An examination of the Degtjareff method for determining organic carbon in soils: Effect of variations in digestion conditions and of inorganic soil constituents. Soil Science, 63: 251-263. Wang X, Wang J, Zhang J (2012). Comparisons of Three Methods for Organic and Inorganic Carbon in Calcareous Soils of Northwestern China. PLoS ONE, 7(8): e44334. doi:10.1371/journal.pone.0044334. Accessed 19/04/2014. Yerokun OA, Chikuta S, Mambwe D (2007). An evaluation of spectroscopic and loss-on-ignition methods for estimating soil organic carbon in Zambian soils. International Journal of Agricultural Research, 2(11): 965970.

254

International Journal of Scientific Research in Knowledge, 2(5), pp. 249-256, 2014

Henry Oppong Tuffour is a Ph.D. candidate in Soil Physics / Soil Hydrology at the Kwame Nkrumah University of Science and Technology, Kumasi, Ghana and Soil Science lecturer at the Anglican University College of Technology, Nkoranza Campus, Ghana. He received his first degree in 2008 with the award of a Bachelor of Science in Agriculture and a Master of Science in Soil Science in 2012 with major in Soil Physics and Geostatistics from the Kwame Nkrumah University of Science and Technology, Ghana. His current research focuses on hydrological modelling of infiltration involving the soil particle phase and groundwater quality.

Israel Boateng Yeboah currently studying for Master of Science degree in Soil Science (Major in Soil physics/ hydrology) at the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. He obtained his first degree in Agriculture in 2010 in the same institution. His current research is focuses on the vulnerability assessment of groundwater to pesticide pollution

Rev. Fr. Professor Mensah Bonsu is a Visiting Professor (Post-retirement from the University of Cape Coast, Ghana) in Soil Science at the Department of Crop and Soil Sciences of the Kwame University of Science and Technology, Kumasi, Ghana. He obtained his first degree in Agricultural Mechanization in 1972 and a Masters degree in Soil Science with Soil Physics major in 1978 from the University of Ghana. He later pursued another Masters degree in Soil Science (Soil Physics / Soil Hydrology Major) in 1984 and Ph.D. in Soil Science with major in Soil Physics / Soil Hydrology Major in 1987 at the University of British Columbia, Vancouver, Canada. He has published numerous refereed articles covering soil physics, soil hydrology, soil conservation and management, climate change and agronomy in professional journals.

Thomas Adjei-Gyapong is a Lecturer in Soil Science at the Department of Crop and Soil Sciences, College of Agriculture and Natural Resources, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana. He holds a Master of Science degree in Physical Land Resource (Analyses Option), University of Ghent, Belgium. He has carried out extensive studies in Soil prospection, disruption and evaluation for Agriculture and Forestry.

255

Tuffour et al. Soil Organic Carbon: Relating the Walkley-Black Wet Oxidation Method to Loss on Ignition and Clay Content

Abdul Aziz Khalid obtained his first degree from the University of Cape Coast, Cape Coast, Ghana in General Agriculture in 2007. He later obtained his Masters degree in Soil Science with focus on soil conservation and management from the Kwame Nkrumah University of Science and Technology (KNUST) in 2010. He is currently studying for a Ph.D. in Soil Science at KNUST. His current research is focused on the impact of crop residue management on soil organic carbon, hydrology and agronomic production in a tropical forest zone.

Awudu Abubakar currently studying for Master of Science degree in Soil Science (Major in soil physics/ hydrology) at the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. He obtained his first degree in Agriculture in 2011 in the same institution. His current research is focuses on chemical transport and sorption.

Caleb Melenya obtained his first degree in Agriculture in 2010 from the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. He recently defended his MSc. Thesis in Soil Science which focused on phosphorous mobilization and transport in the same institution.

Pearl Kpotor is a lecturer in Agronomy at the Anglican University College of Technology, Nkoranza Campus, Nkoranza, Ghana. She obtained her first degree in Agriculture with a major in Plant Breeding in 2009 from the Kwame Nkrumah University of Science and Technology (KNUST), Kumasi, Ghana. In 2012, she obtained her Masters degree in Plant Biotechnology from the same institution. Her research focuses on Global climate Change and food production in the arid and semi-arid tropics.

256