Home

here - ELIOSOFT

1. 50 55 60 frequenza Hz Esplora spettro modellazione diretta picking parametri salva modello seleziona modo 2 selezionare l ultimo punto del modo 3 Rayleigh v utilizzando il tasto destro a mie salva picking cancella picking Esci Full complete Dataset note the energy linked to higher mode over than the fundamental Primo trattamento dati dataset 4022 dat offset minimo 2 m distanza intergeofonica 2 m campionamento 0 5 ms Tracce normalizzate tempo s 30 offset m Utilit Seleziona dati x Invia e mail ver 4 0 Pro Secondo determinazione spettro di velocit e picking calcolo spettro di velocit visualizza curve Spettro di Velocit cs E ee N RB ec co e e e e e m e m e co e velocit di fase m s 3 0 B D 5 i i G a a o frequenza Hz m modellazione diretta picking parametri salva modello seleziona modo v selezionare l ultimo punto del modo 3 Rayleigh v utilizzando il tasto destro salva picking 2 cancella picking Esplora spettro Esci Selection of the signal to mark up the relevant section to higher mode modes NOTE to proceed on with the analyses the new dataset does not need to be saved save button www winmasw com geophysical software amp services winMA
2. layer 4 Poisson p en ce ix layer 5 Poisson phase velocity m s layer 6 Poisson NEES Poisson OM 20 numero di modi da visualizzare 3 min 1 max 20 frequency Hz Analysis explore spectrum Rayleigh l cancel calculate modelling picking movie 2 cancel save parameters save mode fundamental w i use the right button to select the upload model 3 Rayleigh v last point of the considered mode 7 eigen period refresh save picking cancel picking wwWw ellosoft it wi n M A S W ver 4 1 Pro Attenuation analysis exit 1 uploading amp processing MASW analyses 2 velocity spectrum modelling amp picking MASW amp ReMi analyses dataset lamporecchio ASW ertical sgy MASW ReMi VISURE RE ET minimum offset 10 m calculate spectrum Tau v upload ReMi spectrum een geophone spacing 1 m sampling 0 5 ms normalized traces velocity spectrum phase velocity m s 20 frequency Hz explore spectrum spectrum movie select 20 upload save picking cancel save i fundamental w use the right button to select the piosd magi 3 Rayleigh jpg last point of the considered mode inversion www ellosoft it V eigen period refresh save picking cancel picking utilit
3. window 60 min freq Hz 1 max freq Hz 80 Time s Time 10 20 WWW Winmasw com geophysical software amp services Spectrogram 40 50 60 70 80 Frequency Hz page 52 winMASW Picking of the dispersion curve To pick the dispersion curve i e to select those points that according to the user belong to a particular propagation mode of the surface wave it is necessary to 1 choose the mode from the scroll menu 2 click left mouse button the points along a certain coherence that the user identifies and understands as a particular mode see example in picture 2 3 save the picking If there are more datasets in our mode just pass to the other one scrolling down the menu once you ve done and saved the picking of a mode Different modes will be marked in different colours The new data relevant to the second mode will be saved in the formerly indicated file The final inversion will finally consider the whole of data The picking file is a three columns ASCII file cdp curve of dispersion in Italian the first shows the frequencies the second the velocities and the third the mode as chosen by the user scrolling down the menu We suggest to save the dispersion curve in the default directory dispersion curves El winMASW 4 1 Pro Velocity Spectra Modeling Picking GE 1 uploading amp processing MASW analyses 2 velocity spectrum modelling amp picking MASW amp ReMi a
4. www winmasw com geophysical software amp services page 70 winMASW 5 Recommendations and advices General The dispersion of surface waves is a powerful tool to draw the vertical profile of the velocity of transverse waves Vs Nevertheless before using this software or others you need to be confident with the field and with the use and evaluation of picking figures both for inversion and dispersion It is really advisable to read more about and take part to conventions workshops too ELIOSOFT is available to organize such events Visit www winmasw com and or write to winmasw winmasw com One more recommendation before interpreting the velocity spectrum you better thorough check the data youre analyzing traces according to the American acronym GIGO garbage in garbage out if you input garbage you ll get garbage The analysis of the surface waves is not a trick giving a solution even if data quality is low or the user misunderstands them Other signals particularly the guided waves resulting from the presence of layers inside which the waves pass can result in dispersive signals that the user is likely to wrongly read as surface waves Besides the different modes can interfere with each other and give misleading or wrong results Inversion parameters The used inversion procedure as in winMASW is based on the so called genetic algorithms This kind of approach lets the user obtain more reliable results compared t
5. www winmasw com geophysical software amp services winMASW page 63 visualize curves 2 velocity spectrum modelling amp picking MASW ESAC amp ReMi analyses handling the spectra 1 data uploading amp processing MASW compute velocity spectrum input curve FK dataset MASW classicalRay sgy resampled explore spectrum sampling 0 5ms WVRNG high frequency 4096 minimum offset 3 m geophone spacing 3m group velocity save upload phase velocity merge mode separation direct amp refraction travel times picking Rayleigh Dispersion show f k Vsu m s 150 90 135 199 150 400 900 Vs30 210 select mode v to select the last point of the considered mode click the right button im tnn yt save picking cancel picking inversion phase velocity m s inversion Joint DC HV inv 15 20 3 frequency Hz offset m modelling synthetics Poisson thickness m 0 42 0 499 0 47 0 49 0 47 0 351 0 279 0 2 H V modes SW ellipticity 0 0 15 0 0 15 Vs m s Qs 150 17 90 9 135 13 5 199 19 9 150 15 400 40 900 90 190 refr amp ref about Poisson upload mod ZVF v data selection filtering amp spectra filter cancel resampling save model 0 5 v activate refr refl 1 shows DC general setting refresh save show model select 60 uplo
6. El winMasw data upload sing amp modelli a Evs uva x cvm Gu Bgeeetyeu3 r 1 uploading amp processing MASW analyses dataset ZVF dx2 mo4 lowpassfil sqy sampling 0 25 ms minimum offset 4 m geophone spacing 2 m refr refi 100 upload gawa clear 2 ra 3 0 7 time to visualize s done cut flip traces test amplitude zero padding refr amp refl 2 velocity spectrum modelling amp picking MASW amp ReMi analyses MASW compute velocity spectrum phase velocity Fk group velocity handling the spectra save upload merge normal lized traces velocity spectrum OTO T l T 1Eeee4 2 cd n SE 2 aes EIU mode clic 2 cd z Gm ee age n TR 4 lt m m A 03 ED picking a eee ae PITT am or TIST Pr ME E o E E 7 gs Le HMM 2 umm 05rD cebnxeeqee F aA ea ees 4 0 6 F MAE T wife 4 a 5 10 15 20 25 30 35 40 45 50 about Poisson 0 Reference depth V Refraction 484 H V body waves 0 H V modes SW ellipticity www winmesw com Almost useless to remark that the results of your analysis depends first of all by your expertise skill both from the theoretical and practical points of view www winmasw com geophysical software amp services winMASW page 4
7. From the main panels Velocity spectrum a modelling amp picking the user can access to a tool useful for mode separation mode separation button After having uploaded a dataset by clicking the mode separation button the following panel appears E winMASW mode separation D r original save spectrum phase velocity m s La r e e 5 10 15 20 25 30 35 40 45 50 55 offset m On the left side are reported the original data in the f v and x t domains By moving the points of the polygon evident in the velocity spectra f v domain it is possible to draw a polygon that will be successively adopted as filter by clicking the select button only the data within the polygon will be kept while data external to it will be removed the filter actually works in the f k domain An example will clarify the point In the following on the left side are reported the original data while on the right side the data after having removed the portion of data external to the chosen polygon user moves the vertices of the polygon and then click the select button We first decided to put in evidence the data in the area where fundamental mode was supposed to be the very high amplitude of the higher mode s for frequencies higher than 13Hz prevents to see it www winmasw com geophysical software amp services winMASW page 125
8. MFA Multiple Filter Analysis Academy version Version 4 7 improves the method MFA Multiple Filter Analysis method e g Luo et al 2011 previous implementation was more sensitive to noise This method allows to define dispersion curves for group velocities not for phase velocities anyway the procedures are in general absolutely similar to the ones adopted for phase velocity analyses MASW or ReM First as usual upload the common shot gather icon top left Theoretically to perform MFA analyses one trace could be sufficient nevertheless the resulting spectrum will surely result more robust when obtained as the average of several traces we may suggest 3 6 traces at least It is not possible to summarize in a manual all the characteristics and the pros amp cons of a methodology and we will limit the presentation to 2 points only 1 the link between VsuppH phase velocity of the surface wave and Vsuper group velocity of the surface wave and Vs shear wave velocities in depth is different in other words the relationship frequency Vs and frequency Vsuppk is different from the one between frequency Vs and Vsuper 2 with respect to VsuppH Vsuper seem sometimes more sensitive to Vs variation in depth see e g Luo et al 2010 For these reasons we suggest the joint analysis MASW MFA possibly HVSR button Joint Analysis of Phase amp Group Velocities Parameters Alpha0 amp Alpha1 The Gaussia
9. www winmasw com geophysical software amp services winMASW page 57 3C Professional and Academy Versions Joint analysis of Rayleigh and Love waves Since the 4 1 version both dataset relevant to Rayleigh and Love analyses can be jointly analyzed For this reason the main winMASW screenshot displays a group dedicated to a joint analysis of Rayleigh and Love Joint Analysis of Rayleigh amp Love Clicking on Velocity Spectra Modelling amp Picking the window containing uploading picking and modelling of both datasets one relevant to Rayleigh and the other relevant to Love displays i Rayleigh data 0 trace flip n dataset rayleigh 10m SGY minimum offset 10 m geophone spacing 2 m sampling 2 ms 1000 Rayleigh component 0 5 700 select data E activate E 800 select 20 500 gt E cancel save amp 400 i 300 1 5 gt picking 200 select mode x 100 a save A 10 15 320 25 30 35 wg 35 50 E ES frequency Hz input Love data input Love data data trace fip 0 dataset love 10m SGY minimum offset 10 m geophone spacing 2 m 0 5 sampling 2 ms select data E gt activate re o select save E E cancel E ids picking select mode X 2 save cancel 10 us 20 25 30 35 40 45 50 2 et frequency Hz Clicking instea
10. B First shot Jie C A B 2 E Second shot Please notice that this way in the final dataset the geophones are spaced B 2 Mode sequence type 1 move the geophones B First shot la E 1 Second shot X C Mode sequence type 2 move the source B First shot e C A B n LE Second shot n number of channels in a single shot The fastest one is surely the first one a With winMASW you can deal with all of these 3 cases In the following an example of acquisition performed while considering the third type Mode sequence type 2 Anyway remember what is really important is not the number of channels typically 123 are absolutely sufficient see e g Dal Moro et al 2003 but the total length of the array www winmasw com geophysical software amp services page 140 winMASW Two datasets have been acquired according to following features dataset 1 distance between geophones 1 5m min offset 1 5m dataset 2 distance between geophones 1 5m min offset 19 5m It s obvious the distance between geophones has to be the same and the min offset of the second dataset has to be equal to the last offset of dataset 1 plus a value equal to the inter geophone distance EH winMASW Tool per la somma di 2 dataset 5 x Somma dei 2 dataset help leggimi datasett Visualizza e salva il file SGY somma Fi i ruota tracce dataset risultante 0 e gt y
11. CO D D phase velocity m s phase velocity m s h3 e o e e 5 10 15 20 5 10 15 20 5 10 15 20 frequency Hz c frequency Hz r frequency Hz C modify limits save selected spectrum analyze saved spectrum At this stage analyses continue like with MASW but with an only difference as briefly in following scheme principle velocity spectrum velocity spectrum phase velocity m s phase velocity m s frequency Hz frequency Hz In case of MASW data active seismic In case of ReMi data passive seismic Modelling Picking Criteria for both data types active or passive seismic www winmasw com geophysical software amp services winMASW page 35 Suggested procedure If you need to improve the Vs vertical profile you can put a MASW and a ReMi acquisition results together as follows 1 determine the dispersion curve from MASW trial and save the picked curve file cdp 2 Analyze ReMi data save the clearest spectrum see procedure like before 3 upload the picking curve of the MASW data button input curve top right in the section visualize curves 4 proceed with the picking of the Spectrum from ReMi analysis as done with the MASW one the aim is to find some more points at lower frequencies In the file data Lamporecchio you ll find some useful data exercises file for Rayleigh wave MASW analysis Lamporecchio MASW vertical sgy vertical geophones
12. Ji E Rayleigh v D velocity spectra phase amp group velocities l o t E refresh E a 3 S 3 picking 3 some tools select mode v a Time length to visualize sec za 10 15 20 25 30 35 40 45 50 frequency Hz report inversion www winmasw com geophysical software amp services page 50 winMASW Spectral analysis in particular for active seismic datasets A tool for traditional spectral analyses is also available calculation of an amplitude and phase spectra button Spectrum in the group filtering amp spectra Considerations must be done according to the selected component surface waves only refracted waves air waves etc For instance while analyzing surface waves only you ll notice the erosion of the high frequencies i e the disappearing attenuation of high frequencies at large offsets due to geometrical spreading and viscosity attenuation The soil especially if made of unconsolidated materials acts in fact as a filter thus attenuating especially the high frequencies see the chapter dedicated to the attenuation of the Rayleigh waves Here following the pictures relevant to the spectra of the first and last traces of the dataset test attenuation2 sgy 1 due to attenuation phenomena the amplitude of both traces is quite different the peak related to the first trace is over 600 while it decreases to about 100 for the last trace 2 compared to low fre
13. MFA group velocities analyses ReMi and ESAC analyses we highly recommend ESAC HVSR Horizontal to Vertical Spectral Ratio recommended jointly with dispersion analyses 1D P and SH waves refraction travel time modelling generation of synthetic seismograms via modal summation consequently inversion of the Full Velocity Spectra FVS with no need of picking mode interpretation pay attention since this anyway requires a good knowledge of several theoretical aspects and a state of the art PC Moreover Rayleigh wave attenuation analyses for estimating Qs quality factors and much more Automatic inversion is performed via Genetic Algorithms but we would like to underline that forward direct modelling is often preferred see Chapter 4 Genetic algorithms represent an optimization procedure belonging to the classification of heuristic algorithms or also global search methods or soft computing Compared to the traditional linear inversion methods based on gradient methods Jacobian matrix these inversion techniques grant a very reliable result in terms of precision and completeness In fact common linear methods give solutions that heavily depend on the starting method that the user has to supply According to the nature of the matter inversion of dispersion curves the large quantity of local minima draws the initial model to a local minimum that can really be different from the real one or global one In other word
14. shallow bedrock deeper than 18 654 meters deeper Humber of layers amp constrains option 1 option 2 number 5 of layers from reference model force search space 30 30 Inversion genetic menu individuals models min 6 max 7000 generations min 10 max 400 Vp amp density optimization close winMA S Vv ver 4 1 Pro b phase velocity m s Dispersion curve amp search space N N N a oco co N So ST O O O 4 6 8 10 frequency Hz 600 800 Vs m s 1000 dispersion curves demo cdp 12 1200 1400 14 1600 1800 Dispersion curve amp search space phase velocity winMASW re 4 6 8 10 frequency Hz 250 Vs m s dispersion curves demo cdp 12 page 67 Picture 6 Search spaces if a deeper or more near surface bedrock is supposed than the max penetration of the signal In the case of a surface bedrock the velocity of the last layer is defined at higher levels than those indicated in the approximation 2 2 or better 1 2 5 WWW Winmasw com geophysical software amp services page 68 winMASW Inversion parameters Parameter Meaning Advised value 4 6 Number of layers Number of layers used to rebuild the N
15. www winmasw com geophysical software amp services page 58 Joint modelling Rayleigh amp Love anisotropy Vsu Vsy The window where model parameters can be input also displays a column for anisotropy Vs This parameter determines supplied in percentage values how bigger is the Vs that rules the dispersion of the Love waves than the Vsy that determines the dispersion of the Rayleigh waves winMASW Professional amp Academy Versions Bg winMAsw modeling lie is layer 1 layer 2 layer 3 layer 4 layer 5 layer 6 half space Ysy m s 100 112 151 123 260 0 0 thickness m Vs anisotropy Poisson 2 14 Poisson v 1 2 2 Poisson 2 12 Poisson w 42 0 Poisson v 6 0 Poisson v 0 0 Poisson v 0 0 Poisson number of modes to visualize min 1 max 20 cancel calculate www winmasw com geophysical software amp services winMASW page 59 Resonance period see also Appendix D If the option HVSR is activated see general settings of the modelling group both the dispersion curves over a velocity spectrum and the chosen profile by the user with eigen period and frequency are displayed The value of the resonance frequency can be determined by means of two procedures 1 from the equation of resonance Where both thicknesses and velocities Vs of all layers over the bedrock defined as the layer characterized by a Vs 2 800
16. 7 2 NK TE automatically identifies the data format for a seg2 By the way there s always the possibility to force the reading according to different formats should the seismograph mistake on writing E winMASW Determinazione dello Spettro di Velocit El winMASW Determinazione dello Spettro di Velocit primo passo input dei dati primo passo input dei dati N tempo sec PO A Co MARRE LE LN SAT AN m JA Rok Ure SUTIS TT ERT I I tempo sec The velocity domain When calculating the velocity spectrum the data can also be represented in the domain t velocity just activate the option Tau v on the toolbar The parameter is the intercepted time on the time axis and the operation is made by linear Radon transform For very expert users the representation of data in this domain can be useful for more considerations even for teaching purposes For more details we refer back to future versions of winMASW 2 ooje File Edit View Insert Tools Desktop Window Help t velocity 100 150 200 250 300 350 400 450 500 550 600 v m s www winmasw com geophysical software amp services winMASW Possible effects of short datasets primo passo input dei dati input file MC3O6IFP SGY tempo sec ruota le tracce www sliosoft it winMASw primo passo input dei dati salva schermata about winMASyy velocit
17. cancel frequency Hz tn __ modeling emai parameters save model 2 win MAS Ww ver 4 1 Pro veiecty spectra model E Eigen period refresh Elena In the folder remi spectra you ll find a velocity spectrum outcome of a ReMi analysis To exercise upload the relevant dataset to the MASW acquisition in Rayleigh waves and proceed with the picking see as well the file Lamporecchio MASW Rayleigh cdp Upload then one of the velocity spectrum you got from the ReMi analysis Lamporecchio ReMi spectrum 1 mat Lamporecchio ReMi spectrum2 mat Lamporecchio ReMIi spectrum3 mat and lay over it the former saved curve Lamporecchio MASW Rayleigh cdp You ll see how now according to theory the curve touches on the lower edge of the spectrum resulting form the ReMi analysis and notice how you could add a lower frequency data down to about 4Hz 350 m s www winmasw com geophysical software amp services winMASW page 129 ReMi velocity spectrum Lamporecchio ReMi spectrumZ phase velocity m s 5 10 TN 20 26 frequency Hz explore spectrum www winmasw com geophysical software amp services page 130 winMASW Some references AA VV 2005 Procedure Manual Measurements Protocol amp Crew Manual Part A v 1 NATO SfP project 980857 Assessment of Seismic Site Amplification and Seismic Building Vulnerability in the Former Yugoslav Republic o
18. di fase m sec lolx secondo passo determinazione dello spettro di velocit calcolo spettro di velocit spettro di velocit 1600 1400 _ f es mi S mi mi Boc co Exp e 2b e No ce e 20 40 60 90 100 modellazione diretta parametri g refresh inversione picking seleziona modo selezionare l ultimo punto del modo utilizzando il tasto destro salva picking fo fine input file pilat amp tracce segy offset m ruota le tracce www aliosoft it winlVvLASVv salva schermata about winMAsvy velocit di fase m sec lolx secondo passo determinazione dello spettro di velocit calcolo spettro di velocit spettro di velocita 60 80 100 modellazione diretta parametri Bi refresh inversione 1800 1500 1400 ce N ma eo um IE co ce mm 600 400 200 20 40 F Hz picking seleziona modo selezionare l ultimo punto del modo utilizzando il tasto destro salva picking E page 29 Problems related to a short acquisition time in the picture over the dataset it is limited to the first 0 2 seconds while the data below has been zero padded zeros were added to the traces in order to better determine the spectrum In brief gathering the whole trend of surface waves i
19. picking file used to invert File with resulting medium model can be used to Text file visualize the dispersion curves over the velocity model mod spectrum observed in the section direct modelling in its relevant part Output file medium model achieved through inversion of the dispersion curves ready to be read and used by Deepsoil freeware software to estimate the site response to seismic stress see http www uiuc edu deepsoil see further on box The screenshot displaying at the end of the figure inversion observed and calculated data graph misfit generation and vertical profile of S waves velocity Pict 7 fig format comes from Matlab NEM in the stratigraphic column showing Vs mE png values density and thickness of the final model considered the medium model Picture resuming the model and indicating the site eigen period on the basis of the equation resonance period png see paragraph on the site eigen period In the Professional version the same H V spectral ratio is shown its peak is the eigen site period Deepsoil modelLINEAR dp Table 2 Output file automatically saved in the out folder default one is the winMASW output In this out folder there s the considered picking file too Vp density and elastic moduli After the 3 0 version the user can get the estimation of density Vp and of some elastic moduli Since the dispersion of the Rayleigh waves essentially depends on Vs and on the
20. v 2y y tv Vi P Young modulus in Pa pV 4 3k f x Lam modulus in Pa pV i m Compression modulus in Pa pvel P es 4 3 where k Vp Vs dimensionless p density Kg m Vs e Vp velocity of shear and compressional waves in m s Of course to change the Pascal value in Megapascal value just divide the value per 10 Mega 7 1 million www winmasw com geophysical software amp services winMASW page 145 Appendix D Vs30 and site resonance frequency Regulations usually require a Vs30 value from the foundation depth not from the surface From the main panel you can start a tool that allows to re calculate the Vs30 of whatever formerly determined model by modelling or inversion mod extension file starting from a given depth that needs to be more than zero but less than 30m m winMASW VS30 at the foundation depth co nr Upload model amp define depth of the foundation Upload the model mod Vs m s 166 184 516 671 Spessori m 2 3 3 6 2 Depth of the foundation meters 3 Results Vs m s 184 516 671 Thickness m 2 3 6 2 Vs30 at surface m s 431 Foundation depth 3m Vs30 at foundation depth m s 530 Vs30 should be provided from the foundation depth The current tool considers previously determined models mod files from modelling or inversion and calculate vs30 from a user defined depth 30 and 30m Togethe
21. Following picture clarifies the problem As you can see some portions of the traces result cut off because the signal has gone over the registration instrument dynamic range as an excessive gain was defined during acquisition 200 400 600 sample 800 1000 1200 n tu ru un e co wo channel Such problem has to be avoided all times you deal with analysis in the area of frequencies both for the analysis of the dispersion curves and the attenuation It is obvious as well that the gain value mustn t be excessive to avoid saturation and of the same value for all traces if we need to evaluate the amplitude decrease with the offset In winMASW you can verify the real traces amplitude clicking on test amplitude in the section dedicated to attenuation Once uploaded the dataset you wish to analyze click on test amplitude the following reported window opens showing the original data and the corrected one by means of the geometric correction www winmasw com geophysical software amp services page 138 winMASW winMASW 4 0 Pro Testing Data for Rayleigh Wave Attenuation Analysis see winMASW manual Original data no corrections Geometrically corrected Data 30 20 30 offset m offset m Frequency Hz 19 Frequency Hz 50 log amplitude log amplitude As you can see the geometric correction function of the offset slightly increases amplitudes In fact following an
22. X recorded data Vertical time min Use the left button of the mouse for the points that define the portions to remove and use the right button to select the very final point Please see manual amp video tutorial The user can thus select specific events and remove them The selection is done by clicking the start selection button Then the user will identify each event using the left button of the mouse and clicking the right button just for the very last point In the case reported in the next figure three events are selected To do that the user must click 5 points the first 5 points using the left button the final one almost at the end of the dataset using the right button If the user is then satisfy with the choice of the points it will be sufficient to click accept and get back to the main panel otherwise the user can cancel his her choice by clicking the reset button ROR Gr a A gt th GA ii ri iseg gx g ree jS recorded data time min Use the left button of the mouse for the points that define the portions to remove and use the right button to select the very final point Please see manual amp video tutorial ps this tool substitutes the automatic removal present in previous releases of winMASW WWW Winmasw com geophysical software amp services winMASW page 101 Main Panel First step upload the SAF file second icon on the tool bar Set all the
23. active and passive methodologies Chapter 3 Understanding Surface Wave Phenomenology Chapter 4 Horizontal to Vertical Spectral Ratio Chapter 5 Inversion and Joint Inversion concepts and practice Chapter 6 Full Velocity Spectrum inversion and other unconventional approaches Chapter 7 Some final notes References Appendix Case Studies A brief introduction Case study 1 A simple ZVF analysis for geotechnical purposes Case study 2 A simple but interesting dataset Case study 3 Inverse dispersion by the book Case study 4 When the joint analysis of Love amp Rayleigh waves is necessary Case study 5 Joint analysis of Rayleigh wave dispersion and P wave refraction Case study 6 A comprehensive survey in the Swiss Alps Case study 7 Joint analysis of Rayleigh and Love waves via FVS analysis Case study 8 A typical civil engineering job Case study 9 A landslide area Case study 10 Back to the Swiss Alps Case study 11 Modes and components a very tricky site Case study 12 Analyzing phase and group velocities and HVSR Case study 13 Some focus on HVSR computation Case study 14 Surface waves on the Moon How to get it store elsevier com A copy can be otherwise obtained together with winMASW Pro or Academy www winmasw com geophysical software amp services winMASW page 5 winMASW Academy 6 1 and successive releases works only on computers with 64bit
24. frequency Hz modify limits save selected spectrum analyze saved spectrum By analysing the entire dataset more than 10 minutes we get the following data n B winmasw passive seismics linear array estimated D estimated R estimated global 500 500 450 450 400 400 350 350 to e mx velocity m s velocity m s phase velocity m s 8 250 e 200 150 100 2 4 6 8 10 12 14 16 18 frequency Hz modify limits save selected spectrum In both cases the lowest frequency which is actually possible to safely consider for the ESAC dispersion curve is about 3 5Hz below this value the velocity decreases losing its meaning It is very likely that analyzing a non linear i e bidimensional array with longer radii such lowest frequency would decrease thus allowing the reconstruction of the Vs of deeper levels Pay attention these evidences cannot be generalized The aim of these notes is to help the users to get familiar with data analyses and phenomena www winmasw com geophysical software amp services page 46 winMASW Rayleigh and or Love These methodologies can be easily used for determining the dispersive properties of Rayleigh waves by considering the vertical motion which is not influenced by Love waves In order to analyze Love waves in the horizontal plane things become more complex see e g Tokimatsu 1995 Di Giulio et al
25. see further how and where to save the result www winmasw com geophysical software amp services page 86 winMASW oome general considerations during the modelling or inversion procedures the only variables are the Q values not the layers thickness These last are in fact fixed by the model mod we uploaded after we determined it through the analysis of the dispersion curves This makes the modelling of the dispersion curve easier than the analysis of the inversion curves for which we had to play both with the Vs and the thicknesses it is very important to highlight that the attenuation curves in order to be sufficiently detailed should refer to the fundamental mode of the Rayleigh waves if your dataset shows different modes at same frequencies you should avoid the analysis of the attenuation see box dedicated to the section verbose how to read the attenuation curves the rule is traditionally the same valid for the dispersion curves high frequencies refer to the near surface part while the low ones refer to the deepest part Geometric corrections When this option is selected some geometric correction takes place before the analysis of the attenuation For those analyzing data got at site this selection has to stay selected this can be avoided only in case of the elaboration of specific datasets already resulting from detailed academic studies Spectral Ratio SR and Matching The software proposes
26. www winmasw com geophysical software amp services Poisson 0 42 0 50 0 47 0 49 0 47 0 35 0 28 phase velocity m s Vp m s 404 1767 567 1421 631 8341625 O T mehr ma n e ee perverse 00 NN ON IC NIC 15 3 frequency Hz 20 modelling general setting Rayleigh v Reference H V modes SW ellipticity 0 3 H V body waves Poisson thickness m 0 42 2 0 499 0 47 0 49 0 47 0 351 0 279 02 0 15 0 0 15 Vs ms Qs 150 17 90 9 135 13 5 199 19 9 150 15 400 40 900 90 190 r synthetics upload mod zyr v about Poisson save model shows DC refresh show model phase vel just overlap v Refraction synthetics report DC report SS effective the considered mode click the right button save picking cancel picking inversion inversion Joint DC HV inv W WNW Winrnmaesw corn page 64 winMASW 4 2 Inversion of the dispersion curve The second and last step to obtaining the vertical profile of the Vs is the inversion of the formerly picked dispersion curve Remember by the way that we suggest the direct modelling procedure see Dal Moro 2008 Inversion is made by means of an optimization process genetic algorithms that requires the computer a big calculation effort The result is though more reliable and can give an es
27. 2006 Kohler et al 2007 We then suggest to all the newcomers to deal with Rayleigh waves all the registered winMASW users will receive details on the best procedures to analyze Love waves according to the ESAC approach Integrating MASW and ESAC dispersion curves and some recommendations ESAC is useful especially in the low frequency range active methods such as MASW or MFA can instead better perform in the high frequency range Anyway a crucial point to properly image the low frequency dispersion curve is the total length of the array ReMi and ESAC or any other array based passive method cannot be properly used if you consider short arrays As a consequence if your array is less than about 70m but possibly much longer passive techniques are quite useless and it is better to go for a joint MASW HVSR analysis Please remember that ESAC is superior to ReMi fundamentally because thanks to its bidimensional acquisition better handle the directionality of the signal do not forget than to set a bidimensioanl array ESAC analyses performed while considering linear arrays makes little or no sense at all The following snapshot shows a MASW velocity spectrum with overlaying the ESAC dispersion curve it is quite clear that the ESAC analysis allows to see frequencies lower than the ones possible while considering the MASW acquisitions On the other side MASW data see much better the high frequencies So they mutually integrate
28. 6Hz the max wave length will be 83 mt and consequently the max depth of penetration will be between 35and40 mt WWW Winmasw com geophysical software amp services page 74 winMASW velocity spectrum amp dispersion curve V profile picking best model mean model RE e e 300 200 phase velocity m s 5 10 15 20 25 frequency Hz depth m misfit evolution misfit fittest model average value fittest model mean model considered models T 10 20 30 40 50 60 70 D 200 400 600 800 1000 1200 generation Ve m s dataset demo sgy dispersion curve demo cdp www winmasw com Vs30 best model 259 mis Vs30 mean model 264 m s a dispersion curve s Vs profile T picking best model z m madal o 1 1 2 o gt o v Oo E a 4 6 8 10 12 14 16 18 20 22 24 frequency Hz E LL a misfit evolution 5 z t fittest model mean model considered models 10 20 30 40 50 60 0 200 400 600 800 1000 1200 generation Vs m s dispersion curve demo cdp WWwWWw WwInmMasw com Vs30 best model 271 m s b Vs30 mean model 267 m s Picture 7 Final screenshot automatically saved in the output folder If you get in the section inversion of the dispersion curves after the analysis of the dataset in the section determination of the velocity spectrum you ll get a picture similar to the one on top case a In this case you c
29. H V spectral ratio for improving Vs vertical profiling in the deeper layers Synthetic seismograms and Full Velocity Spectra FVS inversion no need of picking interpretation of dispersion curves Wink aS ww In order to receive our video tutorials and case studies please subscribe to our mailing list winmasw winmasw com www winmasw com geophysical software amp services winMASW page 9 Lite HVSR Standard 3C Ge con MASW analyses considering both Rayleigh and Love eee E waves ee their joint inversion LAM mu EN Vsa0calculation calculation a a p o pe o Analysis of Rayleigh wave attenuation to determine Qs quality factors Rem snayses passvesewmis O X X Group velocity Analyses Multiple Filter Analysis for X group velocity determination both for Rayleigh amp Love Computation and modelling of the H V spectral ratio X X X X Nakamura s method to estimate the resonance frequency Spectral analyses computation of amplitude and phase spectra and spectrograms frequency content over time X 1D modelling of refractions only for P X also considering low velocity layers waves 1D modelling of reflections also considering low velocity layers Tool to combine two shots and simulate a dataset with a double number of channels or three times etc see Appendix B Elastic moduli calculation tool Joint inversion of dispersion curve amp HVSR Synthetic seismograms both f
30. Near Surface Applications Dal Moro G 2014 Elsevier 252pp WWW Winmasw com geophysical software amp services page 110 winMASW 12 Dispersion curve amp HVSR Joint inversion Academy Please read the chapter devoted to the HVSR modelling Modelling HVSR for details about HVSR Due to several problems in order to properly perform this joint inversion it is highly recommended to read the following papers Dal Moro G 2011 Some Aspect about Surface Wave and HVSR Analyses a Short Overview and a Case Study BGTA 52 241 259 visit www winmasw com for a draft of it Albarello D and Lunedei E 2010 Alternative interpretations of horizontal to vertical spectral ratios of ambient vibrations new insights from theoretical modeling Bulletin of Earthquake Engineering 8 519 534 Dal Moro G 2010 nsights on Surface Wave Dispersion and HVSR Joint Analysis via Pareto Optimality J Appl Geophysics 72 29 140 Lunedei E Albarello D 2009 On the seismic noise wavefield in a weakly dissipative layered Earth Geophys J Int 177 1001 1014 The panel see picture down here is completely similar to that related to the joint analysis of Rayleigh and Love dispersion thus it does not need any special description EI winMASW Pro Joint Inversion of Surface Wave Dispersion Curves and HVSR Input picked dispersion curves Dispersion curve s and HVSR 1 Input dispersion curve Rayleigh Phase vel 2 Input H
31. Now the user can continue with the analysis it is not necessary to save the new dataset in a new file but it can be done using the save button in the data selection group www winmasw com geophysical software amp services winMASW page 25 El winMASW data upload processing amp modelling ae X TEke AB9 a 11 uploading amp processing MASW analyses dataset somma st1 st2 sgy sampling 0 26 ms minimum offset 1 5 m geophone spacing 1 5 m normalized traces offset m filtering amp spectra refraction refraction 100 clear refraction spectrum r Other tools amp setting 1 02362 Time length to visualize s flip traces zero padding 2 velocity spectrum modelling amp picking MASW amp ReMi analyses phase velocity 7 k group velocity MASW compute velocity spectrum handling the spectra El Velocity Spectrum Paramet Em x velocity spectrum define limits velocity m s 20 r frequency Hz 1 Vs m s Poisson thickness m 2 0 35 0 35 general setting 0 35 Rayleigh v 3 phasevel 0 Reference depth E Refraction HVSR 4 H V modes modelling synthetics r visualize curves picking cancel picking inversion Joint DC
32. Operating Systems OS Windows8 included and recommended Other versions of winMASW 6 x Lite Standard Professional 3C properly work on both 32 and 64bit operating systems included Windows8 Previous winMASW releases up to 5 2 work on any OS but windows8 For all the versions all the releases from the 7 0 work only on 64bit Operating Systems Definition of beta version A development status given to a program or application that contains most of the major features but is not yet complete Sometimes these versions are released only to a select group of people or to the general public The testers are usually expected to report any bugs they encounter or any changes they d like to see before the final release This is the second major stage of development following the alpha version and comes before the release candidate from wikipedia NOTE ON WORKSHOPS AND TRAINING Eliosoft is available for workshops and meetings aimed at explaining in details all the practical and theoretical aspects related to MASW Rayleigh Love MFA ESAC ReMi and HVSR techniques winMASW Academy is a highly sophisticate tool which is impossible to fully properly exploit without attending our training workshops www winmasw com geophysical software amp services page 6 winMASW Increase and Improve your equipment Eliosoft now also provides e4 5Hz vertical and horizontal geophones for MASW and ESAC acquisitions remember that
33. Palacios S Rosa Herranz J Garc a Fern ndez M Jimenez M J 2011 Soil characterization in urban areas of the Bajo Segura Basin Southeast Spain using H V F K and ESAC methods Journal of Applied Geophysics 75 2011 543 557 Roth M Holliger K 1999 Inversion of source generated noise in high resolution seismic data The Leading Edge 18 1402 1406 oafani J O Neill A Matsuoka T Sanada Y 2005 Applications of Love Wave Dispersion for Improved Shear wave Velocity Imaging Journal of Environmental and Engineering Geophysics 10 135 150 otesky R M 1978 Experimental compressional wave velocity measurements in compacting powders under high vacuum Applications to lunar crustal sounding Proc Lunar Sci Conf 9th 3637 3649 Tokimatsu K Tamura S Kojima H 1992 Effects of Multiple Modes on Rayleigh Wave Dispersion Characteristics Journal of Geotechnical Engineering ASCE 118 10 1529 1543 Tokimatsu K 1995 Geotechnical Site Characterization Using Surface Waves Proc First Int Conf on Earthquake Geotechnical Eng pp 1333 1368 Tonn R 1991 The Determination of the Seismic Quality Factor Q from VSP Data a Comparison of Different Computational Methods Geophysical Prospecting 39 1 27 Xia J Miller R D Park C B Tian G 2002 Determining Q of near surface materials from Rayleigh waves J Appl Geophysics 51 121 129 Xia J Miller R D amp Park C B 1999 Estimation of near surface shear
34. ReMi like acquisitions because in that case time and position thus distance of the source s is not known In seismology it is possible to apply MFA technique only once the epicentre is defined thus its time and position is determined with respect to the receiver www winmasw com geophysical software amp services winMASW page 49 The joint modelling is then performed in the usual way see example down here from the uploaded dataset it is computed the velocity spectrum representing the phase velocity MASW technique and the group velocities MFA technique Of course it is also possible to upload a H V curve and perform a triple analysis phase velocities group velocities HVSR phase velocity modelling general settings 0 Ref depth HV body waves select data Ysu mis 100 190 300 500 700 VS30 268 g 0 HV modes t Refractio e 7 Refraction e 5 2 E Vsv Ani S thk 8 m s m rai 100 o 0 33 4 picking 190 0 0 33 5 select mode v vv 300 o 033 6 5 10 15 20 2 30 35 40 45 50 8 VON L3 Ni offset m 700 0 033 0 0 0 0 35 0 0 0 0 35 dataset test attenuation2 sqy minimum offset 20 m Mer i geophone spacing m thickness mj 40 5 0 6 0 7 0 n 1 rode il sampling 2 ms Th Se hie ein Ad Modes group 1 T liti 1 8 1 9 2 1 24 NI TI m eng j i IM Modes phase 2 O U 3
35. are at hand it is advisable to choose a value of 0 25 for healthy rocks and of 0 35 in the other cases 0 45 in the case of loose surface sands Anyway when dealing with the energy distribution among different modes Poisson values i e Ve play major role see appropriate box in the Chapter devoted to Synthetic Seismograms Now it is time to launch the inversion click on run Calculation times are of course a function of the inversion parameters table 1 but that should take about 1 minute any case You can find some more about it in next paragraph advices www winmasw com geophysical software amp services winMASW Dispersion curve input file Bl winMASW 4 1 Pro Dispersion curve inversion Analysis Rayleigh Depth of the bedrock ls the bedrock deeper than 18 654 meters shallow deeper Humber of layers amp constrains option 5 v option 2 number of layers from reference model force search space Inversion 30 30 genetic menu individuals models min 6 max 7000 generations min 10 max 400 Vp amp density optimization Outputting output folder i BJ B winMASW 4 1 Pro Dispersion curve inversion Dispersion curve input file Analysis i v Is the Depth of the bedrock
36. clear Below you can see the graphic misfit generation that gives evidence of the improvement of the model according to advancing generations On the right the Vs profile according to velocity Since version 3 0 we can see the profile up to at least 30 mt The max depth of the signal penetration can anyway be lower such theoretic value is in the file winMASW report txt Finally bottom the average Vs up to 30 mt is given for both final models the best and the medium VS30 is the fundamental technical parameter as requested by regulations Quality of inversion 2 aspects can determine a good or bad quality 1 picking a logic dispersion curve 2 properly setting up the inversion numbers of layers Vs variability range and thicknesses number of models and generations The bad quality of an inversion can really depend on different factors and experience can be of great help Maximum penetration depth This value is the outcome of the relationship between velocity and the frequencies represented in the dispersion curve Always deeper layers influence as a matter of fact always lower frequencies wider wave lengths so that the lowest frequency will determine the higher depth of penetration This value is determined by the approximation 2 5 or 2 2 therefore it only is indicative Example If for your dispersion curve the value of the phase velocity Vg corresponding to the lower frequency is 500m s and the frequency is
37. dense dispersion curve won t give better results but a longer calculation time We generally suggest to concentrate on not more than 10 couples of points couple frequency velocity Basic structure of a picking file cdp FREQUENCY Hz VELOCITY m sec MODE WWW Winmasw com 15 9956 19 1886 17 9342 20 557 22 0395 25 1184 26 4868 28 3114 31 8465 geophysical software amp services 1135 77 815 43 929 45 761 134 733 986 685 12 668 832 641 684 992 818 E2025 soi i i i gt C25 page 53 page 54 winMASW Visualizing different curves picking or output curves Clicking on the button top right of the screen you will visualize and compare different formerly picked dispersion curves cdp and or dispersion output curves cdo They will show over the velocity spectrum in use secondo passo determinazione dello spettro di velocit Input input curva 2 a di velocit Li te fase m sec 23 Button MOVIE From the 2 0 version the data can be seen in their animated version button zi movie in the box 1 i UA uploading amp processing fea In case of a good dataset it is possible to appreciate the propagation of the surface waves front see the supplied dataset soma sgv ni m offset m campioni 5 10 15 20 www winmasw com geophysical software amp services winMASW page 55 Modelling of the dispersion curve Beneath the velocity spect
38. eee IE a i Ja Al a al F Hz _ F Hz _ F Hz C F Hz MCYS2723 dat 47 5 52 5s MCYS2723 dat 50 55s MCYS2723 dat 52 5 57 Ss MCYS2723 dat 55 60s phase velocity m s phase velocity m s phase velocity m s phase velocity m s a W Wa mre ATIS m o EN a LES E I E ao LES LOOPS Aa El F Hz i F Hz F Hz F Hz modify limits analyze saved spectrum The last three plots refer to the average spectra the first 2 related to the 2 possible direction so to say one from left the other from right the last one is the total mean spectrum www winmasw com geophysical software amp services page 34 winMASW If instead of choosing an average spectrum a single event spectrum is preferred when the user will eventually click on next window or save selected spectrum to save the spectrum the event i e the time window of the seismic dataset will also pop up For instance Lamporecchio ReMi sgy 31 5 33s D velocity spectrum gen B time sec phase velocity m s 5 10 15 20 25 30 35 40 offset m frequency Hz On the left the selected event please notice the passage of the surface waves and on the right the selected spectrum In case you choose to visualize the average spectra only just select the show average spectra only check box you will clearly obtain only the 3 average spectra Sf Lumix estimated D estimated R estimated global 500 400 phase velocity m s
39. for active MASW Rayleigh the radial component and Love waves can be recorded using only horizontal geophones e2Hz vertical geophones for ESAC and MAAM acquisitions MAAM is a kind of mini ESAC implemented in our HoliSurface software application see the article Unconventional Optimized Surface Wave Acquisition and Analysis Comparative Tests in a Perilagoon Area Dal Moro et al 2014 e3 component geophones 2Hz eigen frequency for HVSR and optimized joint acquisition of ESAC and MAAM data previous article ga ni HoliSurface 3 component HOLI3C geophone a passive geophone to connect to your seismopragh EP uc Se ME PAN i DAE EI P 4 a eri de oe D M i ua SL n Togli gt SE ee ee a v En Sa Vertical and manel cm and the metallic tripod for WOFRIRE on asphalt cover Your clips when you order your geophones please remember to indicate whether your seismic cable accepts Split Spring or Mueller type clips Split Spring E n aaa Miller or Mueller www winmasw com geophysical software amp services winMASW page 7 Main news in winMASW Academy 7 0 v winMASW now recognizes automatically your CPU Please activate the parallel computing as soon as you launch the software For performing computationally heavy analyses inversions it is important to use a good computer ideally a 12 core CPU which means 24 virtual cores Y Highly improved the Full Velocity Spectrum a FVS performances in te
40. input input file button a previously saved velocity spectrum r B winMAsw Dispersion curve or velocity spectrum inversion NE lt _ lt oo 2 5 i u or velocity spectrum Analysis Dispersion cu number of traces 2 Sampling 1 ms input file Minimum offset 5 m input n j geophone spacing 2 m Depth of the bedrock options2 5 10 15 20 number frequency Hz nce i Inversion genetic menu 30 individuals models min 6 max 7000 ations min 10 max 400 30 gener Vp amp density optimization Outputting output folder RUN WWW WINMesw com velocity spectrum full MONFAspectr rves or Velocity Spectrum amp search space After that similarly to the standard approach the inversion of picked dispersion curves user must set a geologically meaningful search space Some relevant points see also the synthetic seismograms Chapter 1 The method is based on the generation of synthetic seismograms via Modal Summation see Synthetic seismograms Chapter 2 Computational times are necessarily quite heavy It is then suggested a high performance PC e g a 8 core or more workstation It is also highly recommended to start option 2 from a model previously identified and saved via forward modelling Reduce individuals models and generations number of layers amp constrains group to 40 and 40 respective
41. m s are taken in consideration 42 from the spectral H V ratio resulting from the body waves Nakamura approach see Herak 2008 this last being available in the Pro version only Compared to the procedure 1 consider that if the given profile has a last layer with a Vs 2 800m s this will be taken as bedrock in the case this value is not reached then the bedrock i e the layer with Vs 2 800 m s can be determined by the behaviour of the most superficial layers Vs profile amp eigen period reference depth 0m Frequency 6 Hz Period 0 17 s bedrock estimated 100 200 300 400 500 600 700 300 900 Vs m s Case in which the final semi space has a Vs 800 m s the bedrock depth is in this case calculated according to the behaviour of the higher layers Vs in this case the last layer the final semi space as indicated by the user has a Vs of about 680 m s We strongly suggest to always manually give the bedrock depth both whether clear or supposed by the way in the region of a Vs 2 800 m s This will avoid poor meaning results www winmasw com geophysical software amp services page 60 winMASW Spectral ratio H V from body waves In the Professional and Academy versions it is possible to compute the H V spectral ratio and model it both according to body waves Herak 2008 and Surface Waves Lunedei amp Albarello 2009 Click on the third little icon from left on the tool bar to upload the dataset rele
42. o phase velocity m s w o o input curve 200 pcing select mode v 100 5 save cancel phase velocity spe observed velocity spectrum 20 25 Te 30 m ge oe ye wt frequency Hz misfit evolution a ee te et fittest model average value generation NE IM Www WInNMasw com www winmasw com geophysical software amp services group velocty spec Rayleigh component modelling paretomodel5 mod general settings 0 Ref HIV body wav Hv surface Refract 02 sv Anis thk mis QS s Peisson m 9 68630 049 014 g 0 0413 1 1 1 0 0364 1 67 184 18 0 048 1207 195 B16 0 0 28 07 48 0 0336 21 10 15 20 25 30 35 40 317 1320 033 171 frequency Hz 265 8 0231 60 paretomodel5 mod Love component 242 946 0 0456 417 36 38 0 0345 130 1245 904 0 0 112 Rayleigh modes 5 Love 4 phase velocty refresh calculate effect save model upload model synthetics ZVF vert geoph v show v just overlap F 15 20 25 30 35 40 _ show mo f H requency Hz synthetics report DC Www winmasw com report SS inversion Vs profile fittest model mean model considered models 200 400 Ve m s 600 200 1000 dataset ZVF1004spectrumHALF mat velocity spectrum ZYF geofluid mat Vs30 best model 237 mis Vs30 mean model 236 mis page 124 winMASW 15 Putting in evidence specific even hidden modes Academy
43. option 2 from reference model Once fixed other parameters defining the inversion process details see specific sections in the manual launch button RUN the process What we suggest is a mixed approach where the automatic inversion helps inside a modelling directly done by the user on the basis of given known stratigraphic data Due to the non uniqueness issue if you only rely on the automatic inversion you ll risk solutions that can be numerically fine limited misfit but not perfectly matching the local seismic and stratigraphic situation www winmasw com geophysical software amp services page 152 winMASW ReMi Analysis only Rayleigh waves from main window open the section ReMi spectra Once uploaded the data file at least 1 minute long and fixed the few parameters calculate the velocity spectrum launching the procedure with a click on spectra calculation try different lengths of the analysis window from a min of 4 seconds to the whole length of the dataset 3 save the best defined spectrum highlight it through the little button 4 enter the section Velocity spectrum Modelling amp Picking clicking on analyze saved spectrum 4 in this section just upload the just saved spectrum clicking on upload ReMi spectrum N The rest of the operation direct modelling and or inversion is absolutely similar to what before reported re the MASW analyses Remember that the inte
44. original filtered 1000 5 p 900 800 phase velocity m s phase velocity m s n e e save dataset tr nm Y n gt i Dennen a E tr UU i LLLI T ry vy MM LAA LL nmy 4 I L L I I l ININ LNA Umm we mermpr TEE LUI Urt A 111111 MA LALA LA vvv ALLL id EN I tit 5 10 15 20 25 30 35 40 45 50 5 10 15 20 25 30 35 40 45 50 offset m offset m If instead of putting in evidence the fundamental mode we decide to remove such a mode and putting in evidence the higher one s then original filtered 900 900 800 800 700 v 700 E gt 600 gt 600 8 8 2 2 o 500 500 E 9 o 400 c 400 EI Qa a 300 300 200 200 100 100 save dataset 0 1 0 1 0 2 E b i 0 2 gt S2 SZ T Ere ND 0 3 Press Jeet s tt 0 3 gt gt s 22T E 0 4 i gt Pre gt Cee SS 0 4 O gt r D o 0 5 1 3 PDIIEEEREERENEE v 0 5 E gt lt ea E 06 Pro Sse 06 23332 0 7 gt 0 7 0 8 gt i 313 fait 0 8 0 9 0 9 5 10 15 20 25 30 35 40 45 50 5 10 15 20 25 30 35 40 45 50 offset m offset m The ramp represents the value of the ramp of the filter i e how quickly we pass from 0 to 1 in the mask filter too low values might create oscillations in the filtered data in the x t domain while too high values risk not to remove data we actually want to remove It is also p
45. parameters and launch the computation by the compute button Once analyses are finished the 3 single average spectra and the mean HVSR are shown winMASW HVS CES X 3 fe wee MT 0110516 60530 SAF 128Hz Average Spectra Step 1 optional decimate 128Hz new frequency resample Step 2 H V computation remove events 40 window length s 10 tapering 1 z spectral smoothing triangular window E show particle motion raw data E 0 Step 3a optional directivity analysis 10 frequency Hz Step 3b optional directivity over time MT 0110516 60530 SAF HVSR window length 40s directivity in time time step 5 s Save option 1 save HVSR as it is Save H V from 0 5 to 64 Hz save HV curve as it is Save option 2 picking H V curve pick HV curve save picked HV Quick analysis f Vs 4H average Vs m s from surface to bedrock depth of the bedrock m comput frequency Hz WWWw WInNMdasSw com To model the HVSR also jointly with MASW or ReMi data save the HV curve go to the Velocity Spectrum a Modeling amp Picking panels and upload the saved HV curve A series of tool useful for exploring the data are available on the too bar zoom in amp out Data cursor etc You can use the snapshot button showing a photo camera to save the whole
46. s EE locsesoduease Hr save option 1 save HVSR as it is ueslonedi eode n TODD wu tot uorum sewed emcee onde I NO NE cones omm anne v ndo uncle nudis deubuoede oem e ouem bum noc SERENO liio d o ave M V from 045 to 64 H 4 F7 3 PM HMHMI Rem IiLLeee d e d di de bebee Ree 4 Lei edi Leiei e HnedP e ibEF save HV curve as it is oc O a save option 2 picking WV curve JI 24 ee a a at a do a nett ate y 1 0 do2 cbo 0 0o2doohodnpoo ENVIO I pick HV curve save picked MV G U 3 O E C 3 3 3 3 compute SESAME for picked curve 1 quick analysis f Vs 4M erage Vs ms i from surface to bed 0 LLILELE 40 pth of the bedroc 10 10 10 frequency Hz rock gt clean L compute To model the HVSR also jointly wih MASW or ReM ESAC data save the HV curve go to the Velocty Spectrum a Modeling amp Picking panels and upload the saved HV curve This overall approach is designed in order to allow the user to compute the SESAME criteria for each single peak imagine what otherwise can happen when more than a single peak is present See also the winMASW HVSR SESAME rar archive in the Documents folder in the winMASW installation directory It is also highly recommended to attend one of our workshops www winmasw com geophysical software amp services winMASW page 105 11 Modelling HVSR 3C Pro amp Acad
47. shot is internal to the geophone array minimum offset 2 5 By clicking on the lt icon in the toolbar you will get the following panel EJ winMASW removing traces 000000 original dataset offset m traces to keen In the traces to keep box the user must insert the traces to keep by adopting the following syntax first trace to keep step last trace to keep WWW Winmasw com geophysical software amp services page 24 winMASW In the following example we keep all the traces from 6 to 24 all of them since the step is 1 please notice that by writing for instance 6 2 14 we would keep the following traces 6 8 10 12 and 14 2 4 offset m offset m traces io beep B24 We first click on show to see whether the parameters we inserted are the right ones and once we are satisfied with the new dataset we just click accept This way we get back to the main panel 1 uploading amp processing MASW analyses dataset b dat sampling 0 125 ms minimum offset 0 m normalized traces 3393533 offset m resamping data selection fitering amp spectra refr amp refi activate fier cancel refriref 100 select 60 u save cancel save spectrogram clear other tools amp setting 0 511875 time length to visualize s done fiip traces test amplitude zero padding
48. wave velocity by inversion of Rayleigh waves Geophysics 64 691 700 Kohler A Ohrnberger M Scherbaum F Wathelet M Cornou C 2007 Assessing the reliability of the modified three component spatial autocorrelation technique Geophys J Int 2007 168 779 796 White R E 1992 The accuracy of estimating Q from seismic data Geophysics 57 1508 1511 Zhang S X amp Chan L S 2003 Possibile Effects of Misidentified Mode Number on Rayleigh Wave Inversion J Appl Geophysics 53 17 29 www winmasw com geophysical software amp services winMASW page 133 Appendix A Data acquisition for MASW ReMi ESAC and HVSR analyses Please see also Guidelines for MASW ReMi ESAC and HVSR acquisitions pdf enclosed in the winMASW installation folder subfolder documents The acquisition of data aimed to MASW analyses is not that different from similar acquisition aimed to refraction studies You just need to array geophones standard vertical component geophones if we need to analyze Rayleigh waves or traditional horizontal component geophones if we need to analyze Love waves lined up with the source see Figure A1 using a source with vertical hammer impact to generate Rayleigh waves or a transversal one for the Love ones Sorgente Y i LE ey Cy i Figure A1 Seismic array In the case you use a vertical impact source the traditional hammer blow and vertical component geophones you ll acquire useful da
49. while EX for Explosive These letters clearly refer to the kind of source About the receivers these can be vertical Z horizontal radial R or horizontal transversal T As a consequence synthetic seismograms can relate to different acquisitions ZVF Vertical Force e g vertical sledgehammer and vertical geophones for Rayleigh waves RVF Vertical Force e g vertical sledgehammer and radial geophones for Rayleigh waves ZEX Explosive source and vertical geophones for Rayleigh waves REX Explosive source and horizontal radial geophones for Rayleigh waves THF Horizontal Force shear wave source and horizontal transverse geophones for Love waves oee also our guidelines for a good data acquisition in the Documents folder within the winMASW installation folder Down here the same models reported in the Poisson and the Energy Distribution box different Poisson values but in this case for the radial component Please notice the different energy content for the different modes As this dataset was acquired using vertical geophones the correct comparison must be done with the synthetics datasets reported in the previous box related to the ZVF component see the very good agreement between the field dataset on the left on the Poisson and Energy distribution box and the central model characterized by high Poisson values Synthetic traces Rayleigh amm RVF component Synthetic traces Rayleigh an
50. while the other versions can operate on both 32 and 64 bit OS Suggested OS are anyway win7 or even better win8 64bit lf Academy version the user wants to deal with synthetic seismogram computation and related operation such as the Full Velocity Spectrum FVS inversion the heavy computational load requires up to date computers and we recommend a 6 core CPU at least better more www winmasw com geophysical software amp services page 14 winMASW 2 Installation winMASW set up is automatic just click on the appropriate installation file for details please read the README PDF file in the winMASW CD and follow the simple instructions PAY ATTENTION for all Operating Systems and especially for Windows Vista According to some operative systems Windows Vista in particular the management of privileges and writing in some files is quite restrictive It was noticed that especially with Vista installing the software outside the system folders C Program Files C Programmi e C Windows that meaning inside dedicated folders as for instance C winMASW o C geofisica winMASW can help sorting that matter out avoiding consequent problems Should you face any problem when launching winMASW this usually happens with Windows Vista first check the privilege details by clicking the right key of the mouse on the winMASW icon on the desktop It is necessary that the winMASW user has a writing privilege on the folde
51. 00 180 160 shallow deeper phase velocity m s 140 Humber of layers amp constrains 420 optionti option 2 4 10 number frequency Hz v of layers from reference model force search space 2 Inversion genetic menu 30 individuals models min 6 max 7000 2 30 generations min 10 max 400 2 Vp amp density optimization Outputting output folder 250 Vs m s ver 4 1 Pro Figura 5 Defnition of the search space The user can modify it according to own prior knowledge of the geological site see table A2 for typical Vs values of random materials Poisson s ratio As known the dispersion of the surface waves depends primarily on Vs and thickness of layers Since density and Vp play a second role on that try to play with different values of Poisson s ratios in the direct modelling their values only can be approximate ones We generally need to remember that a solid rock can have a Poisson s ratio between 0 15 and 0 30 0 25 is the standard value while soft materials like silt and clays value around 0 35 and 0 4 Sands get variable values from 0 3 up to the theoretic value of 0 5 the issue is the looser and inconsistent the material the higher the Poisson s value as said up to the max of 0 5 valid for fluids Asphalt can be ranged a value around 0 2 Generally speaking if no detailed information
52. 100 25 15 frequency Hz 20 www winmasw com geophysical software amp services mm Rayleigh Dispersion Vsv m s 151 146 129 323 966 Vs30 241 thickness m 2 9 3 2 6 9 6 0 Poisson 0 40 0 49 0 45 0 42 0 25 Vp m s 369 985 414 866 1683 25 15 frequency Hz 20 page 122 winMASW Summary plots upper left panel the observed velocity spectrum lower left panel the misfit evolution on the right the final retrieved models over the search space dispersion curve s V profile 500 g Rayleigh Dispersion Ysu m s 151 146 129 323 966 Vs30 241 400 poisson 0 42 0 48 0 45 0 41 0 25 N ce e phase velocity m s Q9 e ce e 5 10 15 20 25 frequency Hz misfit evolution 1 2 e 1 4 2 E 1 6 ea fittest model O fittest model mean model 2 average value considered models 5 10 15 20 25 30 35 40 45 0 200 400 600 800 1000 1200 generation Va m s dispersion curve fullldONFAspectr mat WWW WInNMasw com VS30 best model 241 m s VS30 mean model 250 m s www winmasw com geophysical software amp services winMASW page 123 Full Velocity Spectrum FVS analysis two examples Background colours report the velocity spectra of the field data while black contour lines overlaying over them are the velocity spectra of the synthetic data the identified model Please notice the very good match between experimental
53. 2 pas0 s ale so iy 3ndjno pue indui ino aJaUyMm J9pJ0J 9u uoissoes juaJuno INOA 104 Jap joj HulysOM Y 3es uonoe jsJlJ 9 IS q9M JNO WOd jenuew AAS Y UIM 2U peojumop aseajd s rejep J9Y INy 104 one E139adS e2119A 03 E3juOZIJO0H s eu1e jeuoisueuuipiq 104 sasAjeue y4 pue 3vy 3 s euue Jeau 103 IN9M S29IUISI9S BAISSEd 9Su324 ayy Agnjase gt peas woo seuuM E AH 9 071 uDrojKey jo UOISJOAUT zuio a Says jou Awapedy Q6 19A LLLA LIE OF MSAH uoisjedsig jo uorssanuj yuo avs orans 1003 A H J X 2vs3 i won al joued ulew MSYNUIM SQUJUIA 0 PLW UR pues oseuound INCA 1391 524 0 79pJ0 U 9 a uoiSJ8AU Burjopoui sisAjeue Buissoooud seuin jo e uonoe4je1 CL pue YSAH Ui Apuiof osje S ALM 9A07 pue ybIajAey jo sisAjeue juiof ay 104 aued 038 uoIsJaAUI Huljapow sisAjeue 6uissado4d seuin jo e1 uonoe4je1 GL pue YSAH YPM Agurof osje ejep 9407 10 uBio Ae3 juauodwo9 oj Buis jo sisAjeue y 104 Gued S9poo 991nos usdo FO 5 SWEIDOWSISS Seu AS TN uonenuasye snem ypiojAey X Aene seau A934 N EXTA Ap90 aA Mo Huliapisuod osje seuil 9AC1 uonoeJje1 Jo BurpepolN amp eure jeuoisueuiig 44 pue 3vS3 mi VAL c Jv ie a ED IM A Ce Y Aes Yum fe E bo Vice M fene SI Cum Sonn uL fin cmt M WP este dip ip A X CICER JV
54. 2D sections Academy In the Academy version a tool for creating 2D velocity section is also available 2D velocity contouring button from the main panel This way after careful analyses for several datasets for the vertical Vs reconstruction it is possible to construct a 2D section E winMASW 2D velocity sections CHA 72 e mm og WY YW AZ nina VA ee upload data input vel thick file file C Users GC Desktopwelthk tet 5 profiles depth B of the uploaded profiles 17 m Maximum depth for the 2D section 19 meters show profiles Vs profile in line position m 20 30 50 80 100 Format The data are organized in a standard ASCII file with txt as extension see as example the example velthk txt file in the IO 2Dprofiles folder Data pertinent to the single vertical profiles are organized according to the following format Vs1 Vs2 Vs3 Vs4 Thk1 thk2 thk3 Vs1 Vs2 Vs3 Thk1 thk2 Vs1 Vs2 Vs3 Vs4 Thk1 thk2 thk3 In this case there are 3 velocity profiles please notice that the second profile has a number of layer different with respect to the first and last profiles Example with 6 profiles file velthk txt 100 200 300 1000 DIN 110 210 288 999 3 3 2 3 6 6 90 199 999 43 123 188 288 1200 2 1 3 3 6 9 100 200 300 1000 3 34 57 www winmasw com geophysical software amp services winMASW page 147 Once the data file is uploaded within the depth group it will ap
55. 3 Synthetic Seismograms AcademMyY nennen 112 14 Full Velocity Spectra FVS Inversion AcadeMy 117 15 Putting in evidence specific even hidden modes Academy 124 16 Demo version and supplied data 126 Some Telel erbe S uos GE P RU De ER era PEU e URINE Bu EP ci i dub Mana up due PvE IMS 130 APPEN ID AAN Lccatamqadcdimaeicnquacmeen qv UPC COMUNI D RUN ED RID OI doIUODCPERONID BUD IDiS 133 Data acquisition for MASW ReMi ESAC and HVSR analyses 133 Appendix B Combine 2 dataset ciuccutcemirtutiian voi rbesetul Lect tbBrCas etof vous eb Dr bor vbUD eL er voee 139 Appendix C Calculation of the elastic MOdUlI 144 Appendix D Vs30 and site resonance frequency 145 Appendix E Creating 2D sections ACademy 146 Appendix F case studies processing etc i 148 Appendix G the Stesky s equation i 150 Appendix H analysis procedures in brief i 151 APPEND o e ba FF Fc 153 APPEendKLi BUGS 155 CON 156 www winmasw com geophysical software amp services page 12 winMASW Introduction The winMASW software allows to analyze seismic data in order to achieve the vertical profile Vs shear wave velocity using the following methods MASW analyses Rayleigh amp Love waves also jointly
56. 4 127 63 24 38 35 66 Zig Su Dyer E9933 8 35 178 92 8 89 72 04 93 37 22 62 21 34 36 89 56 93 77 67 121 35 33 92 171 11 14 00 63 63 33 38 50 66 10 11 15 93 126 33 43 35 159 92 43 11 127 04 0 37 95 62 37 34 01 11 118 07 9 33 43 35 157 92 4411 29 96 116 37 13 62 1 66 91 11 57 93 100 33 228 35 78 92 21411 146 04 139 63 2862 66 66 53 11 91 07 43 33 30 65 139 92 28 89 41 04 56 37 75 62 59 34 4 11 10 93 94 67 124 35 54 08 190 11 3 96 22 37 95 38 74 66 42 11 1 07 11 33 24 65 1 92 37 11 117 04 71 37 9262 40 34 6 Matlab format mat Actually very similar to the former one but written according to the Matlab protocol It deals with a simple matrix and the user has to input relevant data to the sampling interval the min offset and the geophone distance 7 SAC format big endian a common format for seismological data used for 1 channel dataset www winmasw com geophysical software amp services page 28 winMASW Numerical data format When uploading data it is very important the format be correct chose from a scroll menu like Floating point 32 bits Integer 16 bits etc If the format is not correct the visualization will be different from what you see in picture 1 for instance where the seismic trace correctly displays according to the so called wiggle modus Herewith two more examples of the rte evidently wrong reading format are X4 CES TI r provided d s Und AY Seu winMASW
57. 500 600 700 V s m s misfit vs generation Spectrum amp Dispersion Curves Attenuation curves test attenuation2 sgy phase velocity m s SR median Final Best Model best madel model mean 50 100 150 200 250 300 40 50 60 generation Frequency Hz Frequency Hz The number of dispersion curves modes to visualize over the velocity spectrum bottom right is fixed by the parameter Modes to visualize in the panel Inverting Q Qs Qp o Qp 2Qs During the modelling or inversion procedure you re invited to choose whether to adopt a model for which is valid either Qs Qp or Qp 2Qs Qs is the quality factor of the transverse waves S Qp is that of the compressional waves P Several authors disagree with the relationship between Qs e Qp The matter seems purely academic and an only irrevocable answer is impossible Just consider that apart very rare cases in the attenuation of Rayleigh waves the dominant factor is Qs while Qp has a minor role you ll understand if you play a bit with the direct modelling proving both options alternatively This also means that what really can be supplied in detail from the analysis of surface waves is Qs whereas Qp only can be estimated in general the value ranges between Qs and 2Qs www winmasw com geophysical software amp services winMASW page 95 Typical values of Q Quality factors In low quality soils very poorly competent Q values range between 3 a
58. 75 250 480 Vp m s 2Vs density 2 gr cm all layers Rayleigh phase velocity m s 20 40 3 0 20 40 frequency Hz frequency Hz www winmasw com geophysical software amp services winMASW page 81 Model thickness m 4 8 6 Vs m s 100 200 120 330 notice Vs inversion Vp m s 2Vs density 2 gr cm all layers Rayleigh s fh em i ho e So ho c c ho DI c phase velocity m s E a Oo 4 aa aa un m i EL c 40 2 20 40 frequency Hz frequency Hz The model file model mod in case of analysis of Love waves In the case you analyze the dispersion of Love waves no information about compression waves Vp will be available their dispersion only depends on Vs thickness and density In this case the final model file model mod shows an absolutely arbitrary Vp assuming Vp 2Vs corresponding to a Poisson s ratio of about 0 33 an average value typical of tough loose sediments Search space in case of analysis of Love waves In the case you wish to go on with the inversion of Love waves remember though that the direct modelling is the best choice to all surface waves analyses increase of about 30 the superior limit of deeper layers Vs as the software automatically suggests in the inversion section when the number of layers to be used is chosen 8 Joint Inversion of Rayleigh amp Love wave dispersion curves Pro amp Academy If at site both dat
59. According to the ratio between the dataset and the window dimension different windows could show like the one displayed on top The user has to select the best defined spectrum by means of the little button on the right side below of each spectrum The chosen spectrum in the screen will be displayed again on the top left side of following screen achieved clicking on next window or save selected spectrum www winmasw com geophysical software amp services winMASW page 33 When the last window finally is reached you ll save the selected spectrum clicking on that same button Once the clearer spectrum therefore more useful to our analysis is saved we ll get access to the module Velocity spectra Modelling Picking clicking on Analyze saved spectrum Here in section 2 velocity spectrum modelling amp picking MASW amp ReMi analyses we ll upload the just saved spectrum clicking on upload ReMi spectrum Bl winMASW ReMi spectra MCYS2723 dat 20 25s MCYS2723 dat 30 35s MCYS2723 dat 32 5 37 5s MCYS2723 dat 35 408 Velocity m s Velocity m s Velocity m s Velocity m s IST 25 a M a 25 gH a W a A PS a dn 5 2 25 53 F Hz F Hz F Hz F Hz MCYS2723 dat 37 5 42 5s MCYS2723 dat 40 453 MCYS2723 dat 42 5 47 5s MCYS2723 dat 45 50s m m S S 2 2 E E gt gt gt a a a a 2 2 2 2 o o o o gt gt gt gt SEP aed 3 UU 1a 5 mio
60. C datasets using the sequence mode and eventually obtain a datasets useful for MASW and MFA analyses WWW Winmasw com geophysical software amp services page 144 winMASW Appendix C Calculation of the elastic moduli From the main panel an utility to calculate a series of elastic modules from the values of Vs Vp and density Velocities have to be expressed in m s and density in gr cm while the calculated moduli Young modulus shear modulus compression modulus and Lame modulus have to be expressed in MPa MegaPascal Poisson modulus and ratio Vp Vs are dimensionless Remember the estimated Vp from MASW analysis mustn t be considered and largely depends on the Poisson value you set up before launching the inversion The only modulus we can consider as a good estimation of the real value is the shear modulus that doesn t depend on the Vp This utility is very useful because considering the obtained Vs by MASW analysis and the obtained Vp by refraction studies we can get an approximate estimation of the moduli FP winMASYY calcolo moduli elastici L E YS misec 600 Input VP m secy 1000 Densit gr cm3y 21 Moduli elastici Poisson 0 21875 Modulo di Young MPa 1843 Modulo di taglio MPa 756 Modulo di compressione MPa 1082 Modulo di Lam MPa 555 Rapporto WYPAS 1 66667 Formulas to express the moduli according to Vs Vp and density Poisson modulus dimensionless
61. G E E Li amp E Li LI LI Li e Li LI E Li E e E E LI Li B E E B EEEE ee SOS Sos S SS S SOS 9 us X X X WX Ou X a X T WoW ou wow T Tou oW Uu oW Ww ow WS X WS uS vu v v D OD v ovo 0 0 0 uv v LI v v v v LI v v vu vu vv 60 v LI LI v vu y es dv v v LI v v v v v v v v v v v 900070 5 N 5 5 S 5 S 5 S S 5 5 5 5 5 5 5 5 S 5 S S 5 5 5 5 5 5 5 x 5 s 5 5 5 S 5 s S 5 5 5 5 5 5 5 S 5 55555 b S S ooo oo 00 00 o o o o o o o o 0 o S o o o o o o o o o S o o o o S o o o oo O0 o o oooo ooo man Ah HO Oa uv a vw vw v n vw a o o v vw vw vw vw vw Iz v o vw I vw vw o i a LJ u o o u u uo HAAG VHA Ov o o o o O O O 0 00 O0 o o o i o o o o o ir o o eo o o o o o o d c o e o eo o c o o o o o o a o o0 O O Uupo NN NN O4 ON 4 44 NN O4 ON e N N N ex ex e e b e e N N N N ex e e e N N N N N ex ex e e e N N N Dal ex ex NNN NNN o 2 3 approximate depth 0 50 100 150 200 250 shot location m Lateral variations estimated along a 280m long line by means of our ADAM 2D system acquisition and processing www winmasw com geophysical software amp services winMASW page 159 Copyright October 2014 All Rights Reserved For news and updates please visit regularly our web site VAVAVAVAVAVAVAVA nn fi BI ni ee www winmasw com geophysical software amp services winMASW page 160 UIO9 MSEUWIUIMO MSEUWIUIM IEW
62. HV inv WWWw WwInmMasw com 1 uploading amp processing MASW analyses dataset somma st1 st2 5gy sampling 0 26 ms minimum offset 1 5 m geophone spacing 1 5 m normalized traces lectio filtering amp spectra 0 99Hz spectrum resampling refraction T clear refraction ion 60 save E cancel other tools amp setting 1 02362 Time length to visualize s 2 velocity spectrum modelling amp picking MASW amp ReMi analyses spectra handling the velocity spectrum MASW compute velocity spectrum phase velocity m s 3 40 50 frequency Hz 70 Vs m s Poisson thickness m 109 035 general setting 180 0 35 260 Rayleigh v 3 phasevel 9 35 400 0 35 synthetics ZVF v 0 Reference depth Refraction 0 35 0 35 HVSR 4 H V modes 0 35 modelling visualize curves input curve picking inversion WWWw WwInmesw com b Figure 1 Calculation of the velocity Spectrum The limits of the window need to be entered frequencies and velocities min and max www winmasw com geophysical software amp services page 26 winMASW f k spectrum In case the f k option is activated the f k spectrum will also be shown Amp
63. Journal International 179 254 264 Luo Y Xia J Xu Y amp Zeng C 2011 Analysis of group velocity dispersion of high frequency Rayleigh waves for near surface applications Journal of Applied Geophysics 74 157 165 Ohori M Nobata A Wakamatsu K 2002 A Comparison of ESAC and FK Methods of Estimating Phase Velocity Using Arbitrarily Shaped Microtremor Arrays Bulletin of the Seismological Society of America Vol 92 No 6 pp 2323 2332 August 2002 O Neill A Dentith M List R 2003 Full waveform P SV reflectivity inversion of surface waves for shallow engineering applications Exploration Geophysics 34 158 173 Park C B Miller R D amp Xia J 1999 Multichannel analysis of surface waves Geophysics 64 3 800 808 Park C B 2002 Multichannel analysis of surface waves MASW MASW Workshop Notes open file http www terrajp co jp MASW Workshop Tokyo pdf Park D amp Hashash Y M A 2004 Soil Damping Formulation in non linear Time Domain Site Response Analysis Journal of Earthquake Engineering 8 249 274 Rix G J Lai C G Spang A W Jr 2000 In Situ Measurement of Damping Ratio Using Surface Waves Journal of Geothecnical and Geoenvironemanteal Engineering 126 472 480 Robertsson J O A Pugin A Holliger K Green A G 1995 Effects of near surface waveguides on shallow seismic data 65th SEG Meeting Houston USA Expanded Abstracts 1329 1332 Rosa Cintas S Galiana Merino J J Molina
64. M IT Mm d 3 4 si a winmasw com p L E LU d 4 ELIOSOFT geophysical software amp services winMASW 7 0 beta User manual NET aa 21 21 31 CE d i dai psa Ba page 2 winMASW intentionally left blank www winmasw com geophysical software amp services winMASW page 3 Copyright 2014 All Rights Reserved www winmasw com geophysical software amp services page 4 winMASW Surface Wave Analysis for Near Surface Applications presents the foundational tools and techniques necessary to properly analyze surface wave propagation nowadays performed for a number of applications In the last decades geophy from the late of the different c llustrating a numb e pitfalls ima SISATVNV 3AVM 39VS4NS suggesting the way ome them via joint analyses Authored by a geophysicist with nearly 20 years of experience in research consulting and geophysical software development n o D lt m gt D un T gt O m gt U U En O gt o zZ U M SURFACE WAVE ANALYSIS A unique blend of theory and practice the book s concepts are basec exhaustive field so research conducted over the past decades FO R N E A R S U R FAC E ISBN 978 0 1 0128 mum Giancarlo Dal Moro 2 8 s gt E 007709 ELSEVIER Table of Contents A brief foreword Chapter 1 Surface Waves propagation dispersion and attenuation Chapter 2 Data Acquisition
65. Rayleigh Waves we consider a Vertical Force VF user can choose whether to consider the vertical component ZVF or the radial one RVF see Herrmann s terminology While analyzing Love wave we are dealing with a shear source perpendicular to the array THF Transversal Horizontal Force The computed synthetic dataset is automatically saved ion the working folder as SEGY file the name also reports the fact whether the file refers to Rayleigh or Love waves synthRayleigh sgy and or synthLove sgy While considering the visco elastic case thus the elastic check box is inactive for the modelling we will assume Qs Vs 5 and Qp 2Qs as simple rule of thumb For further information see also next chapter on velocity spectra inversion Velocity model Q model depth m depth m IE F r IF L L 0 500 1000 1500 2000 0 velocity m s Otherwise it is assumed that Qs Qp 10000 see following plots www winmasw com geophysical software amp services winMASW page 113 Q model 0 E 0 Eo E E TI Qs 2 2 Qp 4 4 6 6 8 8 E 10 E 10 Q Q S 12 12 14 14 16 16 18 18 20 20 L L L L L L 0 500 1000 1500 2000 0 9999 1 1 1 0001 1 0001 velocity m s Q x 10 WWW Winmasw com geophysical software a
66. Refraction Travel Times by means of Bi Objective Evolutionary Algorithm J Appl Geophysics 66 15 24 Dal Moro G amp Pipan M 2007 Joint Inversion of Surface Wave Dispersion Curves and Reflection Travel Times via Multi Objective Evolutionary Algorithms J Appl Geophysics 61 56 81 Dal Moro G amp Gabrielli P Presentazioni Power Point sull analisi delle onde superficiali e il software winMASW presentate al GEOFLUID2006 Piacenza 4 7 Ottobre 2006 CD di winMASW cartella documenti Dal Moro G Pipan M Forte E Gabrielli P Sugan M Forlin E amp Finetti 1 2005 Shear Wave Profiling via SH Reflection Analysis and Rayleigh Wave Inversion Riassunto Esteso per il convegno SEG Society of Exploration Geophysicists 2005 75 Annual Meeting Huston Texas November 2005 Dal Moro G Pipan M amp Gabrielli P 2007 Rayleigh Wave Dispersion Curve Inversion via Genetic Algorithms and Posterior Probability Density Evaluation J Appl Geophysics 61 39 55 Dal Moro G Forte E Pipan M amp Sugan M 2006 Velocity Spectra and Seismic Signal Identification for Surface Wave Analysis Near Surface Geophysics 4 243 251 Dal Moro G Pipan M Forte E amp Finetti l 2003 Determination of Rayleigh wave dispersion curves for near surface applications in unconsolidated sediments SEG Society of Exploration Geophysicists Proceedings 73 Annual Int Mtg Dallas Texas October 2003 Di Giulio G Cornou C Ohrnberger M
67. SAF file for HVSR analyses and a mat file to upload in the ESAC panel E winMASW HoliSurface o Tools Desktop Window Help Ele Edit View Insert T USMS 5 5 09wss umoud mau a uploaded normalized traces sampling 2ms 500Hz 458752 samples 100 1 1 a 3 3 4 zec2ena NL time s o ACAD M CO OA BUD 2 80 ste trite AGI ARTO DEA PRE fe RH Mia aan atis renee te n En lta oi M EL B LLL MELLE LUE db oll lo LULU DOs SARIEI M SoU AE SRN QUUD UEENB A MICE ree SU is HMMM eh ak A qe ene POS REED te ORIGEN PRE POUR RUE T QI PRA SO W AUA ODN TO 1 A IP O 1 O OA DR ID OR RO Wo EPA OE TA CR PIRA ERE DM DL I E Re iA Uii IDA Pe RI DAL H DR IO DI E I aera JI L3 O IE 12 Ll L Lua BP A 1 A DA O DD I lo reed ned D E AR Ro mf o 1 dba T mi Bub quin pil e AGO Hos iene io PA Ao ai deba SIMILI LI gl HST es tl 0 cc I As S re m ce N it channels www winmasw com geophysical software amp services page 154 winMASW Obtained data files on the left the data of the SAF file for the HVSR analysis on the right the 22 traces related to the vertical component geophones to use for the ESAC analysis Three traces for HYSR UD NS EW actual amplitude O E ese E Li ae m m EL i a A as sa ae e i E I LLLI MIU LLL LLL LLLI EZ i i 7 E d LE LN channals TCEMCD and HoliS
68. SW page 31 Exploring the spectrum The button explore spectrum allows to visualize the velocity spectrum in 3D on the left also having on the right approximate Vs values based on the Steady State Approximation The goodness of such approximation is related to the specific site in case you have strong VS variations especially if already for shallow depths the values given by such approximation are not reliable In other words the approximation is reliable only when the Vs gradient does not vary abruptly velocity spectrum GD view velocity spectrum explore n ce e IL e e LE ce e un e ev t2 ce e 450 ho n e 400 350 phase velocity m s nN ce e frequency Hz 14 0 300 phase velocity m s 143 0 250 depth m amp Ys m s 4 6 157 3 200 150 40 100 phase velocity m s n e ce e n e 30 40 frequency Hz frequency Hz WWW Winmasw com geophysical software amp services page 32 winMASW ReMi Analyses From the main panel by clicking on the ReMi spectra button you can access the section dedicated to the analysis of those spectra obtained by analysis of passive seismic data seg2 and segy format velocity spectrum limits frequency Hz 2 AVATVAVTATVIIBISBI AS OISR uem velocity m s input file s show average spectra only window length in seconds for spectra calculation geophone distance m minimum valu
69. VSR 10 Max allowed Vs anisotropy percentage minimum 0 maximum 50 Love cdp 250 200 150 phase velocity m s 10 Max Poisson variability percentage 8 10 12 14 16 minimum 0 maximum 30 frequency Hz Pur V noheaderlines short txt Number of layers amp constrains options option 2 from a starting model of layers E force search space HVSR modelling according to Body Waves frequency Hz Genetic menu number of individuals models min 6 max 7000 s number of generations min 10 max 400 2 Vp and density optimization Outputting output folder 600 800 1000 1200 1400 WWW VviInmesvv corn Vs m s www winmasw com geophysical software amp services winMASW page 111 oome points must be kept in mind very carefully especially about HVSR 1 If you are using body waves you should consider fundamental period resonance only thus remove from the HVSR curve all the data referred to higher frequencies in the above reported example we should remove the frequencies higher than about 4Hz 2 If you are using surface waves ellipticity you must considering 2 main points a large computational times b possible computational problems which can stop the inversion procedure see box Problems in HVSR computation using Surface Wave ellipticity microtremor exe To get deeper into the topic you can read the tw
70. Vs vertical profile is necessary see then paragraphs 4 1 and 4 2 Once this is done you can step on to analyzing the attenuation in the relevant section attenuation analysis Now you can calculate the attenuation curve i e the curve that for each frequency describes the amplitude drop of the signal according to the offset that is the distance of the signal relevant to the Rayleigh waves in the case of obtained data through the standard acquisition vertical geophones and hammer After that you need to upload the formerly obtained Vs model mod extension file Now it is possible to model or invert the observed attenuation curve obtaining the Q quality factors www winmasw com geophysical software amp services page 20 winMASW The software has the following folder structure Several documents manual examples etc In the sub folder videos you can store the video tutorial you can download from our website www winmasw com you may create serious problems some demo data are stored in this data disp folder data HV Some HVSR data are stored here Here some picked dispersion curves Output folder for attenuation IO analyse Output folder for dispersion Input Output output_HV Default output folder for HVSR data processing Here you might save data useful for 2Dprofiles 2D velocity contouring see Creating 2D sections Appendix PLEASE NOTICE that if you store your field datas
71. Wathelet M Rovelli A 2006 Deriving Wavefield Characteristics and Shear Velocity Profiles from Two Dimensional Small Aperture Arrays Analysis of Ambient Vibrations in a Small Size Alluvial Basin Colfiorito Italy Bulletin of the Seismological Society of America 96 pp 1915 1933 Fah D Kind F Giardini D 2001 A theoretical investigation of average H V ratios Geophys J Int 145 535 549 Gaherty J B 2004 A surface wave analysis of seismic anisotropy beneath eastern North America Geophys J Int 158 1053 1066 Gerstoft P amp Mecklenbrauker C F 1998 Ocean acoustic inversion with estimation of a posteriori probability distributions J Acoust Soc Am 104 808 819 Goldberg D E 1989 Genetic Algorithms in Search Optimization and Machine Learning Addison Wesley Publishing Company Inc 412 pp Herak M 2008 ModelHVSR A Matlab tool to model horizontal to vertical spectral ratio of ambient noise Computers amp Geosciences 34 1514 1526 Herrmann R B 2003 Computer Programs in Seismology Open files http www eas slu edu People RBHerrmann CPS330 html Lunedei E Albarello D 2009 On the seismic noise wavefield in a weakly dissipative layered Earth Geophys J Int 177 1001 1014 www winmasw com geophysical software amp services page 132 winMASW Luo Y Xia J Miller R D Xu Y Liu J Liu Q 2009 Rayleigh wave mode separation by high resolution linear Radon transform Geophysical
72. You should be able to tell the anti virus to ignore its considerations for the winMASW installation folder or you should use a different anti virus e g AVG In case of problems please send an email to winmasw winmasw com clearly describing the problem and indicating the User D UID and Serial Number SN of your USB dongle www winmasw com geophysical software amp services page 156 winMASW Contacts e mail winmasw winmasw com Twitter https twitter com Z winmasw Skype g_dal_ moro phone by appointment only www winmasw com geophysical software amp services winMASW page 157 REGISTRATION Do not forget to register your purchase please provide the name of the dealer seller the date of your purchase and the version Standard Professional or Academy You will receive software updates and advices on MASW ESAC ReMi and HVSR techniques Send an email to winmasw winmasw com If you miss to register in 2 weeks from the purchase no assistance and no free upgrade will be provided during the first year Processing of MASW and MFA Rayleigh amp Love ReMi ESAC MAAM and HVSR data We can analyse your data for you details about proper field procedures necessary to obtain good datasets are provided to all interested customers Vertical vs profila phase vc ocity m5 b sso2oud M a pre nolo a firenze aa SI i normalized traces
73. a are not suitable for attenuation analyses The following plots refer to the test attenuation 50Hz sgy dataset and show coherent trends but the user should consider that for frequencies lower than about 18Hz the dominant mode is the first higher one not the fundamental one That means that the frequency range to consider for good attenuation analysis is only 18 50Hz there is no signal for frequencies higher than 50Hz www winmasw com geophysical software amp services winMASW page 91 Frequency Hz 5 Frequency Hz 15 Frequency Hz 25 log amplitude log amplitude log amplitude 30 40 50 60 offset m offset m offset m Frequency Hz 35 Frequency Hz 45 Frequency Hz 55 6 re 6 6 5 Cy 4 E 4 e 3 e S m Ei 3 2 Ei Spessori m 2 0 2 0 40 8 0 1000 boisson 0 33 0 33 0 33 0 33 0 33 800 600 phase velocity m s 400 200 3 ZO EFEPEGEZLS frequency Hz Quite clearly in case of datasets like the one reported down here it is not possible to analyze the attenuation Frequency Hz 6 Frequency Hz 10 8 Frequency Hz 15 6 8 8 Cy y T 3 sat 3 3 E E E m e o Ei E Ej 30 40 50 60 i 30 40 50 60 offset m offset m Frequency Hz 25 2 Frequency Hz 30 S 219 S 3 Ei 7 82 8 78 30 40 50 60 offset m offset m www winmasw com geophysical software amp services page 92 winMASW To reca racticall Once uploaded the dataset top right visuali
74. according to body and surface waves HVSR Some notes on the ratio H V Regarding the nature of the ratio H V i e those events determining the value as observed in nature the academic debate is quite lively not only because of technical scientific matters but also because of party interests Following some of the most agreed upon general concepts The value of the H V ratio as observed depends on a complex relationship between surface waves both Rayleigh and Love and partly on body waves The relative importance of the different components depending on the site characteristic see for instance Bonnefoy Claudet et al 2008 Despite any chosen modelling don t ignore the role of quality factors Q in our case we assume Qs value to be the same as Vs divided per 10 and Qp 2Qs When adopting different modelling types i e based on only Rayleigh waves or on Rayleigh and Love or body waves or all of them together summed up the spectral ratio changes but the main peak determining the eigen period appears at the same frequency generally speaking From the only H V value it is impossible to get any information on the Vs that s why in this case you need to consider the MASW ReMI values and compare It isn t possible as well to directly compare the value of the H V peak and the value of the lithologic amplification factor to get this last we suggest to refer to the specific relevant regional rules or guidelines To recap i
75. ad flip polarity resample spectrum phase vel spectrogr Rayleigh v 3 clear just overlap accept 0 Reference v Refraction synthetics H V body waves other tools amp setting report DC flip traces test amplitude zero padding 0 report SS time to visualize done cut W WNW Winrmoesw corn 0 3 effective 2 velocity spectrum modelling amp picking MASW ESAC amp ReMi analyses visualize curves handling the spectra 1 data uploading amp processing MASW compute velocity spectrum Fk i explore spectrum pui cavo a dataset MASW classicalRay sgy resampled sampling 0 5ms WVRNG high frequency 4096 minimum offset 3 m geophone spacing 3m THEE i group velocity save upload phase velocity merge mode separation direct amp refraction travel times picking show f k Rayleigh Dispersion Vsu m s 150 90 135 199 150 400 900 Vs30 210 select mode v to select the last point of VISITS 30 40 50 offset m data selection filtering amp spectra filter cancel resampling 0 5 v activate refr refl 0 1 select 60 save upload flip polarity resample spectrum accept spectrogr sione refr amp refl other tools amp setting time to visualize done cut 03 flip traces test amplitude zero padding
76. al model Vs to estimate the Vs30 The result will be the model file mod from where you can start to model invert the attenuation curve Remember the analysis of the Rayleigh waves gives a good esteem of the Vs thickness and density but the Vp value can be estimated only in an approximate way on the basis of the Poisson s value chosen by the user and modified by optimization algorithm In the case of more details on the Vp for example from refraction studies or well data it is possible to modify the value of the model from the file mod that is a common file ASCII You can access the section about the analysis of the attenuation both from the main panel of winMASW and from the section determining the Spectrum Here after uploading the dataset you need to analyze and in case displaying the velocity spectrum to have a view on the present mode you ll analyze button calculate curve the attenuation curve in the proper frequency interval Then upload the previously determined soil model that represents the elastic part files mod Now we could model or invert the observed attenuation curve expression of the viscose part Since even in the case of attenuation same considerations are valid like those for the dispersion curves we always suggest to direct model instead to invert Once you identify a model whose attenuation curve is close enough to the observed one the analysis ends and can give the model of quality factors Q
77. alyses RVF component i eR 203800 ILLI Fer LL ET wisn l ie pe 0 Ut Jie B UU ES n 23331 E FEFE eetibiiite yy 0 ae eee Fereti LII INTESE Me I N rp gt INI TA Hmm B 0010 00 010 nWnid Hina LOLCCCUTE 293 i ASTAL Zh 0 8 E i i i 0 8 l I Uu i f m I j T jr a li t EM r Ii LI UN li nu t 5 10 15 20 25 30 35 40 45 50 55 5 10 15 20 25 30 35 40 45 50 55 offset m offset m Synthetic Dataset Velocity Spectrum Synthetic Dataset Velocity Spectrum Rayleigh Dispersion Vsv mis 152 150 127 335 971 Vs30 250 Vp m s 284 281 238 627 1586 phase velocity m s phase velocity m s 5 10 15 20 25 5 10 15 20 25 frequency Hz frequency Hz www winmasw com geophysical software amp services winMASW 14 Full Velocity Spectra FVS Inversion Academy page 117 This method is meant to invert velocity spectra and not as in the standard approach the picked dispersion curves see e g O Neill et al 2003 Dal Moro amp Moura 2013 That means that no picking must be performer by the user To invert a velocity spectrum it is necessary to
78. alysis aims to measure the value of the amplitude decrease linked to the only viscose component and not to the geometric factor linked to the distribution of energy over an always wider front Such component is called geometric In the bottom of this window you can see the amplitude decrease for both limit frequencies as specified on the main window Min amp Max frequencies fields In the reality what is plotted according to the offset is the amplitude logarithm This way the link between amplitude and offset in itself exponential becomes linear The larger data scattering as observed for low frequencies can be due to different reasons problems related to near field effects bigger for lower frequencies or to components linked to body waves data noise appearing of different modules of the surface wave etc www winmasw com geophysical software amp services winMASW page 139 Appendix B Combine 2 dataset From the main panel you can get access to an utility to sum two separately acquired dataset up The traditional case is that of two dataset each made of 12 channels many users don t dispose of 24 channels seismographs The aim is to obtain an only 24 channels dataset able to generate more detailed velocity spectra To do that make sure you executed both acquisitions correctly There are at least 3 way to acquire data useful for this operation see following picture a Mode alternate move just the source
79. alyze the saved spectrum tabu compute In the upper right corner user can set the parameters to adopt for the ESAC FK analyses In the lower area of the main panel on the left in addition to the resampling option which is absolutely recommended there are some buttons to manage visualize the data and the channel map on the right are reported the buttons to manage the obtained dispersion curves and pseudoi velocity spectrum Options verbose and f k analysis nearby the compute button to launch the computation by default only ESAC analyses are performed but the f k analysis option will activate FK analysis as well by activating the option verbose you will obtain a series of intermediate outputs useful to check the data quality etc see next pages www winmasw com geophysical software amp services winMASW page 39 Data acquisition record length how long should be a dataset to safely perform this kind of analysis There is no magic number for it since it depends on the site its stratigraphic conditions and the level of seismic noise we intend to use As a general rule 5 minutes can be sufficient but it is definitely better to acquire at least 10 minutes number of channels a reasonable number is 16 24 It is essential a good distribution covering different distances wavelength short wavelengths will see the shallow levels while long wavelengths will penetrate deeper geophones do not
80. amp services page 128 winMASW Data for ReMi analysis and joint inversion Rayleigh amp Love In the file data Lamporecchio you ll find 3 datasets that have been acquired during a workshop in Tuscany names are self explanatory Notice that the relevant data to Love waves have been acquired with 10 Hz geophones and are therefore poorer of low frequencies and generally a bit more spurious dirty The user can proceed with the analysis of data to practice considering though that the solution model is really similar to following icon Voy Voy Vp EIE EE m s 110 100 247 580 580 1100 Rayleigh component input Rayleigh data 800 dataset lamporecchio ASA ertical say ie m s 110 131 172 580 VS30 247 minimum offset 10 m 700 thickness m 2 2 3 0 8 7 geophone spacing 1 m Poisson 0 38 0 41 0 35 0 31 sampling 0 5 ms 600 p mis 247 343 355 1100 select data select 20 cancel save phase velocity m s picking select mode Y 5 10 15 20 25 30 35 40 dataset lamporecchio ASVY orizontal say ida nie MEZ seca minim alfa ins 700 Ys anisotropy 13 17 18 0 geophone spacing 1 m sampling 0 5 ms 600 Rho griem 1 7 1 8 1 9 2 1 select data D AA Bu gt time s D phase velocity m s O ceo picking cl select mode 5 10 15 20 25 30 35 40 save
81. an associate a curve to the relevant spectrum If you instead get in the section Inversion of the dispersion curves directly from the main winMASW screenshot uploading a formerly picked curve that won t be possible and you ll get a picture like the one below case b there is no velocity spectrum on the background of the dispersion curve The solution on top is to prefer if not necessary a www winmasw com geophysical software amp services winMASW page 75 e oe 6 H i TO Z O ua O O 0 500 1000 1500 2000 2500 velocita onde S m sec Figure 8 Example of final result mean and best models in green the adopted search space As an example in Figure 9 you can see two examples of inversion obtained on a synthetic 6 layers model Left the result of the inversion when using 6 layers right the result when using 10 www winmasw com geophysical software amp services page 76 winMASW This file in html format to be readable with any internet browser too files all useful information to Results winMASW html evaluate results and classify soils according to the E Italian regulations The full complete report on operations and results winMASW report txt detail information are also present of scarce interest for the final user Dispersion curve of the model medium got by curve mode fundamental cdo inversion The first line reports the name of the and or possibile higher modes
82. and synthetic data 1 input Rayleigh d oe de PP 4 0 1 ber dd Pe gt gt Mme NN SN EN gt he Pe Z5 0 2 gt PX gt phbtowo gt oe P yr i Pb gt i oo PD Pret bb 0 3 dn ae P gt gt gt Pe ky EU ad 4 7 04 gt gt EE Pye 4 b p gt E gt gt b 505 Sy 11411747 gt ROS Py 0 6 12 9 saan Se 0 7 rat yD 0 8 0 9 B 0 10 20 30 40 50 60 70 offset m input Love data gt 0 1 gt gt r o gt 221584 218 02 Bob gt ebp ehhh N I 0 3 i Fi pri TL s 04 etd ae Php PP e Irr gt P gt E Dh Zc gt S05 Addi 0 6 Dopey Ppp Fay ph 08 i DD 09 4 0 10 20 30 40 50 60 70 offset m some tools time length to visuaize done cut decimate misfit dataset ZVFspOFF4 mat sampling 2ms 500Hz 451 samples minimum offset 4 m geophone spacing 3 m fip data cu spectrogram 600 select data activate I 500 8 select 400 cancel save Di a mode separation save spectrum 300 upload spectrum 200 input curve picking auo p selec v 100 5 save cancel 800 fiter dataset THFspOFF4 mat sampling 2ms 500Hz 451 samples minimum offset 4 m geophone spacing 3 m fip data cut spectrogram select data activate select save cancel mode separation save spectrum upload spectrum A n eo o e e e eo o o
83. anel obtained while considering a small a 0 2 and a high a 0 9 amount of Love waves As can be seen the obtained HVSR curves are significantly different since Love waves moves only on the horizontal plane the effect is an overall increase in the H V curve www winmasw com geophysical software amp services winMASW page 109 Two consequences are straightforward 1 The amount of Love waves synthetically expressed by the a factor should be considered as a further variable in the inversion process aimed at determining the Vs profile experience teaches that its value typically ranges from 0 3 to 0 5 2 The HVSR curve alone is insufficient to properly and precisely define a Vs profile even when geological stratigraphical information are available The only viable approach is represented by the joint inversion with further geophysical data typically the dispersion curves of Rayleigh or Love waves a Vs profile 0 T T T T T T E 10 E 20 CI D O 30r 4 40 L 1 1 ji L 1 100 200 300 400 500 600 700 800 900 Vs m s b 6 Alfa 0 9 mmm Alfa 0 2 or 4 c o a 3 z 2 L UD mo 4 1r ia 0 at 10 10 Frequency Hz Figure A Effects of Love waves on the HVSR a considered Vs profile b the HVSR curves obtained while considering a different amount of Love waves the a factor in the microtremor field Further details in Surface Wave Analysis for
84. aset have been registered using therefore the vertical geophones with the shared vertical source to obtain data re the Rayleigh waves and the horizontal geophones with shear wave source to obtain data re the Love waves it is now possible to jointly invert www winmasw com geophysical software amp services page 82 winMASW them at the same time in order to have a more solid model and to highlight possible wrong interpretations through the evaluation of the relationship between both components misfits The module can be uploaded from the main window Rayleigh amp Love Joint Inversion Upload the dispersion curves both Rayleigh and Love already picked in it as previously done with the single dispersion curves and proceed as usual Together with the output we already are confident with among the most relevant we ll also get if activating the option bi objective space a relevant graph to the values of the 2 considered misfit both Rayleigh and Love If the picking was done correctly like the inversion the result in the figure should be quite linear converging to the point 0 0 If instead one of the curves if not both was misinterpreted picked or if the number of layers and the search space is not adequate the result will be quite irregular the more irregular the more mistakes you made Naturally this is also clear when you can t identify a model giving dispersion curves sufficiently fitting both Rayleigh or Love com
85. atio EET WC EET MERECE ETE m c is m a Li o A E ra m P ua ua m d e Fal uni SURFACE WAVE ANALYSIS FOR NEAR SURFACE APPLICATIONS extracted from Gz The HVSR determined from field data is the result of the combined effect of both Rayleigh and Love waves basically according to the following equation aH f air Hn f HSVR f VA Being Hg and Vg the Rayleigh wave contributions in terms of power spectra see Arai and Tokimatsu 2004 on the horizontal H and vertical V axes and H the component related to Love waves the a parameter can be considered as the amount of Love waves in the background microtremor field While some authors forget to include the effect of Love waves and treat the H V spectral ratio as simply representative of the Rayleigh wave ellipticity others properly include their effect In the latter case anyway a problem arises about the proper value of the a parameter which is likely related to the specific overall regime that characterizes the region and the meteorological climatic conditions Incidentally it can be speculated that the differences in the H V spectral ratios observed in different days seasons see Paragraph 4 3 1 of the above mentioned book could be the result of a different amount of Love waves in the background microtremor field An elementary modelling will clarify the point Figure A reports a Vs model upper panel and the two HVSR curves lower p
86. ce 0 5 24 Maxium considered trace 300 Number of generations p i 6 66 Min amp max frequencies 2 Modesto visualize Qp Qs zn er 1 f geometric corrections calculate curve verbose Inversion 1 5 Attenuation curves test attenuation2 mod Qs Qp Activate 0 0 2 0 22 Q 12 D 4 2 Cancel P 0 2 0 18 3 213 D 3 8 Velocity Spectrum l 0 16 0 14 E me 02 a Fani S 0 1 ax Q 40 D 1 3 Im E 0 08 gt 0 06 3 50 D 1 0 Match 0 04 HM SR median 0 02 Model 30 30 10 20 30 40 50 60 10 20 30 40 50 60 200 400 600 800 0 20 40 60 Frequency Hz Frequency Hz Vs m s Qs www winmasw com geophysical software amp services winMASW page 93 Summary plot Vs model test attenuation2 mod vs Poisson Q model Qs Qp 3218 D 2 6 40 D 1 396 50 D 1 0 300 400 500 V s m s Velocity Spectrum Attenuation curves Dispersione Rayleigh i Vs m s 110 180 320 500 700 VS30 277 Spessori m 40 5 0 6 0 7 0 Poisson 0 28 0 36 0 30 0 33 0 33 Vp m s 200 380 600 1000 1400 e c c I ce phase velocity m s SR median Match Model 30 40 frequency Hz frequency Hz Example of screenshot picture resulting when clicking on summary plot after the identification modelling of the Q model that best approximate our attenuation curves The number of dispersion curves modes to visualize ov
87. ce the signal between 15 and 25Hz phase velocity between 100 and 200 m s m which is due to aliasing phenomena and consequently has no relevance for our purposes We can also notice that as remarked by several authors e g Rosa Cintas et al 2011 at the lowest frequencies FK method tends to identify higher velocities with respect to ESAC The following plot reports the kx ky diagrams for some frequencies FK method eee ee top J di FE RAO A aon mL TE Ha The following plots report the results obtained by reducing to 15 the number of channels of the above presented data originally composed of 23 traces used traces distances between pairs of receivers a d ad 2 J channel map 4 time min station distances m x removed channel s i i 14 16 0 10 20 30 40 50 60 0 2 4 6 8 10 12 station distances m channels www winmasw com geophysical software amp services winMASW page 41 winMASW ESAC and FK analyses for passive acquisitios 2D q ac PE lt gt channels to remove 101314 111291516 channel map 25 x m e uploaded traces x removed trace s x m 48 44 40 36 32 28 24 20 16 12 8 4 000 00000 0 0 0 y m 000000000000 3 6 9 12 15 18 21 24 27 30 33 show update channel map 7 resample to 6ms 166 666Hz show data show radius distribution d geometry _upload geometry first dataset muscov remi 1 DAT save geome
88. d that not needed to be put in relation with the propagation of the surface waves would lead to a worse quality of the analysis This can happen if the distance between the first geophone and the source is too little and a serie of events linked to the near field happen We d suggest not to use offsets lower than 10m Therefore when the first traces amplitude are much higher than the general trend due to the presence of direct waves don t consider those traces for the analysis Therefore the following will show total dataset selected dataset amplitude spectrum of the traces and velocity spectrum in the frequency wished range Dataset actual amplitude Selected Traces Maximum considered trace Amplitude Spectra all traces Velocity Spectrum considered portion Frequency Hz eo e ue e Phase Velocity m s 20 5 10 15 Trace Frequency Hz 10 15 20 25 30 35 40 45 50 Following picture reports 6 plots for 6 different frequencies from the min to the max These graphics show how the amplitude logarithm changes according to the offset for a given frequency value These trends help understand in which range we can have good quality data and what frequencies are instead to avoid because for instance at those frequencies the result is more modes or different modes or no signal even Obviously what we aim to demonstrate is the fall of the amplitude according to the offset A different trend means that the dat
89. d on Rayleigh amp Love joint inversion you ll get access to the section aimed to the joint inversion of Rayleigh and Love dispersion curves winMASW 4 1 Pro Joint Inversion of Rayleigh amp Love wave Dispersion Curves Input picked dispersion curves Input Rayleigh wave dispersion curve Input Love wave dispersion curve Max alowed s anisotropy in percentage 3 minimum 0 maximum 50 Humber of layers amp constrains 2 r option 1 number of layers option 2 from a starting model 5 Genetic menu number of individuals models 40 min 6 max 7000 amp number of generations 40 min 10 max 400 E Vp and density optimization Outputting output folder RUN close ver 4 1 Pro winMA SW Dispersion curves 600 re phase velocity m s e ce e phase velocity m s 20 20 25 30 frequency Hz 25 30 frequency Hz 35 35 Rayleigh wave dispersion curve Model 1 Rayleigh cdp 40 Love wave dispersion curve Model 1 Love cdp 40 45 45 50 50 depth m 100 200 300 400 500 600 Vs m s Since the close analogy between the two of them this chapter is going to show the only procedures involved in the standard analysis of one component focusing on some features only
90. e s 1 sampling rate 5 maximum value 5 show data E Resample to amp ms ESAC spectra calculation Clicking on input file s you ll select the relevant file to ReMi acquisition passive seismic see relevant appendix You can simultaneously upload more files typically 65 seconds long by keeping the key CTRL pushed while selecting the wished files to consider These need to have all same acquisition parameters geophone distance sampling interval and length In order not to overload the computer memory we suggest not to upload more than 5 dataset at once for instance 65 seconds long each such value depending anyway on the computer features therefore even longer Once the geophone distance as used in the acquisition phase and the length of the window to consider are set the same sampling rate interval can be modified but the read and proposed value is likely to be correct every time and you can modify it only if the seismograph really has some problems or you are confident you re doing the right thing consider also that the length can be changed as many times as you need to reach better spectra you just need to click on spectra calculation At the end of all operations a window like the one below will show The software splits the dataset or datasets in many windows of equal length like the one indicated by the user the value must rank between 2 seconds and the length of each single dataset
91. e case the geometric correction is off no geometric corrections a window with three graphics will show the amplitude offset one refers to the data not corrected by geometric component winMASW40 Pro Testing Data for Rayleigh Wave Attenuation Analysis see winMASW manual CLE X Original data no corrections Frequency Hz 5 Frequency Hz 65 log amplitude log amplitude 40 50 60 offset m Frequency Hz 48 30 log amplitude log amplitude offset m Example of dataset obtained with different gains for each channel trace and therefore unuseful to analyze the attenuation of the seismic data notice how amplitude doesn t fall according with the offset To clarify the meaning of these graphics refer to the box about the acquisition of data for the analysis of the attenuation In Appendix A WWW Winmasw com geophysical software amp services page 90 winMASW Button verbose When calculating the attenuation curves button calculate curve if you click on verbose you ll get three windows The first window shows the dataset with highlighted the first and the last considered traces for the analysis of the attenuation The reason for this option is such traces need to be defined in the main window of both Reference trace and Maximum considered trace fields that for some acquisitions the effects due to proximity of the source presence of direct waves can be verifie
92. e curves explore spectrum 70 modelling Vs ms Poisson thickness m about Poisson 100 0 35 1 calculate r general setting 180 0 35 2 upload mod 0 35 5 Rayleigh 3 phase vel 260 save model 400 0 35 7 refresh f F i 0 Reference depth Refraction T 2m 3 0 0 35 0 E HVSR 4 HAV modes eport 0 0 35 synthetics ZVF z v elastic shows DC www winmasw com geophysical software amp services picking 2 show t cancel picking inversion Joint DC HV inv Www winmasw com page 142 winMASW Obtained Velocity spectrum considering the only dataset 2 far offsets 1 uploading amp processing MASW analyses 2 velocity spectrum modelling amp picking MASW amp ReMi analyses visualize curves dataset stendimento dat MASW compute velocity spectrum handling the spectra input curve 2 sampling 0 26 ms group velocity save upload merge explore spectrum minimum offset 19 5 m geophone spacing 1 5 m picking normalized traces velocity spectrum 1600 show t select mode v to select the last point of the considered mode click the right button save picking 1400 1200 1000 600 cancel picking phase velocity m s 600 inversion inve
93. ed The origin of the Cartesian system to adopt is absolutely irrelevant so your 0 0 point can be anywhere since these methodologies assume the planewave condition it is important that the main sources are distant enough to meet such condition consequently do not use these methods for analysing datasets acquired considering a nearby active source WWW Winmasw com geophysical software amp services page 38 winMASW Main panel of the ESAC FK module In the upper left area it is possible to insert the x y coordinates which can be actually written saved in a simple ASCII file see next pages for the format In a further line channels to remove it is possible to indicate traces that we want to exclude from the analysis for instance because too noisy winMASW ESAC and FK analyses for p c X velocity spectrum _ ____ FK parameters x m 0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 upload geometry dataset ReMi 6 x3 DAT E min freq 2 max freq 25 512 wavenumbers y m 000000000000 11 19 2027 000 3 0 0 0 0 Save geometry 983mPling 6 ms min vel 100 max vel 1000 15 window length s channels 79 to remove mde channel map ESAC parameters 2 spectral smoothing hold on 15 window lenath s 30 40 x m p uploaded traces x removed trace s H H verbose V resample to 6ms 166 666Hz show data show radius distribution clear save spectrum an
94. ember 2012 Lisbon Portugal Dal Moro G amp Marques Moura R M 2014 Multi component Joint Analysis of Surface Waves submitted to J Appl Geophysics www winmasw com geophysical software amp services winMASW page 131 Dal Moro G 2012 Onde di superficie in Geofisica applicata Acquisizione e analisi di dati secondo tecniche MASW e HVSR Dario Flaccovio Editore 192 pagg www darioflaccovio it in talian Dal Moro G amp Ferigo F 2011 Joint Inversion of Rayleigh and Love Wave Dispersion Curves for Near Surface Studies Criteria and Improvements J Appl Geophysics 75 573 589 Dal Moro G 2011 Some Aspects about Surface Wave and HVSR Analyses a Short Overview and a Case Study BGTA Bollettino di Geofisica Teorica e Applicata invited paper 52 241 259 Dal Moro G 2010 Insights on Surface Wave Dispersion and HVSR Joint Analysis via Pareto Optimality J Appl Geophysics 72 29 140 Dal Moro G 2008 Tre divagazioni il mito dell inversione MASW in Friuli esempi di applicazione congiunta MASW rifrazione Presentazione ad invito per la Giornata di studio Monitoraggio Dinamico per la Valutazione della Sicurezza Sismica dei Ponti Centro Internazionale di Scienze Meccaniche CISM Udine 18 aprile 2007 Scaricabile dal link http studiosisma com uploaded DOC 20SISMICA MASW 20 20ReMi Dal Moro pdf in Italian Dal Moro G 2008 Vs and Vp Vertical Profiling via Joint Inversion of Rayleigh Waves and
95. emy In the Pro amp Academy versions winMASW allows to model the HVSR both according to body waves Herak 2008 and surface wave elliptity Lunedei amp Albarello 2009 About this last possibility winMASW works as a graphical interface for the executable by Albarello amp Lunedei 2009 The executable can by downloaded from the following link http nato gfz hr SW zip Once downloaded save microtremor exe in the winMASW Application folder If you activate the HVSR modelling and set a number of modes higher then zero you will get the modelled HVSR both according to body waves and surface waves in case the number of modes is set to zero you will get HVSR only according to body waves In order to approach this method please read at least the following papers Dal Moro G 2011 Some Aspect about Surface Wave and HVSR Analyses a Short Overview and a Case Study BGTA 52 241 259 visit www winmasw com for a draft of it Albarello D and Lunedei E 2010 Alternative interpretations of horizontal to vertical spectral ratios of ambient vibrations new insights from theoretical modelling Bulletin of Earthquake Engineering 8 519 534 Dal Moro G 2010 nsights on Surface Wave Dispersion and HVSR Joint Analysis via Pareto Optimality J Appl Geophysics 72 29 140 Lunedei E and Albarello D 2009 On the seismic noise wavefield in a weakly dissipative layered Earth Geophys J Int 177 1001 1014 By studying the
96. ended Abstract for the IAEG International Association for Engineering Geology and the Environment XII CONGRESS Turin September 15 19 2014 In Engineering Geology for Society and Territory Volume 5 Springer International Publishing 1177 1182 Dal Moro G 2014 Joint Inversion of Rayleigh Wave Dispersion and HVSR of Lunar Seismic Data from the Apollo 14 and 16 sites submitted to ICARUS Dal Moro G amp Keller L 2013 Unambiguous determination of the Vs profile via joint analysis of multi component active and passive seismic data EAGE Near Surface 2013 Proceedings 19th European Meeting of Environmental and Engineering Geophysics EAGE Near Surface 2013 Bochum Germany 9 11 September 2013 having received a very high score from the reviewers during its presentation the work was invited to be published as full paper in Geophysical Prospecting Dal Moro G 2013 Joint Analysis of Lunar Surface Waves the Apollo 16 dataset Proceedings 19th European Meeting of Environmental and Engineering Geophysics EAGE Near Surface 2013 Bochum Germany 9 11 September 2013 having received a very high score from the reviewers during its presentation the work was invited to be published as full paper in Geophysical Prospecting Moura R M Noronha F Almeida F amp Dal Moro G 2012 Vs measurements through dispersive wave methods in the urban environment of Porto North Portugal 15th World Conference on Earthquake Engineering 24 28 Sept
97. eo co 0 9 a111 dat 2i 0 St 1 y n 0 1 enna a 0 2 aaa 9 04 ZIE 05 0 3 0 8 0 4 3 D 5 10 15 20 Sale offset m E dataset 2 us Fler ruota tracce 0 5 20 3 10 15 offset m winMA SW offset m www winmasw com geophysical software amp services winMASW page 141 To point out the importance of the use of 24 channels datasets we ve reported windows relevant to velocity spectra obtained when considering the spectra as separated and finally windows referring to the final total 24 channels dataset outcome of the 2 initial 12 traces each datasets Obtained Velocity spectrum considering the only dataset 1 near offset 1 uploading amp processing MASW analyses dataset stendimento1 dat sampling 0 26 ms minimum offset 1 5 m geophone spacing 1 5 m normalized traces offset m resampling data selection filtering amp spectra refraction 100 Gea e AME Ccancei save r other tools amp setting 0 8 Time length to visualize s 2 velocity spectrum modelling amp picking MASW amp ReMi analyses MASW compute velocity spectrum ty EAk group velociy handling the spectra remp velocity spectrum 1500 1400 1200 1000 900 phase velocity m s 600 400 200 10 20 30 40 50 60 frequency Hz visualiz
98. epth of the bedrock shallow deeper N ce e phase velocity m s ce e ce e Number of layers amp constrains option 1 option 2 2 X IL force search space 2 preliminary vertical V profile and search space Inversion genetic menu 10 individuals models min 6 max 7000 34 generations min 10 max 400 J Vp amp density optimization depth m n Outputting I 1000 1200 WwWww wiNnmasw com Inversion is finally launch by clicking the RUN button as usual user can set the output folder www winmasw com geophysical software amp services winMASW page 121 Results At the end of the inversion procedure the following plots will be shown they refer to the best model B Velocity model depth m 1000 velocity m s 0 500 1500 est identified model velocities and quality facotrs Q depth m 2000 Q model seismic traces and velocity spectrum observed data on the left best identified model on the right Uploaded Dataset Rayleigh analyses 0 0 SIT d GG 0 2 s n 0 2 04 T zx 0 4 S eL 7 06 Mis 7 06 0 8 i 0 8 5 15 20 25 30 35 40 45 50 55 offset rn Uploaded Dataset Velocity Spectrum 500 500 450 450 400 400 4 Vp m s 369 985 414 866 1683 o 350 350 gt 3 300 8 300 oa Pu 3 250 gt 250 5 200 200 150 150 100
99. er the velocity spectrum bottom left is fixed by the parameter Modes to visualize in the panel Inverting Q b Inversion Both parameters Number of models and Number of generations have the same meaning of those met in the inversion of dispersion curves so there s no need to recap Since modelling modalities the suggested numbers are respectively 100 and 300 in seconds calculation times Before launching the inversion you need to choose a search space fix it after some direct modelling and bearing in mind the traditional values of Q for different materials from a min of 3 for highly dissipative soils to over 100 for healthy rocks In order to be sure an inversion gives satisfactory results you can compare the final misfit value with the attenuation curve medium one a value about 10 is typical of a good result Example if the final misfit value is 0 004 while the medium value of the attenuation curve is around 0 006 then the final misfit is about the 7 of the medium value that meaning the model is acceptable www winmasw com geophysical software amp services page 94 winMASW Final screenshot of the attenuation curve inversion automatically saved as png file in the folder winMASW outputattn but can be saved wherever and in other graphic format Vs model test attenuation2 mod Final Q model Qs Qp 9211 D 4 5 Q 12 D24 296 winMASW es Q 66 D 0 8 30 100 200 300 400
100. ervices page 150 winMASW Appendix G the Stesky s equation Very often mixed materials such as the gravels can have very different Vs values approximately ranging from 300 to 500 m s depending on the relative amount of cobbles and fine matrix usually made of sand or clay see Figure 1 16 In this respect it can be interesting to report the equation describing the resulting effective velocity when two different materials are mixed up Stesky 1978 1 s me ai v9 9 eq 1 2 where Vm 1s the velocity of the mixed media V4 and Vg the velocities of the A and B materials the volume fraction of the material A in the mixed media www winmasw com geophysical software amp services winMASW page 151 Appendix H analysis procedures in brief Here following you ll find the sequence of the procedures to follow re the different available types of analysis Refer to each relevant section to deepen needed details Independently on the kind of analysis you intend to perform the very first action to do is the setting of the working folder from the main winMASW panel button working folder This is a folder previously created by the user where all the field datasets were stored copied and where the outputs of the performed analyses will be stored MASW Analysis in Rayleigh or Love waves 1 from the main window open the section Velocity spectrum Modelling amp Picking 2 upload the site datum first b
101. es winMASW page 27 Seismic data format winMASW can currently read the following formats 1 segy format segy or sgy This is a standard format suggested by the Society of Exploration Geophysicists SEG 2 seg2 dat or sg2 This is a well known common gathering format Many of the current available seismometers use this data writing format typically in the integer 32 bits or Floating Point 32 bits formats From the 3 0 version the software can automatically recognize the format but in order to correct possible mistakes of the seismometer the user is allowed to force the format 3 SU format seismic Unix Traditional format although not frequently used 4 ORG format org traditionally used by many OYO seismographs integer 32 bits traditional format 5 ASCII format txt or asc formats Some devices and software use the ASCII format can be read with all text editors where the columns stand for the different traces in other words this is a m x n matrix with m being the number of data and n the number of traces Please notice that the first line reports dt sampling rate in seconds dx geophone distance in meters mo minimum offset in meters Example dt 1millisecond minimum offset 6m geophone distance 3m 0 001 3 6 2 89 40 07 125 33 27 35 76 92 29 89 26 04 55 37 117 62 66 34 92 11 86 07 16 67 135 35 62 92 93 11 119 04 3863 124 38 57 66 90 11 93 93 130 33 161 35 9492 170 11 111 0
102. est to use 7 8 layers You can use more under specific complex stratigraphic situations when low velocity channels are present velocity inversion It isn t advisable to use less than 4 even if in front of a very simple situation as for instance when a debris alluvial layer lays over rock because 1 the rock however healthy has an altered surface with lower geo mechanical qualities then the deeper part inside 2 in such situation also a modest debris stratification of the surface can be important when it comes to analyze the dispersion of the surface waves To end up we cant forget to focus on the fact that in those situations limited scarce materials over others with a really higher Vs guided waves of wide amplitude can take place that can be mistaken for surface waves since their dispersive features Be suspicious of any unexpectedly high propagation velocity a value about the double you d expect that needs to be related to guided waves whose propagation depends on the Ve but not on Vs see Robertsson et al 1995 Roth amp Holliger 1999 Since the 4 0 version you can use a former tested and saved mod model whose curves are in accordance with the velocity spectrum you observed as starting model to get the inversion of the dispersion curve If you use this method option 2 for the inversion that model can become the starting point for the optimization and solutions in a Vs interval and thicknesses modifiable by the use
103. ets in a certain folder anyfolder e g Desktop Berlin the output folder will be automatically created within your working folder that you can set from the main or any other panel of winMASW From any panel it is also possible to launch if properly installed Google Earth icon r T _ _ e ou File Edit View Tools Add Help w Search RyTo Find Businesses I Directions Fly to e g Reservor Rd Clayvile NY v Places E vis My Places d NO r Verifica che il livello Edifici 3D sia selezionato Temporary Places Google Eye alt 2801 16 km We suggest to download our video tutorials available since January 2011 in the winMASW Documents videos folder then accessible from any panel icon Ex Manuals are also accessible from any panel icons 9 9 for the English amp Italian version See video tutorials from www winmasw com publications area WWW Winmasw com geophysical software amp services winMASW page 21 4 1 Velocity Spectra picking modelling dispersion and refraction MASW ESAC ReMi ESAC amp MFA analyses phase and group velocities Important notice just model do not invert Techniques based on the analysis of surface waves MASW ReMi and MFA methods are more and more common but some aspects related to their generation and propagation impose to proceed with caution as it means determination of vel
104. f Hz Picture 11 Analysis of the demo sgy supplied with winMASW Fortunately not all cases are that difficult A further supplied dataset is somma stist2 sgy intergeophonic distance and minimum offset both equal to 1 5m the segy files only manages whole values and the user has to input the right data when uploading This dataset is the result of the two datasets stendimento1 dat and stendimento2 dat and it actually is the easiest solution both files stendimento1 dat and stendimento2 dat have the format integer16 bits Note the wide amplitude of the frequency band from about 8 to 60Hz Should a velocity inversion take place the dataset somma st st2 sgy has to be modelled inverted using at least 5 layers www winmasw com geophysical software amp services winMASW page 127 The dataset test attenuation 50Hz sgy is supplied in order to gain confidence with the analysis of the attenuation The min acquisition offset is 10m the automatic value is instead 5 note how the geometric corrections and the attenuation curves change when you differently input the right value or leave the wrong one The right Vs model that you get from the analysis of the dispersion curve and that you need to upload before proceeding with the inversion modelling of the attenuation curve is saved in the file test attenuation 5strata ok mod folder output As seen in the chapter about the analysis of the attenuation for this dataset the max usable f
105. f Macedonia Croatia and Slovenia http nato gfz hr SfP 20 20980857 20manual 20part 20a 20v1 pdf last access Oct 2012 AA VV 2005 SESAME Guidelines for the Implementation of the H V Spectral Ratio Technique on Ambient Vibrations Measurements Processing and Interpretation Open file http sesame fp5 obs ujf grenoble fr Papers HV User Guidelines pdf 62 pp last access Oct 2012 Albarello D amp Lunedei E 2010 Alternative interpretations of horizontal to vertical spectral ratios of ambient vibrations new insights from theoretical modeling Bulletin of Earthquake Engineering 8 519 534 Bonnefoy Claudet S Kohler A Cornou C Wathelet M Bard P Y 2008 Effects of Love Waves on Microtremor H V Ratio Bulletin of the Seismological Society of America 98 288 300 Dal Moro G Ponta R Mauro R 2014 Unconventional Optimized Surface Wave Acquisition and Analysis a Geotechnical Application in a Perilagoon Area Submitted to J Appl Geophysics MEZIVANE VERNE RE Surface Wave Analysis for Near Surface Applications Dal Moro G 2014 Elsevier 252pp ISBN 978 0 12 800770 9 SURFACE WAVE ANALYSIS FOR NEAR SURFACE APPLICATIONS m O D z m gt a en A m gt n m gt v v E n gt r4 un Dal Moro G Coviello V Del Carlo G 2014 Shear Wave Velocity Reconstruction via Unconventional Joint Analysis of Seismic Data Two Case Studies in the light of Some Theoretical Aspects Ext
106. f in possession of an experimental H V value and needing to find a model considering both the dispersion of the surface waves and the H V observed curve we suggest to focus on chasing the right peak frequency without trying to find the perfect balance with the amplitude the H V value Note as well picture on top that the H V values allow to identify possible strong contrasts of Vs deep underground remember though that to constrain the Vs you need MASW ReMi values that can give clear details about the Vs and that it is most important to reproduce the correct frequencies when the peak shows In the case on top for instance a strong discontinuity at about 60 cm depth is clear it represents an alpine valley covered in a large quantity of fluvio glacial sediments WWW Winmasw com geophysical software amp services page 62 winMASW Modelling refraction Pro amp Academy amp reflection Academy This section allows the user to model the arrival times of refraction events by means of very similar procedures to those used in the modelling of dispersion curves section Before starting consider that The adopted calculation tool allows the insertion of channels with inversion of velocity a thing that the traditional refraction method utilized as only method can t evaluate this refers to the known problem of low velocity channel In our case making the most of all information from the dispersion curves and refraction we can eval
107. fessional The Routine to the calculation of the spectrum ratio H V from body waves as available in the 3C Professional and Academy version is bestowed to ELIOSOFT under license of Prof Herak University of Zagreb Croatia further technical details are reported in the bibliography www winmasw com geophysical software amp services page 18 winMASW 4 Use procedure basics When winMASW is launched a second DOS window more than the main one background displays containing all information about analysis and operational proceedings B winMASW Surface Wave Analyses amp Beyond Lor en U 99 a8 ST Internet sites vw r_ parallel computing Parallel Computing Off please activate 4 activate passive seismics H V tools combine 2 datasets compute HIV En mpute H V concatenate data SEG to SAF conversion data resampling elastic moduli Vs30 at foundation convert accelerograms computing synthetics vertical stack joint analysis of Rayleigh amp Love HVSR joint inversion of dispersion HVSR combine 3C data 20 olocittcnatoian Rayleigh amp Love Joint Inversion 3D block velocity In order to register your purchase please send an email to winmasw 2winmas w com read carefully the license Academy ver 5 1beta The very first thing to do is the activation of the Parallel Computing see upper right part of the main wind
108. file to Love wave MASW analysis Lamporecchio MASW horizontal sgy file to ReMi analysis Lamporecchio ReMi sgy Following two snapshot The first is relevant to MASW analyses with picked curve due to the spectrum type we ve picked up to about 6Hz as at lower frequencies a trend reveals an intervention of the first higher mode The second below is relevant to ReMi analyses laying over the picked dispersion curve on MASW data You ll notice that compared to the ReMi spectrum this is now laying on the inferior limit of the signal and not on the peak as it happens with MASW data On this spectrum another point at about 4Hz frequency is now added and phase velocity of about 320m s normalized traces velocity spectrum ae e 4 co e ce uu ce ce e e O e time s phase velocity m s M ara peN ce ce iex e 5 10 15 20 25 30 35 offset m frequency Hz MASW data traces velocity spectrum and picked curve Please consider that considering passive acquisitions we recommend ESAC rather than ReMi thanks to its bidimensional acquisition array ESAC is in fact capable of handling the directionality of the signal and its mathematics has in general better performances compared to linear ReMi WWW Winmasw com geophysical software amp services page 36 winMASW ReMi velocity spectrum Lamporecchio ReMi spectrum2 zx exi em e LI e ce e 400 phase veloci
109. ful to the ReMi analysis Aim of the ReMi analysis is the registration of events which we don t know the place of origin and the moment when they are likely to happen Therefore we switch on the seismograph and through the 24 geophones we register what happens in the hope a useful signal reaches us This means that the registrations need to be long Many seismographs are limited though We consequently advise to make different one by one registrations in order not to weigh too much on the software as well in order to get at least 5 minutes registrations the length of each dataset not being less than 30 seconds If for instance our seismograph allows to take a data for a minute we ll make 5 acquisitions of 1 minute each Thanks to winMASW we then will be able to upload and analyse all data and choose the most defined velocity spectrum Geophone spacing m Array must be as large as possible Common vertical component geophones basically apart from some academic application in passive Type of geophones seismic you can only analyze the dispersion of Rayleigh waves Eigen period 4 5 Hz or less Registration time s 10 20 minutes we suggest to acquire different files each 1 minute long for a total of about 10 to 20 minutes Number of channels traces minimum 16 geophones dt sampling interval 0 002 s 2 ms 2 milliseconds No AGC Automatic Gain Control Recommendations No filter Table A3 Summary data regarding the
110. gher modes Uploaded Dataset Rayleigh analyses ynthetic traces Rayleigh analyses ZVF component Synthetic traces Rayleigh TET ZVF component I s 0 EL M lt r pl x n Id GF J va OO Oeo Eo FE E 4 T pi 5 4 alg A J i to i s l S ub T T a mei 0 2 H i EH HET EFFI 25 30 l 25 offset m offset m offset m J Uploaded Dataset Velocity Spectrum Synthetic Dataset Velocity Spectrum Synthetic Dataset Velocity Spectrum Pol 10 I ue m s 411 719 435 844 1683 a nis 411 719 435 844 1683 Vp m s 284 281 238 627 1586 phase velocity m s phase velocity m s phase velocity m s 15 20 15 15 frequency Hz frequency Hz frequency Hz New in winMASW 6 x Now if while computing synthetic seismograms the just overlap option is selected what you get is a plot where the contour lines of velocity spectrum of the synthetic seismograms is actually overlapped with the velocity spectrum of the field dataset the background colors Down here an example from the joint phase and group velocity analysis panel please notice how the blue contour lines of the synthetic model reproduce quite well the actual field data colored velocity
111. h new parameters by clicking again on phase velocity for MASW analyses or group velocity for MFA analyses Since the input data here also will be considered for the spectrum as shown in the results window see Picture 7a we suggest to use such limits that can give a clear visualization without need to zoom in Button flip traces Dominio x t niil Wrong lay out An erroneous disposition of the channels can give following common shot gather arrival times decrease as the offset increases Click on the button rotate traces _rustaletracce to obtain below lay out Right lay out Right layout of traces offset increasing rightwards Tempo sec 40 Offset m WWW Winmasw com geophysical software amp services winMASW page 23 Tool for removing traces from the uploaded dataset Upload a dataset 1 uploading amp processing MAS analyses dataset b dat sampling 0 125 ms minimum offset 2 5 m geophone spacing 0 5 m normalized traces TERI 01 TRATTA DIS 0 15 ALLL NUES 02 ELLE 0 3 0 4 0 45 i 0 5 offset in data selection Timing spectra activate mern cancel raian 100 fms Come 8 ey ose on e Ga am eese ocio other took amp setting 0 511875 time length to visualize s done fip traces test amplitude zero padding In this example the
112. have to be equally spaced The following plot report an example of a possible and easy to manage geometry L shaped used traces distances between pairs of receheens Hi di T T T T 1 m 1 i d 2 channel map i al 7 a z x i i e E ii E i E H m if H LU n E 10 i Bi 20 4 9 99 9 0 9 9 9 0 0 in 40 an i From winMASW 5 0 Academy some procedures ESAC and FK are among these exploit parallel computation so using a multi core computer a significant reduction of computational times is obtained Very roughly the reduction is proportional to the number of cores if you have a 4 core computer your computational times will be reduced to about 1 4 of the times necessary to a single core machine oince these techniques are particularly useful for retrieving the dispersive properties at the lowest frequencies it is important to underline that the equipment should be properly set good seismograph and low frequency geophones minimum 4 5Hz In the following an example of dataset acquired adopting a L shaped array characterized by a limited length maximum radius distance between geophones only 58 25m dnce EL nad Prices lr pie sogar CO eophore ay e M US i i i i a LU IG Asus raw He mea maim mpacinum Wis T kid RIT um WWW Winmasw com geophysical software amp services page 40 winMASW Please noti
113. ies data selection refraction O flip traces activate quick refraction exit b win MA S VV ver 4 1 Pro Attenuation analysis Picture 3 Direct modelling calculation of the dispersion curve laid on the observed velocity spectrum of a supplied model by the user www winmasw com geophysical software amp services page 56 winMASW About the final deepest layer in the picture example below only 5 layers have been input the last one being a semi infinite layer and therefore without any thickness value leave the value equal to zero azione Spettri di Velocit Primo trattamento dati _ _____ Secondo determinazione spettro di velocit e picking visualizza curve dataset 4022 dat a anne calcolo spettro di velocit rr B distanza intergeofonica 2 m campionamento 0 5 ms Tracce normalizzate Spettro di Velocita goo VS mis 132 235 220 180 620 vS30 271 Poisson spessore 2 m 1 strato 0 45 23 2 strato 0 45 43 3 strato 0 35 velocit di fase m s 4 strato 0 35 TER 0 35 6 strato zu 20 30 40 frequenza Hz semispazio numero di modi da visualizzare picking m 3 min 1 max 20 Tipo di analisi i seleziona modo X 2 selezionare l ultimo punto del modo Rayleigh i i i Rayleigh utilizzando il tasto destro vwrinivia SV
114. inates Ycoordinates 48 44 40 36 32 28 24 20 16 12 8 4 0 0 W00 500000000600 O 00000000500 N N e Outputs by default it is computed and saved only the ESAC dispersion curves and its respective pseudo velocity spectrum this is just the misfit between the observed correlations and the values obtained by the Besse function computed for the different phase velocities If you activate the f k option the respective dispersion curve cdp file will also be computed and saved The verbose option will provide further intermediate outcomes such for instance the k k plots etcetera www winmasw com geophysical software amp services page 44 winMASW Outputs Log file images and files automatically saved in the working folder When the analyses are accomplished in the working folder you will find the following files automatically saved logeSACFK txt log file summarizing the performed operations ESACdc cdp dispersion curve saved as effective curve Tokimatsu et al 1992 since compared to modal dispersion curves the computation of the effective curve requires heavier computational load its automatic inversion requires state of the art PC recommended an esacore CPU ESACpsvelspe mat pseudo velocity spectrum determined through the ESAC analyses snapESACFK png snapshot of the main panel with the resulting spectrum and the dispersion curves If you activated the f k a
115. ione microtrem ex iN A fi taskmgr exe pr percorso file WEWOUDE G Termina processo ePowerTray exe G Apoint exe d Termina albero processi csrss exe S Debug winMASW ex C Virtualizzazione controllo dell account utente firefox exe 32 d Crea file di dettact hoswp dipbo d rea file di agli Capture exe Q Imposta priorit Cap exe 32 d i dia aliena AdobeARM e d cmd exe 32 G Propriet plugin contain G l hpagp0 Lex d Vai ai servizi Mostra i processi di tutti gli utenti Termina processo Processi 101 Utilizzo CPU 52 Memoria fisica 63 Select the microtremor exe button and terminate the process To go head it is now necessary to modify the model as the previous one was responsible for some problematic event in the HVSR computation www winmasw com geophysical software amp services winMASW page 107 Site resonance frequency The topic is surely complex and this is not meant as a text book Here it is possible just where to recall that Vs mean Vs down to the bedrock H thickness of the sediments overlying the bedrock For instance if the bedrock is 37m deep the Vs that we have to consideri s a Vs37 The peak in the observed HVSR gives the resonance frequency of the site the value of the peak cannot be considered as the amplification factor www winmasw com geophysical software amp services page 108 winMASW Love waves and the Horizontal to Vertical Spectral R
116. is looked for are different according to a constant or not constant velocity gradient Once the limits of the search space are fixed search space or parameters space the parameters of the genetic menu can be input table 1 In order not to give the user irrelevant details some standard genetic parameters have already been fixed crossover and mutation rates www winmasw com geophysical software amp services winMASW Bl winMASW 4 1 Pro Dispersion curve inversion Dispersion curve or Analysis Rayleigh Y Depth of the bedrock 5 shallow deeper Number of layers amp constrains number option option 2 force search space 2 of layers from reference model Inversion genetic menu 30 individuals models min 6 F Vp amp density optimization max 7000 30 generations min 10 max 400 Outputting output folder close winMASW ae Dispersion curve amp search space www eliosoft it Picture 4 Starting window of the section inversion WWW Winmasw com geophysical software amp services page 65 page 66 winMASW 2 winMASW 4 1 Pro Dispersion curve inversion Dispersion curve Analysis N Dispersion curve amp search space Rayleigh dispersion curves demo cdp 240 220 Depth of the bedrock Is the bedrock deeper than 18 554 meters 2
117. it is necessary to perform some operation that must be explicitly allowed So do not worry and allow that Possible problems with some anti virus systems Some anti virus systems might be unable to manage the software protection based on the envelope system that protects winMASW In this case it is possible that your anti virus will warn you about a trojan in the winMASW exe binary application In that case just tell the anti virus to ignore the folder where you are installing winMASW Smarter anti virus systems we might suggest the free and excellent AVG do not have such a problem which so far occurred with AVAST AVIRA and Panda Once the set up is done the icon winMASW will display on the desktop the usual double click will launch the application winMA Sy Also a group winMASW is created on the Windows programs main menu see snapshot www winmasw com geophysical software amp services page 16 winMASW E r Preferiti Nome om l Nero L Norton Internet Security Norton Online Backup L NVIDIA Corporation Packard Bell Sicurezza e supporto Documenti L Packard Bell GameZone i Real Rockey4ND SDK V1 30 E sesarray 2 0 0 beta Skype n Computer Suite grafica CorelDRAW X4 Video Web Camera Windows Live winMASW 4 1 DEMO winMASW 4 3 Pro Immagini Giochi Pannello di controllo Dispositivi e stampanti Programmi predefiniti ri Connect to winMASW internet site di Documents Guida e support
118. iven when uploading the data chart of the amplitude logarithm compared to the offset in this case the graphics refer to corrected data in its geometric component winMASW 4 0 Pro Testing Data for Rayleigh Wave Attenuation Analysis see winMASW manual Original data no corrections 20 30 offset m offset m Frequency Hz 11 Frequency Hz 48 log amplitude Example of dataset acquired keeping the same gain for all the channels amplitude falls according with the offset therefore theoretically suitable to analyze the attenuation Notice however some amplitude changes jumps compare for instance amplitude of the traces at offset 26 and 30 that can be due to various problems i e to a bad coupling geophone soil or the use of different quality geophones Bl vinMasw 40 Pro Testing Data for Rayleigh Wave Attenuation Analysis see winMASW manual Original data no corrections Geometrically corrected Data aT ST pe r oloa m Frequency Hz 11 Frequency Hz 48 log amplitude log amplitude 25 30 20 20 offset m offset m Example of dataset acquired using different gains for each channel trace therefore unsuitable to analyze the attenuation of the seismic data notice how the amplitude doesn fall according with the offset because the distant channels have higher gain than the near ones WWW Winmasw com geophysical software amp services winMASW page 89 Geometric corrections off In th
119. litude k 1 m 10 20 30 50 60 70 40 f Hz 80 log Amplitude k 1 m 1 20 3 50 60 70 80 40 f Hz Thus in case of picking or modelling it is possible to see the curve in both the domains f k and f v B winMASW data upload processing amp modelling URS Gla e x Tl 4 BO OF k 1 m 1 uploading amp processing MASW analyses dataset somma st1 st2 sgy sampling 0 26 ms minimum offset 1 5 m geophone spacing 1 5 m normalized traces orf 02 03 a4 osti 07 as ol 60 ther tools amp setting 0 a es Lusi Amplitude f Hz 2 velocity spectrum modelling amp picking MASW amp ReMi analyses MASW compute velocity spectrum r phase velocity m s velocity spectrum 1600 1400 1200 1000 800 600 400 200 30 40 frequency Hz 50 handling the spectra fundamental ow to select the last point of the considered mode click the right button Vs m s Poisson thickness m 0 35 general setting 0 35 0 35 Rayleigh v phase vel v 0 35 0 Reference depth Refraction oss 0 035 O HVSR 4 H V modes 0 35 W W W Winmosw cornm k 1 m log Amplitude f Hz WWW Winmasw com geophysical software amp servic
120. ly 3 It is essentials to consider i e save and upload a velocity spectrum which contains only information signals related to surface waves As a consequence before calculating and saving the velocity spectrum clean carefully your data e g by removing refraction events and by removing useless data and re sampling the traces try to obtain a dataset with 512 samples Fix velocity and frequency limits within which signals are related to surface waves thus avoid too high or too low frequencies and velocities A correct determination of the velocity spectrum is clearly vital for a meaningful inversion www winmasw com geophysical software amp services page 118 winMASW Example raw data 24 traces po dt 0 125ms incidentally too much 1 msec is absolutely sufficient TER acquisition length 2 024 seconds number of samples 16384 normalized traces me T reduce dt to 1 msec and keep the data only down to about 1 2 seconds Data optimized for the FVS inversion cut the data DONE and CUT buttons and re sample RE SAMPLE button dt 2ms acquisition time kept 1 14 seconds number of samples 571 1 ipload processing amp modelling a E VSA d may uer ol eyes Bll Ble o gt Oy S a A r 1 data uploading amp processing 2 velocity spectrum modelling amp picking MASW ESAC amp ReMi analyses dataset MASW classicalRay sgy MASW compute velocity spectru
121. m handling the spectra explore spectrum sampling 2ms 571 i phase velocity C FK group velocity save upload merae minimum offset 3 m geophone spacing 3m mode ration normalized traces 15 20 25 frequency Hz offset m Furthermore let s remove the first 2 traces scissor button in the toolbar This way we avoid near source effects also reducing the computational load amp amvse S 1 data uploading amp processing 2 velocity spectrum modelling amp picking MASW ESAC amp ReMi analyses dataset MASW classicalRay sgy RARE Cpu wn p handling the spectra explore spectrum sampling 2ms 571 i i 7 Fk group velocity save upload merae minimum offset 8 m geophone spacing 3 m mode ration normalized traces 20 25 frequency Hz offset m WWW Winmasw com geophysical software amp services winMASW page 119 4 Even if it is not necessary to pick any dispersion curves it is anyway necessary to perfectly understand the procedures and all the relevant points e g be careful when higher modes are heavily dominating the dataset 5 Poisson moduli are essential in determining the energy distribution among different modes see the box Poisson and the Energy Distribution We therefore recommend to fix their values while setting the search space according to realistic con
122. me directivity in time ime step co save option t save HVSR as it is save HV curve as it is save option 2 picking WV curve pick HV curve save picked HV compute SESAME for picked curve quick analysis f Vs 4H average V erage Vs mis 180 trom surface to bedrock 20 depth of the bedrock m 1000 Vs of the bedrock clean compute WWwWw winmdasw com truma Modeling amp Picking panels and upload the saved HV curve User can must modify the fc limits fc button on the toolbar in order to center the peak he she wants to evaluate with respect to the SESAME criteria In the above reported example we modified the fc limits and by activating the full output option obtained the plot reported down here Now the adopted fc limits are reported more discreetly and point with a H V value minor than 1 are not marked show data reset MT 0130527 84759CL EAN SAF 128Hz Average Spectra steps1 optional decimate 128Hz new frequency resample 10 remove events 59 Ra lean ax 20 wind length 3 10 tapering 10 5 pectral smo particle motio ti J full cuput compute step3a optional directivity analysis compute max freq 32 MHz step amp 3b optional directivity over time MT 0130527 84759CLEAN SAF HVSR window length 20s T T T directivity in time 8 amp 1 I I LIII E EEEE PNO E ENE BEF S P E E d WAREN REEE SU Nn mm m La
123. me min www winmasw com geophysical software amp services page 98 winMASW Show location button from the 5 2 version If your 3 component geophone is equipped with a GPS antenna and geographical data are properly reported in the SAF file you can obtain a localization map on Goggle Earth with a simple click Geographical data are reported in the SAF according to the following format EVT X longitude EVT Y latitude e EVT Z altitude Example of geographical data in the SAF file m Xk 1022 EVT Y RN 0 e em In case these data are present the show location button activates and you can see your data location on Goggle Earth which must be clearly installed on your File Modifica Visualizza Strumenti Aggiungi Guida camerno Q liess mia E www winmasw com geophysical software amp services winMASW page 99 Particle motion By activating the show particle motion of the original data option you will obtain a plot of the soil motion along 3 sections vertical NS vertical EW and horizontal NS EW clearly related to the original raw data Particle motion horizontal plane Density function www winmasw com geophysical software amp services page 100 winMASW Removing specific events before computing HVSR By clicking the remove events button in the main panel the eee window will pop up Y E winasw cleaning microtremor data before HVSR analyses Ro l
124. mentioned papers it will be possible to understand under which conditions the different approaches must be considered valid In a nutshell usually body waves properly explain HVSR only for the fundamental frequency while higher frequencies are better explained in terms of SW ellipticity the problem is sometimes to understand the appropriate number of modes to adopt also always considering the attenuation In winMASW while modelling HVSR the quality factors are set according to the following rule of thumb Qs Vsg 5 and Qp 2Qs www winmasw com geophysical software amp services page 106 winMASW Problems in HVSR computation using Surface Wave ellipticity microtremor exe Computing HVSR considering Surface Wave ellipticity is a very heavy task Under some circumstances it is possible that microtremor exe so to say stalls If after few seconds the HVSR window does not react See image down here imati i o X B winmasw vs profile for estimation of eigen frequency http www lamoglieofferta com profilo php profilo_id 9414 Vsv profile amp eigen period reference depth Om Frequency 9 5 Hz It is then necessary to interrupt the process mictrotremor exe in the Activity Manager of the windows operating system Ctrl Alt Canc buttons r File Opzioni Visualizza Applicazioni Processi Servizi Prestazioni Rete Utenti Nome immagine Nome u CPU Memoria Descriz
125. mp services page 114 winMASW Example Rayleigh amp Love III Dataset Rayleigh analyses Synthetic Dataset Rayleigh analyses Mc HMM TTT LUT 40 40 is m E m Uploaded Dataset Velocity Spectrum Synthetic Dataset Velocity Spectrum phase velocity m s phase velocity m s 5 10 15 20 25 30 35 5 10 15 20 25 30 35 frequency Hz frequency Hz Rayleigh on the left the field data on the right the synthetics LITT Dataset Love analyses EERTIIT Dataset TI analyses usc HE TR NESTOR ud ae offset m Uploaded Dataset Velocity Spectrum Synthetic Dataset Velocity Spectrum 500 v v 400 2 300 gt gt o o n n 200 2 a 100 5 10 15 20 5 10 15 20 frequency Hz frequency Hz Love on the left the field data on the right the synthetics www winmasw com geophysical software amp services winMASW page 115 Poisson and the Energy Distribution Vp are surely not very important in determining the dispersion curves but have a profound effect on the energy distribution among different modes That means that modifying the Poisson values i e Vp you will modify the energy distribution among different modes Down here an example on the left a real dataset on the left and in the central panels two synthetic models with the same Vs thicknesses but different Vp values Notice the different energy distribution quite often higher Poisson values determine more energetic hi
126. mp services winMASW page 153 Appendix I Tool TCEMCD Using our HOLI3C geophone and winMASW Academy tool TCEMCD you can efficiently acquire your passive data to process for ESAC and HVSR analyses In the following picture it is reported a typical acquisition setting 21 vertical component geophones and our HOLI3C 3 component 2Hz geophone are connected to a common 24 channel seismic cable Using the TCEMCD tool in the Academy version you can then extract the data for the ESAC and HVSR analyses for their joint analysis inversion you will obtain a standard SAF file for HVSR and a mat file for the ESAC analyses aeni channel trace 1 ESAC HVSR joint acquisition 4 5Hz vertical geophone 3 component geophone for HVSR ESAC and MAAM 2 12 amp eo e e e ES the three components of our 3C geophone 24 E 16 13 UD 14 NS 15 EW HOLI3C are connected to three channels of the seismic cable by means of the provided cable This way the vertical geophone of our HOLI3C is exploited also for the ESAC analyses In the following snapshot are shown 960 seconds about 15 minutes of data In this case the channels 13 14 and 15 refer to the data collected using our HOLI3C geophone channel 13 refers to the vertical geophone channel 14 refers to the NS geophone and channel 15 refers to the EW geophone By clicking convert amp re arrange you will obtain 2 separate files the
127. n r 5 l 15 2 25 3U 35 12 frequency 72 E E n wie fil ever d ian soe mm RI Cr V i e i VPN V i I TA ORAN ul offset rr e mail winmasw winmasw com WWW Winmasw com geophysical software amp services page 158 winMASW ADAM 2D Apparent Dispersion Analysis of Multi component Data 2D Acquisition and processing of multi component data for the characterization of vast 2D areas Thanks to the opportunities which originated from the synergies with some European partner companies and our procedures based on multi component FVS Full Velocity Spectrum Dal Moro et al 2014 approach we are now capable of acquiring and or processing multi component data aimed at characterizing large areas 2D and 3D from the geotechnical point of view 3 CSF ZVF dx2 mo3 26 1400 Z T Lm TET 1200 See competed spectrum 35 4000 7 5 E 4800 E rr 600 E 400 200 n amp 40 15 20 25 5 10 15 20 25 frequency Hz frequency Hz Observed and processed velocity spectra analyzed according to the procedure described in Dal Moro et al 2004 01 4 e ow Nn n un m L ex ce o_o o r m m CI oe CI gt o 2 n m 7 7 7 7 7 7 ee e wv wv wv wv v9 v vw v v 3 9 c T Cc bl T T err erre eo 0000000000 0 o o o o e o o o o o o o o te o o o o e o o o o o o o o o o o o o o o oo coco GE E E E E E E E E
128. n filter used during MFA depends on 2 parameters Alpha0 and Alpha1 There are no universal values suitable for any dataset as these values depend on the specific dispersive properties of the considered dataset Anyway thanks to some improvements present in the 4 7 version we would suggest 2 fix values 120 and 0 01 respectively It is also recommended a re sampling to 1msec see panel down here em Velocity Spectrum Parameters qroup velocity analyses velocity spectrum define limits velocity m s 20 600 Spectral analysis number of samples minimum maximum Traces to consider First trace Increment Last Trace frequency Hz 1 Alpha0 see manual minimum maximum Alpha see manual verbose 4 resample to 1msec www winmasw com geophysical software amp services page 48 winMASW By activating the verbose option you will also obtain the 2 following windows The first one reports the delays as a function of the offset n n gt m e ua o E b group velocity m s offset m while the second one presents the last considered trace on the left and the obtained velocity spectrum on the right r File Edit View Insert Tools Desktop Window Help UgGdsfAagGT9L 2 DE D offset last trace 66m 0 0827068 LA e e group velocity m s Q9 ce e 25 30 frequency Hz Please notice that MFA analyses cannot be applied to
129. nalyses MASY Remi visualize curves dataset stendimento1 dat minimum offset 1 5 m ene spectrum Tau v upload ReMi spectrum 55 geophone spacing 1 5 m sampling 0 26 ms normalized traces velocity spectrum EEITIIILII 0 1 X E m E 0 2 0 3 0 4 0 5 06 D 7 0 6 0 9 10 l F 4 offset rm frequency Hz utilities data selection refraction flip traces activate phase velocity m s spectrum select 20 modelling picking movie 2 cancer save fundamental i use the right button to select the ad model 3 Rayleigh v last point of the considered mode www ellosoft it winivia S Ww are Attenuation analysis Picture 2 Picking of the dispersion curve www winmasw com geophysical software amp services winMASW Identifying and picking the dispersion curve is of course an operation of crucial relevance for the final result The user should have the necessary confidence to do that operation A misunderstanding curve profile or assigning the mode could lead to inconsistent results In order to have a clearer view on the dangers following article is suggested Velocity Spectra and Seismic Signal Identification for Surface Wave Analysis Dal Moro et al 2006 e Possible Effects of Misidentified Mode Number on Rayleigh Wave Inversion Zhang amp Chan 2003 Picking a too
130. nalysis option you will get FKdc cdp dispersion curve determined through the FK analyses Moreover if you activated the verbose option you will also get the following files CrSpVsFr png fig Cross spectra versus frequency ESACdata png fig 2Dpassivedata png fig data channel positions radii fk8frequencies png fig 8 k ky plots for 8 frequencies FK method Remember that the cdp files are simple ASCII files which can be uploaded in winMASW AL input curve button see next pages www winmasw com geophysical software amp services winMASW page 45 ReMi spectra and ESAC dispersion curves In the ReMi panel where purely linear arrays can be analyzed it is possible to activate the ESAC option By doing that we will obtain an overlap of the ReMi spectra and ESAC dispersion curve Dataset Purgessimo provided as example dataset together with the software winMASW 5 0 and later Analyzing just a couple of files total length 2 minutes we obtain a clear evidence of a mode jump which is absolutely not so common while analyzing passive data but which is absolutely good for our analysis n RE vinMASW passive seismics linear array EJS E x estimated D estimated R estimated global 500 450 400 350 hase velocity m s phase velocity m s phase velocity m s 200 150 100 2 4 6 8 10 12 14 15 18 20 2 4 6 8 10 12 14 16 18 20 frequency Hz
131. nd 20 melted soils with better cohesion can reach up to 60 Rock materials can range about 100 up to 200 Consider that the mistake in determining the Q value is proportional to the same value high Q values are therefore less precisely determined than lower values so that given a Q gt 100 the mistake can have a value same the Q value in itself see White 1992 Incidentally a damping factor often is quoted to quantify the viscose component instead of the quality factor Q They are connected through the following formula DEE 20 www winmasw com geophysical software amp services page 96 winMASW 10 Analyzing HVSR 3C Pro amp Academy Parameters to set are clearly indicated please refer e g to SESAME 2005 The data format is the one adopted by the SESAME project SAF SESAME ASCII data Format SAF file format The SAF SESAME ASCII data Format format is a common ASCII file which can be open and edited with any text editor winMASW assumes that the 3 columns represent first column vertical component first column NS component first column EW component Moreover There are 2 relevant parameters in the header 1 The sampling frequency SAMP_FREQ 2 The date START_TIME In case of errors while reading your SAF file obtained from your triaxial geophone you may try to correct the header lines and the columns the order of the columns and create a SAF file which is consistent with the abo
132. nd theory dispersion curves from a model which the user tests and modifies starting from prior stratigraphic knowledge and data through the direct modelling procedures www winmasw com geophysical software amp services page 22 winMASW MASW Analysis First upload the common shot gather Execute it clicking on the icon top left allowing to open the needed file see the box regarding the supported format Due to frequent and possible common shots mistakes the user always need to verify that the input data min offset distances between traces and so on are correct and if necessary correct them According to the execution mode order of the traces it can be possible that the dataset be inverted In this case you only need to click on rotate traces see box Rotate traces giving consequently an order to the data Once the data are uploaded the next step is the calculation of the Velocity Spectrum through the relevant homonymous button The user has to input velocity and max and min frequencies in other words the limits of the velocity spectrum see picture 1a The result will be similar to what you can see in picture 1b In the case you are not happy with the spectrum window frequency velocity interval and you prefer to have a better vision of the area where the signal of interest is actually concentrated the dispersion curve s related to surface wave propagation you can recalculate the velocity spectrum wit
133. nd type of 3 component geophone for HVSR measurements for analyzing surface waves please write us winmasw winmasw com and we will give you our recommendations REGISTRATION AND MAINTENANCE Do not forget to register your purchase please provide the name of the reseller the date of your purchase and the version Standard 3C Professional or Academy You will receive software updates and our newsletter To register send an email to winmasw winmasw com If you miss to register in 2 weeks from the purchase no assistance and no free upgrades will be provided during the first year during which new releases of the software are provided for free an annual maintenance fee is required if you wish to receive successive updates VIDEO TUTORIALS Updates of the present manual can be downloaded from our web site In addition it is also possible to download a series of video tutorials aimed at properly using the main winMASW features The tutorials are focused on specific kinds of analyses standard MASW analysis Rayleigh waves joint analysis of Rayleigh amp Love dispersion joint analysis of phase amp group velocities ReMi and ESAC analyses HVSR analyses etc Several helps are displayed just pointing the cursor on the button a notice on yellow background will display providing basic information Some of the pictures in the manual might refer to former versions of the software All upgrades keep their same features adding new functio
134. nmasw com geophysical software amp services winMASW page 103 Directivity Once computed the mean HVSR it is also possible to evaluate possible directivity of the recorded signal up to a specified frequency HAF HVSR directivity 20 40 60 80 100 120 140 160 180 Azimuth degree Please notice that the plot on the left reports frequencies in logarithmic scale while on the right plot frequencies follow a linear scale The image is automatically saved in the working directory Saving the H V curve At the end you can decide to save the resulting H V curve the adopted format is the one suggested by the SESAME project according to 2 possible options 1 option 1 save HVSR as it is 2 option 2 picking H V curve In the first case the user will save the H V curve in the indicated frequency range In the second case it will possible to pick the HVSR curve in order to remove possible anthropic artefacts signals not related to lithology registered users receive a mailing list with examples etc www winmasw com geophysical software amp services page 104 winMASW show data reset MT_0130527 84759CL EAN SAF 128Hz Average Spectra stepi optional decimate samph step H V computation remove events 59535 iT y cleanaxes 2 20 ra g 10 taper s moothing thang ti ti data t comput stepla optional directivity analysis compute max freq 3 steps3b optional directivity over ti
135. nown and layers thickness is as well these can be set In order to give Vs a wider range This is a way to reduce the freedom of the system and the numbers of individuals and generations to consider too In a typical case 7 layers with search space similar to the one the software automatically proposes the pair 60 60 can be a reasonable choice Advice 2 search space The search space has to be fixed on the basis of the prior geological knowledge If these are limited a wide search space is needed together with a higher number of individuals and generations If you start the option force search space the programme can start searching for the best model apart from the user s given limits advised in case of non expert users Advice 3 get acquaintance with the method its potentials and limits Upload a dataset and calculate its velocity spectrum Try to reproduce the observed data introducing models in the section direct modelling Modify a parameter at once and note how curves change What happens if you increase the depth of a fast layer What happens when inserting a velocity inversion in depth What is the ratio between the Vs layer its depth and phase velocity of the wave observed on surface These kind of trials and exercises see the direct modelling are very helpful to get confident with the methods www winmasw com geophysical software amp services page 72 winMASW Advice 4 number of layers to use We sugg
136. ns NOTICE Remember that a software is just a tool Accuracy of results always depends on the users skills and experience We highly recommend to attend to one of our workshops and or visit us for the 1day training included in our offer www winmasw com geophysical software amp services winMASW page 11 Index dniigore Ureilo g temer TM 12 T oyster requirements ele 13 ZSC ER RT ORT 14 3 User License KEY and assistance is 17 4 Use Drocedure DaslCSsiusssit sidentisieuten bin Eo PtbRu deni e Peut eat uut ente Potente Pot ep b tal satiate acct 18 4 1 Velocity Spectra picking modelling dispersion and refraction MASW ESAC ReMi ESAC amp MFA analyses phase and group velocities 21 4 2 Inversion of the dispersion curve 64 5 Recommendations and advices iii 70 6 Results of the inversion of the dispersion CUFVESs n 73 7 Analysis of Love waves Pro amp AcadeMy cece eeseeaeeeeeeseeeeeeees 78 8 Joint Inversion of Rayleigh amp Love wave dispersion curves Pro amp Academy 81 9 Analysis of the attenuation of Rayleigh waves Pro amp Academy 85 10 Analyzing HVSR 3C Pro amp Academy i 96 11 Modelling HVSR 3C Pro amp AcadeMY aaraa rat 105 12 Dispersion curve amp HVSR joint inversion Academy 110 1
137. o following paper Alternative interpretations of horizontal to vertical spectral ratios of ambient vibrations new insights from theoretical modelling Albarello D and Lunedei E 2010 Bulletin of Earthquake Engineering 8 519 534 Insights on Surface Wave Dispersion and HVSR Joint Analysis via Pareto Optimality Dal Moro G 2010 J Appl Geophysics 72 29 140 For these and other reasons that we recall during our workshops it is always preferable to adopt the forward modelling approach from the Velocity Spectrum Modelling amp Picking panel rather than the automatic inversion WWW Winmasw com geophysical software amp services page 112 winMASW 13 Synthetic Seismograms Academy In the Academy version winMASW also generates synthetic seismograms computed via modal summation Quite clearly to properly use these tools it is necessary to be familiar with several aspects related to seismics also including signal processing this is why the tools are available only for the Academy version Modelling main points e Minimum and maximum frequencies considered while generating synthetics are the same of the currently considered velocity spectrum phase velocity button in the MASW compute velocity spectrum group e The adopted number of modes is the same as the number indicated in the general setting sub group see modelling group by default 3 modes are considered e Source while analyzing
138. o tecnico i Uninstall winMASW 4 4 Academy s winMASW 4 4 Acd WinRAR Wisdom soft AutoScreenRecorder 3 Fre 4 Indietro Cerca programmi e file Arresta il sistema m WWW Winmasw com geophysical software amp services winMASW page 17 3 User License key and assistance The winMASW software works by means of a hardware key USB dongle To know more about costs please ask the distributor or look up the web pages dedicated to winMASW in the ELIOSOFT web site www winmasw com In case the key gets damaged a new one only can be obtained after receipt of the damaged one and after the new key its delivery expenses are settled ELIOSOFT as well as any software distributor can t be responsible for any improper use of the software due to misuse or misunderstanding or bad knowledge of the MASW MFA ESAC ReMi methods Possible advices alternative interpretations of the software the purchase of the license actually gives the possibility to achieve two different interpretations forwarded by users are allowed to be used by ELIOSOFT for further teaching and exemplification purposes Technical support is provided by e mail winmasw winmasw com or on the phone visit the website www winmasw com for updated information Free of charge upgrades are available within the first year of purchase new software versions That time expired a fix quote is requested according to both the date of purchase and to the version standard or pro
139. o the traditional inversion methods based on the Jacobian Matrix and needs longer calculation times Such times are evidently proportional to the number of individuals i e models and of generations the user chose in the genetic menu see Table 1 and Picture 4 The calculation times also proportionally depend on the number of used layers and on the data quantity frequency velocity couple in the picking file where the user saved the identified dispersion curve Increasing the quantity of dispersion points to a number of about 10 won t bring an effective better result but a longer and useless calculation time Regarding the number of layers it should be given on the basis of prior geological knowledge In normal conditions 7 layers are enough while more strata are sometimes necessary www winmasw com geophysical software amp services winMASW page 71 Advice 1 number of models and generations The number of individuals models and generations to adopt has to be proportional to the algorithm effort to achieve a good solution Parameters then have to balance the number of considered layers more layers more freedom for the system and higher computing effort and the amplitude of the parameters space The search space can be set by the user according to prior geological and stratigraphic information and to the known Vs of the most common lithological types see Table 2 in relevant appendix If a site stratigraphy is k
140. ocity spectra and picking of the dispersion curve we want to invert This topic is considered for instance in the following papers Dal Moro G amp Marques Moura R M 2014 Multi component Acquisition and Joint Analysis of Surface Waves Two Case Studies for Two Possible Inversion Strategies submitted to J Appl Geophysics Dal Moro G amp Ferigo F 2011 Joint Inversion of Rayleigh and Love Wave Dispersion Curves for Near Surface Studies Criteria and Improvements J Appl Geophysics 75 573 589 Dal Moro G 2011 Some Aspects about Surface Wave and HVSR Analyses a Short Overview and a Case Study BGTA Bollettino di Geofisica Teorica e Applicata invited paper 52 241 259 Please also regularly visit the www winmasw com site section publications From a practical point of view please consider the following 1 It is absolutely untrue that the maximum energy related to the surface waves is necessarily associated to the propagation of the fundamental mode many modes can coexist in a same dataset 2 It is absolutely untrue that higher modes only appear at high frequencies They can instead exist at low frequencies as well and several mode jumps can take place As a consequence it is often advisable to replace the picking gt inversion procedure with the direct modelling see relevant section searching the more logic coherence between energy distribution in the velocity spectrum so far the only objective data a
141. of anisotropy due to lithological stratifications and or tectonic breaks see again Safani et al 2005 and Gaherty 2004 Finally note as well how Love waves only can be used with active seismic data MASW and not passive ReMi as in this last case you won t be able to tell the transversal component of micro quakes linked to Love waves from the radial component instead of the Rayleigh waves Please carefully read Dal Moro amp Ferigo 2011 Dal Moro amp Moura 2014 and all the case studies available from our web site www winmasw com www winmasw com geophysical software amp services winMASW page 79 Model thickness m 2 5 3 5 5 Vs m s 100 230 380 520 Vp m s 2Vs density 2 gr cm all layers Rayleigh phase velocity m s phase velocity m s 20 40 40 frequency Hz frequency Hz Model thickness m 2 5 3 5 5 Vs m s 100 230 180 420 notice the Vs dispersion deep Vp m s 2Vs density 2 gr cm all layers Rayleigh Lu c c phase velocity m s 40 a 20 40 frequency Hz frequency Hz www winmasw com geophysical software amp services page 80 winMASW Model Thickness m 2 5 3 5 5 Vs m s 100 75 250 480 notice the Vs inversion in the second layer Vp m s 2Vs density 2 gr cm all layers Rayleigh phase velocity m s phase velocity m s 20 40 frequency Hz frequency Hz Model thickness m 2 5 3 5 5 Vs m s 75
142. ons than in the search space Vs and min and max thickness as indicated by the user This allows a better fitting shouldn t the user have set right parameters in the search space Non expert users should make use of this option Vp and density optimisation Setting this option up the software after the search for the best Vs and thickness will modify the Poisson s ratio compared to those indicated by the user in order to improve the fitting Allow it www winmasw com geophysical software amp services winMASW page 69 Bi objective space in case of joint inversion Rayleigh Love If this option is active a full complete joint inversion is processed This means that during the optimization process each model misfit is calculated both for Rayleigh and Love In the opposite case optimizations come separately This option slows calculation times a bit but allows the expert user to check the distribution of models in the so called bi objective space refer to relevant literature on how to proceed Dal Moro amp Pipan 2007 Dal Moro 2008 Dal Moro 2009 Once launched the inversion a state bar will show both the passed and the remaining time B 30 Avanzamento Inversione x Per interrompere cliccare la x Tempotrascorso 0 00 12 Tempo stimato rimanente 0 00 28 The procedure ends when following window shows up E FINE PROCEDURA INVERSIONE x Risultati salvati nella cartella C send SW outputs
143. or Rayleigh amp Love waves Velocity spectra inversion via synthetic seismogram I no need of dispersion curve picking but longer computational times Tool for the vertical stacking Tool for creating 2D sections Tool for putting in evidence specific even hidden modes ME ESAC and FK analyses bidimensional arrays NEW Tool TCEMCD Three Component Extraction from Multi Channel Data for efficient passive joint ESAC HVSR acquisitions connect your vertical geophones and our HOLI3C 3 component geophone to your seismic cable and with this tool you ll then be able to extract the data for the joint analysis of dispersion via ESAC and HVSR see TCEMCD Appendix Tool for combining several traces acquired by a single 3 component geophone using different offsets and obtaining datasets useful for MASW analysis considering both Rayleigh radial and vertical component Love waves See the GEMINI2 geophone http www holisystem com on line from January 2015 www winmasw com geophysical software amp services page 10 winMASW WARNING 1 Buy and use horizontal geophones and jointly analyze Rayleigh radial component and Love waves Please acquire the data strictly following our guidelines and download and read the following collection of case studies http www winmasw com brochure_winMASW_ 20EAGE pdf If you are not completely sure about the best type of equipment to buy type and number of geophones a
144. ormally 4 5 layers but 5 6 if vertical profile a layer with inversion of velocity is suspected Once the number of layers has been given winMASW proposes a wide range of Search space Value limits that each variable can values which the user can have thickness and Vs of each modify generally limiting layer them according to specific knowledge of the area Number of models constituting the Number of number of people evolving to better individuals models solutions The higher the number of layers the more numerous the model 30 to be considered Number of generations through which new models are explored that 30 get better and better with time Number of generations Table 1 Inversion parameters Max allowed Vs anisotropy in case of joint Rayleigh Love inversion In case of joint inversion Rayleigh Love you can let the algorithm free considering Vs from Love different from Vsy from Rayleigh the number we herewith consider is the possible percentage difference between these 2 parameters This may allow to mark some anisotropies up see for instance Safani et al 2005 or easier for less expert users to consider a sort of tolerance in the optimization inversion process Possible anisotropies only have a value according to the users experience Therefore consider them in this view Force the search space Setting this option up the software is allowed to search for other soluti
145. ossible not necessary To save or upload a filter save filter amp upload filter buttons To save a filtered spectrum save spectrum button To save a filtered dataset save dataset button as segy file If we like the action of the adopted filter and want to get back to the main panel considering the filtered dataset it is sufficient to click the button Accept Otherwise by clicking the cancel button we will close this panel and get back to the original unfiltered dataset Please use such a a tool only if you perfectly know what you are doing Artefacts due to improper use of it might appear See also Luo et al 2009 WWW Winmasw com geophysical software amp services page 126 winMASW 16 Demo version and supplied data Demo The very last demo version available is the 4 1 which is absolutely inadequate to represent the current features of winMASW The potentiality of winMASW are now presented in a series of video tutorials that can be downloaded from the internet site www winmasw com publications area Supplied data Some datasets are supplied in the folder data together with the software The first dataset demo sgy supplies a velocity spectrum and therefore a dispersion curve typical of a sequence of non consolidated sediments over a bedrock the first 13 m consist of fine sediments silt and sands then 5 m of conglomerates and at a depth of about 18m the rocky s
146. ou ll have to stick to some cautions The basic principle actually valid for all acquisitions is that the field dataset has to be most reliable In other words it must represents experimental conditions The introduction of filters gains etc alters the originality of a data It is during the analysis of data at home in the office that some more operations can be introduced filtering gains corrections etc in order to highlight some features instead of others When analyzing seismic wave attenuation there are few crucial points to consider 1 the gain value has to be the same for all channels 2 no trace can have signal saturation 3 make sure you have a good geophone soil coupling The reason for that prudence to be followed in whatever acquisition lays on the fact that the analysis of the seismic attenuation are aimed to quantify the decrease of the seismic signal amplitude according to distance and to the different frequencies making the signal It should be clear that alterations can damage the analysis of signal attenuation Since attenuation varies according to frequency as for dispersion lower frequencies are sensitive to deeper soils features any filtering is not allowed unless you are aware of the consequences during the analysis Finally since bad coupled geophones lead to a worse data quality in the altered amplitudes make sure the geophones are permanently coupled to the soil What is the signal saturation
147. ould be avoided avoiding as well to get more than useful data In the actual case as in most as a matter of fact 1 second is more than enough excessively increasing the registration time would give an increase of the relevance of the data without a relevant enhancement of the dataset quality Material __ Vs m s Incompetent soils and peat 60 180 Competent soils 180 350 Very competent soils and 350 600 gravels Weathered rock 600 1000 Solid rock gt 1000 Table A2 indicative values of Vs according to different materials WWW Winmasw com geophysical software amp services winMASW page 135 Passive seismic acquisition ESAC and ReMi In a ReMi acquisition our interest is registering the dispersion of surface waves generated by micro tremors i e events of whatever nature even human coming from unknown sources but typically far and hopefully containing lower frequencies than those generated by acquisitions got in active seismic sessions Regarding these low frequencies some points should be highlighted In order to register with enough quality the signal coherence at low frequencies two aspects should be considered 1 Necessity of long arrays that allow to better define the coherence linked to the signal dispersion at low frequencies 2 use of particularly sensitive geophones to low frequencies make sure you have good quality 4 5Hz or less geophones A last consideration on the dataset length use
148. ow This will allow a full exploitation of multi core CPUs thus reducing the computational times required by some operations this is particularly important if you deal with the computation of synthetic seismograms and related procedures such as the full velocity spectrum inversion In the following the snapshot of the main winMASW window after the activation of the parallel computing compare with previous snapshot In this case we activated 7 workers but if your computer has a standard i7 CPU you can activate up to 4 cores i e 8 threads to know the number of cores available in your computer please have a look at its manual WWW Winmasw com geophysical software amp services winMASW page 19 Internet sies Parallel Computing On f workers working folder utilities 7 When determining the Vs vertical profile therefore the Vs30 the first step is the determination of the velocity spectrum Then it is the turn of the direct modelling Suggested procedure or of the automatic inversion of the interpreted picked by the user of course dispersion curve To do this first go to the section determination of the velocity Spectrum see paragraph 4 1 from the main winMASW window shown below In the Professional version it is possible to analyze the signal attenuation of Rayleigh waves in order to get the Q quality factors In order to achieve it first of all the reconstruction of the
149. panel At this point you can also decide whether to analyze or not the directivity see related box or save the obtained H V curve see the Saving the H V curve box for the 2 possible options In case the user wants to re perform the analyses using different parameters e g smoothing etc it is recommendable to re upload the data It is also shown the continuity of the HVSR during the acquisition in this case about 12min that means the HVSR for each considered window HVSR vs Time 2D view HVSR vs Time 3D view frequency Hz 2 4 6 8 10 12 14 16 time min frequency Hz time min WWW Winmasw com geophysical software amp services page 102 winMASW HVSR Report and SESAME criteria Once computation is done a html file is produced summarizing main points file is automatically saved in the working directory SESAME criteria regarding the reliability of the H V curve and the possible peaks are also evaluated By default these are computed while considering the 0 5 15Hz frequency range but this interval can be modified by using the f button in the tool bar Please also read the following two documents http nato gfz hr SfP 20 20980857 20manual 20part 20a 20v1 pdf http sesame fp5 0bs ujf grenoble fr Papers HV_User Guidelines pdf In the following the results considering the data in the 0 5 15Hz frequency range Peak frequency Hz 3 2 0 3 Peak HVSR value 5 5 0 5 Criteria for a
150. pear the depth of the last horizon interface User must then specify the maximum depth to visualize a default value is proposed by adding 2meter to the depth of the deepest horizon It is also possible to specify whether to visualize or not the single profiles on the 2D section just activate or not the show profiles checkbox see plots down here It is possible to specify the kind of data Vs Vp or none Actually it is possible to contour any kind of dataset just respect the format see Format box Vs section 500 400 300 200 100 O 30 40 50 60 70 80 90 100 distance m depth m Section obtained single vertical profiles are not visualized Vs section 60 70 80 90 100 distance m depth m 20 30 40 50 Section obtained while also visualizing single vertical profiles About the position of the single vertical profiles in line position There are 2 possible situations 1 the distance between the profiles is constant In this case it is sufficient to specify only such a constant distance for instance 30m 2 The distance between the profiles is not constant In this case it is necessary to specify the positions of all the profiles within square brackets In the reported example the positions are 20 30 50 80 100 Obtained section can be saved according to several possible formats In the toolbar File gt Save www winmasw com geophysical software amp se
151. performing a joint inversion e g Rayleigh Love or Surface Waves HVSR one of the final plots reports the evacuate models in the bi objective space also indicating a quantitative value S ex pressing the symmetry of the Pareto front models with respect to the universe of the models 1 means perfect symmetry 0 completely asymmetric Quite clearly a simple visual evaluation is more than sufficient and user should not stick to much to SI Incidentally the symmetry is not the unique parameter to consider to evaluate the overall congruency of the performer inversion process For more information about Pareto symmetry Dal Moro G amp Ferigo F 2011 Joint Inversion of Rayleigh and Love Wave Dispersion Curves for Near Surface Studies Criteria and Improvements J Appl Geophysics 75 573 589 Dal Moro G 2011 Some Aspect about Surface Wave and HVSR Analyses a Short Overview and a Case Study BGTA Bollettino di Geofisica Teorica e Applicata invited paper 52 241 259 Dal Moro G 2010 Insights on Surface Wave Dispersion and HVSR Joint Analysis via Pareto Optimality J Appl Geophysics 72 29 140 Dal Moro G 2008 Vs and Vp Vertical Profiling via Joint Inversion of Rayleigh Waves and Refraction Travel Times by means of Bi Objective Evolutionary Algorithm J Appl Geophysics 66 15 24 Dal Moro G amp Pipan M 2007 Joint Inversion of Surface Wave Dispersion Curves and Reflection Travel Times via Multi Objective Evolutiona
152. pesci I MALI K r be Y A ui MI LWUOUIMSCUIJUIW MAAM WWW Winmasw com geophysical software amp services
153. ponent The anisotropy of Vs waves is calculated as follows percentage values V PS ss SH Notice that the percentage of anisotropy can also be considered as a tolerance as to wrong values in dispersion curves picking This means that imprecise picking values can be kept under control by means of introducing a difference between the Vsy relevant to the dispersion of the Rayleigh waves and the Vs relevant to the dispersion of the Rayleigh Waves We don t need to mention that the evaluation of anisotropies requires an expert user WWW Winmasw com geophysical software amp services winMASW page 83 Distribution of models in the bi objective space and Pareto front symmetry 1201 n 100 ae n s obj 2 60 80 Model distribution in case of consistency between the two objectives the front if pointy and the symmetry of the Pareto front models with respect to the rest of the models is quite high 20 40 B ol 100 120 Model distribution in case of some inconsistency between the two objectives the front if quite large and the symmetry relatively modest www winmasw com geophysical software amp services page 84 winMASW Symmetry Index SI Pareto Front Symmetry e at m SMIRNE TY RIETI Sete een c een e seme rigree een DOT PNE P 1 s 9 ales Mi aie ol ta i 0 50 100 150 200 250 300 While
154. quencies the highest result more attenuated H winMASW spectral analysis trace 1 Amplitude spectrum Phase spectrum unwrapped 0 trace 1 frequencies Hz spectral analysis 30 30 frequency Hz frequency Hz trace 24 Amplitude spectrum Phase spectrum unwrapped trace 24 z 0 2 frequencies Hz 0 4 p 0 6 0 8 1 uk 14 1 6 1 8 compute 2 30 30 frequency Hz frequency Hz www winmasw com geophysical software amp services winMASW page 51 Here below the amplitude and phase spectra of the air wave note the high frequency characterizing the signal peak around 145 Hz Bl winMasw spectral analysis trace 24 Amplitude spectrum frequencies Hz mine 1 max 300 50 100 150 200 250 frequency Hz Phase spectrum unwrapped 50 100 150 frequency Hz Consider the potentiality of this tool when selecting a data portion if in that portion there are frequencies incompatible with the component you aim to isolate investigate that can be useful to redefine the selection polygon Spectrograms In the utilities section the button spectrogram will give the possibility of computing the spectrograms of the uploaded dataset trace by trace This will allow the user to evaluate the Spatial and temporal change in the frequency content due to seismic wave propagation and attenuation trace 12 42m 0 trace 12
155. r winMASW output if you set up the software as Administrator and then launch it with a different user this might not happen If using Windows Vista it is strongly suggested to install the software in an external folder as C winMASW ATTENTION If you install winMASW in a folder where former versions of the same software are kept all data files and analysis data will be deleted Use therefore a different folder and or make sure you previously run a back up 15 Installazione di winMASW x Benvenuti nel programma di installazione di winMASW winMASW versione 21 sar installato sul computer S consiglia di chiudere tutte le applicazioni attive prima di procedere Premere Avanti per continuare o Annulla per uscire TN The winMASW application refers to Matlab libraries simultaneously installed with the software choose English as set up language Choose Setup Language x E Select the language for this installation from the choices below L Inglese Stati Uniti zl Should the graphic libraries Matlab already be installed in your Computer winMASW 4 8 refers to the 7 15 or later version of MCRInstaller of Matlab choose the option modify in following window www winmasw com geophysical software amp services winMASW page 15 5 MATLAB Component Runtime 7 5 InstallShield Wizard Program Maintenance Modify repair or remove the program Modify i Change which p
156. r with value around that one This method results very useful when in front of a stratigraphy with strong acoustic impedance contrasts for which the traditional approximation 1 2 would be a wrong starting model on which the automatically suggested search space is based WWW Winmasw com geophysical software amp services winMASW page 73 6 Results of the inversion of the dispersion curves Results are supplied both graphically and as text files txt and html format see table 2 and are automatically saved in the folder output or any other file as chosen by the user Every time the user launches the programme all files in the output folder go deleted Therefore if you keep your results in the folder output save the inversion files in another one named as wished You ll notice two models the best model in terms of lower misfit i e the discrepancy between the observed and the calculated curve and a medium model calculated by means of MPPD Marginal Posterior Probability Density see ref Rayleigh Wave Dispersion Curve Inversion via Genetic Algorithms and Posterior Probability Density Evaluation Dal Moro et al 2006 Consider this as your final result however the two models don t differ significantly The final screenshot Picture 7 displays different information Top left the observed data the curve of the best model and the medium one often the two curves are one over the other therefore not
157. r is also provided the eigen period frequency Bonnefoy Claudet et al 2008 show how the observed H V Spectral Ratio cannot be attributed to a single and simple phenomenon and that depending on the local geological situation the observed HVSR carries different modes of the Rayleigh and Love waves partially together with the body waves too The whole in a very unpredictable and complex Way What is agreed upon is that the peak value of the H V ratio if any represents the resonance frequency of the site but there still isn t a universal agreement on the meaning of the picked value which does not straightforward represents the amplification factor In other words if we record see e g image below a peak at 6Hz and a H V value of 2 5 we will be able to claim that the resonance frequency of the site is 6Hz but the estimation of the amplification factor won t be an immediate consequence Vs profile amp eigen period reference depth 0m HVSR 3 Frequency 6 Hz Period 0 17 s 0 2 4 6 8 Depth m P 0 5 100 200 300 400 P Fari A 600 700 800 900 5 10 pares T 20 25 30 Depending on the type of analysis we choose and on the needed accuracy we can stick to the main recommendations issued by the local authorities or work in details referring to the guidelines suggested by software applications like SHAKE or Deepsoll WWW Winmasw com geophysical software amp services page 146 winMASW Appendix E Creating
158. reliable H V curve 1 f0 gt 10 Lw 3 2 gt 0 25 OK 2 nc gt 200 6314 gt 200 OK 3 f0 gt 0 5Hz sigmaA f lt 2 for 0 5f0 lt f lt 2f0 OK Criteria for a clear H V peak at least 5 should be fulfilled 1 exists f in the range f0 4 f0 AH V f A0 2 yes at frequency 2 0Hz OK 2 exists f in the range f0 4f0 AH V f A0 2 ves at frequency 4 2Hz OK 3 A0 gt 2 5 4 gt 2 OK 4 fpeak Ah v f sigmaA f f0 5 OK 5 sigmaf lt epsilon f0 0 313 gt 0 157 NO 6 sigmaA f0 lt theta f0 0 642 lt 1 581 OK Please be aware of possible industrial man induced peaks or spurious peaks due to meaningless numerical instabilities Remember that SESAME criteria should be considered in a flexible perspective and that if you modify the processing parameters they can change Some recommendations about it SESAME criteria are not a chapter of the Bible if you modify the processing parameters length of the analysis window transient event removal tapering etc the outcome will change also with respect the SESAME criteria the criterium 5 among the Criteria for a clear H V peak is often the strictest one Flexibility is thus necessary while considering its value please be aware of possible industrial peaks AA VV Report SESAME Project 2005 which cannot be identified automatically and that only an human eye can properly recognize www wi
159. requency is 50 Hz there is no signal connected to the Rayleigh waves over this value the reason is the presence of both 2 modes actually you shouldn t consider frequencies lower than 18Hz The solution to this site is roughly starting from the more superficial layer and down considering Qs Qp 10 10 7 15 50 A further useful dataset to analyze the attenuation is the test attenuation2 sgy min offset 20m whose solution is following model also see the sample file test attenuation2 mod Vs Ves2Vs Qs Qr 6 30 19 half space Note as in both these cases it s the soil first 10 m that characterize the attenuation curve Deeper Q values are higher and the grade of indeterminateness is high also see box typical value of the quality factor Q Re the analysis of the attenuation note as both the file demo sgy and somma st1st2 sgy are not suitable for the analysis of the attenuation the user can try to understand why according to what read in the chapter re attenuation In the sub folder IO within the winMASW installation folder it is also present a folder named Data Purgessimo containing a complete dataset MASW data both Rayleigh and Love waves HVSR and passive acquired considering a linear array The solution is discussed presented in the pdf available at the following link http www winmasw com italian Joint 20Analyses 20in 20winMASW 20 20case 20study 20P pdf www winmasw com geophysical software
160. rms of computational load and precision of the solution see the system requirements paragraph the case studies available from our web site and the following paper Surface Waves Joint Inversion via FVS Dal Moro et al 2014 Y Definitive implementation of apparent effective dispersion curves e g Tokimatsu et al 1992 for both Rayleigh radial and vertical components and Love waves Use them for passive data only ESAC method v HVSR modeling considering both Rayleigh and Love waves also considering attenuation and the a factor Arai amp Tokimatsu 2004 see Surface Wave Analysis for Near Surface Applications Dal Moro Elsevier SISATYNYV SAVM RE RE m o D z m gt e u D n gt Lan m 2 gt TU D O gt z rr SURFACE WAVE ANALYSIS M FOR NEAR SURFACE EET APPLICATIONS Giancarlo Dal Moro bed pel ELSEVIER v Implementation of the joint inversion of HVSR and dispersion according to FVS or apparent dispersion curve www winmasw com geophysical software amp services page 8 winMASW jew YV WINimMmasw com WinMASW eo OT ll I MA v V vsurrace W Software for Surface Wave analysis MASW ReMi ESAC FK and MFA modelling and inversion of Rayleigh and Love waves dispersion analysis of Rayleigh wave attenuation for the estimation of Qs quality factors determination of the resonance frequency from microtremor analyses HVSR Nakamura s method and modelling of the
161. rogram features are installed This option displays the is Custom Selection dialog in which you can change the way features are installed C Repair E Repair installation errors in the program This option fixes missing or corrupt files shortcuts and registry entries C Remove ex Remove MATLAB Component Runtime 7 5 from your computer During the set up of libraries a message as follows might display MATLAB Component Runtime 7 5 Installer NET Framework is not installed If you require it select Cancel and install NET Framework first Otherwise select OK to continue Cancel Just click on Ok ignoring it Firewalls and parallel computing If your firewall is particularly strict it is possible that when you lunch the software a message like the following appears Avviso di protezione Windows e Windows Firewall ha bloccato alcune funzionalit del programma Windows Firewall non consente al programma di accettare connessioni di rete in entrata Se si sblocca il programma lo sblocco avr effetto in tutte le reti pubblica a cui ci si connette Rischi Herivanti dallo sblocco di un programma e T Nome mpiexec exe Autore Sconosciuto Percorso C program Filesimatlabir201 1b bin win32 mpiexec exe Percorso di rete Rete pubblica Che cosa sono i percorsi di rete Continua a bloccare W Sblocca This happens cause in order to exploit all the workers cores of your machine
162. roup velocity save upload merge explore spectrum minimum offset 1 5 m geophone spacing 1 5 m lized t locity a picking normalized traces velocity spectrum 1600 show f k select mode v 1400 to select the last point of the considered 1200 mode click the right button w 1000 save picking 5 i 800 a n x 600 inversion x 400 200 10 20 30 40 50 60 0 E offset m frequency Hz T modelling Vs m s Poisson thickness m 3 thetics cvi ES cz rit refraction about Poisson 100 035 1 calculate O n 100 generat setin 159 035 2 ueedmea ZV Tl z 0 99Hz l l 60 Rayleigh 3 phasevel 28 0 35 5 save model 7 elastic fe cancel save clear refraction 400 0 35 7 refresh F shows DC 3 0 Reference depth Refraction 800 035 0 rnd models 2 other tools amp setting 0 0 35 0 repot 3 report HVSR 4 H V modes 1102362 Time length to visualize s 7 GS In this case the velocity spectrum is better defined compare the amplitude of the red coherence bar with the former ones For relative amplitude reasons low frequencies don t have the same strength of high ones but we are now allowed to better identify the velocity relevant to 10 Hz approximately 600m s Combining two SAC datasets SAC datasets are made of 1 trace only such a format is usually adopted in passive seismology It is possible to combine two or more SA
163. rpretation picking modalities for a ReMi datum is different from that of a MASW datum see section 4 1 Attenuation analysis In order to analyze the attenuation of Rayleigh waves you first need to determine the vertical shear wave velocity profile e g by means of MASW dispersion analysis in an ASCII mod file format where thicknesses Vs Vp and densities and preliminary Qs and Qp values are reported 1 open the section Attenuation analysis this can be done from several panels the main one and the single component analysis or the Joint Analysis of Rayleigh amp Love panels 2 upload the seismic file active seismic data clicking on the button top left 3 in the case of strong refractions or other noises clean the datum activate and select 4 visualize the velocity spectrum button calculate update to define the frequency interval inside which it makes sense to determine the attenuation curve 5 calculate the attenuation curve clicking on calculate curve in section step 2 calculate attenuation curve 6 upload the model file mod with formerly determined Vs profile button input Vs model in section step 3 attenuation curve modelling inversion 7 proceed with modelling section Modelling Q or inversion section Inverting Q The same considerations as for the dispersion curves are valid here too with regards to modelling inversion WWW Winmasw com geophysical software a
164. rrelation correlation correlation correlation correlation correlation WWW Winmasw com geophysical software amp services winMASW page 43 Data input and parameters It is possible to upload several datasets together as common in Windows to do it is sufficient to use the ctr button In fact some seismographs can record at the most 1 minute by recording several datasets each for instance 1 minte long it is then possible to acquire a sufficient amount of data to perform all the analysis based on passive acquisitions ReMi ESAC FK most of the times 15 minutes are enough Some points e The Resampling option to reduce the computational times it is absolutely recommended this option is active by default but can be de activated e Computational load for the FK method are higher than for ESAC and strongly depend on the adopted parameters We recommend to consider frequencies not higher than about 30Hz High frequencies are in fact better imaged via active MASW acquisitions and analyses The lowest useful frequency depends on the array length dimensions the characteristics of the site and on the eigen frequency of the used geophones e The positions of the channels geophones can be reported manually in the pertinent windows or can be written in a simple ASCII file according to the syntax indicated in the following box the file must have a header line and the pos extension File ASCII pos extension Xcoord
165. rsion Joint DC HV inv 400 200 10 20 30 40 50 60 70 z frequency Hz Y modelling Vs m s Poisson thickness m 3 Ee data selection fitering amp spectra refraction about Poisson 100 0 35 1 calculate 1 X 100 general setting 180 0 35 2 uploadmod ZVF E 0 99H ry Rayleigh v 3 phasevel v 260 035 3 save model V elastic E accept cancel save 400 035 7 Lretmesh f shows oc 2 0 Reference depth Refraction 900 035 0 rnd models 2 r other tools amp setting E 035 0 an 2 n 102262 Time length to visualize s e m o 038 3 With distanced offsets low frequencies are evident but the most proper velocity value to be picked results quite unclear since the scarce number of traces and depending on the user s choices what velocity has to be picked at 10 Hz www winmasw com geophysical software amp services winMASW page 143 Obtained velocity spectrum considering the whole dataset sum of two single datasets 1 uploading amp processing MASW analyses 2 velocity spectrum modelling amp picking MASW amp ReMi analyses visualize curves dataset somma st1st2 sgy P sampling 0 26 ms phase velocity Fk g
166. rum there is the section direct modelling that allows the user to calculate the dispersion curves of a model up to seven layers whose parameters can be fixed by the same user clicking on the displaying menu on parameters The calculated curves are screen shown and saved on an ASCII file Frequency Vp in the file modelladiretta txt in the file winMASW output The aim is to get the evaluation of a possible mode in respect of an observed spectrum In the case of geological complex situations i e seismic data of hard interpretation this is a very useful method and generally the interpretation is enough as an inversion is no longer necessary It is clearly possible to fix different values of the Poisson ratio in order to modify the ratio Vp Vs you ll notice that the Vp values once the value of the Vs has been fixed as modifying the Poisson ratio means modifying the Vp are not that relevant if compared to the role of the thickness and shear wave velocities Vs El winMASW 4 1 Pro Velocity Spectra Modeling Picking g om 1 uploading amp processing MASW analyses 2 velocity spectrum modelling amp picking MASW amp ReMi analyses MASIN ReMi visualize curves calculate spectrum Tau v upload ReMi spectrum 2 El winMASW forward modeling y mis di E m velocity spectrum layers 110 0 40 Y co e m layer 2 130 0 40 X d e layer 3 170 Poisson X
167. rvices page 148 winMASW Appendix F case studies processing etc Some case studies are included in the Documents sub folder within the winMASW installation directory and we suggest to download all the case studies and papers available from our web site www winmasw com area publications More case studies are sent to our users though our newsletter Joint analysis forward modelling of Rayleigh amp Love waves Example Vs vertical profile WW W W inmoasw com 300 400 Vs m s Rayleigh velocity spectrum Love velocity spectrum Rayleigh Dispersion 500 sv m s 132 170 126 210 496 516 481 Vs30 336 phase velocity m s phase velocity m s 25 30 35 40 45 25 30 35 40 45 50 frequency Hz frequency Hz www winmasw com geophysical software amp services winMASW page 149 Examples of useless dafasets winMASW Determinazione dello Spettro di Velocit primo passo input dei dati Note as apart from the high noise the sudden change in character slope and frequency from the meren offset 20 m on maswl DAT Fo lii m m Ej winMASW Determinazione dello Spettro di Velocit primo passo input dei dati Note the various change in slope and signal eue character testifying the probable presence of strong lateral variations that make this dataset useless tempo sec li ruota le tracce www winmasw com geophysical software amp s
168. ry Algorithms J Appl Geophysics 61 56 81 www winmasw com geophysical software amp services winMASW page 85 9 Analysis of the attenuation of Rayleigh waves Pro amp Academy Leaving theoretical details and underlying that the best way to improve your knowledge is attending to one of our workshops it is anyway useful to remember that seismic waves decrease in amplitude while propagating for two reasons 1 because energy distributes on an always wider area this reduces energy therefore amplitude 2 because of complex energy absorption conversion depending on the soil they re passing through The first aspect is called geometric component while with the second we refer to the Q quality factors describing the viscose quality of a mean high Q values represent a good elastic mean that doesn t absorb seismic energy that much The traditional method to analyze seismic attenuation develops in 3 phases a determination of the soil elastic method Vs Vp density and thickness b determination of the attenuation curve attenuation coefficient depending on the frequency c inversion or modelling of the attenuation curve or modelling of the observed attenuation curve once fixed see point a the elastic part i e the soil model Vs Vp density and thickness From a practical point of view first proceed with the analysis of the dispersion curve of the Rayleigh waves that is the procedure used to determine the vertic
169. s The only big difference is that these waves appear only in the case of a positive velocity gradient i e only if between layer n and layer n 7 deeper velocity increases Besides Love waves only respond to Vs thickness and density of the mean Vp already of scarce importance to Rayleigh waves is now fully pointless It is obvious that achieving and elaborating dataset from both see Appendix A is highl recommended see Dal Moro amp Ferigo 2011 Do not forget by the way that nature is not likely to follow simple standardized rules It is possible you find sites with great Love waves and very bad Rayleigh ones Note as when examining the Love waves the only given elastic modulus by winMASW is the shear modulus other in fact depend on the Vp see Appendix C that has no importance to the dispersion of Love waves Why analyzing Love waves There are two reasons As pointed out by Safani et al 2005 the superior modes often are less present and velocity spectra often are better defined compared to those given by the Rayleigh waves as the relevant wave area of the first is less subject to scattering Love waves only depend besides on Vs no longer on Vp as the Rayleigh ones Note that the dispersion of Love waves as far as refraction depends on the horizontal component of the Vs Vsy while Rayleigh waves Vs depends on its vertical component Vsv So if you can do a good joint analysis of both waves you ll be able to find cases
170. s linear methods require the starting method to be as close as possible to the real solution If not the risk is to be mistaken and give wrong solutions Genetic algorithms offer instead a wider search of possible solutions Some details on foundations and functioning of genetic algorithms are available below http en wikipedia org wiki Genetic algorithm http www talkorigins org fags genalg genalg html http www doc ic ac uk nd surprise 96 journal vol4 tcw2 report html The key point of winMASW is anyway represented by the possibility of joint analysis of several datasets components acquired according the guidelines briefly reported in the Appendix A and explained in a series of articles and white papers indicated in the manual www winmasw com geophysical software amp services winMASW page 13 1 System requirements The winMASW software application works with all the recent 64bit Operating Systems OS It is important to frequently update the operating system click on windows update from the menu Start on Windows in order to let winMASW properly use updated functionalities A minimum of 6 Giga RAM and 3 GHz CPU are recommended see also paragraphs 4 2 and 5 about software performances in term of computing times Operating systems 6 x releases may have problems with windows XP an old operating system already dismissed by Microsoft itself winMASW Academy and HoliSurface work only on 64 bit operating systems
171. s not enough Numerical factors impose long enough datasets www winmasw com geophysical software amp services page 30 adapt to most situations Primo trattamento dati dataset 4022 dat offset minimo 2 m distanza intergeofonica 2 m campionamento 0 5 ms Tracce normalizzate Sete LLL mum E y Y nl 40 RE ee tempo s 1 4 1 6 1 8 2 D 30 50 roe Litilit Seleziona dati CA www eliosoft it winMASW ee winMASW Section Select data This section allows to select a specific group of datasets to further analysis You only need to do the start up start up button a polygon will show Stretching its angles you ll select a specific area of data Click on Select to cut the portion out outside the polygon The number appearing on the button right side is the number of the samples of the used ramp to cut the external data of the studied polygon out In other words the values inside the polygon stay the same the external are set to zero The number of samples of the ramp actually is the number of samples in the transit area If too limited different problems related to a sharp signal cut off might happen The default value 20 Secondo determinazione spettro di velocit e picking 1600 1600 1400 za at ex N e co ej L velocit di fase m s 5 0 15 20 25 3 3 40 4 visualizza curve fa
172. sideration consider the presence of water etc clearly also allowing the Vp and density optimization 6 As we are dealing with attenuation as well it must fixed a K factor Qs Vs K see also next box If you are not an expert please avoid this approach or ask for our help www winmasw com geophysical software amp services page 120 winMASW Example of discordance between the spectrum got from a synthetic dataset computed considered also the attenuation and the dispersion curves computed to the simple elastic Traces ZVF Velocity Spectrum 20 25 30 offset m frequency Hz Q model Velocity model 80 100 300 400 500 600 700 800 900 Q velocity m s Please notice how especially in the 12 22Hz frequency range the dispersion curve computed considered the elastic case is slightly slower than the spectrum computed considering synthetic seismic traces calculated considering attenuation as well Also notice that the signal between 8 and 10Hz is related to the first higher mode Fr Bs peroni Gr seu eso INN X ee Dispersion curve or velocity spectrum Analysis Dispersion curves or Velocity Spectrum amp search space number of traces 24 sampling 1 ms input file minimum offset 5 m geophone spacing 2 m Rayleigh v E velocity spectrum fullMONFAspectr 5 10 15 20 25 number frequency Hz of layers from reference model un ce e Phase vel ce e D
173. spectra in the background this clearly means that our tentative model is good phase velocity modelling H general settings select data 0 Ref depth activate 0 05 H V body waves Lsees 60 0 H V modes SW ell 0 1 cancel save an x Dr ra Refraction EEFE flip data Ve thk m s S Poisson m 532 3 0 24 0 1 phase velocity a 02 r1 mode separation 025 0 3 0 35 0 4 picking select mode v 10 20 30 40 50 80 5 60 oe e frequency Hz 674 67 03 a 355 33 03 215 group velocity 676 55 0 3 55 dataset ZVF24 mat 800 68 0 15 E minimum offset 10 m Modes phase B geophone spacing 2 m Modes group 5 U sampling 1 046 ms Rayleigh v 3 velocity spectra phase amp group velocities refresh calculate 0 A model 2 save mode D gt upload model E 2 __upioad mode zi 5 2 synthetics ZWF v 11 picking select mode v o HE ice 3 some tools 0 449592 time to visualize sec done save cancel 10 15 20 25 30 35 40 45 50 55 60 frequency Hz As a matter of facts this approach is similar and for some reasons even better than the effective dispersion curve approach e g Tokimatsu et al 1992 www winmasw com geophysical software amp services page 116 winMASW ZVF RVF ZEX REX or THF In the terminology adopted by Prof Herrmann VF stands for Vertical Force HF for Horizontal Force
174. suggested parameters for passive seismic acquisitions ReMi or ESAC www winmasw com geophysical software amp services page 136 winMASW Note to ReMi acquisitions The reason why ReMi measures are sometime requested instead of or together with the MASW ones is that they can catch lower frequencies making the users investigations deeper This is not wrong But some problems inside the same analysis of the surface waves dispersion make this a bit unrealistic practically speaking To cut it short we remind you that to have good results at low frequencies you should need quite long arrays and very sensitive low frequency geophones About passive array measurements we highly recommend ESAC being based on bidimensional acquisition geometry the directivity problem is solved and the overall robustness of the mathematics behind is such to provide much better results If you are not completely sure about the best type of equipment to buy for analyzing surface waves type and number of geophones and type of 3 component geophone for HVSR measurements please write us and we will give you our recommendations winmasw winmasw com www winmasw com geophysical software amp services winMASW page 137 Special recommendations about acquisitions aimed at analyzing attenuation MASW data If you wish to analyze the attenuation of the seismic signal in order to estimate the Q quality factors when acquiring the field data y
175. t spettro di velocit Spessore m 2000 1 strato 0 8 1800 2 strato 12 1600 3 strato 5 5 fe 1200 semispazio 1000 Bn numero di modi da visualizzare prirni 2 modi E annulla calcola velocit di fase m sec eo RM 400 200 10 20 30 40 50 60 F iHz ruota le tracce picking _ _ _ _ _ _________ __ Modellazione diretta iseleziona modo E E parametri e selezionare l ultimo punto del modo a i refresh 2l NP ER utilizzando il tasto destro refresh n M S VV salva picking inverzione fine winhMtA about winMASYW To be considered only up to 28 Hz after that value the first superior mode shows www winmasw com geophysical software amp services page 88 winMASW Button Test amplitude The button test amplitude opens a window with reported seismic data and the graphic of the amplitude logarithm compared to the offset for both the two indicated frequencies in fields Min amp max frequencies on panel Step 2 calculate attenuation curve Depending on the geometric correction being active on or inactive off two different windows will display Geometric corrections on Starting from the corner top left and following clockwise you can read seismic traces in their original recording amplitude corrected seismic traces thanks to the geometric correction therefore on the basis of offsets g
176. t sampling 6 ms reverse phase velocity m s cmn m m velocity spectrum FK parameters min freq 4 max freq 25 512 wavenumbers min vel 80 max vel 1200 3 window lenath s r ESAC parameters 2 vw spectralsmoothing T hold on 2 window lenath s EESAC pseudo spectrum o ESAC dispersion curve x FK dispersion curve STO 15 20 frequency Hz 7 verbose clear save spectrum analyze the saved spectrum 7 f k analysis Please notice that the reduction of the adopted traces reflects in an increase in the aliasing phenomena compare the plot with the one presented while using all the 23 traces As for any passive acquisition where the signal amplitude is quite low those who have an old seismograph with a limited dynamic range it is important to set a sufficiently high gain clearly avoiding signal clipping Sampling interval rate 2ms 500Hz are absolutely sufficient we assume that your seismograph has a good anti alias filter If you acquire with a higher sampling frequency in order to avoid uselessly higher computational times it is recommended a re sampling by default winMASW does propose does that see option in the lower left corner www winmasw com geophysical software amp services page 42 winMASW correlation correlation correlation Hi Quality Data E E 8 8 8 correlation correlation co
177. ta to the analysis of the Rayleigh waves Using horizontal component geophones with axis perpendicular to the array and transversal shear source you ll get a dataset useful for the analysis of the Love waves Remember that horizontal geophones can be used to record both Love waves and Rayleigh waves radial component thus optimizing the acquisition procedures see also Dal Moro amp Ferigo 2011 and all our case studies availabe in the publication area of our web site Please use them instead of using the vertical geophones capable of recording onl the vertical component of Rayleigh waves often very problematic in terms of complexity of the velocity spectrum www winmasw com geophysical software amp services page 134 winMASW dataset risultante tempo sec th LR e n e ZUUI ill 15 offset m Figure A2 Example of two Common shot gather The widening of the initial signal due to the dispersion of the surface waves is clear in this case Rayleigh waves In the acquisition phase it is important to make sure that the surface waves trend is not cut by a reduced acquisition time In the shown case an acquisition time lower than 0 40 seconds would have caused the loss of the useful signal to analysis In the case as reported in Figure A2 an acquisition time of half second could have been enough By the way problems coming from the cut of the data see box effect of data cut off sh
178. thickness of layers i e on geometry such density and Vp values and consequent elastic moduli can be viewed as approximate only Nonetheless they don t depend on Vp but on density and Vs only the value of the shear modulus see appendix C is of course a good estimation Notice that if using Love waves you ll get that modulus only as the others depend on the V that has nothing to share with the dispersion of Love waves www winmasw com geophysical software amp services winMASW page 77 depth m depth m 0 500 1000 1500 2000 2500 0 500 1000 1500 2000 2500 Vs m s V m s Miglior modello Modello medio Modello reale Figure 9 Example of results for a synthetic model For the inversion on the left 6 layers were used for the one on the right 10 in pale grey the search space Subsurface model E T pa cL qi Eu Output file stratigra png medium model stratigraphic column Vs density and thickness WWW Winmasw com geophysical software amp services page 78 winMASW 7 Analysis of Love waves Pro amp Academy Love waves are generated by a shear wave source the traditional transversal hammer blow to a layer of geophones and since the transversal movement they need the use of horizontal geophones for SH waves transversal to the former mentioned layer to be recorded It is another kind of surface wave that generally responds to the same rules of the Rayleigh wave
179. timation of the outcome reliability too standard deviations In picture 4 the main window of inversion shows directly available from the main menu or the section determination of the velocity Spectrum ale step select the dispersion curve formerly achieved you intend to invert clicking on see picture 4 Of course if you get in this section from the section determination of the velocity Spectrum after the picking of the dispersion curve the programme will directly upload the curve without therefore uploading the file where it was saved Once data are uploaded the dispersion curve will show in the window top on right The dispersion curve you need to invert can t be in the output file you ll choose A The second step aims to optimize the inversion procedure In fact assumptions made in the automatic definition of the search space Vs values and thicknesses within which the best solution is looked for can be different if the bedrock is readable or not deep down underground If the velocity gradient doesn t dramatically change as it can happen in alluvial plains where the rocky substratum is quite deep or when the layering directly takes place on the rocks the so called assumption 2 2 or as recent studies demonstrate A 2 5 see relevant box is valid while in the case there is a significative variation in the gradient when the bedrock is at scarce depth let s say between 5 and 30 mt this is no longer
180. two attenuation curves one is achieved through the Spectral Ratio SR method the average value is shown considering different couples of traces the other through the Matching method amplitude versus offset see window with 6 plots generated by the option verbose see Tonn 1991 the obtained values are by the way similar In the case you go on with the inversion of the attenuation curve take as reference the resulting curve from the Matching method WWW Winmasw com geophysical software amp services winMASW page 87 Suitable Dataset to attenuation analysis primo passo input dei dati _ secondo passo determinazione dello spettro di velocit input file calcolo spettro di velocit MASVYO01 DAT spettro di velocit 1000 900 Bu 700 600 tempo sec 500 velocita m sec 400 300 200 100 10 20 30 40 5 60 F Hz LIII picking modellazione diretta modo fondamentale zu fe parametri gt selezionare l ultimo punto del moda n o cai utilizzando il tasto destro MAS Vy D inversione fine Wir about winMASVW One only mode is visible and very clear the fundamental mode primo passo input dei dati secondo passo determinazione dello spettro di velocit calcolo spettro di veloci
181. ty m s ho Q9 e explore spectrum Velocity Spectrum from ReMi data note how the formerly picked curve on MASW data stays along the inferior edge of ReMi spectrum See also Appendix A for advices about the criteria for data acquisition Note We do not recommend MASW ReMi as decisive tool to retrieve a perfect Vs profile The best procedure for obtaining a robust Vs profile is represented by the joint analysis of Rayleigh amp Love wave MASW HVSR www winmasw com geophysical software amp services winMASW page 37 ESAC and F K analyses These methods are particularly suited for the analysis of passive datasets acquired while adopting bidimensional 2D geometry That means that geophones are planted not in a straight line but following 2D geometries circles L shaped crosses or random distributions a X Ww AEN D 2 geophone arrays in urban areas Some relevant points the possibility of analysing data acquired by means of a 2D geometry allows to overcome the directionality problems which afflicts ReMi methodology there s no ideal geometry The way you are able to disseminate your geophones depends on local conditions The easiest geometries are clearly those that require the minimum effort L shaped or crosses often the circular one is the most difficult please consider that the location coordinates x y of the geophones must be accurately report
182. uate inversions of velocity even according to Vp remember you can t get solid information about it from the dispersion of surface waves Consider that in order to get a refraction effect of noteworthy amplitude the acoustic impedance jump i e the result of Vp multiplied for the density must be significant a Vp decrease according to depth doesn t result in a refraction event the file of the saved or uploaded model has the same format of the one used to model the dispersion curves If a modelling of the dispersion has preliminarily been executed and therefore some Vs can be predicted Vs as well as Vsy and Vs will be considered as same when writing the refraction model button save In the opposite case in the mod file Vs values according to an assumed Poisson value around 0 35 will be reported As an example look below at a joint modelling The same model is considered from a dispersion curves and first arrivals times point of view velocity spectrum Rayleigh Dispersion phase velocity m s Dispersion of the first three modes of the Rayleigh waves for the model top left Vp vertical profile Data amp refraction travel times direct pro 400 500 600 700 800 Vp m s LE E m In red the arrival times of two refraction events relevant to the 2 only interfaces having a positive impedance jump in green the arrival times of a direct wave In the Academy version also reflection travel times are shown lii
183. ubstratum This situation appears quite complex as in the case a strong contrast in velocity happens here represented by the bedrock the approximation 2 2 5 see box on the max penetration depth is not enough and the user will have to define a search space winMASW allows by the way to use the option surface bedrock see second step of the inversion procedure where the user can force the programme to properly modify the procedures in the definition of the automatic search space The dataset represents an extreme critical event lack of high frequencies because of a very dissipative soil and sediments very low over a massive bedrock at a depth just on the limit of the signal capacity of penetration according to this geologic asset We again remind that high velocities and low frequencies assure a deeper penetration if under given conditions the expression linking velocity and wave length given A further wrong side of this dataset depends on the fact that the third layer underground beneath the two first superficial ones shows a light velocity inversion Vs about 90 m s that further unbalances the approximation 2 2 5 A possible solution would be the joint use of the higher mode if correctly interpreted The user can proceed accordingly eo Miglior modello Miglior modello Modello medio 7 Modello medio o e Profondit m A E re a E 2 e e Velocit di fase m s I 10 20 30 50 offset m
184. urface G3 ve MS MAAM HVSR joint acquisition Vs E PS 3 E ae d i 3C G1 e 7 62 ied E 2Hz vertical geophone rt 3 component geophone for HVSR and MAAM p P E pense 3 is i ri G5 d 963 a 3C EER a ae nig ee homo PAT RIS onde E Pr A Ioa ND al Ale RO oa lisa di fi gal al pr Pe L a 4 EIE lE DL I P E E a i Hy Ale i a a PP al A d e iq i I Sab i i channels The same tool TCEMCD is also present in our software HoliSurface and can be used for the joint acquisition and analysis of data according to MAAM Miniature Array Analysis of Microtremors HVSR Horizontal to Vertical Spectral Ratio WWW Winmasw com geophysical software amp services winMASW page 155 Appendix L Bugs Any software inevitably has some problem which is typically due to a usage different than the one planned by the programmers In the following some possible problems that can be easily solved by the user 1 It is very important to proceed with a windows update including ALL the components also those which are not considered essential It was sometimes observed that the final images resulted black just due to missing graphical components which are included in the windows update 2 Some anti viruses it happened only with AVAST AVIRA and PANDA so far might not be able to handle understand the hardware protection FEITIAN used for protecting winMASW
185. utton top left checking that acquisition parameters are correct 3 determine their velocity spectrum clicking on calculate spectrum in section 2 right of the window remember to choose the limits of the spectrum on the basis of the datum characteristics Now there are 2 possibilities to be used together in order to lead the user to define a final reasoned solution on the basis of the previously known geologic data a proceed with a direct modelling of the dispersion curves b execute the picking of the dispersion curve that is our interpretation always save the curve and proceed with the inversion section We strongly advise the first approach direct modelling as it allows forces a full control over the process the user can interpret the data on the basis of his stratigraphic knowledge In order to follow this mode click on parameters in the section modelling and insert the relevant data to thickness and layers Vs The aim is to obtain dispersion curves that can perfectly lay over coherently the distribution of energy shown by the calculated velocity spectrum If instead you wish to render this process search for the best solution automatic proceed with the picking of the dispersion curve and pass on to the section inversion Here you ll choose the numbers of layers to use for the inversion or in the case a rough model has previously been detected in the direct modelling and saved you can upload it
186. valid and we need a different strategy to identify the search space If the velocity gradient is constant we can assume an approximate 2 2 5 that as mentioned suggests that at a A 2 5 depth a Vs about 1 1V where V is the velocity of the Rayleigh wave at a wave length 1 X v f displays actually a correcting factor balances according to the considered depth The blue curve Picture 5 represents that approximate value while the two red ones are to be read as max and min values These values then depend on the number of layers the user considers to define the search space in green see picture 6 It is pretty obvious that on the basis of the site knowledge the user can modify the assigned values in order to get a faster and better research In fact the wider the space of the parameters the higher the number of the individuals and generations to set in the genetic menu while if we can limit the research space these can be kept low reducing calculation times Then the user needs to choose the numbers of layers to use through the scroll down menu Once done the window parameters picture 5 shows where the user has to put the search space For each layer you need to input thickness and min and max Vs you d consider as possible for each of them As in picture 5 the software can propose some values Made assumptions in the automatic definition of search space values of Vs and thicknesses within which a solution
187. vant to that observed ratio H V the traditional yellow folder named HV The software reads ASCII files and assumes that the first column reports the frequency and the second the spectral ratio H V possible header lines to be ignored and can t affect the reading analysis In other words the file format is line 1 example of site name line 2 example of frequency sampling line n Vs profile amp eigen period reference depth 0m Frequency 6 Hz Period 0 17 s bedrock estimated 100 200 300 400 500 600 TOO 800 900 5 10 15 20 25 30 Vs m s Frequency Hz On the left the Vs profile of the considered model with site eigen period as from the resonance equation on the right the H V ratio according to Herak 2008 if no relevant velocity inversion happens the value determined by the equation and the one determined by the spectral ratio H V are typically very close If a formerly observed HV is uploaded and a model for forward modelling is introduced and its dispersion curves are calculated having selected the option eigen period both the theoretical HVSR curve in blue according to body waves and in magenta according to Surface Waves and the observed curve in green are displayed www winmasw com geophysical software amp services winMASW page 61 Vsh profile amp resonance frequency reference depth Om body waves CE ht i ci SW ellipticity 3 modes Example of observed and modelled both
188. ve reported format It is clearly absolutely necessary to have the correct sampling frequency reported in the header the assumed syntax is SAMP_FREQ In case the starting time START TIME is not properly set a simple warning message is visualized but you can still proceed with the analyses as this parameter is not vital Important Notice The software assumes that the data were already corrected for possible amplitude compensation related to the geophone outputs and thus that the used SAF file is free from instrumental bias which might alter the frequency response Please contact the triaxial geophone manufacturer about this point www winmasw com geophysical software amp services winMASW page 97 Visualizing the data By activating the show data option the 2 following windows will pop up Raw data Original data 1000 500 0 500 1000 1500 0 2 4 6 8 10 12 14 15 18 time min NS original data EW original data UD original data 1000 1000 500 500 0 500 400 1000 500 1500 i 1000 0 5 10 15 O 5 10 15 0 5 10 15 time min time min time min Data after basic processing zero mean and detrending zero mean amp detrended data 0 2 4 6 8 10 12 14 15 18 time min NS zero mean amp detrended data EW zero mean amp detrended data UD zero mean amp detrended data 600 1000 500 O 500 1000 600 O 5 10 15 O 5 10 15 O 5 10 15 time min time min ti
189. zed the velocity spectrum bottom left and calculated the dispersion curves box in the middle upload the model of the soil that was previously calculated through the analysis of the dispersion curves the obtained mod file To do that click on Input Vs model and search for the proper mod file in the output folder of the analysis of dispersion curves At this stage the Vs profile will display in the dedicated window a Modelling If we choose the direct modelling always suggested just click on Q model As soon as we identify the wished model because in accordance with the observed curved click on summary plot to visualize a summary screenshot of the model see further box That screenshot is automatically saved in png format in the folder winMASW outputattn but can be saved wherever and in another format as usual File Save As Step upload data amp visualize velocity spectrum Velocity Spectrum step 3 attenuation curve modelling inversion fli dataset test attenuation2 sgy min amp max frequencies 1 60 data flip minimum offset 20 m caicdatol pata Modelling G eophone interval 2 m ES Test ampitude Sampling 2ms ee EE npa vs model Emodei summary pit winMA m Ww WIRE traces D Step 2 calculate attenuation curve Inverting ver 4 0 Pro 1 Reference trace clean theoretical 100 Number of models Search spa

here - ELIOSOFT

Contents

Download Pdf Manuals

Related Search

Related Contents