Dr. Giovanni Angelo Meles

Giovanni Angelo Meles received a M.Sc. degree in physics from the Università Statale di Milano, Milan, Italy, in 2004. In 2007 he moved to Zurich, where in 2011 he earned a Ph.D. degree from the Swiss Federal Institute of Technology (ETH) under the supervision of Professors Alan Green, Stewart Greenhalgh, and Jan van der Kruk.

His Ph.D. project focused on full-waveform inversion of ground-penetrating radar, with special interest in algorithmic developments. During his studies, he developed the first fully vectorial waveform inversion scheme capable of simultaneously updating permittivity and conductivity distributions in any recording configuration. To tame the non-linearity problem associated with high contrast media, he subsequently introduced a frequency-time domain inversion algorithm based on progressive bandwidth expansion. He devoted the last part of his Ph.D. to sensitivity and resolution analysis, also setting the basis for Gauss-Newton inversion of GPR data.

From September 2011 to December 2016, he has been within the Edinburgh interferometry Project (EIP) at The University of Edinburgh, Edinburgh, U.K, where he has conducted research within the context of seismic interferometry and Marchenko redatuming in close contact with Prof. Andrew Curtis.

Since February 2017 he has been working as part of Prof. Kees Wapenaar's group at Delft University of Technology, Delft, The Netherlands.

Research Interests:

Giovanni's research interests comprise wave scattering and diffraction, imaging, full-waveform inversion and tomography.

While in Edinburgh, he has been focusing mainly on source-receiver interferometry (Meles and Curtis, 2013) and properties of pseudo-physical energy (Löer et al., 2013).

Within the context of source-receiver interferometry, he has proposed algorithms to analyze multiply diffracted waves based on move-out invariants (Meles and Curtis, 2014b).

More recently, he has devised a new methods based on convolutional representation theorems and Marchenko autofocusing (Meles et al., 2015, Meles et al., 2016) to estimate internal multiples or primary reflections .


Wavefield finite time focusing with reduced spatial exposure.

Wavefield focusing is often achieved by time-reversal mirrors, where wavefields emitted by a source located at the focal point are evaluated at a closed boundary and sent back, after time-reversal, into the medium from that boundary. Mathematically, time-reversal mirrors are derived from closed-boundary integral representations of reciprocity theorems. In heterogeneous media, time-reversal focusing theoretically involves in- and output signals that are infinite in time and the resulting waves propagate through the entire medium. Recently, integral representations have been derived for single-sided wavefield focusing. Although the required input signals for this approach are finite in time, the output signals are not and, similar to time-reversal mirroring, the resulting waves propagate through the entire medium. Here, an alternative solution for double-sided wavefield focusing is derived. This solution is based on an integral representation where in- and output signals are finite in time, and where the energy of the waves propagating in the layer embedding the focal point is smaller than with time-reversal focusing. The potential of the proposed method is explored with numerical experiments involving a head model consisting of a skull enclosing a brain.
Normalized L2 norm of the pressure wavefields associated with standard time-reversal focusing (a), standard (double-sided) Marchenko focusing (b), and finite time focusing (c), respectively, plotted as functions of space. In standard time-reversal focusing (a), the norm of the pressure wavefield exhibits a peak at the focal point [blue arrow in (a)], and significant values are almost homogeneously distributed throughout the model [red arrows in (a)]. A similar distribution, with large values along the focal plane, is obtained when standard (double-sided) Marchenko focusing is used (b). In finite time focusing, the wavefield is still exhibiting a peak at the focal point [blue arrow in (c)] while being somehow confined into a double cone centered at the focal point [blue cones in (c)]. Black and green arrows point at regions of the brain with minimal wavefield propagation and large amplitude spots associated with the propagation of the coda of the focusing functions, respectively. Red and blue dashed lines indicate horizontal and vertical sections used in (d)?(e), respectively. Horizontal (d) and vertical (e) slices of the maps in (a)?(c), plotted in decibel scale (20?log10(?p?)). Black arrows in (d) indicate large portions of the focal plane [red dashed lines in (a)?(c)] where wavefield propagation in finite time focusing is significantly reduced as opposed to time-reversal and standard (double-sided) Marchenko focusing. The red and black arrows in (e) indicate zones along the green dashed lines in (a)?(c) where finite time focusing and time-reversal focusing involves slightly larger and slightly smaller wavefield intensity, respectively. Green arrows point at zones outside of the skull where standard (double-sided) Marchenko and finite time focusing involve propagation of coda exhibiting large amplitudes [see green arrows in (c)]. 

Virtual plane-wave imaging via Marchenko redatuming.

Marchenko redatuming is a novel scheme used to retrieve up- and downgoing Green?s functions in an unknown medium. Marchenko equations are based on reciprocity theorems and are derived on the assumption of the existence of functions exhibiting space?time focusing properties once injected in the subsurface. In contrast to interferometry but similarly to standard migration methods, Marchenko redatuming only requires an estimate of the direct wave from the virtual source (or to the virtual receiver), illumination from only one side of the medium and no physical sources (or receivers) inside the medium. We propose to consider a different time-focusing condition within the frame of Marchenko redatuming that leads to the retrieval of virtual plane-wave responses. As a result, we obtain multiple-free imaging using only a 1-D sampling of the targeted model at a fraction of the computational cost of standard Marchenko schemes. The potential of the new method is demonstrated on 2-D synthetic models.

(a) Migration result using the imaging condition of and Marchenko redatumed virtual-plane wavefields. The red arrows point at low amplitude artefacts, whereas the blue arrows point at resolved structures not visible in the standard migration image. The red box encircles an area where dipping interfaces are not imaged. (b) Migration result using standard one-way extrapolation of virtual-plane wavefields. Red arrows point at multiple-related artefacts. (c) Migration result using plane-wave Marchenko wavefields associated with the tilted planes in Fig. 8(c). Blue arrows indicate dipping interfaces now properly imaged. . 

Beyond adaptive subtraction of internal multiples: direct reconstruction of primaries in seismic data sets.

Whereas advanced methods of seismic data processing such as recursive imaging or full-waveform inversion can properly take into account data that includes multiply scattered waves, many current standard processing steps including reverse-time migration (RTM) are based on the so-called Born approximation. This approximation assumes that waves have only scattered from heterogeneities in the medium once, thus requiring that data consist only of primaries – singly scattered energy. A variety of methods are therefore deployed as pre-processing to predict multiples (waves reflected several times); however, accurate removal of those predicted multiples from recorded data using adaptive subtraction techniques proves challenging, even in cases where they can be predicted with reasonable accuracy. To overcome this problem, we propose a new, alternative strategy: instead of synthetizing and removing multiples, we construct a parallel data set consisting of only primaries, which is calculated directly from recorded data. This approach obviates the need for both multiple prediction and removal methods. We show how primaries can be constructed using convolutional interferometry to combine first arriving events of up-going and direct-wave down-going Green’s functions to virtual receivers in the subsurface. The required up-going wavefields to virtual receivers are constructed by Marchenko redatuming, a novel technique that estimates up- and down-going components of Green’s functions between an arbitrary location inside a medium such as the Earth’s subsurface where no sources (or receivers) are placed, and real receivers (or sources) located at the surface. 

Internal multiples prediction using seismic interferometry and Marchenko Autofocusing

Standard seismic processing steps such as velocity analysis and reverse time migration (imaging) usually assume that all reflections are primaries: Multiples represent a source of coherent noise and must be suppressed to avoid imaging artifacts. Many suppression methods are relatively ineffective for internal multiples. Here Meles et al. show how to predict and remove internal multiples using Marchenko autofocusing and seismic interferometry. We first show how internal multiples can theoretically be reconstructed in convolutional interferometry by combining purely reflected, up- and downgoing Green’s functions from virtual sources in the subsurface. We then generate the relevant up- and downgoing wavefields at virtual sources along discrete subsurface boundaries using autofocusing. Then, we convolve purely scattered components of up- and downgoing Green’s functions to reconstruct only the internal multiple field, which is adaptively subtracted from the measured data. Crucially, this is all possible without detailed modeled information about the earth’s subsurface. The method only requires surface reflection data and estimates of direct (nonreflected) arrivals between subsurface virtual sources and the acquisition surface. The method is demostrated on a stratified synclinal model and shown to be particularly robust against errors in the reference velocity model used.


Finger-printing ordered diffractions in multiply-diffracted waves

We present a method to classify diffractors based on the variation of acoustic wave travel time variations (their so-called move-outs) across arrays of sources and receivers. We show that this information is sufficient to allow the diffraction path of any recorded multiply-diffracted wave to be determined: each recorded wave can be associated with the concatenation of an ordered series of known, irreducible, inter-diffractors paths, or equivalently by an ordered series of single-diffraction interactions. These are determined purely by data analysis through inspection and comparison of common-source and common-receiver gathers, without the need for synthetic wavefield computation, or for modelling of the medium through which energy propagates. The method is effected by a new algorithm that identifies diffraction paths by wavefield analysis. Applications of the proposed algorithm within the various fields above range from interpreting reverberating wave energy associated with multiply-diffracted waves in terms of the contributions of its individual diffractors, improved location or characterisation of diffractors or energy sources, removal of multiply-diffracted energy by muting or filtering to improve the performance of methods designed only for singly-diffracted energy, and all of these may lead to improved imaging of the inter-diffractors medium.


Publications  (Google Scholar)

[26] Meles, G.A. , Zhang, L., Thorbecke, J., Wapenaar, K., Slob, E., Data-driven retrieval of primary plane-wave responses. Submitted to Geophysical Prospecting.

[25] Meles, G.A. , van der Neut, J., van Dongen, K., Wapenaar, K., Wavefield finite time focusing with reduced spatial exposure. Journal of the Acoustical Society of America, Volume 145, Pages: 3521-3530.

[24] Meles, G.A., Wapenaar, K., Thorbecke, J., Virtual plane-wave imaging via Marchenko redatuming. Geophysical Journal International, Volume: 214, Pages: 508-519.

[23] da Costa Filho, CA., Meles, G.A., Curtis, A., Ravasi M., Kritski A., Imaging strategies using focusing functions with applications to a North Sea field. Geophysical Journal International, Volume: 213, Pages: 561-573.

[22] da Costa Filho, C., Meles, G.A., Curtis, A., Elastic internal multiple analysis and attenuation using Marchenko and interferometric methods. Geophysics 82 (2), Q1-Q12

[21] Löer, K., Curtis, A., Meles, G.A., Relating source-receiver interferometry to an inverse-scattering series to derive a new method to estimate internal multiples. Geophysics 81 (3), Q27-Q40.

[20] Ravasi, M., Vasconcelos, I., Kritski, A., Curtis, A., da Costa, C., Meles, G. A., Target-oriented marchenko imaging of a North Sea field. Geophysical Journal International, in press.

[19] Meles, G.A., Wapenaar K., Curtis, A. Synthesising primary reflections by Marchenko redatuming and convolutional interferometry. Geophysics, in press.

[18] Meles, G.A., Löer, K., Ravasi, M., Curtis, A., da Costa, C., Internal multiple prediction and removal using Marchenko autofocusing and seismic  interferometry. Geophysics, Volume: 80, 2015, Pages: A7-A11.

[17] Galetti, E., Curtis, A., Meles, G.A., Baptie, B., Uncertainty Loops in Travel-Time Tomography from Nonlinear Wave Physics. Physical Review Letter, Volume: 114, 2015, Pages: 148501/1-5

[16] Ravasi, M., Meles, G.A., Curtis, A., Rawlinson, Z., Liu, Y., Seismic interferometry by multi-dimensional deconvolution without wave field separation. Geophysical Journal International, Volume: 221, 2015, Pages: 1-16.

[15] Ravasi, M., Vasconcelos, I., Curtis, A., Meles, G.A., Elastic extended images and velocity-sensitive objective functions using multiple reflections and  transmissions. Geophysical Journal International, Volume: 202, 2015,  Pages: 943-960.

[14] Löer, K., Meles, G.A., Curtis, A., Automatic identi cation of multiply diffracted waves and their ordered scattering paths. Journal of the Acoustical  Society of America, Volume: 137, 2015, Pages: 1834-1845.

[13] Entwistle, E., Curtis, A., Galetti, E., Baptie, B., Meles, G.A., Constructing new seismograms from old earthquakes: Retrospective seismology at multiple length scales. Journal of Geophysical Research: Solid Earth, Volume: 120, 2015, Pages: 2466-2490.

[12] Meles, G.A., Curtis, A.Discriminating physical and non-physical diffracted energy in sourcereceiver interferometry. Geophysical Journal International, Volume: 197, 2014, Pages: 1642-1659.

[11] Meles, G.A., Curtis, A.Fingerprinting ordered diffractions in multiply diffracted waves. Geophysical Journal International, Volume: 198, 2014, Pages: 1701-1713.

[10] da Costa, C.A., Ravasi, M., Curtis, A., Meles, G.A. Elastodynamic Green's function retrieval through single-sided Marchenko inverse scattering. Physical Review E, Volume: 90, 2014.

[9] Löer, K., Meles, G.A., Curtis, A., Vasconcelos, I. Diffracted and pseudo-physical waves from spatially limited arrays using sourcereceiver interferometry (SRI). Geophysical Journal International, Volume: 196, 2013, Pages: 1043-1059.

[8] Yang, X., Klotzsche, A., Meles, G.A., Vereecken, H., van der Kruk, J. Improvements in crosshole GPR full-waveform inversion and application on data measured at the Boise Hydrogeophysics Research Site. Journal of Applied Geophysics , Volume: 99, 2013, Pages:114-124.

[7] Meles, G.A., Curtis, A. Physical and non-physical energy in scattered wave source-receiver interferometry. Journal of the Acoustical Society of America, Volume: 133, 2013, Pages: 3790-3801.

[6] Klotzsche, A., van der Kruk, J., Meles, G., et al. Crosshole GPR full-waveform inversion of waveguides acting as preferential  ow paths within aquifer systems. Geophysics, Volume: 77, 2012, Pages: H57-H62.

[5] Meles, G.A., Greenhalgh, S.A. Green, A.G., Maurer, H. and Van der Kruk J. GPR Full Waveform Sensitivity and Resolution Analysis using an FDTD Adjoint Method. IEEE Transactions on Geosciences and Remote Sensing, Volume: 50, 2012, Pages: 1881-1896.

[4] Meles, G.A., Greenhalgh, S.A., Van der Kruk, J., Maurer, H. Green, A.G. Taming the non-linearity problem in GPR full-waveform inversion for high contrast media. Journal of Applied Geophysics, Volume 73, 2011, Pages: 174-186.

[3] Klotzsche, A., van der Kruk, J., Meles, G.A., Doetsch, J., Maurer, H., Linde, N. Full-waveform inversion of cross-hole ground-penetrating radar data to characterize a gravel aquifer close to the Thur River, Switzerland. Near surface geophysics 8 (6), 635-649

[2] Meles, G.A. Van der Kruk, J. Greenhalgh, S.A. Ernst, J.R. Maurer, H. Green, A.G. A New Vector Waveform Inversion Algorithm for Simultaneous Updating of Conductivity and Permittivity Parameters From Combination Crosshole/Borehole-to-Surface GPR Data. IEEE Transactions on Geosciences and Remote Sensing, Volume: 48, 2010, Pages:. 3391-3407.

[1] Vassena, C., Giudici, M., Ponzini, G., Parravicini, G., Meles, G.A., Tomographic Approach to Identify Transmissivity with Differential System Method, Journal of Hydrologic Engineering, Volume: 12, 2007, Pages: 617-625.


- Ph.D., Institute of Geophysics, ETH Zurich.
Field of Research: GPR Full waveform inversion
Project title: New developments in Full Waveform inversion of GPR data
Supervisors: Prof. Alan Green, Prof. Stewart Greenhalgh, Prof. Jan van der Kruk

- M.Sc., Computational physics, Dipartimento di Fisica, Universita' Statale di Milano, Milan
Field of Research: Hydrology, Groundwater, Inversion
Title: An inverse problem for phreatic aquifers.


2nd-5th September 2016: Marie Curie WAVES workshop in Doorn, The Netherlands.

17th-21st September 2015: Marie Curie WAVES workshop in Pitlochry, Scotland (UK).

April 2009: 26 Lessons Course on Writing Research Papers for Publication held by Thomas Armstrong in Zurich (Switzerland).

12th-16th May 2008: Course on New Geophysical tools for Hydrological Investigations, held in Zurich (Switzerland).

September 2007: 42 Hours Course on inverse problems, held by Albert Tarantola in Neuchatel (Switzerland). Focus on basic inversion theory, nonlinear inversion, full-waveform inversion.