Adjacency selection in Markov Random Fields for high spatial resolution hyperspectral data

 Markov Random Fields, implemented for the analysis of remote sensing images, capture the natural spatial dependence between band wavelengths taken at each pixel, through a suitable adjacency relationship between pixels, to be defined a priori. In most cases several adjacency definitions seem viable and a model selection problem arises. A BIC-penalized Pseudo-Likelihood criterion is suggested which combines good distributional properties and computational feasibility for analysis of high spatial resolution hyperspectral images. Its performance is compared with that of the BIC-penalized Likelihood criterion for detecting spatial structures in a high spatial resolution hyperspectral image for the Lamar area in Yellowstone National Park. Received: 9 March 2001 / Accepted: 2 August 2001     

Spectral properties of volcanic materials from hyperspectral field and satellite data compared with LiDAR data at Mt. Etna

Spectral properties of volcanic materials in the optical region (350–2500 nm) of the electromagnetic spectrum are analyzed. The goal is to characterize air-fall deposits, recent lava flows, and old lava flows based on their spectral reflectance properties and on the textural characteristics (grain size) of pyroclastic deposits at an active basaltic volcano. Data were acquired during a spectroradiometric field survey at Mt. Etna (Italy) in summer 2003 and combined with hyperspectral satellite (Hyperion) and airborne LiDAR (Light Detection and Ranging) data. In addition, air-fall deposits produced by the highly explosive 2002–2003 eruption have been sampled and spectrally characterized at different distances from the new vents. The spectral analysis shows that air-fall deposits are characterized by low reflectance values besides variations in grain size. This distinguishes them from other surface materials. Old lava flows show highest reflectance values due to weathering and vegetation cover. The spectral data set derived from the field survey has been compared to corrected satellite hyperspectral data in order to investigate the Hyperion capabilities to differentiate the surface cover using the reflectance properties. This has allowed us to identify the 2002–2003 air-fall deposits in a thematic image just few months after their emplacement. Moreover, the observed differences in the field spectra of volcanic surfaces have been compared with differences in the signal intensity detected by airborne LiDAR survey showing the possibility to include information on the texture of volcanic surfaces at Mt. Etna. The approach presented here may be particularly useful for remote and inaccessible volcanic areas and also represents a potentially powerful tool for the exploration of extraterrestrial volcanic surfaces.     

Spatial Relations between Classes as Integrity Constraints

The quality and integrity of spatial data is very important to support interoperability among different systems. To reach this aim integrity rules defined by the application play an important role (for example, constraints between object classes). In this article, we propose a methodology to define integrity constraints using user level spatial relations between classes of individuals. We will also provide mapping rules from user level relations to geometric level operators to allow the computation of relations. As a case study, we will define the constraints for the class of rivers and some of its specializations.     

Eruption style and petrology of a new carbonatitic suite from the Mt. Vulture Southern Italy/: The Monticchio Lakes Formation

The Monticchio Lakes Formation MLF is a newly identified carbonatite-melilitite tuff sequence which is exposed in the southwestern sector of the Vulture volcano. It is the youngest example ca. 0.13 m.y. of this type of volcanism in Italy, although other carbonatites of smaller volume, but with similar characteristics, have been discovered recently. This volcanic event occurred in isolation after a 0.35 m.y. period of inactivity at Vulture. The eruption produced two maar-type vents and formed tuff aprons mainly composed of dune beds of lapilli. Depositional features suggest that a dry surge mechanism, possibly triggered by CO₂ expansion, was dominant during tuff emplacement. The MLF event involved a mixture of carbonatite and melilitite liquids which were physically separated before the eruption. Abundant mantle xenoliths are direct evidence of the deep-seated origin of the parental magma and its high velocity of propagation towards the surface. Often, these nodules form the core of lapilli composed of concentric shells of melilitite andror porphyritic carbonatite. Coarse-ash beds alternate with lapilli beds and consist of abundant lumps and spherulae of very fine-grained calcite immersed in a welded, highly compacted carbonatite matrix. Porphyritic carbonatite shells of the lapilli and fine-grained spherulae of calcite in the tuff matrix suggest incipient crystallisation of a carbonatite liquid in subvolcanic conditions and eruption of carbonatite-spray droplets. Dark coloured juvenile fragments mainly consist of melilite, phlogopite, calcite, apatite, perovskite, and häuyne crystals in a carbonatite or melilitite matrix. The rocks have an extremely primitive, ultramafic composition with very high Mga) 85. and Cr and Ni content 1500 ppm-. The calcite contains high SrO, BaO and REE of up to 1.5 wt.%. Similar compositions are typical of primary, magmatic carbonates which are found in both intrusive and extrusive carbonatites. The high modal Sr-Ba-REE-rich calcite, the typical mineralogy, and the high amount of Sr-group elements identify the carbonate component as a carbonatite. The very high Mga, mantle debris and C, O, He isotope ratios in the range of mantle values indicate a near-primary character for the carbonatite which is distinctive of a restricted group of extrusive carbonatites only found in continental rift areas.     

A physically-based method for removing pits in digital elevation models

Spurious pits in digital elevation models (DEMs) are traditionally removed by filling depressions, often creating flat regions that lead to inaccurate estimation of landscape flow directions. In this study, a physical approach based on a simple landscape evolution model is proposed for DEM pit removal. This method, an alternative to traditional geometrical procedures, enforces more realistic slopes and flow directions on topography. The procedure is compared with the method most commonly used in the literature and distributed with commercial GIS software where, generally, elevations of a depression are increased up to the lowest value among neighbouring cells. Several tests are performed and parameters sensitivity is carried out in order to demonstrate the performance of the proposed model as compared to traditional methods.     

Bridge pier scour mitigation under steady and unsteady flow conditions

Watercourse morphology is affected by local scouring when the flow interferes with anthropic structures. Controlling the scour hole size is of predominant importance to guarantee bridge safety as well as to limit the variations of river morphology. A combined countermeasure against bridge pier scour is proposed and tested in order to reduce the maximum scour depth and deviate it away from the bridge foundation. In the first part of the laboratory campaign, combination of two countermeasures (bed-sill and collar) was evaluated for a circular pier under clear-water and live-bed steady flow conditions. The proposed combined countermeasure exhibited an efficiency of about 64% in terms of scour depth reduction. Afterwards, it was tested in unsteady flow conditions, first for a circular pier, then in the case of a rectangular pier with round nose and tail, two circular in-line piers and two rectangular in-line piers, under a hydrograph with a peak flow velocity slightly above the threshold condition of sediment motion. Results showed that the combined countermeasure had an efficiency of about 63% for a single circular pier; however, higher efficiency (about 75%) was obtained in applications to rectangular pier and two in-line circular or rectangular piers.     

Trends and variability of ocean waves under RCP8.5 emission scenario in the Mediterranean Sea

Ocean Dynamics - Wind-generated ocean waves are key inputs for several studies and applications, both near the coast (coastal vulnerability assessment, coastal structures design, harbor...     

Assessment of the dynamic response of unreinforced masonry structures using a macroelement modeling approach

The seismic performance of unreinforced masonry structures is strongly associated with the interaction between in‐plane and out‐of‐plane mechanisms. The seismic response of these structures has been thoroughly investigated by means of experimental testing, analytical procedures, and computational approaches. Within the framework of the numerical simulations, models based on the finite element method provide a good prediction of the seismic performance of unreinforced masonry structures. However, they usually require a high computational cost and advanced user expertise to define appropriate mechanical properties and to interpret the numerical results. Because of these limitations, simplified models for practical applications have been developed during the last decades. Despite this, a great number of these models focus mostly on the evaluation of the in‐plane response, assuming box (or integral) behavior of the structure. In this paper, a simplified macroelement modeling approach is used to simulate the seismic response of 2 masonry prototypes taking into consideration the combined in‐plane and out‐of‐plane action. The numerical investigations were performed in the static and dynamic fields by using pushover analyses and nonlinear dynamic analyses respectively. The latter is a novel implementation of a model previously developed for static analysis. The results obtained from this study are in good agreement with those provided by a detailed nonlinear continuum FE approach, demonstrating the applicability of this macroelement model with a significant reduction of the computational cost.