Data from: Detection of tephra layers in Antarctic sediment cores with hyperspectral imaging
Data_IsmaelAymerich_PLOSONE_2015 Excel file containing hyperspectral signatures of tephra and No-tephra samples extracted from an Antarctic sediment core. Tephra signatures are labelled 1, while No-tephra spectra are labelled 0. Tephrochronology uses recognizable volcanic ash layers (from airborne p...
| Main Authors: | , , , |
|---|---|
| Format: | Dataset |
| Language: | English |
| Published: |
Dryad
2016
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10261/281800 https://doi.org/10.5061/dryad.v2st1 |
| Summary: | Data_IsmaelAymerich_PLOSONE_2015 Excel file containing hyperspectral signatures of tephra and No-tephra samples extracted from an Antarctic sediment core. Tephra signatures are labelled 1, while No-tephra spectra are labelled 0. Tephrochronology uses recognizable volcanic ash layers (from airborne pyroclastic deposits, or tephras) in geological strata to set unique time references for paleoenvironmental events across wide geographic areas. This involves the detection of tephra layers which sometimes are not evident to the naked eye, including the so-called cryptotephras. Tests that are expensive, time-consuming, and/or destructive are often required. Destructive testing for tephra layers of cores from difficult regions, such as Antarctica, which are useful sources of other kinds of information beyond tephras, is always undesirable. Here we propose hyperspectral imaging of cores, Self-Organizing Map (SOM) clustering of the preprocessed spectral signatures, and spatial analysis of the classified images as a convenient, fast, non-destructive method for tephra detection. We test the method in five sediment cores from three Antarctic lakes, and show its potential for detection of tephras and cryptotephras. Peer reviewed |
|---|