Quantitative relationships between Munsell colour attributes and organic carbon in highly weathered tropical soils

Abstract

Soil organic carbon (SOC) is a very important parameter for assessing the quality of agricultural soils. However, the availability and use of such data by resource-poor farmers, especially in Sub-Saharan Africa, remains a major challenge due to the financial and time constrains involved. Thus, there is need to assess and adopt reliable methods for the rapid estimation of soil organic carbon content by indigenous farmers and field users. The objective of this study was to evaluate the quantitative relationships between soil organic carbon and Munsell colour attributes (value and chroma) in highly weathered tropical soils of the Northwestern Highlands of Cameroon. Forty-six soil profiles (28 Acrisols and 18 Ferralsols), including 46 surface (A) horizons and 181 subsurface horizons (Bo, Bt) were used. Soil organic carbon data and Munsell colour attributes, obtained through standard procedures, were subjected to descriptive statistical, correlation, regression and principal components analyses, in order to evaluate the relationships existing between SOC and Munsell colour attributes. In general, there were negative and significant (p < 0.001) correlations between SOC and all Munsell colour attributes (chroma, value, value + chroma, and value +0.5 chroma). The best models relating SOC and Munsell colour attributes were logarithmic models, with soil colour explaining >70 % of the variance. The results indicate that SOC in highly weathered tropical soils can be conveniently estimated using Munsell soil colour attributes (value + chroma). Better estimates were obtained using logarithmic models for surface (A horizon) soil samples having a sand content >50 %.