Dries Raymaekers

A Seasonally Robust Empirical Algorithm to Retrieve Suspended Sediment Concentrations in the Scheldt River

A seasonally robust algorithm for the retrieval of Suspended Particulate Matter (SPM) in the Scheldt River from hyperspectral images is presented. This algorithm can be applied without the need to simultaneously acquire samples (from vessels and pontoons). Especially in dynamic environments such as estuaries, this leads to a large reduction of costs, both in equipment and personnel. The algorithm was established empirically using in situ data of the water-leaving reflectance obtained over the tidal cycle during different seasons and different years. Different bands and band combinations were tested. Strong correlations were obtained for exponential relationships between band ratios and SPM concentration. The best performing relationships are validated using airborne hyperspectral data acquired in June 2005 and October 2007 at different moments in the tidal cycle. A band ratio algorithm (710 nm/596 nm) was successfully applied to a hyperspectral AHS image of the Scheldt River to obtain an SPM concentration map.

A Wavelet-Enhanced Inversion Method for Water Quality Retrieval From High Spectral Resolution Data for Complex Waters

Optical remote sensing in complex waters is challenging because the optically active constituents may vary independently and have a combined and interacting influence on the remote sensing signal. Additionally, the remote sensing signal is influenced by noise and spectral contamination by confounding factors, resulting in ill-posedness and ill-conditionedness in the inversion of the model. There is a need for inversion methods that are less sensitive to these changing or shifting spectral features. We propose WaveIN, a wavelet-enhanced inversion method, specifically designed for complex waters. It integrates wavelet-transformed high-spectral resolution reflectance spectra in a multiscale analysis tool. Wavelets are less sensitive to a bias in the spectra and can avoid the changing or shifting spectral features by selecting specific wavelet scales. This paper applied WaveIN to simulated reflectance spectra for the Scheldt River. We tested different scenarios, where we added specific noise or confounding factors, specifically uncorrelated noise, contamination due to spectral mixing, a different sun zenith angle, and specific inherent optical property (SIOP) variation. WaveIN improved the constituent estimation in case of the reference scenario, contamination due to spectral mixing, and a different sun zenith angle. WaveIN could reduce, but not overcome, the influence of variation in SIOPs. Furthermore, it is sensitive to wavelet edge effects. In addition, it still requires in situ data for the wavelet scale selection. Future research should therefore improve the wavelet scale selection.

The SeaSWIR dataset

The SeaSWIR dataset consists of 137 ASD (Analytical Spectral Devices, Inc.) marine reflectances, 137 total suspended matter (TSM) measurements and 97 turbidity measurements gathered at three turbid estuarine sites (Gironde, La Plata, Scheldt). The dataset is valuable because of the high-quality measurements of the marine reflectance in the Short Wave InfraRed I region (SWIR-I: 1000–1200 nm) and SWIR-II (1200–1300 nm) and because of the wide range of TSM concentrations from 48 up to 1400 mgL−1. The ASD measurements were gathered using a detailed measurement protocol and were subjected to a strict quality control. The SeaSWIR marine reflectance is characterized by low reflectance at short wavelengths (< 450 nm), peak reflectance values between 600 and 720 nm and significant contributions in the near-infrared (NIR) and SWIR-I parts of the spectrum. Comparison of the ASD water reflectance with simultaneously acquired reflectance from a three-radiometer system revealed a correlation of 0.98 for short wavelengths (412, 490 and 555 nm) and 0.93 for long wavelengths (686, 780 and 865 nm). The relationship between TSM and turbidity (for all sites) is linear, with a correlation coefficient of 0.96. The SeaSWIR dataset has been made publicly available (https://doi.org/10.1594/PANGAEA.886287).

High spatial and spectral remote sensing for detailed mapping of potato plant parameters

This preliminary study shows the potential of highly flexible drones and hyperspectral technology to make detailed chlorophyll maps of an experimental potato field. A novel, innovative hyperspectral frame camera (Rikola Ltd) was employed to gather the spectral information (24 bands) at 5 cm spatial resolution. A first challenge therefore was to setup a dedicated preprocessing chain for the images coming from this novel sensor. Coregistration of the images was successful resulting in an image displacement of only 1–2 pixels. The chlorophyll map created from the Rikola data corresponded well to the field measurements. R2 values of 0.70 were found for a linear relation between the averaged field chlorophyll measurements and the mean of the (R780-R550)/(R780+R550) index calculated for all vegetated Rikola pixels within an experimental potato cultivar plot. These chlorophyll maps which are directly linked to the vegetation status of the crops can be used by the farmer for better management decision making.