Teza, G.

SURMODERR is a MATLAB toolbox intended for the estimation of reliable velocity uncertainties of a non-permanent GPS station (NPS), i.e. a GPS receiver used in campaign-style measurements. The implemented method is based on the subsampling of daily coordinate time series of one or more continuous GPS stations located inside or close to the area where the NPSs are installed. The continuous time series are subsampled according to real or planned occupation tables and random errors occurring in antenna replacement on different surveys are taken into account. In order to overcome the uncertainty underestimation that typically characterizes short duration GPS time series, statistical analysis of the simulated data is performed to estimate the velocity uncertainties of this real NPS. The basic hypotheses needed to apply the method are: i) the signal must be a long-term linear trend plus seasonal and colored noise for each coordinate; ii) the standard data processing should have already been performed to provide daily data series; iii) if the method is applied to survey planning, the future behavior should not be significantly different from the past behavior. In order to show the strength of the approach, two case studies with real data are presented and discussed (Central Apennine and Panarea Island, Italy). Keywords: Non-permanent GPS; Episodic GPS; Campaign-style GPS; Velocity Field; Strain Rate; Survey Optimization.

A continuous global positioning system station (CGPS) provides accurate coordinate time series, while episodic GPS stations (EGPSs), which operated throughout short measurement sessions, are generally used to improve the monitoring spatial density. In an urban environment, EGPSs are typically equipped with removable mounts (topographical tripod or bipod). In this paper, a method is proposed in order to evaluate vertical surface motions by means of differential measurements of removable mount EGPSs with respect to a nearby reference CGPS. For each day, the correct position of this CGPS is used as reference for the quick differential EGPS measurements to allow the correction of their positions. The method is applied to evaluate the subsidence in the centre of Bologna, which is characterized by significant vertical movements, probably related to seasonal climatic effects, and where these movements differ significantly even among closely spaced locations

The terrestrial laser scanner is an efficient topographical instrumentation used to acquire a redundant number of points distributed over a physical surface. The goal of laser scanning is the definition of very accurate models of the studied areas. In this way, deformations or changes can be monitored by means of repeated surveys in different epochs [Pesci et al., 2005; 2007]. The laser signal is characterized by highly collimated, monochromatic, and coherent radiation that is well suitable for very short impulse generation in the nanosecond scale. The operating methodology of a time-of-flight laser scanner is similar to a laser range-finder, measuring the time it takes a laser pulse to travel from a transmitter to the surface surveyed, and back to a detector device. The range d is computed using the relation d = ct / 2, where t is the time of flight and c is the speed of light. The advantage of this instruments is the laser beam deflection over a very accurate angular grid, that can be obtained by oscillating and rotating mirrors, thus providing a wide coverage area between adjacent points. Each point is collected into a local reference system consisting of the origin at the instrument sensor, well-known angular parameters, and very accurate measurements of range. Together with point coordinates (x, y, z) , radiometric values related to the surveyed object’s reflectivity can be calculated from returned signal energy. The maximum measurable range depends on the illuminated material roughness and color, and the laser wavelength [Fidera et al. 2004, Pesci and Teza, 2008]. Divergence values for new generation long-range scanners are extremely reduced, illuminating very small surface elements for each shot. The spot dimension increases linearly with the distance, and is always greater than the lower limit of the instantaneous field of view (IFOV) due to physical diffraction. Effective laser scanner characteristics are defined by a set of parameters, including: range resolution (depending on telemeter efficiency), single point measurement accuracy (depending on the internal electronic device, signal-to-noise ratio and critical time needed for pulse recognition), beam divergence (which defines the IFOV, depending on laser wavelength), and minimum angular step (depending on the internal mirrors calibrated system) [Wehr and Lohr 1999]. Overlap is the laser scanning strategy that can reduce errors, because redundant points are acquired belonging to the same illuminated area. A common overlap is obtained by fixing the ratio between spot dimension (the area illuminated by a single pulse with a given divergence) and angular step so that a given point is measured 10 times. For instance, if the divergence is 3 mrad and angular variation about 0.3 mrad, at 100 m distance, an element included in a 3 cm area is observed 10 times. The final result of a laser scanner application is a very dense point cloud, with radiometric reflectivity data for each point.

Two MatlabTM software packages for strain field computation, starting from displacements of experimental points (EPs), are here presented. In particular, grid_strain estimates the strain on the nodes of a regular planar grid, whereas grid_strain3 operates on the points of a digital terrain model (DTM). In both cases, the computations are performed in a modified least square approach, emphasizing the effects of nearest points. This approach allows users to operate at different scales of analysis by introducing a scale factor to reduce or also exclude too far points from grid nodes. The input data are displacements (or velocities) that can be provided by several techniques (e.g. GPS, total topographical station, terrestrial laser scanner). The analysis can be applied to both regional- and local-scale phenomena, to study tectonic crustal deformations or rapid landslide collapses, and to characterize the kinematics of the studied system. Errors on strains and geometric significance of the results are also provided.

A time-of-flight terrestrial laser scanner (TLS) works like a laser rangefinder but is also able to scan a surface with a selected sampling step, leading to a point cloud defined into an internal reference frame. Moreover, radiometric information about signal returns intensity is provided, computed in terms of the ratio between emitted and received pulse intensities. Besides the geometric and radiometric data, some TLS instruments, equipped by an internal or an external calibrated camera, are able to provide also the RGB information by mean of a texturing procedure, i.e. a procedure of point cloud coloring. In general, the mapping of a 2D RGB image onto a digital surface model during 3D rendering, increases the realism of the resulting model. The point cloud texturing is currently used in TLS observation aimed to cultural heritage study and preservation (see e.g. Beraldin et al., 2002). The texturing procedure consists in camera calibration and orientation. The camera calibration models the true parameters of the camera (i.e. focal length, format size, principal point, and lens distortion, see e.g. Tsai 1986, 1987), whereas the orientation process leads to the position and angles of the camera with respect to the point cloud reference frame. Pairs of corresponding points, visible both in the captured RGB image and in the corresponding point cloud can be selected and used to solve for the calibration parameters of the camera in a the least square approach. Depending on the unknown quantities that must be obtained, in particular the availability of calibration parameters of the camera for the considered field, the procedure requires the detection of some or several pairs of corresponding points. In some cases, 13 parameters could be necessary, and at least 20 pair of points are therefore needed to have an adequate redundancy in the least square computation. The Optech ILRIS 3D (Optech, 2010) is a TLS instrument characterized by good performance in long range acquisition (up to 1200 m), which is equipped with an internal integrated camera (3 megapixel) to provide the RGB information and therefore allow the point cloud texturing. The format of the data directly provided by a TLS instrument is specific for the instrument. In order to allow their importation by the standard data processing software packages (e.g. Innovmetric PolyWorks, I-Site Studio, Leica Cyclone. For a list of available packages see e.g. GIM, 2009), a parsing procedure is necessary. In the case of ILRIS instrument, the parsing procedure is implemented in Optech Parser software and is well described also in the short manual provided by Pesci et al. (2009), where significant information about the planning and the definition of a reasonable acquisition protocol of a TLS measurement is also provided. A new feature available in Optech Parser allows the point cloud texturing, by means of 2D images captured at the scanning epoch by the internal camera by using the camera calibration certificate of the integrated camera. In particular, to perform the computations, the "texture image" and the “texture calibration parameter” files have to be used. The first file is the image automatically stored by the scanner and the second one is provided by Optech (Optech, inc). If these two files are directly used, the texturing procedure is completely automatic. Nevertheless, in some cases a high accuracy in RGB data treatment is necessary, and the recognition of an adequate number of pairs of corresponding point must be carried out by means of Optech Matching Viewer software package (Optech, inc). The Matching Viewer package is able to provide a texture calibration parameter file for a fine texturing. This technical report describes in detail the procedure for point cloud texturing using images from an external calibrated digital camera and Matching Viewer. In particular, the Nikon reflex D50, which is characterized by 28-mm focal length and 8.0 megapixel resolution. A TLS survey of the Santo Stefano historical square of Bologna city, aimed to provide the high resolution acquisition of Corte Isolani Palace, is performed. Figure 1 shows the external camera mounted on the ILRIS instrument during the survey. The scanner handle is shaped to allow the high precision installation of a camera on a calibrated position. Anyway the mechanical components lead to alignment errors to be corrected by means of a specific matching procedure described in the following.

The global positioning system (GPS), in both static and kinematic modes, allows a highly accurate measurement of point coordinates and therefore is widely used for monitoring both slow and fast surface deformations. The information provided by a GPS network can be used at the regional scale, to evaluate tectonic and seismogenic structure evolutions [Hunstad et al., 1999; Pietrantonio and Riguzzi, 2004], such as the estimation of deformation rates in the central Apennine chain [Pesci and Teza, 2007], or at larger scale, to monitor gravitational macroscopic effects due to, for example, rock-mass collapses, landslide activations or other instabilities [Mora et al., 2003; Tzenkov and Gospodinov, 2003; Squarzoni et al., 2005]. The accuracies of GPS measurements are generally a few millimeters for the horizontal coordinate components and sub-centimeters for the vertical ones. In fact, the elevation is highly influenced by atmospheric perturbations, involving zenith delays, which are difficult to be completely removed by means of data modeling. When referring to high accuracy, GPS surveying implies the precise measurements of the vectors between two or more receivers (baselines), the so-called relative positioning: data can be acquired on static and rapid-static conditions, which require GPS stations to be stationary. Several permanent GPS stations continuously operate on the Italian territory, belonging to different institutes like IGS (International GPS Service), EUREF (European Reference Frame), ASI (Agenzia Spaziale Italiana), INGV (Istituto Nazionale di Geofisica e Vulcanologia) and others [Serpelloni et al. 2006; Falco et al., 2007; Devoti et al., 2008]. Due to the high efficiency of this surveying methodology, in the last few years, the number of GPS permanent stations has rapidly increased and continues to expand; the Earth Science Department of Siena University, for example, installed 8 new stations in 2003 to study the tectonic processes in the Central-Northern Apennines [Cenni et al., 2004]. Also private GPS networks planned for commercial civil proposal exist; in particular the ASSOGEO s.r.l (Italian Trimble provider), established a dense GPS network for real time positioning by means of the VRS (Virtual Reference Station) concept [Hu et al., 2003] and work is still in progress to cover the whole Italian territory with a mean size of about 20-50 km.

It is the user guide of a Matlab software intended to calculation of the strain (or strain-rate) on a regular grid, using GPS data. The described software can be freely obtained from the authors (for scientific use only)

The OptecTM ILRIS-3D Terrestrial Laser Scanner (TLS) has been recently acquired by the Istituto Nazionale di Geofisica e Vulcanologia (INGV) Sezione di Bolgna in the context of a TTC program for volcanoes monitoring (TTC 1.3 Controllo Geodetico delle Aree Vulcaniche Attive) supported by the Dipartimento di Protezione Civile. Several experiments were performed by INGV since 2004 to study the level of precision for surface modelling by means of laser scanner long range instruments, in order to detect the best suitable standard for rapid and simple acquisition in volcanic area (Pesci et al., 2007). In particular, during the MESIMEX experiment (October 2006), a national exercitation organized by the Dipartimento di Protezione Civile (DPC) exploited to simulate a volcanic eruption in Naples, the second TLS survey of the whole Vesuvius crater was executed and a large mass variations were estimated revealing the collapse of a portion of the crater. The alignment and comparison of point clouds (2006–2005) show high variations over a large portion of the NE slope and a volume variation of about 6850 m3 was computed. The analysis was performed in almost real time by means of direct comparisons between scans, indicating the laser scanning as one of the most reliable technique for fast monitoring in crisis time (Pesci et al. 2008a). The main characteristics recommended for surveying in volcanic areas were the laser device eyes safety, the achievable very long range (> 1 km), the precision of measurements and final accuracy in data modelling, the acquisition velocity, the instrument portability in terms of weight and size and the ability to manage scanner by means of PC pocket. The ILRIS-3D scanner was chosen based on the previously described recommended points. The simple operation needed for scan execution and the possibility to plan and realize a complete survey by means of only two operators confirmed ILRIS-3D as the best choice for volcanic applications. This technical report is a simple and effective user guide for laser scanner management providing all the necessary instruction from instrument settings, remote connection, data storage, downloading and preprocessing. Authors proposal is to make operators independent enough to scan and carry out survey in interested areas also without a specific experiences in LIDAR (Light Detection And Ranging) monitoring.

Two MatlabTM software packages for strain field computation, starting from displacements of experimental points (EPs), are here presented. In particular, grid_strain estimates the strain on the nodes of a regular planar grid, whereas grid_strain3 operates on the points of a digital terrain model (DTM). In both cases, the computations are performed in a modified least square approach, emphasizing the effects of nearest points. This approach allows users to operate at different scales of analysis by introducing a scale factor to reduce or also exclude too far points from grid nodes. The input data are displacements (or velocities) that can be provided by several techniques (e.g. GPS, total topographical station, terrestrial laser scanner). The analysis can be applied to both regional- and local-scale phenomena, to study tectonic crustal deformations (strain 6 8 10 10 − ≈ − ) or rapid landslide collapses ( 2 4 10 10 − − ), and to characterize the kinematics of the studied system. Errors on strains and geometric significance of the results are also provided.

The availability of a GPS network of 10-20 km mean size, provides good topographical support for the measurement of ground displacements, even at a local scale such as a landslide. In particular, a series of multitemporal kinematic or rapid-static GPS acquisitions of a landslide allows a good characterization of its displacements if the measurements are referred to a GPS reference network. Nevertheless, a wider network formed by stations located at long distances, for example at several tens of kilometers, characterized by large spacing, can lead to results affected by high noise, degrading the accuracy of final point positions. In order to obtain an adequate GPS reference network, some virtual reference stations (VRSs) can be introduced, even if a network refinement based on VRS cannot reach the same accuracy of a real local network. Some experiments, including measurements on a real landslide, have been performed in order to evaluate the performance of this technique. The results point out that the standard deviation of the obtained solutions is about two or three times larger than those which can be reached using a real local network.