405905 research outputs found
Sort by
How accurate are soil moisture profile sensors? – Results from a multi-sensor evaluation using a sandbox experiment.
No full textMany precision farming applications rely increasingly on the near-real time provisioning of accurate root zone soil moisture measurements to enable the efficient and economical use of limited freshwater resources. Besides the established sensor manufacturers who have been around for decades, new companies are entering the market, often with a portfolio of sensors especially designed for agricultural applications. These so-called soil moisture profile sensors (SMPS) exhibit a high potential for agricultural use. Their elongated shape and the ability to measure simultaneously in different depths make them especially suitable for frequent changes of location as required during cultivation of field crops. These sensors measure the volumetric soil water content (VWC) by exploiting the highly different dielectric permittivity of the solid and liquid soil compounds. I this study we use a sandbox experiment to determine the measurement accuracy of different SMPS under controlled moisture conditions. The sandbox is a 2 x 2 x 1.5 m container filled with well-characterized fine sand which is sealed watertight to all sides. The sandbox is equipped with a 20 cm drainage layer and the water level inside the sandbox can be controlled by pumping water in or out using piezometer tubes, which are open at the bottom in the drainage layer. The SMPS were installed into the sandbox and the measurements were compared against reference measurements using CS610 TDR probes connected to a TDR100 (Campbell Scientific) and SMT100 (TRUEBNER) measurements installed in triplicate at six different depths. The measurement accuracy of 10 different sensors were evaluated, with each sensor being tested in triplicate. Most SMPS performed with reasonable accuracy under very dry and very wet conditions. However, strong variation was observed with respect to slope, offset and spread of the measurements and non-linear behavior was observed with some SMPS in the intermediate soil moisture range. The high variability of the measurement accuracy (RMSE: 1.4 – 9.8 vol. %) highlights the importance of choosing a suitable sensor, especially for precision farming applications, where it is crucial to have accurate field data to make the best management decisions without the need for soil specific calibration
Thin Nickel Coatings on Stainless Steel for Enhanced Oxygen Evolution and Reduced Iron Leaching in Alkaline Water Electrolysis
No full text
13CFLUX ecosystem: Lessons from three generations of software development for a sustainable bioeconomy
No full textHow Topological Polymer Loops on the Nanoparticle Surface Control the Mechanical Properties of Nanocomposites
No full text
Synthesis and Investigation of Ce-based surrogate mixed oxide fuel (MOX)
No full textMOX (mixed oxide Fuel) is a topical nuclear fuel that has been used in nuclear power plants worldwide, in particular in Germany and France. MOX fuel is a ceramic commonly consisting of a mix of UO2 and PuO2 with varying Pu concentrations. Amongst other routes, MOX has been commonlyproduced by what is known as the “MIMAS route” (Micronized Master Blend) [1]. This route is well known to result in a heterogenous material that has microstructural regions of different Pu enrichments [2]. This usually expresses itself in a Pu rich, Pu poor and an indermediate phase [2]. Depending on processing conditions different regions in MOX vary in their Pu enrichment and in their size and distribution. Understanding their chemical properties, in particular the local chemistry and redox states of these regions is critical to ensuring safe and correct eventual disposal of MOX when occuring as spent nuclear fuel (SNF).In this work, the industrial MIMAS process has been down-scaled to the laboratory level. That way, surrogate MIMAS MOX Pellets have been synthesized using CeO2 instead of PuO2. Emphasis has been placed on the influence of the feed UO2 powder origin on the structure and chemistry of the final MIMAS MOX ceramic, namely via the precipitation of either Ammonia diuranate (ADU) or Ammnia uranyle carbonate (AUC). These two routes are inspired by the industrial synthesis of MIMAS MOX[1]. High resolution powder syncrotron X-ray diffraction (S-PXRD) measurements performed at the BM20 of the ESRF unveiled discrete but significant phase differences in the surrogate MOX cermaics. SEMEDS further reveals the contrasting MOX materials that used different precursor materials, highlightnig significant heterogeneity. Remarkebly, high-energy-resolution fluorescence detected Xray absorption near edge structure (HERFD-XANES) measurements performed on the U M4-edge and Ce L3-edge indicate considerable differences in the redox states of the MOX materials, in particular their simultaneous posession of both oxidised and reduced U and Ce respectively. These results suqsequently suggest that in actual Pu based MIMAS MOX considerable diversity in redox states are found, but moreover highlight the siginificance of different synthesis routes in influencing the bulk chemistry of synthesized materials.References[1] D. Haas, A. Vandergheynst, J. van Vliet, R. Lorenzelli, J.-L. Nigon, Nuclear Technology, 106, 60 (1994).[2] R. Delville, M. Verwerft, Microscopy and Microanalysis, 29, 78 (2023)
Deep learning based individual tree crown delineation from panchromatic aerial imagery
No full textAccurate delineation of individual tree crowns (ITC) enables a better understanding of tree-level growth dynamics and evaluating tree vitality. In recent year, researches have introduced deep learning techniques in this field. However, the precise segmentation relies on high quality annotated dataset and test images with limited domain gaps between the training data. Under the framework of the Helmholtz project, panchromatic airborne images are captured over a mixed European forest. In this research, we adopt a UAV benchmark dataset as training data. To close the domain gaps, a deep learning based colorization step is added, for which two deep learning frameworks are compared to achieve an improved ITC delineation result in a dense forest area