Day 4

Detailed paper information

Back to list

Paper title Characterisation of soil properties with drone-borne Hyperspectral Imaging and laboratory spectral data
  1. Richard Mommertz Federal Institute for Geosciences and Natural Resources (BGR), Germany Speaker
  2. Lars Konen Bundesanstalt für Geowissenschaften und Rohstoffe
  3. Martin C. Schodlok Federal Institute for Geosciences and Natural Resources (BGR)
  4. Daniel Rückamp Bundesanstalt für Geowissenschaften und Rohstoffe
Form of presentation Poster
  • C4. HAPs/UAVs
    • C4.01 Innovative UAV applications
Abstract text Soil is one of the world’s most important natural resources for human livelihood as it provides food and clean water. Therefore, its preservation is of huge importance. Detailed soil information can provide the required means to aid the process of soil preservation. The project “ReCharBo” (Regional Characterisation of Soil Properties) has the objective to combine remote sensing, geophysical and pedological methods to derive soil characteristics and map soils on a regional scale. Its aim is to characterise soils non-invasive, time and cost efficient and with a minimal number of soil samples to calibrate the measurements. Hyperspectral remote sensing is a powerful and well known technique to characterise near surface soil properties. Depending on the sensor technology and the data quality, a wide variety of soil properties is derivable with remotely sensed data. Properties such as iron, clay, soil organic carbon and CaCO3 can be detected. In this study drone-borne hyperspectral imaging data in the VNIR-SWIR spectral region (400-2500 nm) was acquired over non-vegetated agricultural fields in Germany. In addition, field spectra were taken at several sample locations throughout extensive field campaigns. Soil samples from these locations were used for pedological analyses and spectral measurements in the laboratory following a proposed Internal Soil Standard measurement protocols by IEEE P4005 activities. The laboratory spectra is used to develop methods to predict soil properties to transfer these method to the field and drone-borne data. The prediction methods incorporate the analysis of spectral features and therefore the physical relationships between the reflectance spectra and the soil properties as well as Partial Least Square Regression (PLS) which are widely used to quantify soil properties from hyperspectral data. A further objective is to investigate uncertainties regarding soil parameter retrieval depending on the scale and method of measurement. For the spectral measurements in the laboratory the soil samples are dried, crushed and sieved. The UAV borne data however is influenced by soil moisture, surface roughness, atmospheric and illumination effects. These effects lead to differences in the accuracy for the estimation of soil parameters. The results are presented and critically discussed in the context of soil mapping.