No abstract

Sewage disposal onto agricultural land may result in the high accumulation of organic wastes, which questions the applicability of typical elemental analysis used for the soil components. To monitor the contamination status of agricultural soils at a former sedimentation basin, after the long‐term cessation of wastewater irrigation, 110 locations (15–20 cm depth) and 4 boreholes (up to 100 cm depth) were sampled to determine pH, loss on ignition, and concentration of Ni, Cu, Pb, Zn, and Cr. Additionally, the applicability of portable X‐ray fluorescence (pXRF) for the soil samples highly influenced by the organic wastes was evaluated. The study revealed the presence of a relatively homogenous sewage waste layer (depth of 20 cm), characterized by slightly acidic to neutral pH (6.3–7.5), high organic matter (OM) accumulation (up to 49%), and elevated concentration (mg kg −1) ranges between: Pb (5–321), Cu (31–2828), Ni (10–193), Cr (14–966), and Zn (76–6639). The pXRF analysis revealed metal concentration increase in mineral samples (up to 50%). The regression models and correction factors demonstrated high correlation and significance of pXRF measurement with response to increasing OM content, with the lowest r 2 = 0.86 obtained for Ni. Correlation of pXRF and AES measurement illustrated element‐dependent response for soils high in organics. Zn, Cu, and Cr pXRF analysis led to a slight underestimation in lower values, but overall good correlations (0.87; 0.89; and 0.88 respectively). Pb and Ni pXRF measurement revealed higher deviation from the reference in both lower and higher concentrations (0.74 and 0.70, respectively).

No abstract

Volcanic flanks subject to hydrothermal alteration become mechanically weak and gravitationally unstable, which may collapse and develop far-reaching landslides. The dynamics and trajectories of volcanic landslides are hardly preserved and challenging to determine, which is due to the steep slopes and the inherent instability. Here we analyze the proximal deposits of the 21 July 2014, landslide at Askja (Iceland), by combining high-resolution imagery from satellites and Unoccupied Aircraft Systems. We performed a Principal Component Analysis in combination with supervised classification to identify different material classes and altered rocks. We trained a maximum-likelihood classifier and were able to distinguish 7 different material classes and compare these to ground-based hyperspectral measurements that we conducted on different rock types found in the field. Results underline that the Northern part of the landslide source region is a hydrothermally altered material class, which bifurcates halfway downslope and then extends to the lake. We find that a large portion of this material is originating from a lava body at the landslide headwall, which is the persistent site of intense hydrothermal activity. By comparing the classification result to in-situ hyperspectral measurements, we were able to further identify the involved types of rocks and the degree of hydrothermal alteration. We further discuss associated effects of mechanical weakening and the relevance of the heterogeneous materials for the dynamics and processes of the landslide. As the study demonstrates the success of our approach for identification of altered and less altered materials, important implications for hazard assessment in the Askja caldera and elsewhere can be drawn.

<p>Across Europe there are 2.5 million potentially contaminated sites due to natural and anthropogenic activities. In this regard, phytoremediation approaches are need as a cost-effective and ecosystem-friendly technique to rehabilitate soil compared to conventional methods. Hyperspectral imaging provides an ideal method to improve and monitor existing bioremediation methods, using hyperaccumulator plants. In our study, the hyperaccumulator plant <em>Brassica juncea</em> showed a high tolerance to the accumulation of Cu, Zn and Ni. Hyperspectral measurements were conducted with a HySpex VNIR-SWIR hyperspectral sensor (408-2500 nm) in-situ and in the laboratory. To monitor and optimize the process of accumulation with hyperspectral imaging, we calculated different vegetation indices, related to metal-induced plant stress, such as TCARI/OSAVI, Chlorophyll Vegetation Index (CVI), Red-Edge Stress Vegetation Index (RSVI), Normalized Pigments Chlorophyll Index (NPCI), Red-Edge Inflection Point (REIP) and Disease Water Stress Index (DWSI), using various pre-processing steps (raw, smoothed and brightness corrected data). In addition, the relation between the different indices and the measured heavy metal content in the samples were tested with a multivariate technique using Partial Least Squares Regression (PLSR). Our results revealed, even with no pre-processed image data, changes in chlorophyll- and red-egde-related indices with increasing PTE concentration. With hyperspectral imaging we are already able to monitor differences of the PTE accumulation within the hyperaccumulator plant <em>Brassica juncea</em>.</p>