Modeling light scattering at soil surfaces

Structures visible on soil profiles contain transport-relevant information. The ultimate goal of this study is to map regions of similar texture on excavated soil profiles. The shape and especially the connectivity of such structures is a first-order approximation for quantifying the transport domains that can be used to predict flow and transport. With this study, we try to understand the spectral information of soil profile images to quantify the spatial arrangement of such structural features. The first and important step is to produce a texture map of the soil profiles from reflectance measurements. Thus, we evaluate reflectance and transmittance spectra measured on small areas of soil surfaces. Reflectance and transmittance depend on particle size. Measurements and simulations show that the influence of texture on reflectance is measurable but small. In this study, we describe first light scattering by idealized particles. They shall represent natural soil particles. We calculate the light absorption and the directional distribution of the scattered light. Second, we use these properties to describe the radiative transfer of visible light and infrared radiation using a four-flux model. The four-flux model combined with the idealized particle model yields a model with which we can calculate the relation between average particle size and reflectance. To test our radiative transfer model, the reflectance and transmittance of three soil materials were measured. The three soil materials differ in color. Each material was sieved into seven fractions to prepare samples of the same material but different texture. Fitting the complex refractive indexes to the measured spectra indicates that these materials can be differentiated and classified according to their light absorption properties.