2142-4 advanced conference on seismic risk mitigation and...

2142-4

Advanced Conference on Seismic Risk Mitigation and Sustainable Development

Gheorghe Marmureanu

10 - 14 May 2010

National Institute for Earth Physics Bucharest Romania

Seismic hazard maps for Romania and large populated areas by probabilistic and deterministic approaches, linear and nonlinear methods

ICTP Advanced Conference on Seismic Risk Mitigation and Sustainable Development

SEISMIC HAZARD MAPS FOR ROMANIA AND LARGE POPULATED AREAS

BY PROBABILISTIC AND DETERMINISTIC

APPROACHES, LINEAR AND NONLINEAR METHODS

by

Gheorghe MARMUREANU,Carmen Ortanza CIOFLAN

Trieste-Italy. May 10-14,2010


The scenario‐based methodology developed at this point is strictly based on observable facts and data and complemented by physical modeling techniques, probabilistic and deterministic approaches, linear and nonlinear methods, which can be submitted to a formalized validation process.

Earthquake in Romania have been known since Roman times, when Traian’s legionnaires begun the colonization of the rich plains stretching from the Carpathian Mountains to the Danube river.



Trieste-Italy. May 10-14,2010Romania-South East view



The Vrancea seismogenic zone in Romania denotes a peculiar source of seismic hazard, which represents a major concern in Europe, especially to neighboring regions from Bulgaria, Serbia, Republic of Modova etc. The strong seismic events originating from Vrancea area can generate the most destructive effects experienced in Romania, and may seriously affect high risk man‐made structures such as nuclear power plants (Cernavoda, Kosloduj etc.), chemical plants, large dams, and pipelines located within a wide area from Central Europe to Moscow to Rome.

In plan view, the earthquakes are localized to a restricted area in the bending zone between the Eastern and Southern Carpathians at least three units in contact: the East European plate, Intra‐Alpine and Moesian sub‐plates.



Strain transfer from the active Adriatic, Aegean and Vrancea deformation fronts throughout the ALCADI – Pannonian system

VranceaVrancea

AdrianAdrian

AegeanAegeanStrain transfer from the active Adriatic, Aegean and Vrancea deformation fronts

through the ALCADI- Pannonian System



Topograhy and Moho layer and the isosurface of the 2.2% p-wave velocity anomaly of the Vrancea area from the seismic tomography work



Present day geodynamics of the Pannonian Basin and its surroundings



•Earthquakes in the Carpathian‐Pannonian region are confined to the crust, except the Vrancea zone, where earthquakes with focaldepth down to 200 km occurs. •The ruptured area migrated from 150 km to 180 km (Nov. 10,1940 Vrancea earthquake, Mw =7.7), from 90 to 110 km (March 4,1977 earthquake, Mw =7.4), from 130 to 150 km (August 30,1986, Mw =7.1) and from 70 to 90 km (May 30,1990, Mw =6.9) depth.

The depth interval between 110 km and 130 km remains not ruptured since 1802,October 26,when it was the strongest earthquake occurred in this part of Central Europe. The magnitude is assumed to be Mw =7.9 ‐ 8.0 and this depth interval is a natural candidate for the next strong Vrancea event…



The maximum intensity for strong deep Vrancea earthquakes is function of earthquake depth. In 1977 strong earthquake (MW =7.4 and h=109 km) at its epicenter, in the Vrancea region, the estimated intensity was only VI (MMI scale), while some 170 km away in the capital city of Bucharest, the estimated maximum intensity was IX - IX½ (MMI). In 1940, the effects were in the epicenter area. The intensely deforming Vrancea zone shows a quite enigmatic seismic pattern (peak ground accelerations/intensity one, characteristic response spectra with large periods of 1.5 seconds etc.).



Observed distribution of macroseismic intensity during of large Vrancea earthquakesNovember 10,1940, MGR = 7.4 March 4,1977, MGR = 7.2.



The last seismic map, existing since 1993, has areas where seismic intensities are sub-evaluated (e.g. Dobrogea, Banat etc.), and other areas are over-evaluated. Intensities I=IX on Mercalli scale, at which corresponds a 0.4 g level of acceleration, and 0.8 g at rupture/yielding, make Vrancea county to become an unsustainabledevelopment area for the future. A special situation is represented by the Banat area where the last recorded earthquakes from Banloc, Voitec requires a change in the seismic intensities of this area. The fundamental unacceptable point of view is that this last design code is in peak ground accelerations which generates a lot of drawbacks to civil structural designers and to insurance companies which are paying all damages and life loses in function of earthquake intensity.



The concept of seismic intensity (or severity of earthquakeground motion) is at present a common concept for seismologists, structural engineers and other specialists, or even non-specialists. Persons working with this concept are recognizing at the same time its importance and some current shortcomings of it. The most important shortcomings that must be emphasized at this place consist of the imperfect definition of the concept, of the fact that the main criteria for intensity estimate are based on vulnerabilitycharacteristics which, at their turn, are defined conditionally upon the intensity (building thus a source of bias and tautology), of the lack of satisfactory correlation between the macroseimic intensity on one hand and the instrumental criteria on the other hand .

Also: MSK, MMI…MMII scale for → Vrancea



The complexity of the geology of the extra-Carpathian region is obvious. The geology of main cities from this area , like Iasi, Bacau, Buzau and Craiova is complex. The scientific problems regarding the waves propagation on source crystalline fundament-free soil path taking into account the nonlinear behavior are the most recent researches conducted the NIEP in this domain. This was important especially for the city of Buzau where sedimentary layers thickness is up to 5.0 km and constitutive laws concerning shear and strain dependence are viscoelastic nonlinear. Almost similar case is the city of Iasi where we were dealing with very different sitestopography, consisting in hills and plane areas. In next Figures are given the Bucharest City area isobaths and, respectively, the geological section through Quaternary layers from Ploiesti to Giurgiu and Danube river .



Isobaths are generally oriented from east to west, with a slope of about 8‰dipping from south to north and layers become thicker and thicker



VIII6.96.79026. 9045. 8210:40:06May 30, 1990

4

VIII½7.17.013526. 4745. 5321:28:37Aug. 30, 1986

3

IX7.47.210926. 3045. 3419:22:15 March 4, 1977

2

IX½7.77.415026. 7045. 8001:39:07Nov. 10, 1940

1

IoMw M

Richterh(km)Λ - EΦ- NTime Date Nr

The “etalon” earthquake concept



PGA recorded on August 30,1986 earthquake Mw=7.1128120Galati (GLT)18

144/188141/162Vrancioaia (VRI)35312297Focsani (FOC)17

205193Valeni Munte (VLM)345151Dochia (DCH)16

215202Vaslui (VSL)3398Deva (DEV)15

6460Tr.Magurele (TRM)328681Craiova (CVR)14

7268Tulcea (TLC)313634Constanta (CNT)13

7570Surduc (SDR)309690Carcaliu (CFR)12

2119Roznov (RZN)296463Cernavoda (CVD)11

163153RamnicuSarat(RMS)288277Campulung(CMP) 10

232218Ploiesti (PLS)2775-17171-161Bucuresti (BUC)9

1211Piatra Neamt (PTT)269892Branesti (BRN)8

228215Otopeni (OTP)25117110Braila (BRL)7

168158Onesti (ONS)242523Botosani (BTS)6

8079Muntele Rosu (MLR)239388Bolintin (BLV)5

1514Lotru (LOT)22175164Barlad (BIR)4

111/14 5109/127Istrita (ISR)213634Baia (BAA)3

108/216100/181Iasi (IAS)20110/11989/92,7Bacau (BAC)2

6460Giurgiu (GRG)192624Arges (ARR)1

amax.res.(cm/s2 )

amax(cm/s2 )

Seismic stationNr.amax.res.(cm/s2 )

amax(cm/s2 )

Seismic stationNr



The earthquake on August 30,1986 is used by us as “controlearthquake” in all studies as it fulfils the following states:

(i)-it was strong (Mw =7.1) ;(ii)-it was recorded in a lot of seismic station on Romanian territory ; (iii)-the fall plan solutions are very closed (almost identically) to

other Vrancea stronger earthquakes (Nov. 10,1940; Mw =7.7 ; March 4,1977; Mw =7.4; May 30,1990; Mw=6.9) and with majority of earthquakes with moderate magnitudes (6.7 < Mw <7.1) ;

(iv)-the depth of hypocenter (h≈135 km) is very close to medium value of all strong Vrancea earthquakes.



25.5 26.0 26.5 27.0 27.5

Lon (E )

45.0

45.5

46.0

46.5

Lat(

N)

VR

1940 11 10

1977 03 04

1990 05 311990 05 30

1986 08 30

Scara de magnitudinedupa 1900

3.0-3.43.5-3.94.0-4.4 4.5-4.9 5.0-5.4 5.5-5.9 6.0-6.4 6.5-69 >7

2004 10 27

The fall plan solutions of Vrancea strong earthquakes (Nov. 10,1940-Mw =7.7; March 4,1977-Mw =7.4 ;August 30,1986-Mw =7.1 and May 30,1990-Mw =6.9)



Epicenters E (line AB) and Io points (line A'B') at distance Do =23 km, corresponding to the four strong and major earthquakes (Mw ≥ 6.9)

occurred in the last 70 years


Using the peak ground accelerations (PGA), recorded during strong Vrancea earthquakes on March 4,1977(Mw =7.4), August 30,1986(Mw =7.1), May 30,1990(Mw =6.9) and May 31, 1990(Mw =6.4) and the corresponding macroseimic intensities, there are the following relations[6]:

Log amax (cm/s2) =0.2712 I +0.1814Log amax..res (cm/s2)= 0.2714 I +0.2085 for V ≤ I ≤ IX

From Table , the isoseismal map of the ”etalon”(reference) earthquake is given in Figure and by using the attenuation curves for 30 azimuths (Az).



Site local amplifications were evaluated by using Very Hard Rock(VHR), Joint Source Site Determination (JSSD) methods or by using down-hole seismic measurements performed in Bucharest(Table 3), Iasi, Timisoara etc.

Also, deterministic evaluations of local effects in metropolitan area Bucharest were made by using hybrid and analitical methods.

The amplifications of PGA during of Vrancea earthquake on Oct. 27, 2004 (Mw =6.00) in other 6 boreholes made in Bucharest City with deep sensors at -153m,-78m,-70m,-68m,-66m,- 52m and shallow sensors, respectively, at -24m,-28m,-30m,-28m,-28m and -28 m (Table) are more representative . All of them are real data. There is a large scattering of the amplifications between PGA recorded by deep ,shallow and surface/free field K2 sensors. The ratio is between 2.115 and 5.708 (average 3.912).




Attenuation curves(30) corresponding for the “etalon”earthquake for Az=0, 15, 30,45,60,75… x-axis is epicenter distance D, respectively, hypocenter distance, R and on y-axis is de ration between macroseismic intensity I in a point on the Earth's surface and the maximum macroseismic intensity, Io (named “Io point”) on line A'B'.



Isoseismal map of the „etalon “earthquake



In order to prepare the results obtained for the „etalon (reference) earthquake”to other intermediary-depth Vrancea earthquakes, we have to take into consi-deration the depth factor of the focus, which is different from one earthquake to another. Firstly, we tried to use a relation containing a function f(h):

Log I – Log Io = a-b Log [R/f(h)] Log (I/Io) =a'– bLog R

a'=a +b Log f(h)and using the data attenuation curves ,we computed by using the least square method the coefficients a' and b from relation and the coefficient a=a' –b Log f(h) for different azimuths Az. The family of attenuation curves for the “etalon” earthquake ,in other form, has the following equation:

Log(Ie/Ioe)(Az) =a(Az)-b(Az Log [Re /f(he )] Log(Ie/Ioe)(Az) = a'(Az)-b(Az) Log Re

where: Ioe is the maximum intensity of the “etalon” earthquake; Ie is the intensity of the “etalon” earthquake at a hypocenter distance Re (along de direction defined by azimuth AZ): a' (Az) = a(Az) +b (Az) Log f(he) ; a & b are constant coefficients for a given Az; h=135 km is focal depth of the “etalon”earthquake.



To obtain the “banana” shape of the attenuation curves of the macroseimic intensity I along the direction defined by azimuth Az ,in the case of Vrancea deep earth-quakes at a depth 80 ≤ x < 160 km, has the following equation :

Log Ix,Az = Log Iox+Log (Ie /Ioe)Az +b(AZ) Log{[Re /f(he)] / [Rx/f(x)]}Az +[cLog δ]Az

where : Iox is the maximum intensity of an earthquake at a depth h= x; δ is the directivity factor of the rupture propaga-tion in the focus ; „c” is the way how the directivity factor may influence the directivity of the seismic source;

Re=(D2+he2 )1/2 ; Rx =(D2+ x2 )1/2 ;D is distance between the observation point and the point of maximum macroseismic intensity.



The isoseismal map of the maximum credible Vrancea earthquake or maximum probable Vrancea earthquake (MW = 7.7)



The isoseismal map of the maximum credible Vrancea earthquake or maximum probable Vrancea earthquake and crustal ones (Mw = 7.7)


At local level it is asked to any city to make microzoningseismic maps (local hazard maps) to put into evidence the differences which occurs during a strong earthquake.

The sustainable design of structures asks a local hazard map (microzonation map). If we refer to the specific objectives of this field regarding the Earth physics research, the natural disasters and the environment, the evaluation, prognosis and monitoring methods of the earth phenomena, in Romania there is a

Gov. Law no.372/2004 – National Program of the Seismic Risk Management

where is specified (pg.5) the objective number 10, named “Macrozoning of the Romanian territory and seismic hazard zoning in densely populated urban cities.



Bucharest,Timisoara,Arad,Sibiu,Craiova,Ploiesti,Buzau,IasiBacau, Galati & Tulcea Cities, the analyses were carried out by using the two alternative approaches for the seismic hazard assessment:

(i)-the probabilistic approach and,(ii)-the (neo) deterministic one, linear and nonlinear

methods by developing the nonlinear seismology conceptIn any site, first steps were to make: (i)-identification of local and Vrancea events that can

occur and have adverse consequences; (ii)-estimation of the likelihood of those events occurring;(iii)-estimation of the potential consequences.



In the (neo)deterministic analysis there are determined the maximum credible Vrancea earthquake, the etalon or controlling earthquake effects at the site. At high distances from the source (far field), wavelength like, and the main contribution at the solution is given by Rayleigh and Love modes. An original method, called EERA, created by NIEP and University of Trieste,Department of Earth Sciences, , especially for Vrancea particular strong earthquakes, consists in realistic and quick evaluation of seismic effects in the selected site.

The seismic signal coming from the bedrock was computed by using modal summation method until it reaches “soft” layers; after that, a 1D analysis was carried out on vertical propagation through local geological structure by using various modifications on software.



These analyses were made by using a linear viscoelasticmodel or by using a nonlinear viscoelastic one (the equivalent solid approach), case in which are necessary the parameters: dynamic torsion functions, G=G(γ) and damping function, D=D(γ) for each of the rock/soil sample, γ=shear strain determined in Hardin and Drnevichresonant columns at NIEP. Even more, deterministic evaluation used on the path focus-bedrock-free surface of the site, after a “calibration” for a recorded earthquake in each site where it was possible. This calibration helped us to “building and correction of structures” on the path focus-bedrock-free surface of the site, even if nonlinear effects for maximum possible earthquake were reevaluated by taking into account the nonlinear relations between dynamic amplification factors and magnitude , determined by NIEP.




Aspects on Fuzzy Set Theory concept which occurs on interpretation of seismic hazard maps in probabilistic analysis

A problem that arises in such representations is that the uncertainties wherewith are evaluated quantities which are subjec-ted to research, inherent uncertainties to physical nature of them.

To approach these issues propagation phase indeterminacy of the "input" to the final outcome had become possible in last time on account of „Fuzzy Set Theory” extension to this field of research which have been relatively few attempts so far.

The fundamental concept of the „”Fuzzy Set Theory is that it replaces the relationship of affiliation from classical theory of sets by a „affiliation function", and the consequences of such fundamen-tal changes proved to be extraordinary not only on action plan of the mathematical operators but also in mathematical logic performed



The models which have been developed to assess the likelihood that some elements of soil particle movement beyond the level considered critical (or dangerous) for a given area requires processing of informations such as: (i)-depths of focus ;(ii)-geometry and focal mechanism; (iii)-the direction of active faults/fault style ;(iv)- earthquake magnitudes in the area of interest; (v)-relationships between magnitude and rupture length ;(vi)-frequency-magnitude distr ibution of earthquakes ; (vi i)-the average return periods etc.

All these information contributes to the assessment of probability P (Y> y) for a given location (A) and in a given period of time (T), the value of parameter Y (which may be: macroseismic intensity, acceleration, velocity ,ground displacement etc.) to exceed a given value „y” as a result of an earth-quake no matter where they are its source. In terms of classical theory of sets, to determine whether a point x is on the fault lies in making a statement P (x) on the relation of affiliation between point „x” and the set of all fault points:

F = { x / P(x)} where: P(x) is the proposition: „ point x belongs to the fault plane”. This sentence can take only two values: A (true) or F (false).



In various fields of Geophysics are studied geographical distributions of values of certain geophysical quantities (as is the case of seismic hazardvalues) and in practice it is a representation of these distributions in the form of „Maps with isolines”.

According to the classical definition currently applied in all scientific papers, „The izoline Γv for parameter (for example : intensity,accelerationsetc) M ” is the locus of points Pi ( ϕi ;λi ) on a certain area ( Σ ) for which the respective parameter (M ) has the same value (v):

Γv = { Pi ( ϕi ; λi ) | Pi ( ϕi ; λi ) ∈Σ şi M (ϕi ; λi ) = v = constant} (i)

where: ϕi and λi are geographical coordinates of point Pi (i = 1, 2, ..., n), all the n points belongs to surface Σ (whose dimensions can be as high to those where the local maps are built , at continental and even global scales )

In all such representations, indifferent of studied parameter from geophysics (acceleration, intensity ,displacement etc.), appears as a common definition that the relationship itself (i) for the "izoline" concept has to be interpreted as such fuzzy concepts, that is, as „Fuzzy Set Theory”.



Let’s consider a general case in which the values of the studied parameter have measurable errors ε. This means that the measured value (V) can have any value in the range (k - ε , k + ε) where k is the average size measured (in the meaning of the theory of errors of measurement):

V ∈ (k - ε ; k + ε ) (ii)It follows that the set of points which can be attributed „k”

measured value to that parameter consists not in a curve Γv given by definition (i) ,but a "band" centered on Γv curve, and bounded by the envelope curves (Figure ) corresponding to the interval edges (ii). This confirms the nature of fuzzy (that is, fuzzy, vague…) of sets of points and their associated values meaning that the affiliation a point to isoline has not to be considered as a relation of affiliation in clas-sical sense but as a series of transitional situations, degrees of affiliations with values between 0 and 1.



-8 -4 0 4 80

2

4

6

8

I = K

I = K - 0,5

I = K + 0,5

Fuzzy nature of maps with isolines

ICTP Advanced Conference on Seismic Risk Mitigation

Trieste-Italy. May 10-14,20106.1955.7080.0508 g0.0082 g0.0089 gV

2.4052.4590.0445 g0.0185 g0.0181 gEW

4.7072.4590.0546 g0.0116 g-30 m0.0126 g-70 mNSMunicipal Hospital

6

2.8053.4480.0331 g0.0118 g0.0096 gV

2.0953.5590.0790 g0.0377 g0.0222 gEW

1.7952.2580.0298 g0.0166 g-28 m0.0132 g-52 mNSCity Hall5

3.0363.8640.0340 g0.0112 g0.0088 gV

3.7562.5360.0492 g0.0131 g0.0194 gEW

2.2832.2830.0290 g0.0203 g-28 m0.0127 g-68 mNSCivil Prot.Headquarter

4

3.0003.7160.0249 g0.0083 g0.0067 gV

2.3652.5960.0296 g0.0125 g0.0114 gEW

2.1372.6280.0297 g0.0139 g-24 m0.0113 g-153 mNSNCSRR/INCERC

3

2.1573.5430.0248 g0.0115 g0.0070 gV

2.4351.7400.0409 g0.0168 g0.0235 gEW

1.9262.6670.0416 g0.0216 g-28 m0.0156 g-66 mNSUTCB2

3.0993.5100.0344 g0.0111 g0.0098 gV

4.000 2.5280.0584 g0.0146 g0.0231 gEW

1.2252.1150.0349 g0.0285 g-28 m0.0165 g-78 mNSUTCB1

3:23:1321

The amplificationPGAPGADepthPGADepth

Surface sensor/free fieldShallow sensor Deep sensorStation N0


Site local amplifications were evaluated by using Very Hard Rock(VHR), Joint Source Site Determination (JSSD) methods or by using down-hole seismic measurements performed in Bucharest(Table 3), Iasi, Timisoara etc. Also, deterministicestimation of local effects in metropolitan area Bucharest weremade by using hybrid and analitical methods.

The amplifications of PGA during of Vrancea earthquake on Oct. 27, 2004 (Mw =6.00) in other 6 boreholes made in Bucharest City with deep sensors at -153m,-78m,-70m,-68m,-66m,- 52m and shallow sensors, respectively, at -24m,-28m,-30m,-28m,-28m and -28 m (Table) are more representative . All of them are real data. There is a large scattering of the amplifications between PGA recorded by deep ,shallow and surface/free field K2 sensors. The ratio is between 2.115 and 5.708 (average 3.912).




600-7001830-230050-200Gravel and sandFratesti gravel7

375-4601660-200060-140Marl and fine sands

Marl Complex M6

350-4001200-19008-20Fine sandsMostistea sand5

320-4501650-20500-20ClaysIntermediate clay4

300-3501200-16954-18Gravel and sandColentina gravel3

210-320485-7503-20Silty clayLoesslike deposits2

90-180200-4000-8-Backfill, topsoil1

VS (m/s)VP (m/s)

Interval velocityThickness(m)

Soil typeLithologicalComplex

No.Layer

Interval velocities of the VP and VS waves for the lithological complexes, determined through seismic profile, up-hole and down-hole seismic

measurements



Boreholes and seismic stations on Bucharest metropolitan area



Bucharest. Local geological structure until Marl Complex- 4 layers-GIS modeling


Bucharest.Local geological structure until Fratesti gravel -7 layers -GIS modellig




Values of seismic waves VP in layer 1 of superficially geological structure of Bucharest metropolitan area -GIS modeling



Values of seismic waves VS in layer 1 of superficially geological structure of Bucharest metropolitan area-GIS modeling



Values of seismic waves VP in layer 7 of superficially geological structure of Bucharest metropolitan area-GIS modeling



Values of seismic waves VS in layer 7 of superficially geological structure of Bucharest metropolitan area-GIS modeling



Aki wrote:„Nonlinear amplification at sediments sites appears to be more pervasive than seismologists used to think…Any attempt at seismic zonation must take into account the local site condition and this nonlinear amplification”. In other words, the seismological detection of the nonlinear site effects requires a simultaneous understanding of the effects of earthquake source, propagation path and local geological site conditions. The difficulty for seismologists in demonstrating the nonlinear site effects has been due to the effect being overshadowed by the overall patterns of shock generation and propagation.



Nonlinear normalized curves for sand with gravel from Hardin andDrnevich resonant columns (NIEP)



-0. 06430.37901. 0005.89420.3790.064331.05.1990MGR = 6.1

14.160. 15400.79571.14165.16290.6970.135030.05.1990MGR = 6.7

45.570. 10700.43381.45574.04890.2980.073630.08.1 986MGR = 7.0

%a*(g)maxSa* (g)c=

(SAF)1990/(SAF) later

Samax/amax

(SAF)

Samax(g)amax(g)

(recorded)Earthquake

If we maintain the same amplification factor (SAF=5.8942) as for relatively strong earthquake on May 31,1990 with MGR = 6,1 then at Bacau station for earthquake on May 30, 1990 (MGR =6.7) the peak acceleration has to be a*max = 0.154g(+14.16%) and the actual recorded was only, amax =0.135g.Also, for Vrancea earthquake on August 30,1986, the peak acceleration has to be a*max = 0.107g (+45,57%) instead of real value of 0.0736 g recorded at Bacau seismic station. 1.Bacau Seismic Station



14.160. 09200.3301.0003.58690.3300.09230.05.1990MGR = 6.7

32.940. 15420.41691.32942.69820.3130.11630.08.1 986MGR = 7.0

%a*(g)maxSa* (g)cSa

max/amax

(SAF)

Samax(g)amax(g)

(recorded)Earthquake

Sa* (g) and a*(g) are the maximum spectral acceleration, respectively, maximum acceleration if the system would have a linear response (behavior) to fundamental period. Vrancea earthquake on May 31,1990 (MGR =6.1) could be assumed that the response is in elastic domain and then we have the possibility to compare to it. Coefficient „c” is the ration between SAF for May 31,1990 Vrancea earthquakes and SAF for each earthquake before.

2.Bucharest-Magurele Seismic Station



The effect of nonlinearity in Bucharest - Magurele seismic station



1.862.131.581.821.501.7820%2.692.922.142.562.992.4310%3.484.162.953.632.693.265%4.846.223.725.583.614.742%

Svmax /vmaxSa

max/amaxSv

max /vmaxSamax/amaxSv

max /vmaxSamax/amax

May 31,1990,MGR= 6.1May 30,1990,MGR =6.7Aug 30,1986,MGR=7.0ξ =%

The median values of the SAF of the last strong Vrancea earthquakes

The median values of the spectral amplification factors of the last strong Vranceaearthquakes for damping ξ =5% are: 4.16; 3.63 and 3.26 corresponding to May 31, 1990, May 30,1990, respectively, August 30, 1986. At the same seismic station, for example at Bacau, for 5% damping, SAF for accelerations is 5.22 for May 31,1990 (M =6.1);4.32 for May 30, 1990 (MGR =6.7) and 3,94 for August 30,1986(MGR =7.0) earthquakes.



The dependence of the spectral amplification factor (SAF) of Vranceaearthquake magnitude



26.04 26.06 26.08 26.1 26.12 26.14 26.16

44.36

44.38

44.4

44.42

44.44

44.46

44.48

44.5

44.52

44.54

1.81.922.12.22.32.42.52.62.72.82.933.13.23.33.43.53.63.73.83.9

INC1

MET1MTR1

MLT1PND1

TIT1

BUC

OTP1

26.04 26.06 26.08 26.1 26.12 26.14 26.16

44.36

44.38

44.4

44.42

44.44

44.46

44.48

44.5

44.52

44.54

3.73.753.83.853.93.9544.054.14.154.24.254.34.354.44.454.54.554.64.654.74.754.84.85

INC1

MET1MTR1

MLT1PND1

TIT1

BUC

OTP1

Non-filtered (OTO 1) Otopeni Seismic Station 0.005-1.0 Hz



26.04 26.06 26.08 26.1 26.12 26.14 26.16

44.36

44.38

44.4

44.42

44.44

44.46

44.48

44.5

44.52

44.54

3.053.13.153.23.253.33.353.43.453.53.553.63.653.73.753.83.853.9

INC1

MET1MTR1

MLT1PND1

TIT1

BUC

OTP1

26.04 26.06 26.08 26.1 26.12 26.14 26.16

44.36

44.38

44.4

44.42

44.44

44.46

44.48

44.5

44.52

44.54

2.32.42.52.62.72.82.933.13.23.33.43.53.63.73.83.944.14.2

INC1

MET1MTR1

MLT1PND1

TIT1

BUC

OTP1

Space distribution of the SAF values for each frequency domain in Bucharest area from Vrancea earthquake on August 30,1986 (Mw =7.1

0.05- 2 Hz Otopeni Seismic Station 0.05- 5 Hz



During of last stronger Vrancea earthquakes (August 1986,Mw= 7.1 and May 30,1990,Mw = 6.9) the highest values of acceleration were recorded in the north of Bucharest and the smallest one in the south-east of it. More, PGA in epicenter area (Vrancioaia) during of August 30, 1986 Vrancea earthquake (Mw =7.0) was 162,6 cm/s2 , in the north of Bucha-rest was 156,2 cm/s2-EREN Station and 219,8 cm/s2 Otopenisite , while in the south-east of it was only 76,3 cm/s2 ( Metalurgieisite).All time we were looking to get “banana” shape for distribution of macroseismic intensity Vrancea earthquakes. Also, in our attention was to make the frequency dependent PGA/PGV analysis in generation of attenuation laws. For rapid assessment of the seismic ground motion level in Bucharest, we used the equation:Ln(amax )= 4.726 + 0.976 MGR – 1.146 ln R - 0.0066 h + 0.353 P

and setting P=0 (average curve) in Equation → Table



102.05 cm/s298.70 cm/s2195.48173.53906.7May 30, 1990

95.55 cm/s295.30 cm/s2188.78123.251437.0Aug. 30, 1986

199.00 cm/s2198.00 cm/s2148.91100.371107.2March 4, 1977

(amax) computed by using Equation

(amax ) recorded

R(km)Δ(km)Depth(km)

MGREarthquake

The predictions for this station show the smallest error with respect to the observations for all stations in the Bucharest area, hence, INCERC is used as a “reference station in all studies[21]. In the microzoning map (local seismic hazard) for Bucharest for maxi-mum possible earthquake in Romania (MGR=7.5).There are 14 zones and each zone is characterized by maximum acceleration (a-max ), maximum MMI intensity (Imax ) and fundamental period for soil (T). Each parameter for each zone is obtained from “reference station” INCERC



Otopeni 300cm/s I=VIII ½T=0.67s

2 PERIS300cm/s I=VIII ½T=0.90s

2

TITULESCU240cm/s I=VIIIT=1.10s

2 1/5

CARLTON240cm/s I=VIII

T=1.00s2

1/5

ISPH240cm/s I=VIII T=1.05s

2 1/5

INCERC300cm/s I=VIII ½T=1.56s

2

BALTA ALBA250cm/s I=VIII 1/4 T=0.8s

2

ABATOR200cm/s I=VIII T=0.9s

2

METALURGIEI200cm/s I=VIII T=1.5s

2 METROU200cm/s I=VIII T=1.52s

2

PANDURI280cm/sI=VIII

2

4/5 Dr. Sarii

280cm/sI=VIII

T=0.90s

2

4/5

MILITARI280cm/sI=VIII 4/5T=0.90s

2

MAGURELE300cm/s I=VIII ½T=1.3s

2

Bolintin Vale300cm/s I=VIII 2 ½T=1.2s

Branesti260cm/s I=VIII 3/5T=1.10s

2

EREN300cm/s I=VIII½T=0.8s

2

-50 cm/s 2T=0.8 1.07s÷

Izone = IINCERC ± Δ I (MMI scale)

(a-max )zone = (a-max) INCERC ± Δ a (cm/s2)

(Tsoil )zone = ( Tsoil )INCERC ± Δ Tsoil

(seconds )

Local seismic hazard (microzonation) map of Bucharest with:(i)- MMI intensities (I); (ii)-PGA(cm/s2) and, (iii)- fundamental periods of soils (T) for maximum credible Vrancea

earthquake (MGR =7.5) in Romania



0 1 2 3 4 5Period [s]

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

1400

1500

1600

1700

1800

1900

SA (0

%) [

g/10

]

VR77VR86VR901

0 1 2 3 4 5 6 7Period [s]

0

100

200

300

400

500

600

700

SA 5

%da

mpi

ng [g

/10]

INC comp NSVR77VR86VR901

Response acceleration spectra for last Vrancea earthquakes : March 4,1977(Mw =7.4) T=1.54 s), August 30,1986(Mw =7.1)-T=1,22 s) and May 30,1990 (Mw =6.9) -T=0.47 s

recorded at INCERC -Bucharest (N-S component) for ξ =0% (left).For ζ =5% (right) all periods are smaller.



August 30,1986(Mw=7.1) recorded at Bucharest-Magurele. Response spectra on N-S component has 2 “locks” with periods T1=0,38 seconds and T2=1,5 seconds for ζ=5% and T1=0,32 seconds and T2=1,65 seconds for ζ= 0,0%. Second “lock” is

developing more when Vrancea earthquake magnitude is increasing.



Distribution of the fundamental soil periods in Bucharest forstrong Vrancea earthquakes.



21.0 21.5 22.0 22.5 23.0 23.5 24.0 24.5 25.0 25.5

Longitudine E

47.0

47.5

48.0

48.5

49.0

Latitu

din

e N

4.1 5.2 6.1 5.8 5.1 4.5 4.0 3.4 2.9 2.1

4.8 6.5 7.6 7.6 7.4 6.7 5.9 5.1 4.1 3.1

4.2 5.2 6.1 6.9 7.5 7.6 7.5 6.3 4.6 3.4

3.0 3.6 4.2 4.7 5.4 6.1 6.4 5.2 4.1 3.0

1.7 2.2 2.7 3.2 3.6 4.1 4.1 3.7 3.0 2.22.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0

Seismic hazard map for local and Vrancea deep earthquakes -Crisana-Maramures-Baia Mare area for a recurrence period , T=100 years (Intensities MMI)



7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.39 7.39

7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.39

7.37 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38

7.37 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38

7.37 7.37 7.37 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38

7.37 7.37 7.37 7.37 7.38 7.38 7.38 7.38 7.38 7.38 7.38 7.38

7.37 7.37 7.37 7.37 7.37 7.37 7.37 7.38 7.38 7.38 7.38 7.38

7.37 7.37 7.37 7.37 7.37 7.37 7.37 7.37 7.37 7.37 7.38 7.38

23.54 23.55 23.56 23.57 23.58 23.59 23.60 23.61 23.62 23.63 23.64 23.65

Long(E)

47.63

47.64

47.65

47.66

47.67

47.68

47.69

47.70La

t (N

)

Local seismic hazard(microzonation) for Baia Mare City for a recurrence period T=100 years (MMI) by using classical methods and „Fuzzy Set Theory “



20.5 21.0 21.5 22.0 22.5 23.0

longitudine E

44.0

44.5

45.0

45.5

46.0

46.5

latit

udin

e N

4.00 4.10

4.18 4.64

4.67 5.53

4.99 5.86

4.99 5.86

5.17 4.88 4.99

4.82 5.33

6.03 5.24 5.50

5.92 5.73 5.42

5.02 5.31 5.27

4.00

4.25

4.53

4.67

4.72

4.71

4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

20.5 21.0 21.5 22.0 22.5 23.0

longitudine E

44.0

44.5

45.0

45.5

46.0

46.5

latit

udin

e N

4.07 4.33

4.59 5.11

5.13 6.02

5.44 6.36

5.47 6.37

5.65 5.36 5.49

5.30 5.80

6.50 5.71 5.97

6.40 6.19 5.89

5.46 5.77 5.77

4.18

4.69

4.99

5.12

5.18

5.19

a. H1 an =0.02; Tr=50 years b.H1 an =0.01; Tr=100 years

Seismic hazard map for local and Vrancea deep earthquakes –Banat countyfor a recurrence periods : T= 50 and 100 years( MMI).Probablistic anaysis.



5.10 5.45

5.68 6.25

6.25 7.43

6.58 7.71

6.73 7.76

6.97 6.67 6.87

6.54 7.11

7.76 6.89 7.30

7.73 7.40 7.21

6.59 7.06 7.16

5.37

5.87

6.07

6.19

6.32

6.45

Seismic hazard map for Banat zone for recurrence period : TR = 1000 years(Intensities MMI). Probabilistic analysis.



Local seismic hazard(microzonation) map (in frequencies) of Timisoara City



5.3 5.7

6.0 6.7

6.6 8.2

7.0 8.4

7.3 8.4

7.6 7.3 7.5

7.0 6.9 7.8

8.4 7.3 8.1

8.4 8.0 7.9

7.1 7.7 7.8

6.5 6.4 6.8 7.3 7.8 7.3

7.3 8.1 8.4 8.0

8.0 8.4 8.0 7.3

7.3 7.5 7.1 6.7

6.9 6.5 6.3 6.1

6.6

6.8

6.9

7.1

Seismic hazard map for Banat, Transilvania and North West part of Muntenia area for a recurrence period , T=1000 years (MMI) to local and deep Vrancea eartquakes



26.00 26.02 26.04 26.06 26.08 26.1044.90

44.92

44.94

44.96

44.98

45.00

8.50 8.50 8.50 8.50 8.50 8.51 8.51 8.51 8.51 8.51 8.51

8.51 8.51 8.51 8.51 8.51 8.52 8.52 8.52 8.52 8.52 8.52

8.52 8.52 8.52 8.52 8.52 8.53 8.53 8.53 8.53 8.53 8.53

8.53 8.53 8.53 8.53 8.53 8.54 8.54 8.54 8.54 8.54 8.54

8.54 8.54 8.54 8.54 8.54 8.55 8.55 8.55 8.55 8.55 8.55

8.55 8.55 8.55 8.55 8.55 8.56 8.56 8.56 8.56 8.56 8.56

Local seismic hazard (microzonation) for Ploiesti City for a recurrence period , T=100 years (MMI). Probabilistic analysis. Local and deep Vrancea

eartquakes



0

100

200

300

400

22 23 24 25 26 27 28 29 30

44

45

46

Craiova

Rosiori

Pitesti

Alexandria

Zimnicea

Giurgiu

OltenitaCalarasi

UrziceniSlobozia

Ploiesti

Buzau

Braila

Galati

Focsani

BucurestiCernavoda

Carpatii Meridionali

Platforma Moesica de Vest

Platforma Moesica

Moesia Sud-Vest

M.S-E

Platforma Scitica

Bulgar

Seismic hazard map for Muntenia and south of Moldova (in design accelerations) to strong Vrancea earthquake (h=90 km ,Mw =7.2)



6.52

6.50

6.49

6.48

6.46

6.45

6.43

6.41

6.39

6.38

6.35

6.54

6.53

6.52

6.50

6.49

6.47

6.45

6.43

6.42

6.39

6.37

6.56

6.55

6.53

6.52

6.50

6.49

6.47

6.45

6.43

6.41

6.38

6.58

6.56

6.55

6.54

6.52

6.51

6.49

6.47

6.45

6.42

6.41

6.60

6.59

6.57

6.56

6.54

6.52

6.50

6.49

6.46

6.44

6.42

6.62

6.60

6.59

6.57

6.56

6.54

6.52

6.50

6.48

6.45

6.43

6.63

6.62

6.60

6.59

6.57

6.56

6.53

6.52

6.49

6.47

6.44

6.65

6.63

6.62

6.60

6.59

6.57

6.55

6.53

6.51

6.48

6.46

6.66

6.65

6.63

6.62

6.60

6.58

6.56

6.54

6.52

6.49

6.47

6.68

6.66

6.65

6.63

6.61

6.59

6.57

6.55

6.53

6.50

6.48

6.69

6.67

6.66

6.64

6.63

6.60

6.59

6.56

6.54

6.52

6.49

24.10 24.11 24.12 24.13 24.14 24.15 24.16 24.17 24.18 24.19 24.2045.70

45.71

45.72

45.73

45.74

45.75

45.76

45.77

45.78

45.79

45.80

Local seismic hazard(microzonation) for Sibiu City for a recurrence period , T=100 years (Intensities )-Probabilistic analysis. Fagaras local eartquakes .



13 13 13 26 36 37 37 37 37 26

20 37 47 58

7 21 21 52 192

6 24 24 36 39

32 23 79 84 84 84 84 84 31 8 24 21

17 86 86 109 110 110 110 110 110 77 77 5 6 6 6 22 25 12

50 72 75 75 75 75 75 75 75 52 8 6 22 22 22 25 25

50 51 68 76 75 75 75 75 75 76 52 34 7 28 22 22 22 25

47 50 51 69 76 76 76 76 76 53 53 45 7 28 28 22 22 25

46 59 67 67 67 67 67 53 53 45 4 28 22 22 26

2 22 22 20 59 59 59 59 59 54 29 21 28 28

Seismic hazard map for Sibiu zone to crustal eartquakes from Fagaras-Campulung .All data are in units of gravitational accelerations(g)



26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5 30.0

Longitude

43.5

44.0

44.5

45.0

45.5

46.0Latitude

2.8 4.2 5.5 5.6 6.4 7.4 6.3 4.7 3.4

2.8 4.0 5.0 5.1 5.2 5.6 5.2 4.2 3.1

3.6 4.7 5.1 4.3 3.6 3.7 3.5 3.0 2.2

4.3 5.0 4.5 3.5 2.8 3.6 4.4 2.9 1.6

4.0 4.4 4.4 3.9 4.3 4.5 3.6 2.4 1.2

2.7 3.7 4.2 3.9 3.3 2.5 1.8 1.0 0.4

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

Seismic hazard map for crustal earthquakes in Dobrogea-Tulcea area for a recurrence period , T=200 years (Intensities MMI)



28.4 28.45 28.5 28.55 28.6 28.65 28.7 28.75 28.8

RAD

45.1

45.15

45.2

170175180185190195200205210215220225230235240245250

28.4 28.45 28.5 28.55 28.6 28.65 28.7 28.75 28.8

VER

45.1

45.15

45.2

62

66

70

74

78

82

86

90

94

98

102

106

110

114

118

28.4 28.45 28.5 28.55 28.6 28.65 28.7 28.75 28.8

TRA

45.1

45.15

45.2

142144146148150152154156158160162164166168170172174176178180182184

Local seismic hazard map with space distribution of PGA in Tulcea City and surroundings to strong deep Vrancea earthquake of magnitude Mw =7.5.

Deterministic analysis.



28 28.02 28.04 28.06 28.08 28.1 28.12 28.14 28.1645.6

45.62

45.64

45.66

45.68

45.7

45.72

45.74

300305310315320325330335340345350355360365370375380385390395400405

Seismic hazard map –Galati City in accelerations to Vrancea deep earthquakes with magnitude MW =7.7 .Deterministic analysis



4.3 4.7 5.1 5.4 5.4 5.1 4.5

4.0 4.3 4.7 5.2 5.3 5.1 4.4

3.7 4.0 4.2 4.3 4.4 4.3 4.1

3.4 3.6 3.7 3.8 3.8 3.7 3.6

3.2 3.3 3.3 3.4 3.4 3.3 3.3

46.0

46.5

47.0

47.5

48.0

Latit

udi

ne

N

25.5 26.0 26.5 27.0 27.5 28.0 28.5Longitudine E

3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0

Seismic hazard map –Moldova county to crustal earthquakes and recurrence period, T=100 years (Intensities MMI)



0.0704

0.0706

0.0708

0.0710

0.0712

0.0700

0.0702

0.0704

0.0706

0.0707

0.0696

0.0698

0.0700

0.0702

0.0703

0.0693

0.0694

0.0696

0.0698

0.0699

0.0689

0.0690

0.0692

0.0694

0.0695

0.0685

0.0686

0.0688

0.0690

0.0691

0.0681

0.0683

0.0684

0.0686

0.0688

0.0677

0.0679

0.0680

0.0682

0.0684

28.77 28.78 28.79 28.80 28.81 28.82 28.83

Longintudine

45.16

45.17

45.18

45.19

Latitudine

Local probabilistic hazard map –Tulcea City. Exceed probabilities - I=VII



26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5

longitude

43.5

44.0

44.5

45.0

45.5

latitu

de

4.03 4.18 4.55 5.06 5.91 6.72 6.14 5.19

4.03 4.39 4.94 5.06 5.28 5.57 5.36 4.91

4.22 4.76 5.26 4.79 4.65 4.68 4.61 4.41

4.49 5.00 4.83 4.45 4.34 4.45 4.71 4.19

4.46 4.80 4.80 4.52 4.59 4.65 4.25 4.03

4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

26.0 26.5 27.0 27.5 28.0 28.5 29.0 29.5

longitude

43.5

44.0

44.5

45.0

45.5

lati

tud

e

4.60 4.94 5.32 5.85 6.71 7.53 6.95 6.00

4.61 4.94 5.44 5.67 6.07 6.36 6.15 5.69

4.69 5.17 5.64 5.30 5.38 5.48 5.41 5.21

5.01 5.44 5.26 4.98 4.94 5.05 5.29 4.84

5.01 5.33 5.33 5.06 5.12 5.17 4.76 4.48

4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0

H1 an =0.005; Tr=200 ani H1 an =0.002; Tr=500 ani

Seismic hazard maps –Dobrogea countyfor all crustal earthquakes and recurrence periods: T=200 and 500 years (Intensities MMI).



Conclusions1.The Carpathians- great variations in lithosphere structure have

led to pronounced weakness and the localization of strains in adjacent basins. The strong deep Vrancea earthquakes are localized to a restricted area in the bending zone between the Eastern and Southern Carpathians at least three units in contact: the East European plate, Intra- Alpine and Moesian sub-plates. The strain transfer path is coming from the active Adriatic, Aegean and Vrancea deformation fronts through the ALCADI-Pannonian System;

2.The seismicity of Romania comes from the energy that is released by crustal earthquakes , which have a depth not more than 40 km, and by the intermediate earthquakes coming from Vrancea region (unique case in Europe) with a depth between 60 and 200 km;



3.The last zonation seismic map, existing since 1993, has areas where seismic intensities are sub-evaluated (e.g. Dobrogea, Banat etc.), and other areas are over-evaluated. The chapter regarding „Seismic action” from „Norms for construction building-2004”, developed by UTCB-Dept. of Concrete Structures and unaccepted by National Institute for Earth Physics, proposed design acceleration values far from credible ones;

4.The fundamental unacceptable point of view is that this design code is in peak ground accelerations (PGA) which generates a lot of drawbacks to civil structural designers and to insurance companies which are paying all damages and life loses in function of earthquake intensity;



5.National Institute for Earth Physics from Bucharest developed the concept of “control” earthquake. The earthquake on August 30, 1986 is used by us as “control” earthquake in all next studies as it fulfils the following states:(i)-it was strong (MGR =7.0); (ii)-it was recorded in a lot of seismic station on Romanian territory; (iii)-the fall plan solutions are very closed (almost identically) to other Vrancea stronger earthquakes (November 10,1940; MGR =7.4 and March 4, 1977; MGR =7.2) and with majority of earthquakes with moderate magnitudes (6.5 < MGR <7.0);(iv)-the depth of oh hypocenter (h≈135 km) is very close to medium value of all strong Vrancea earthquakes.(v)-the shape of attenuation curves are like „concave banana”



6.The maximum credible Vrancea earthquake has the Richter magnitude , MGR =7.5 and this is the magnitude used by us for our Cernavoda Nuclear Plant. The isoseismal map of the maximum possible Vrancea earthquake is obtained finally by using all data known so far and used o combination of probabalistic and deterministic approaches. Also, nonlinear amplification at sediments sites appears to be more important and any attempt at seismic zonation must take into account the local site condition and this nonlinear amplification. It is important in microzonation studies to have the space distribution of the spectral amplification factors values for each frequency domain in Bucharest area for strong Vrancea earthquakes ;



7.Site local amplifications were evaluated by using Very Hard Rock (VHR), Joint Source Site Determination (JSSD) methods or by using down-hole seismic measurements performed in Bucharest, Iasi, Timisoara etc. Also, deterministic estimation of local effects in metro-politan area Bucharest ,Timisoara, Baia Mare, Sibiu,Tulcea, Moldova,Galati etc.were made by using hybrid and annalytical methodsdeveloped at University of Trieste,Department of Earth Sciences;

8.In the microzoning map (local seismic hazard) for Bucharest for maximum possible earthquake in Romania (MGR = 7.5) there are 14 zones and each zone is characterized by maximum acceleration (a-max ), maximum MMI intensity (I-max ) and fundamental period for soil (T). Each parameter for each zone is obtained from “reference station” INCERC. The fundamental periods of soils in Bucharest have values between 1.3 seconds and 1.9 seconds;



9.For probabilistic seismic hazard evaluation for Crisana-BaiaMare area and Baia Mare City we used „Fuzzy Set Theory “concept. The fundamental concept of the „”Fuzzy Set Theory is that it replaces the relationship of affiliation from classical theory of sets by a „affili-ation function", and the consequences of such fundamental changes proved to be extraordinary not only on action plan of the mathematical operators but also in mathematical logic performed.

10.The unsolved problem is in connection to intensity scale for intermediary-depth Vrancea earthquakes… There is not consistency/concordance between physical parameters (PGA,PGV, PGD) of earthquake and them effects to buildings, people etc.

→MMII scale for intermediary-depth



Thank you for your attention !

2142-4 advanced conference on seismic risk mitigation and...

Documents