Shadow analysis of soil surface roughness compared to the chain set method and direct measurement of micro-relief

Abstract

Abstract. Soil surface roughness (SSR) expresses soil susceptibility to wind and water erosion and plays an important role in the development and the maintenance of soil biota. Several methods have been developed to characterise SSR based on different methods of acquiring data. Because the main problems related to these methods involve the use and handling of equipment in the field, the present study aims to fill the need for a method for measuring SSR that is more reliable, low-cost and convenient in the field than traditional field methods. Shadow analysis, which interprets micro-topographic shadows, is based on the principle that there is a direct relationship between the soil surface roughness and the shadows cast by soil structures under fixed sunlight conditions. SSR was calculated with shadows analysis in the laboratory using hemispheres of different diameter with a diverse distribution of known altitudes and a surface area of 1 m2. Data obtained from the shadow analysis were compared to data obtained with the chain method and simulation of the micro-relief. The results show a relationship among the SSR calculated using the different methods. To further improve the method, shadow analysis was used to measure the SSR in a sandy clay loam field using different tillage tools (chisel, tiller and roller) and in a control of 4 m2 surface plots divided into subplots of 1 m2. The measurements were compared to the data obtained using the chain set and pin meter methods. The SSR measured was the highest when the chisel was used, followed by the tiller and the roller, and finally the control, for each of the three methods. Shadow analysis is shown to be a reliable method that does not disturb the measured surface, is easy to handle and analyse, and shortens the time involved in field operations by a factor ranging from 4 to 20 compared to well known techniques such as the chain set and pin meter methods.