Sensitivity analysis of automatic landslide mapping: numerical experiments towards the best solution

The automatic detection of landslides after major events is a crucial issue for public agencies to support disaster response. Pixel-based approaches (PBAs) are widely used in the literature for various applications. However, the accuracy of PBAs in the case of automatic landslide mapping (ALM) is affected by several issues. In this study, we investigated the sensitivity of ALM using PBA through digital terrain models (DTMs). The analysis, carried out in a study area of Poland, consisted of the following steps: (1) testing the influence of selected DTM resolutions for ALM, (2) assessing the relevance of diverse landslide morphological indicators for ALM, and (3) assessing the sensitivity to landslide features for a selected size of moving window (kernel) calculations for ALM. Ultimately, we assessed the performance of three classification methods: maximum likelihood (ML), feed-forward neural network (FFNN), and support vector machine (SVM). This broad analysis, as combination of grid cell resolution, surface derivatives calculation, and performance classification methods, is the challenging aspect of the research. The results of almost 500 experimental tests provide valuable guidelines for experts performing ALM. The most important findings indicate that feature sensitivity in the case of kernel size increases with coarser DTM resolution; however, the peak of the optimal feature performance for the selected study area and landslide type was demonstrated for a resolution of 20 m. Another finding indicated that in combining a set of topographic variables, the optimal performance was acquired for a DTM resolution of 30 m and the support vector machine classification. Moreover, the best performance of the identification is represented for SVM classification.     

Experimental calibration of vanadium partitioning and stable isotope fractionation between hydrous granitic melt and magnetite at 800 °C and 0.5 GPa

Vanadium has multiple oxidation states in silicate melts and minerals, a property that also promotes fractionation of its isotopes. As a result, vanadium isotopes vary during magmatic differentiation, and can be powerful indicators of redox processes at high temperatures if their partitioning behaviour can be determined. To quantify the partitioning and isotope fractionation factor of V between magnetite and melt, piston cylinder experiments were performed in which magnetite and a hydrous, haplogranitic melt were equilibrated at 800 °C and 0.5 GPa over a range of oxygen fugacities (\({f_{{{\text{O}}_{\text{2}}}}}\)), bracketing those of terrestrial magmas. Magnetite is isotopically light with respect to the coexisting melt, a tendency ascribed to the VI-fold V³⁺ and V⁴⁺ in magnetite, and a mixture of IV- and VI-fold V⁵⁺ and V⁴⁺ in the melt. The magnitude of the fractionation factor systematically increases with increasing log\({f_{{{\text{O}}_{\text{2}}}}}\) relative to the Fayalite–Magnetite–Quartz buffer (FMQ), from ?⁵¹V_mag-gl = ? 0.63?±?0.09‰ at FMQ ? 1 to ? 0.92?±?0.11‰ (SD) at ≈?FMQ?+?5, reflecting constant V³⁺/V⁴⁺ in magnetite but increasing V⁵⁺/V⁴⁺ in the melt with increasing log\({f_{{{\text{O}}_{\text{2}}}}}\). These first mineral-melt measurements of V isotope fractionation factors underline the importance of both oxidation state and co-ordination environment in controlling isotopic fractionation. The fractionation factors determined experimentally are in excellent agreement with those needed to explain natural isotope variations in magmatic suites. Furthermore, these experiments provide a useful framework in which to interpret vanadium isotope variations in natural rocks and magnetites, and may be used as a potential fingerprint the redox state of the magma from which they crystallise.     

Geologic hazards of the Ocoña river valley,Peru and the influence of small-scale mining

Natural Hazards - The Ocoña River Valley, in the Arequipa Department in southern Peru, extends over 150 km from the Pacific coast to its headwaters in the Andes Mountains. While...     

Towards new applications of underwater photogrammetry for investigating coral reef morphology and habitat complexity in the Myeik Archipelago,Myanmar

Photogrammetry represents a non-destructive, cost-effective tool for coral reef monitoring, able to integrate traditional remote sensing techniques and support researchers’ work. However, its application to submerged habitats is still in early stage. We present new ways to employ Structure from Motion techniques to infer properties of reef habitats. In particular, we propose the use of Digital Surface Models and Digital Terrain Models for assessing coral colonies extension and height and discriminating between seabed and coral cover. Such information can be coupled with digital rugosity estimates to improve habitat characterization. DTM, DSM and orthophotos were derived and used to compute a series of metrics like coral morphologies, reef topography, coral cover and structural complexity. We show the potentialities offered by underwater photogrammetry and derived products to provide useful basic information for marine habitat mapping, opening the possibility to extend these methods for large-scale assessment and monitoring of coral reefs.     

Implementation of the European Water Framework Directive: Natural variability associated with the CARLIT method on the rocky shores of the Ligurian Sea (Italy)

The Water Framework Directive (WFD) 2000/60/EC, adopted by the European Community in 2000 with the goal of maintaining and improving the aquatic environments, requires that member states achieve and maintain a good ecological status of all water bodies by 2015. In the marine context, the ecological status has to be quantified applying indexes based on appropriate key biological elements, which allow the categorization of water bodies into five Ecological Status (ES) classes. The CARLIT index is a cartographic monitoring tool enabling the Ecological Quality Ratio (EQR) to be calculated using macroalgae in coastal hard bottoms as a key biological element; at present it is being applied in Spain, France and Italy. To detect actual changes of water quality and, consequently, rely on the application of indexes for the assessment of the ecological status in the marine environment, it is necessary to evaluate their sensitivity to natural variability at different temporal and spatial scales. In this study we present a first quantification of natural (spatial and temporal) variability of EQR‐CARLIT quality assessment in 2006 and 2007 along the Ligurian coast (North‐Western Mediterranean, Italy). The EQR‐CARLIT values recorded along the Ligurian coastline show that natural variability of EQR‐CARLIT is low and that it does not affect the attribution of a given stretch of coast to a particular ES class, in agreement with the major human pressures acting in the area (urbanization, river load, sea‐farming). A small‐scale variability was detected, strengthening the use of cartography of the whole rocky shore, whenever possible, or, alternatively, the assessment of the ecological status for long stretches of coast, to encompass the small‐scale variability due to local factors.     

The role of water mass dynamics in controlling bacterial abundance and production in the middle Adriatic Sea

Month-to-month fluctuations in the abundance of bacteria and heterotrophic nanoflagellates (HNF) and bacterial production, as well as various chemical (nutrients, oxygen) and physical (salinity, temperature) parameters were analysed at a station located in the open middle Adriatic Sea during one decade (1997-2006). Being influenced by both coastal waters and open Adriatic circulation in the surface layer, and by the deep Adriatic water masses in the deep layers (100 m), this station is quite suitable for detecting the environmental changes occurring in the open Adriatic Sea with respect to the circulation of its water masses and their long-term changes and anomalies. Multivariate methods were used to identify seasonal and inter-annual changes of the investigated parameters, associating observed changes to the changes in Adriatic water masses and circulation regimes. The analyses showed that bacterial abundance and production were controlled by different water mass dynamics during 1997-2001 compared to 2002-2006 period, particularly noticeable in different seasonal patterns of biological parameters. The interplay between North Adriatic Dense Water (NAdDW) and Levantine Intermediate Water (LIW) resulted in a change in the available nutrients (NAdDW is poor in orthophosphates), and as a consequence different bacterial abundance and production. A few periods were examined in detail, such as 2004, when LIW inflow was particularly strong and was accompanied by an increase of bacterial and HNF abundances, as well as of bacterial production.     

Geomorphological and historical changes on the Argentina delta,Ligurian coast,Italy

A preliminary investigation of the geomorphological evolution of the small delta of the Argentina river (Riviera di Ponente, Italy) shows that it has largely been affected by anthropic processes. Erosion is undermining land and structural developments not only in the delta, but also along the nearby shores and a thorough knowledge of them seems necessary prior to management applications. In fact, processes affecting the evolution of the investigated area could well be regarded as a good example of the evolution of many other deltaic areas in the Northern Mediterranean.