s:

Studia Quaternaria, vol. 31, no. 2 (2014): 91 99. DOI: 10.2478/squa-2014-0009 THE ROLE OF GEO PHYS I CAL ERT METHOD TO EVAL U ATE THE LEAKPROOFNESS OF DIAPRAGM WALL OF DEEP FOUN DA TION TRENCHES ON THE EX AM PLE OF THE CON STRUC TION OF RE TAIL AND OF FICE COM PLEX IN LUBLIN, PO LAND Grzegorz Pacanowski 1, Pawe³ Czarniak 1, Anna B¹kowska 2, Rados³aw Mieszkowski 2, Fa bian Welc 3 1 Pol ish Geo log i cal In sti tute Na tional Re search In sti tute, Rakowiecka 4, 00-975 War saw, Poland, e-mails: grzegorz.pacanowski@pgi.gov.pl, pawel.czarniak@pgi.gov.pl 2 Uni ver sity of War saw, Fac ulty of Ge ol ogy, In sti tute of Hydrogeology and En gi neer ing Ge ol ogy, wirki i Wigury 93, 02-089 War saw, Po land, e-mail: anna.bakowska@uw.edu.pl, r.mieszkowski@uw.edu.pl 3 Instytut Archeologii Uniwersytet Kardyna³a Stefana Wyszyñskiego, ul. Woycickiego 1/3, no. 23, 01-938 Warszawa, Poland, e-mail: f.welc@uksw.edu.pl. Ab stract This pa per ad dresses the prob lem of as sess ing the leakproofness of the bot tom of a deep foun da tion trench, se cured by cav ity wall, us ing geo phys i cal meth ods of elec tri cal re sis tiv ity to mog ra phy. The study was con ducted on a large con - struc tion pro ject in Lublin, in a place where there are com pli cated soil-wa ter con di tions: the ground wa ter level is above the pro posed depth of foun da tion trench, the sub soil is het er o ge neous, and there are karsted and weath ered car - bon ate sed i ments with con fined aqui fer be low the bot tom of the trench. A hy drau lic frac ture oc curred at the bot tom of the trench dur ing the en gi neer ing works, which caused the wa ter flow into the trench. In or der to rec og nize the soil-wa ter con di tions the first stage of geo phys i cal mea sure ments of elec tri cal re sis tiv ity to mog ra phy (ERT) was made. The ap plied meth od ol ogy al lowed to de ter mine the ex tent of the hy drau lic frac ture zone within the bot tom of foun da tion trench. In or der to as sess the leakproofness of Di a phragm Wall the geo phys i cal ERT mea sure ments were re peated (stage 2) A clear re duc tion in the value of the elec tri cal re sis tiv ity of soils in the area of hy drau lic frac ture was caused by clay in jec tion. The re sults of ERT mea sure ments are dis cussed and graph i cally pre sented. Key words: Electrical Resistivity Tomography (ERT), hydraulic fracture, clay injection, foundation trench, complicated soil-water conditions. Manu script re ceived 5 May 2014, accepted 6 November 2014 INTRODUCTION Geo phys i cal meth ods be gan to be used to solve geotech ni cal prob lems in the 60s and 70s of the twen ti eth cen - tury (Keller & Frischknecht 1966, Bogoslovsky & Ogilvy 1977), mainly the seis mic method and the elec tri cal re sis tiv - ity method. Cur rently non-in va sive geo phys i cal sur veys are per formed to iden tify soil-wa ter con di tions at the de sign stage of any large con struc tion pro ject. Geo phys i cal in ves ti - ga tion can de ter mine or elab o rate on: geo log i cal struc ture (Keller & Frischknecht, 1966), elas tic soils pa ram e ters and de gree of con sol i da tion of soils (Foti & Lancellotta, 2003), iden ti fi ca tion of bur ied tech ni cal in fra struc ture (Jol et al. 2009), shal low cav i ties in rock mass (Martínez-Pagán et al., 2013), mon i tor ing of the en vi ron ment (Sharma, 2002), hy - drogeolog i cal con di tions (Kirsch, 2009; Bru net et al., 2010), risk of land slides pro cesses (Göktürkler et al., 2008; Hack, 2000), haz ards of min ing ar eas (Cham bers et al., 2007) and many oth ers (e.g. Kowalczyk, Mieszkowski, 2011 and Barski, Mieszkowski, 2014). In the case of con struc tion pro jects that are, or will be sited in com pli cated soil-wa ter con di tions (e.g. shal low groundwa ter level or the im pact of soils of low bear ing ca - pac ity) geo phys i cal mon i tor ing is con ducted. Its pur pose is to con trol the soils be fore and af ter sta bi li za tion works, such as the seal (in jec tion) of clayey slurry or ce ment ing. Geo - phys i cal in ves ti ga tions on struc tures of this type are de - scribed, in the works of Cardarelli et al. (2007), Farooq et al. (2007), Ramirez et al. (1993), Daily & Ramirez (2000), Santarato et al. (2011) and Slat er et al. (2000). The pa per fo cuses on the pre sen ta tion of the meth od ol - ogy and re sults of geo phys i cal ERT method used to de ter -

92 G. PACANOWSKI et al. Fig. 1. View of the NW part of the foun da tion trench. Fig. 3. Sche matic lay out of in jec tion holes around the dam aged piezometer (area of fail ure). Fig. 2. Lo ca tion of the ERT geo phys i cal re search area. mine the hy drau lic frac ture zone at the bot tom of the se lected foun da tion trench and to con trol the leakproofness of soils in the bot tom of the trench af ter the sealed works. CASE-HISTORY In 2012, gi ant of fice and ser vice com plex started to be built in Lublin (south-east ern part of Po land). The area of deep ex ca va tion was ap prox i mately 2.5 ha (Fig. 1). Lo ca tion of the in vest ment is shown in Fig. 2. Level of the foun da tion plate was de signed at a depth of ap prox i mately 10 m be low ground sur face (161.2 m above the sea level). The trench was pro tected by Di a phragm Wall. At the stage of de sign ing the con struc tion a wide range of geotechnical and geological measurements were conducted to iden tify the geo log i cal struc ture and to de ter mine the phys - i cal and me chan i cal pa ram e ters of soils: drillings, static prob - ing (CPT) and ba sic lab o ra tory tests of soil sam ples col lected dur ing drill ing. The in ter pre ta tion of the geo log i cal struc ture was shown in the form of geo log i cal cross-sec tions (Fig. 4) the un ex plored rock and soil mass be tween the bore holes was in ter po lated. On the ba sis of this in ter pre ta tion it was con - cluded that the deep en ing of the foun da tion trench and the per for mance of the bot tom plate will not en coun ter any en vi - ron men tal dif fi cul ties. In the spring 2013, while dredg ing in the north-west ern part of the ex ca va tion, the old pre vi ously un iden ti fied piezometer was dam aged by the ex ca va tor. This piezometer was used to mon i tor the fis sure wa ter level in Cretaceous sediments. Fissure water level in the Cretaceous sediments was under considerable piezometric pressure: it is level was drilled ca at 155 m above sea level, while sta bi lized at ca 166 me ters above sea level and about 5 m above the pro - jected bot tom of the foun da tion trench. As a re sult of fail ure the wa ter rap idly be gan to flood the trench. Dam aged piezome ter was sealed quickly, then large em bank ment was piled high around it to se cure the bot tom of the trench from ex - pand ing hy drau lic frac ture zone and con tin ued flood ing. This ac tion has stopped the flow of wa ter for a pe riod of time but also stopped con struc tion works for sev eral months. It was de cided that the area of fail ure will be sealed by di - a phragm wall made of clayey slurry in jec tions. Sche matic lay out of in jec tion holes around the dam aged piezometer is shown in Fig. 3. In or der to de ter mine the ex tent of hy drau lic frac ture zone it was de cided to con duct geo phys i cal mea sure - ments of sys tem 2D and quasi 3D of elec tri cal re sis tiv ity to - mog ra phy (study of quasi 3D re sults were gen er ated from 2D lines). Af ter clayey slurry in jec tions whithin these zones the con struc tion works have been com pleted with out ma jor problems in 2013. THE GEOLOGY The ge ol ogy of the in vest ment area was iden ti fied by drill ing and static prob ing to a depth of ap prox i mately 24 m (to 148.2 m above sea level). Drill ing and prob ing (in the to - tal amount of 23) was lo cated mainly around the per im e ter of the pro jected di a phragm walls (Fig. 4).The geo log i cal pro file is (from top): anthropogenic em bank ments (of 4 m thick - ness), Qua ter nary or ganic soils (of 4 m thick ness), Qua ter - nary river sands interbedded by al lu vial soils (silty clay and silt of a thick ness from 3 to 6 m), Neo gene clays and silts (of 9 10 m thick ness) and Cre ta ceous car bon ate rocks. The roof

GEO PHYS I CAL ERT METHOD 93 Fig. 4. Sim pli fied geo log i cal cross-sec tion in the area of fail ure (above) and geo phys i cal pic ture of ERT mea sure ment (be low). Geo phys i - cal in ves ti ga tions were per formed af ter re mov ing the top layer of em bank ments and or ganic soils. of Cre ta ceous de pos its is at depth ca 25 m be low ground sur - face (ca 155 m above sea level). The floor of the Cre ta ceous rocks wasn t reached. There are two aqui fers in this pro file: first ground wa ter level, un con fined aqui fer, in Qua ter - nary river sands, at a depth of 7.5 8 m (ap prox i mately 164 m above sea level), sec ond ground wa ter level, con fined aqui fer, in the Cre ta ceous rocks, level oc curred at a depth of about 20 22 m (150 152 m above sea level), while sta bi lized at a depth of 5 6 m (166 m above sea level) Sim pli fied geo log i cal cross-sec tion in the area of fail ure is pre sented in Fig. 4. It should be noted that the lo ca tion of geo log i cal bound aries is al most hor i zon tal, and the dif fer ent geo log i cal lay ers have a sim i lar thick ness. There were no ero sion cuts de tected. THE METHODOLOGY OF ERT MEASUREMENTS Geophysical measurements of electrical resistivity to - mog ra phy (ERT) were con ducted within the foun da tion trench. Elec tri cal re sis tiv ity to mog ra phy method was de vel - oped in the late twen ti eth cen tury and is widely de scribed by many re search ers, for ex am ple by Griffiths & Barker (1993) and Loke a (2012). This method was cho sen be cause of the expected contrasts in the electrical resistivity of soils. Mea sure ments were car ried out in two stages: Stage 1: Ex tent of hy drau lic frac ture zone within the bot tom of the foun da tion trench was spec i fied. Stage 2: Leakproofness of di a phragm wall (clayey slurry in jec tions) made in the bot tom of the trench (on the ba - sis of step 1) was spec i fied.

94 G. PACANOWSKI et al. Fig. 5. Doc u men tary map of ERT mea sure ments. The lo ca tion of mea sure ment pro files are shown in Fig. 5. The frac ture zone is sit u ated in the NW part of the trench (Fig. 5). The mea sure ments were made us ing Terameter LS ap pa - ra tus (man u fac tured by ABEM, Swe den). The di pole-di pole ar ray was used. This ar ray was cho sen for two rea sons: firstly it gives high den sity of mea sure ment points, sec ondly it al - lows the use of multi-chan nel mea sure ments. Rel a tively short time of mea sure ments was very im por tant. In the hy - drau lic frac ture zone (north-west ern part of the trench) there was 2 m in ter val be tween elec trodes and a length of mea sur - ing pro files was lim ited by the length of the foun da tion trench to the 80 m. Thus the depth of rec og ni tion of dis tri bu - tion of elec tri cal re sis tiv ity in the sub soil was up to ap prox i - mately 13 14 m.

GEO PHYS I CAL ERT METHOD 95 Fig. 6. Elec tri cal re sis tiv ity cross-sec tion- stage 1, pro file 14. Fig. 7. Elec tri cal re sis tiv ity cross-sec tion- stage 2, pro file 14 (Ex pla na tions as Fig. 6). Ad di tional mea sure ments were car ried out in the en tire trench area to de velop a map of the roof of Neo gene co he sive deposits. This soil layer was particularly important, because it pro tects the trench from the con fined aqui fer. Ad di tional ERT pro files were also made us ing di pole-di pole ar ray, but the in ter val of the elec trodes was in creased to 3 4 m which gives a slightly greater depth of pros pect ing.

96 G. PACANOWSKI et al. Fig. 8. Cross-sec tion of dif fer ences be tween fig. 6 and fig. 7. RE SULTS RES2Dinv pro gram (Loke, 2012) was used to pro cess the data re ceived. There are clear con trasts in soils re sis tiv ity in the area cov ered by the ERT mea sure ments: a com plex of river sands interbedded by al lu vial soils (silty clay and silt) with resistivities above 50 m, co he sive soils com plex Neo gene clays and silts with low resistivities of 5 to 40 m, hy drau lic frac ture zone within the co he sive soils com - plex of resitivities from 40 to 70 m. Stage I ERT mea sure ments al low to clearly iden tify the hy drau - lic frac ture zone (layer IV Fig. 6). Re fers to pro file 14, at the depth of bot tom of foun da tion trench soils of low re sis tiv ity dom i nate co he sive soils of very low per me abil ity. The hy - drau lic frac ture zone crosses the foun da tion depth in the 40 42 m of pro file length. The value of re sis tiv ity are in - creased in this zone to 60 80 W m and con trasts to the co he - sive soils (5 40 m). In the range of 30 32 m and 65 68 m of pro file length there were also dis tin guished soils of be in creased value of electrical resistivity. There are probably saturated sands. Their pres ence in the bot tom of the foun da tion trench may cause some prob lems dur ing the con struc tion works. Stage II ERT mea sure ments showed that di a phragm wall cov - ered the en tire em bank ment and cohesionless soils founded in the bot tom of the trench (layer II and the sur face layer of V Fig. 6). The re duc tion of the elec tric re sis tiv ity in the area of hy drau lic frac ture zone, en tire area of em bank ment and cohesionless soils in the bot tom of the trench was noted as a re sults of clayey slurry in jec tions (Fig. 7.). Clayey di a phragm wall re duced anom aly IV at 30 32 m

GEO PHYS I CAL ERT METHOD 97 Fig. 9. Elec tri cal re sis tiv ity cross-sec tion- stage 2, pro file 3. Fig. 10. Elec tri cal re sis tiv ity cross-sec tion- stage 2, pro file 4. of pro file length, but in creased re sis tiv ity zone at the 65 68 m of pro file length re mained al most un changed. The comparison of the distribution of electrical resistiv - ity for stage 1 and stage 2, and re sis tiv ity de crease af ter in jec - tion works for se lected pro file is pre sented in Fig. 8. Red color marks the places where there was the great est change in per cent age value of re sis tiv ity, white color where there is the small est change. The ef fect of clayey slurry in jec tions is also noted in the other ERT pro files due to the con trast to the re sis tiv ity of the cohesionless soil. Significant decrease of electical resistivity is noted in the lo ca tion of clay injetions (Figs 9, 10).

98 G. PACANOWSKI et al. Fig. 11. Map of the roof of low-re sis tiv ity soils (co he sive) made on the ba sis of ERT cross-sec tion and ar chi val bore holes. The map of the roof of low-re sis tiv ity soil In ter pre ta tion of all the ERT pro files and anal y sis of bore holes and static probes al lowed to pre pare the map of the roof of low-resistivity sediments (Fig. 11). Low resistivity soil layer cor re sponds to the co he sive soils of very low per - me abil ity. This layer pro tects the trench from the con fined aqui fer. The depth of the roof of low-re sis tiv ity layer is vari - able and it s rang ing from 152 m to 165 m above sea level. Two ar eas where the roof of low-re sis tiv ity sed i ments clearly

GEO PHYS I CAL ERT METHOD 99 de creases were high lighted in Fig. 11. In the area 1 the roof oc curs at ap prox i mately 155 m above sea level this is the place where the hy drau lic frac ture oc curred and the wa ter flowed to the trench. In the area 2 the roof is ap prox i mately 152 m above sea level, at this area a slight flow of wa ter into the trench was ob served, but it was man aged to be con trolled the lo cal dewatering. CONCLUSIONS The res o lu tion of pre sented meth od ol ogy of ERT mea - sure ments was 2 m hor i zon tally, while ver ti cally it was de - pends on the depth and var ies from 1 m (near the surface) to 2 m (at the bot tom part of the cross-sec tion). Due to the accuracy of distribution in electrical resistiv - ity even mi nor anom a lies, slight vari a tion of li thol ogy, hy - drogeo log i cal con di tions, anthropogenic de pos its or other non-doc u mented ef fects of hu man ac tiv ity can be de tected. Thus more de tailed and re li able de scrip tion of geo log i cal struc ture and geo log i cal bound aries may be achieved. The ERT method is rel a tively quick method that en abled the pre cise non-in va sive iden ti fi ca tion of soil and rock lay ers in the ar eas of com pli cated ge ol ogy, where the con ven tional di rect in va sive re search meth ods could not guar an tee the full rec og ni tion of soil-wa ter con di tions. The in ter pre ta tion of the un ex plored rock and soil mass be tween the bore holes may lead to the con struc tion fail ure both at the stage of con - struc tion works (as in the pre sented ex am ple), but also mak ing it even more dan ger ous at the stage of use the com - pleted build ing. REFERENCES Barski, M., Mieszkowski R., 2014. Up per Ju ras sic large-scale de - bris flow de pos its in interbiohermal bas ins of the sponge megafacies in Po land new in sights, Neues Jahrbuch Fur Geologie Und Palaontologie-Abhandlungen Vol. 272/1, pp 47 59 Bogoslovsky, V.A., Ogilvy, A.A., 1977. Geo phys i cal meth ods for the in ves ti ga tion of land slides. Geo phys ics 42, 562 571. Bru net, P., Clément, R., Bouvier Ch., 2010. Mon i tor ing soil wa ter con tent and def i cit us ing Elec tri cal Re sis tiv ity, To mog ra phy (ERT) A case study in the Cevennes area, France Jour nal of Hy drol ogy 380 (2010) 146 153. DOI: 10.1016/j.jhydrol. 2009.10.032 Cardarelli, E., Cercato, M., Di Filippo, G., 2007. As sess ing foun da - tion sta bil ity and soilstructure in ter ac tion through in te grated geo phys i cal tech niques: a case his tory in Rome (It aly). Near Sur face Geo phys ics 5, 141 147. DOI: 10.3997/ 1873-0604.2006026 Cham bers, J. E, Wilkinson, P. B., Weller, A. L., Meldrum, P. I., Ogilvy, R. D, Caunt, S., 2007. Mineshaft im ag ing us ing sur - face and crosshole 3D elec tri cal re sis tiv ity to mog ra phy: A case his tory from the East Pennine Coal field, UK, Jour nal of Ap - plied Geo phys ics 62, pp 324 337. DOI: 10.1016/j.jappgeo. 2007.03.004 Daily, W., Ramirez, A. L., 2000. Elec tri cal im ag ing of en gi neered hydraulic barriers. Geophysics 65(1):83 94Farooq, M., Kim, J.H., Park, S., Song, Y.S., 2007. Non-de struc tive eval u a tion of cementgrout by sur face elec tri cal re sis tiv ity method. Advanced Nondestructive Evaluation II 1, 599 604. Foti, S., Lancellotta, R., 2003. Ca pa bil i ties of seis mic tests in soil char ac ter iza tion. In: Maugeri, M., Nova, R. (Eds.). Gotechnical analysis of seismic vulnerability ofmonuments and histor i cal sites, pp. 83 98. P tron Ed i tor, It aly. Göktürkler, G., Balk Aya, Ç. & Erhan, Z., 2008. Geo phys i cal in ves - ti ga tion of a land slide: The Alt nda land slide Site, zmir (west - ern Tur key). Jour nal of Ap plied Geo phys ics, 65 (2): 84 96. DOI: 10.1016/j.jappgeo.2008.05.008 Griffiths, D. H., Barker, R. D., 1993. Two-di men sional re sis tiv ity im ag ing and mod el ling in ar eas of com plex ge ol ogy. Jour nal of Ap plied Geopgysics, no. 29, pp. 211 226. Hack, R., 2000. Geo phys ics for slope sta bil ity. Sur veys in Geo - phys ics 21, 423 448. Jol, H.M. (Ed.), 2009. Ground Pen e trat ing Ra dar: The ory and Ap - pli ca tion, 1st ed., Elsevier. Kirsch, R., (Ed.), 2009. Ground wa ter Geophisics, A Tool for Hy - dro-ge ol ogy, 2en ed., Springer. Keller, G.V. & Frischknecht F.C., 1966. Elec tri cal meth ods in geo - phys i cal pros pect ing. Pergamon Press Inc., Ox ford. Kowalczyk S., Mieszkowski, 2011. Okreœlanie sp¹gu gruntów organicznych metodami geofizycznymi na przyk³adzie dwóch poligonów badawczych na Ni u polskim, Biuletyn Pañstwowego Instytutu Geologicznego Tom 446/1 r. 2011, str. 191 198 Loke, M. H., 2012. Tu to rial: 2-D and 3-D elec tri cal imag - ining surveys, Geotomo Software, Malaysia. Martínez-Pagán, P., Gómez-Ortiz, D., Martín-Crespo, T., Manteca, J.I., Rosique, M., 2013. The elec tri cal re sis tiv ity to mog ra phy method in the de tec tion of shal low min ing cav i ties. A case study on the Vic to ria Cave, Cartagena (SE Spain), En gi neer ing Geology 156, 1 10. DOI: 10.1016/j.enggeo.2013.01.013 Ramirez, A, Daily, W, Labrecque, D, Owen, E, Chesnut, D.,1993. Monitoring an underground steam injection process using elec tri cal re sis tance to mog ra phy. Wa ter Resour Res 29(1): 73 87Schueremans, L., Van Rickstal, F., Verderickx, K., Van Gemert, D., 2003. Evaluation of masonry consolidation by geo-elec tri cal rel a tive dif fer ence re sis tiv ity map ping.rilem Ma te ri als & Struc tures 36, 46 50. Santarato, G., Ranieri, G., Occhi, M., Morelli, G., Fischanger, F., Gualerzi, D., 2011. Three-di men sional Elec tri cal Re sis tiv ity Tomography to control the injection of expanding resins for the treatment and stabilization of foundation soils, Engineer - ing Ge ol ogy 119, 18 30 pp. DOI: 10.1016/J.ENGGEO.2011. 01.009 Sharma, P.V., 2002. Environmental and engineering geophysics. Cam bridge Uni ver sity Press, Cam bridge, UK. 475 pp. Slat er, L., Binley, A., Daily, W., John son, R., 2000. Cross-hole elec tri cal im ag ing of a con trolled sa line tracer in jec tion. Jour - nal of Ap plied Geo phys ics 44, 85 102.