Solution and use of chronofunctions in studying soil development

Abstract

Thirty-two chronosequences from 27 areas were selected from the literature for constructing chronofunctions and for correlating rates of change in soil properties with variables representing climate and parent material. The chronosequences originate from areas situated between 66°N and 78°S latitude and represent seven climatic regions, ranging from tropical rainy to cold desert, and seven types of parent materials, including till, aeolian sand, alluvium, mine spoil, volcanic ash, raised beach deposits, and mudflows. Fourteen of the chronosequences contain soils which range in age from 0 to 500 yr; seven span a 12,000-yr period, three a 100,000-yr period, and eight a period of greater than one million yr. Three linear and non-linear models were tested on 15 soil properties. The single-logarithmic ( Y = a + b log X) model yielded the best correlation coefficients, when soil property ( Y) was correlated with time ( X), using linear regression techniques. The dates and equations allow for the following conclusions: 1. (1) The rates of decrease in pH and in base saturation are similar, regardless of the nature of the parent material or climate. 2. (2) The rates of increase in clay content of the B horizon and solum thickness are positively correlated with clay content of the parent material. 3. (3) The rates of increase in solum thickness, oxidation depth, soluble salt content of the salt-enriched horizon, and clay content of the B horizon are positively correlated with mean annual temperature whereas the rate of increase in total N in the surface mineral soil is negatively correlated with present-day mean annual temperature. 4. (4) The increase in bulk density of the surface soil is positively correlated with present mean annual precipitation. 5. (5) The rate of change in C:N is not correlated with variables representing climate and parent material.