Erosion and nitrogen loss from forest soils in relation to soil structure

Soil erosion by rain and runoff causes the loss of potential productivity from natural landscapes and the deterioration of the quality of surface water. Most studies of soil erosion have focused on agricultural soils, but little work has been done on forest soils. Understanding the causes and effects of soil erosion and the resultant loss of nutrients is essential to the development of sustainable soil management practices. This study considers the soil properties that influence the susceptibility of soils to erosion, as well as the external factors that promote erosion. The combined effects of these factors on soil and nitrogen (N) loss were examined. Experiments were conducted in field and laboratory plots using simulated rainfall. In the laboratory, three forest soils of contrasting structure were subjected to simulated rainfall of constant intensity. For each soil, there were four erosion treatments. These permitted comparison of erosion due to varying kinetic energy (KE) of rain at a constant slope and varying slope at a constant KE, and due to the presence or absence of a drying cycle. Erosion of soil and size-distribution of sediment were strongly influenced by the mechanical stability of soil aggregates. Indices based on the Mean Weight Diameter of uneroded soils provided a good indication of the susceptibility of soils to erosion and the characteristics of sediment. As direct measurement of erosion is expensive, information on erodibility can be useful in prediction of erosion. The possibility of inferring soil erodibility from physical properties of soil was explored with the erosion model GUEST (Griffith University Erosion Systems Template). This model estimates parameters that describe erodibility due to rainfall- or runotf-only erosion processes. Results showed dependence of sediment concentration on soil strength. However, the relationship between erodibility parameters and soil strength indicated that information on other physical properties may be also required in prediction of erosion. The bonding strength of soil aggregates provides a good measure of the susceptibility of soils to aggregate breakdown and erosion. The process of aggregate breakdown was studied through the application of a known amount of disruptive energy to soil-water suspensions using ultrasound. A simple dynamic model was used to estimate the rate of aggregate breakdown and dispersion from the data on sizedistribution of soil aggregates. The rates of aggregate breakdown for various soils were well related with their mechanical stability and the erosion rates. It was shown that this method was sufficiently robust to indicate the mechanical stability of soils and erodibility. The loss of N due to erosion was directly related with the amount of soil loss. Irrespective of the soils and erosion treatments, the N-concentration of sediment was found to be greater than that of the uneroded soil. Production of N-enriched sediment presumably arose from a combination of uneven distribution of N within aggregates, aggregate breakdown including raindrop stripping, abundance of organic matter and residue, and deposition, rather than due to a single mechanism and/ or specific.soil condition promoting nutrient enrichment of sediment. A strong dependence of measured N-loss with soil loss indicated that N-loss can be predicted from soil loss, without requiring information about N in soils or sediment.