Thomas Udelhoven

Quantitative analysis of soil chemical properties with diffuse reflectance spectrometry and partial least-square regression: A feasibility study

Soil chemical properties from different locations in the Trier region, Rhineland-Palatinate, SW Germany were evaluated using VIS/NIR reflectance spectrometry (ASD FieldSpec-II spectrometer, 0.4–2.5 μm) and partial least-square regression (PLS). Generally, laboratory spectrometry performed better than field spectrometry probably due to strong interferences of soil structure. In a plot experiment reliable estimations were obtained for total amounts of Ca, Mg, Fe, Mn and K but not for organic carbon and nitrogen. In the landscape-scale context the estimations for organic carbon could be significantly improved but it was also concluded that the development of statistical prediction models is limited to geologically homogeneous areas. In both experiments CAL extractable nutrients could not be satisfactorily estimated. This excludes diffuse VIS/NIR spectrometry as a diagnosis tool of short- or medium-term changes of the soils nutrient status. However, the method can be used as a quick screening method in questions where the spatial distribution of organic carbon and total metal contents is addressed, as in soil development and soil degradation monitoring, and when time or laboratory costs are critical factors.

Development of a Hierarchical Classification System with Artificial Neural Networks and FT-IR Spectra for the Identification of Bacteria

The practical value of elaborated vibrational spectroscopic techniques in medical and microbiological biodiagnostics depends strongly on the reliability, the speed, the ease of use, and the evaluation procedures of the acquired data. In the present study, artificial neural networks (ANNs) were used to establish a hierarchical classification system for microbial Fourier transform infrared (FT-IR) spectra suitable for identification purposes in a routine microbiological laboratory. A radial basis function network (RBF) proved to be superior for a top-level classification of the FT-IR spectra at the genus level. Species within these genera were sequentially further classified by using multilayer perceptrons (MLPs), which achieved a larger differentiation depth than RBF networks. The MLPs were trained with several learning algorithms. Best performance was achieved with the cascade correlation (CC) approach to determine the network topology combined with resilient propagation (Rprop) as the training algorithm. The final hierarchically organized model was able to discriminate between four genera of microorganisms comprising 42 different strains of Pseudomonacae, 33 strains of Bacillus, 46 strains of Staphylococcus, and 6 species and 24 strains of yeast genera Candida. Altogether, 145 strains from international microbial strain collections are comprised in 971 spectra. The species Candida albicans could be further classified with respect to susceptibility against the antibiotic drug fluconazole, which is of therapeutic relevance. Key factors for the classification results of the bacterial FT-IR spectra were the data pretreatment, the number of wavelengths selected by a feature extraction algorithm, the type of network, and the learning function used for the ANN training.

How Normalized Difference Vegetation Index (NDVI) Trendsfrom Advanced Very High Resolution Radiometer (AVHRR) and Système Probatoire d’Observation de la Terre VEGETATION (SPOT VGT) Time Series Differ in Agricultural Areas: An Inner Mongolian Case Study

Detailed information from global remote sensing has greatly advanced our understanding of Earth as a system in general and of agricultural processes in particular. Vegetation monitoring with global remote sensing systems over long time periods is critical to gain a better understanding of processes related to agricultural change over long time periods. This specifically relates to sub-humid to semi-arid ecosystems, where agricultural change in grazing lands can only be detected based on long time series. By integrating data from different sensors it is theoretically possible to construct NDVI time series back to the early 1980s. However, such integration is hampered by uncertainties in the comparability between different sensor products. To be able to rely on vegetation trends derived from integrated time series it is therefore crucial to investigate whether vegetation trends derived from NDVI and phenological parameters are consistent across products. In this paper we analyzed several indicators of vegetation change for a range of agricultural systems in Inner Mongolia, China, and compared the results across different satellite archives. Specifically, we compared two of the prime NDVI archives—AVHRR

Long term data fusion for a dense time series analysis with MODIS and Landsat imagery in an Australian Savanna

The spatial resolution of Landsat imagery has proven to be well suited for the analysis of vegetation patterns and dynamics at regional scale; however, the low temporal frequency is often a limitation for the quantification of vegetation dynamics. The spatial and temporal adaptive reflectance fusion model (STARFM) combines moderate resolution imaging spectrometer (MODIS) and Landsat thematic mapper/enhanced thematic mapper plus (TM/ETM+) imagery to a high spatiotemporal resolution dataset. A time series of 333 STARFM images was generated between February 2000 and September 2007 (8-day interval) at Landsat spatial and spectral resolution for a 12 × 10     km heterogeneous test area within the North Queensland Savannas. Time series of observed Landsat and predicted STARFM images correlated high for each spectral band (0.89 to 0.99). The STARFM algorithm was tested in a regionalization study where sudden change events were analyzed for a pallustrine wetland. A MODIS subpixel analysis showed a very close relationship between STARFM normalized difference vegetation index (NDVI) data and MODIS NDVI data (root mean square error of 0.027). A phenological description of the major vegetation classes within the region revealed distinct differences and lag times within the ecosystem. The 2004 dry season NDVI minimum-map correlated highly with the validated 2004 foliage projective cover product ( r 2 = 0.92 ) from the Queensland Department of Environment and Resource Management.

The use of fine sediment fractal dimensions and colour to determine sediment sources in a small watershed

As many particle-associated contaminants and nutrients are supply controlled, the question of particle source is crucial. In addition, sediment storage has important implications for the delivery and fate of pollutants. Increased accumulations of fine sediment (<63 μm) in gravel beds not only modify benthic habitat but also increase the retention time of sediment-associated contaminants in these biologically active areas of river systems. It is of overriding importance to determine the origin of the fines and the amount, location and process of storage. There is little doubt that the characteristics of particles can be used to derive this information. There is no general agreement, however, about the characteristics that should be considered in such investigations. Furthermore, scientists have a great demand for simple methods, especially for routine use under dry weather conditions when suspended particle concentrations are low. This investigation shows that, in addition to loss on ignition, the determination of fractal dimension and particle colour also provide a fast and easy approach. After filtering the suspensions through glass microfibre filters, the dried filter residues are scanned by a colour scanner. Particle-bound cations and heavy metals were analysed by atomic absorption spectrometry after material decomposition with nitric acid. Fractal dimensions were obtained from measurement of the digital pictures using standard methods of digital image analysis. The fractal dimension decreases, indicating an increase in the regularity of particle morphology as one moves from gravel-stored sediment to surface-stored sediment to suspended sediment. Generally, flocs with high organic content exhibit more irregular morphology while single mineral particles have a more regular shape. In this study, the dominant factor for variable particle morphology was determined to be the mechanical forces influencing the flocs. An increase in shear associated with high flows and the impact of colliding with the channel bed appears to result in less flocculated, more regular particles at the surface. In addition, the morphology of particles shows that the exchange of particles in the gravel bed takes place during and shortly after flood events when the armoured layer is broken up and the sediment down to a depth of several centimetres is disturbed. Deposition onto the sediment surface was observed during the falling limb of every event. Surface sediment and suspended sediment show similarities in loss on ignition and particle morphology, particularly in low flow conditions when the stored amount of 25 mg/cm2 is not exceeded. The exchange between suspended sediment sources over the course of a year can be described using colour variation of the fine sediment. Along with particle-bound manganese, it can be shown that during both the winter month, which exhibit a high baseflow, and flood events, distant sediment sources predominate. In summer low flow conditions, in-channel sources are more important. Furthermore, the use of colour values can allow prediction of the sediment chemical properties as, for example a significant linear correlation between particle colour and particle-bound manganese was found.

The NeuroDeveloper®: a tool for modular neural classification of spectroscopic data

The NeuroDeveloper® is a high-performance software tool for the classification of spectroscopic data using artificial neural networks (ANNs). The software consists of four specialized modules, each constructed for one step within the classification procedure. The FeatureDeveloper offers a user-friendly visualization of the raw-spectra, uni- or multivariate methods for wavelength selection and compression, as well as standard tools for spectra preprocessing. The NeuroSimulator performs the neural network training. The identification of a suited network architecture is supported by an automation tool. The development of modular and hierarchically organized neural networks of any degree of complexity is done in the ModuleDeveloper. Finally, the evaluation of new data by trained ANNs is achieved by the Classification module with a complete documentation via printout and log-files.

TimeStats: A Software Tool for the Retrieval of Temporal Patterns From Global Satellite Archives

TimeStats is a free tool for the analysis of multitemporal equidistant georeferenced remote sensing data archives, such as MODIS, AVHRR, MERIS and SPOT-Vegetation. Key features include parametric and non-parametric methods for trend detection, generalized-least square regression, distributed lag models, cross spectra analysis, windowed trend and frequency analysis, continuous wavelet transform, empirical mode decomposition and extraction of phenological indexes (peaking times and magnitudes). The intension of this paper is to demonstrate how these methods can be used for data mining in long-term remote sensing data archives to retrieve transient, cyclic and stochastic components and to regress autocorrelated series in a statistical meaningful way to each other. TimeStats is programmed in the Interactive Data Language® (IDL) and freely distributed with the IDL virtual machine®. Generated raster output files are saved in the standard ENVI® format with appropriate header files and are portable to common geospatial satellite imaging processing software packages. Software binaries and an extended user manual can be obtained from the author.

A Hyperspectral Thermal Infrared Imaging Instrument for Natural Resources Applications

Abstract: A new instrument has been setup at the Centre de Recherche Public-Gabriel Lippmann to measure spectral emissivity values of typical earth surface samples in the 8 to 12 μm range at a spectral resolution of up to 0.25 cm −1 . The instrument is based on a Hyper-Cam-LW built by Telops with a modified fore-optic for vertical measurements at ground level and a platform for airborne acquisitions. A processing chain has been developed to convert calibrated radiances into emissivity spectra. Repeat measurements taken on samples of sandstone show a high repeatability of the system with a wavelength dependent standard deviation of less than 0.01 (1.25% of the mean emissivity). Evaluation of retrieved emissivity spectra indicates good agreement with reference measurements. The new instrument facilitates the assessment of the spatial variability of emissivity spectra of material surfaces—at present still largely unknown—at various scales from ground and airborne platforms and thus will provide new opportunities in environmental remote sensing.

Capability of feed-forward neural networks for a chemical evaluation of sediments with diffuse reflectance spectroscopy

Abstract Diffuse reflectance spectroscopy (0.4–2.5 μm) is evaluated as fast and non-destructive method for the analysis of sediments, characterised by a wide range of mineral constituents. Combined with feed-forward artificial neural networks (ANNs) this technique is used to estimate quantitatively the chemical composition from the sediments based on a supervised training with one model. The examined characteristics include contents of inorganic carbon, Fe, S, Al, Si, Ca, K, Mg and calcite. The efficiency of several learning algorithms (Backpropagation, Quickprop, Resilient propagation (Rprop), Cascade Correlation (CC)) is investigated. All learning algorithms perform well using principal component (PC) scores of the first derivative spectra as input for the supervised training. ANNs trained with Quickprop and Rprop produced most accurate estimations of the chemical characteristics and the performance was better than for standard multivariate statistical tools (stepwise multiple linear regression (SMLR), principal component analysis (PCA)). An interpretation of the results is given by a detailed consideration of the correlation structure among the chemical constituents.

Uptake visualization of deltamethrin by NanoSIMS and acute toxicity to the water flea Daphnia magna

The objective of this study was to investigate the uptake of deltamethrin, an insecticide, by Daphnia magna neonates by SIMS and to compare these findings with results based on established toxicity tests. Young daphnids (aged <24 h) were exposed to 0, 50 and 200 microg L(-1) (ppb) deltamethrin. Mobile, immobile and dead animals were enumerated after 24 and 48 h following OECD 202 [OECD 202, 2004. Daphnia sp., acute immobilisation test, guideline for testing of chemicals] guidelines. The animals were embedded in epoxy resin, cut into semi-thin sections (500 nm) and placed on silicon supporters. NanoSIMS 50 (Cameca) images were made from tissues of the intestine for carbon, nitrogen (measured as CN), phosphorus and bromine. To distinguish between relative concentrations of bromine in the guts from different exposure concentrations of deltamethrin, a carbon normalization method was carried out. Both deltamethrin concentrations and time showed a significant effect on immobilization and mortality of the daphnids (P<0.0001). Bromine from deltamethrin could be visualized by NanoSIMS in all exposed gut tissues (gut wall, microvilli layer, perithropic membrane). Highest deltamethrin concentrations following (12)C normalization were found in animals exposed to 200 microg L(-1) deltamethrin, followed by 50 microg L(-1) and the control. NanoSIMS 50 was successfully used as a supplemental technique for elucidating the relation between the uptake and localization of deltamethrin and its toxicity to D. magna. These results highlight the potential usefulness of NanoSIMS to detect marker elements of xenobiotic compounds within exposed organisms, to compare relative exposure concentrations, and to locate these compounds at their original tissue location.