Airborne Imaging Spectroscopy Time Series Analysis
Imaging spectroscopy enables detailed studies of land surface based on its optical properties acquired in many, usually a few hundred, narrow spectral bands. Resulting hyperspectral imagery collected over a selected area and time span can be used, e.g., for studying the responses of vegetation to climate change or other environmental disturbances caused by human activities. In situ and laboratory spectroscopy provide point-based measurements of selected samples, and they are useful for the derivation and assessment of theoretical concepts and for calibration and validation measurements from airborne or satellite sensors. In this module, you will learn about:
- retrieval of land surface properties from spectroscopic measurements
- relating optical properties of vegetation/leaves to their biophysical and biochemical parameters
- acquisition of airborne and RPAS hyperspectral imagery and its radiometric and geometric pre-processing
- machine learning classification applied to hyperspectral imagery
- importance of spectral, spatial, and temporal resolution for vegetation monitoring
- in situ and laboratory measurements with a spectroradiometer
- upscaling spectroscopic measurements (from in situ to RPAS/airborne and satellite)
- processing of image and laboratory spectroscopic datasets in case studies over the arctic-alpine tundra in the Krkonoše Mts. and the Norway Spruce forest in the Ore Mts., Czechia
Structure
The module is structured into the following themes:
- Principles of imaging and laboratory spectroscopy
- Airborne hyperspectral data acquisition and pre-processing
- In situ and laboratory spectroscopy of vegetation
- Machine learning in imaging spectroscopy
- Temporal vs. spatial and spectral resolution
- Case study: Seasonal spectral separability of selected grass species of the Krkonoše Mts. tundra ecosystem
- Case study: Discrimination of selected grass species from time series of RPAS hyperspectral imagery
- Case study: Seasonal dynamics of flood-plain forests
Prerequisites to perform this module
The following skills and background knowledge are required for this module.
- Basics of statistics
- Basics of geoinformation systems and handling raster/vector data
- Principles of remote sensing
- Basic programming skills (Python, R)
Follow this link for an overview of the listed prerequisites and recommendations on external material for preparation.
Software
For this module, you will need the software listed below. Follow the links to the individual software or tools, for help in setting them up.
Use Cases and Data
In the research-oriented case studies, this module uses the sataset Tundra vegetation monitoring in Krkonoše Mountains.
Start the module
... by proceeding to the first theme on Principles of imaging and laboratory spectroscopy.