8–9 Sept 2026
Europe/Berlin timezone

How to Identify Patches of Hyperspectral Entropy in Forest Soil Profiles from Fläming, Germany

Not scheduled
20m
Poster Complexity

Speaker

Dr Riffat Rahim (German Centre for Integrative Biodiversity Research Halle-Jena-Leipzig – iDiv , Johannisallee 19a 04103 Leipzig)

Description

Forest soils exhibit strong depth-related spatial heterogeneities, with localized hotspot zones that regulate key ecosystem processes and contribute to soil resilience. Conventional bulk soil measurements are limited in their ability to capture this fine-scale variability. This study aims to explore the potential of hyperspectral imaging (HSI) to quantify spatial heterogeneity and by this to identify potential hotspot zones in forest soil profiles from the Fläming region, Germany, showing a great heterogeneity of the geological substrate due to different thicknesses of a sand loess cover over coarser-grained Late Saalian moraine material.
Intact soil cores were collected from field sites and analyzed using a HySpex hyperspectral imaging system covering the spectral range of 400–2500 nm. The acquired hyperspectral data were preprocessed and analyzed to extract relevant spectral information. Dimensionality reduction techniques, including Principal Component Analysis (PCA) and Minimum Noise Fraction (MNF), were applied to identify the most informative spectral features.
To assess spatial heterogeneity, pixel-wise spectral divergence was computed using the Kullback–Leibler pseudo-divergence (KLPD), incorporating both spectral shape and intensity differences. Each pixel spectrum was compared to a global median reference spectrum derived from the entire soil core. Based on this, spatial maps and entropy-based classifications were generated to identify zones of increased spectral variability
Preliminary results revealed a highly skewed distribution of spectral divergence, with most pixels exhibiting low variability and a limited number of pixels showing strong deviations from the global median reference. Spatial analysis indicated that these high-divergence zones occurred as discrete patches, particularly in the upper sections of the soil profile. Classification into low, medium, and high entropy levels highlighted these localized regions of increased spectral complexity.
These findings suggest that HSI is capable of resolving fine-scale soil heterogeneity and identifying candidate hotspot zones. Further integration with independent soil biological and physico-chemical measurements will be used to evaluate the functional relevance of these regions.

Status Group Postdoctoral Researcher
FOR TALKS: Poster Presentation Option Yes, I’m willing to present as a poster.

Author

Dr Riffat Rahim (German Centre for Integrative Biodiversity Research Halle-Jena-Leipzig – iDiv , Johannisallee 19a 04103 Leipzig)

Co-authors

Prof. Bruno Glaser (Department of Soil Biogeochemistry Martin Luther University Halle Wittenberg) Dr Hans Von Suchodoletz (Institute of Geography, Jena University, Institute of Geography, Leipzig University) Mr Marco Pohle (Department of Monitoring and Exploration Technologies , Helmholtz Centre for Environmental Research – UFZ) Mr Michael Seidel (Institute of Geography, Leipzig University) Prof. Michael Vohland (Geoinformatics and Remote Sensing, Institute of Geography, Leipzig University, German Centre for Integrative Biodiversity Research Halle-Jena-Leipzig – iDiv) Prof. Peter Dietrich (Department of Monitoring and Exploration Technologies, Helmholtz Centre for Environmental Research - UFZ Leipzig) Dr Ulrike Werban (Department of Monitoring and Exploration Technologies)

Presentation materials

There are no materials yet.