As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method
Rob ert DUDA1, * and Pawe³ KOWALCZYK1
1 AGH Uni ver sity of Sci ence and Tech nol ogy, Fac ulty of Ge ol ogy, Geo phys ics and En vi ron men tal Pro tec tion, al. A. Mickiewicza 30, 30-059 Kraków, Po land
Duda, R., Kowalczyk, P., 2019. As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method. Geo log i cal Quar terly, 63 (1): 126–138, doi: 10.7306/gq.1454
As so ci ate ed i tor: Tatjana Solovey
We ana lyse the fac tors used for as sess ing ground wa ter in trin sic vul ner a bil ity to pol lu tion in the mean res i dence time es ti ma - tion method, pro vid ing a fi nal vul ner a bil ity eval u a tion. The fol low ing fac tors were ana lysed: depth to shal low ground wa ter, ef - fec tive pre cip i ta tion in fil tra tion co ef fi cient, ter rain in cli na tion, vol u met ric wa ter con tent of soils and rocks in the un sat u rated zone and vol u met ric wa ter con tent of the top soil. GIS sur veys were per formed for two geomorphologically di verse re gions: a high land piedmont and a low land plain in Po land (Cen tral Eu rope). In both cases, ground wa ter had spa tially di verse vul ner a - bil ity to con tam i na tion. The re search method used re lied on de ter min ing the per cent age par tic i pa tion of the area with par tic u - lar val ues of the pa ram e ters ana lysed in ar eas of dif fer ent de grees of vul ner a bil ity. Knowl edge of the ex tent and dis tri bu tion of vari abil ity of the pa ram e ters ana lysed in ar eas of par tic u lar de grees of vul ner a bil ity helps ex plain the causes of spa tial vari - a tion in ground wa ter in trin sic vul ner a bil ity to con tam i na tion in given ar eas.
Key words: ground wa ter vul ner a bil ity, in trin sic vul ner a bil ity, res i dence time, travel time, MRT, GIS.
INTRODUCTION
The in trin sic vul ner a bil ity of ground wa ter to con tam i na tion is af fected by the depth to ground wa ter, net re charge (ef fec tive in - fil tra tion), im pact of the vadose zone (li thol ogy), soil me dia, the vol u met ric wa ter con tent of the soils or rocks in the un sat u rated zone, to pog ra phy, aqui fer me dia and hy drau lic con duc tiv ity of soils or rock in aqui fers. When ground wa ter oc curs in karst and frac tured-karst aqui fers, the vul ner a bil ity value de pends on the epikarst, pro tec tive cover and prop er ties of over ly ing lay ers above the wa ter ta ble, in fil tra tion con di tions, pre cip i ta tion, karst-net work de vel op ment and the con cen tra tion of flow (Doerfliger et al., 1999; Daly et al., 2002; Vias et al., 2006).
The par tic i pa tion of these pa ram e ters in the as sess ment of ground wa ter in trin sic vul ner a bil ity var ies, there fore it is as - sumed in nu mer ous rat ing meth ods, also called para met ric meth ods or in dex meth ods of vul ner a bil ity eval u a tion, i.e. qual i - ta tive meth ods. In these meth ods, the in flu ence of par tic u lar fac tors on the fi nal vul ner a bil ity value re sults from in di vid ual weights and prop erly as sumed rat ing scores. Al though the re - sults ob tained with the in dex meth ods are qual i ta tive, they are
still com monly used world wide, es pe cially via the very ef fec tive DRASTIC method (e.g., Krogulec, 2006; Ró¿kowski, 2007;
Saidi et al., 2011, 2017; Khemiri et al., 2013; Krogulec and Trzeciak, 2016). This method has been im proved and ad justed to com plex hydrogeological con di tions (Witkowski et al., 2003;
Yu et al., 2012; Jiménez-Ma drid et al., 2013; Hernández-Espriú et al., 2014; Kazakis and Voudouris, 2015; Bonfanti et al., 2016;
He et al., 2018). Com bined meth ods gen er at ing a qual i ta tive re - sult are also used (Gemitzi et al., 2006; Civita, 2010).
Apart from qual i ta tive meth ods, a purely quan ti ta tive method can be ap plied. It is based on the eval u a tion of a phys i - cal pa ram e ter, i.e. mean res i dent time (MRT), also called mean travel time (MTT), i.e. time in which wa ter per co lates with con - ser va tive sol utes through the un sat u rated zone. Con ser va tive sol utes nei ther change their con cen tra tion dur ing their trans port from the sur face down wards to ground wa ter (e.g., in the course of biodegradation) nor un dergo re tar da tion (e.g., as a re sult of ad sorp tion). The trans port time is eval u ated by as sum ing the
“pis ton-flow” model, in which the con ser va tive pol lut ant moves down wards at the av er age ve loc ity of the per co lat ing wa ter (Bachmat and Collin, 1987; Hennings, 2000; Witczak et al., 2007; Wachniew et al., 2016). Mean de notes that the time value was not de ter mined as a point but rep re sents math e mat i cally and spa tially av er aged val ues of par tic u lar fac tors ac counted for in the es ti ma tion of time. The av er ag ing is ne ces si tated by the need to gen er al ize the 3D vari abil ity of the un sat u rated zone, i.e. het er o ge ne ity and ani so tropy, and also the hy dro dy - namic dis per sion of the pol lut ant mass, which al ways oc cur in
* Cor re spond ing au thor, e-mail: duda@agh.edu.pl
Re ceived: August 24, 2018; ac cepted: January 8, 2019; first pub lished on line: March 14, 2019
real con di tions, and which vary spa tially. More im por tantly, the ground wa ter vul ner a bil ity is spa tially eval u ated with GIS on con - tour maps, de lin eated on the ba sis of var i ous data orig i nally ob - tained of dif fer ent scale, res o lu tion and ac cu racy of mea sure - ment. The un cer tainty level is con sid er able; there fore, we can - not be sure to de ter mine the time, and so we only es ti mate this time. Anal o gous to the rat ing meth ods, this method also ac - counts for the spa tial vari abil ity of pa ram e ters, and the use of GIS is in ev i ta ble.
The eval u a tion of ground wa ter vul ner a bil ity based on a time scale, i.e. a mea sur able phys i cal pa ram e ter, al lows one to eas - ily and clearly as sess or val i date the re sult by us ing en vi ron - men tal trac ers (Zuber et al., 2005, 2011; Newman et al., 2010;
Wachniew, 2015; Hagedorn et al., 2018). As so ci ated with land-use, in di cat ing po ten tial and real sources of con tam i na - tion, this ap proach also shows the time in which the pre ven tion mea sures can be planned and im ple mented (Bachmat and Collin, 1987). How ever, it is strongly bur dened with sub jec tiv ity and es sen tially dif fi cult to prop erly as sume the MTR in ter val for par tic u lar in trin sic vul ner a bil ity de grees. This is sue has not been fi nally agreed upon yet. Try ing to be ob jec tive, Witczak et al. (2007, 2011) con cen trated on the qual i ta tive, de scrip tive clas si fi ca tion of in trin sic vul ner a bil ity given by Fos ter et al.
(2002), as sum ing the fol low ing clas si fi ca tion of in trin sic vul ner a - bil ity: very high for MRT <5 years, high for MRT 5–25 years, mod er ate for MRT 25–50 years, low and very low for MRT >50 years. There was no ev i dence of fit ting par tic u lar MRT in ter vals to the clas si fi ca tion pro vided by Foster et al. (2002). Witczak et al. (2011) as sessed MRT with:
MRT = MRTs + MRT1 + MRT2 [1]
where: MRT – the travel time of con ser va tive sol utes through of the un sat u rated zone [years]; MRTs – the travel time of con ser va tive sol - utes through top soil [years]; MRT1 – the travel time of con ser va tive sol utes through per me able soils and rocks [years]; MRT2 – the travel time of con ser va tive sol utes through low-per me able soils [years];
and:
MRTS R
=1000 15´. ´qS [2]
( ) ( )
[ ]
MRT d F
R
lp p
1
1000 15 1
=
´ - . ´ - ´q [3]
( )
[ ]
MRT d F
R
lp lp
2
1000 15
= ´ - . ´ ´q [4]
where: qs, qp, qlp – vol u met ric wa ter con tent of: soil pro file, per me - able rocks and low-per me able rocks [dimensionless], re spec tively (see Ta ble 1); d – depth to ground wa ter [m], Flp – par tic i pa tion of low-per me able strata in the un sat u rated zone, R – av er age an nual ef fec tive pre cip i ta tion in fil tra tion (i.e. re charge rate) [mm/a], 1000 – re cal cu la tion mea sure co ef fi cient, 1.5 – as sumed top soil pro file thick ness in metres.
The av er age an nual ef fec tive pre cip i ta tion in fil tra tion, i.e.
the re charge rate in the area ana lysed, can be de ter mined with any method pro duc ing a spa tially vari able re sult, e.g. the in fil tra - tion co ef fi cient method (Staœko et al., 2012; Tarka et al., 2017;
Gumu³a-Kawêcka et al., 2018). Witczak et al. (2011) es ti mated the re charge rate tak ing into ac count its spa tial vari abil ity rep re - sented by a few pa ram e ters:
R=P´a b g´ ´ ´d [5]
where: P — av er age an nual rain fall [mm/a]; a – ef fec tive pre cip i ta - tion in fil tra tion co ef fi cient de pend ing on the li thol ogy of near-sur face strata [–] (see Ta ble 1); b – co ef fi cient de pend ing on the ter rain in cli - na tion [–]; g – co ef fi cient de pend ing on the land cover [–]; d – co ef fi - cient de pend ing on the depth to ground wa ter [–].
A method based on the res i dence time (travel time) was also used in other ground wa ter in trin sic vul ner a bil ity as sess ments (Maxe and Johansson, 1998; Krogulec, 2006; Herrmann et al., 2012; Potrykus et al., 2018). De spite the dif fer ences in the vul - ner a bil ity clas si fi ca tion and the ef fect of par tic u lar fac tors taken into ac count in the meth ods com pared, at tempts were made to di rectly com pare vul ner a bil ity re sults ob tained with a quan ti ta tive method and the re sults of some rat ing meth ods (Ravbar and Goldscheider, 2009; Hermanowski and Ignaszak, 2017). There As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 127
Ef fec tive pre cip i ta tion in fil tra tion co ef fi cient [%]
Vol u met ric wa ter con tent of the soils and rocks
in the un sat u rated zone [%] Vol u met ric wa ter con tent of the top soil [%]
shales, siltstones, mudstones 9
elb ae mrep
gran ites, meta mor phic
rocks, etc. 1 loose sand, silty loose sand,
weakly loamy and silty sand 12 loams, land slide colluvia;
frac tured rocks 13 lime stones, dolomites 2 light loamy sand, silty sand,
sandy silt 17
loess, silts, sandy loams; sand -
stones, siltstones, mudstones 15 sand stones, marls*, chalk* 5 light and silty loam, loamy silt 24
sand stones 17 sandy-gravel sed i ments 10 me dium and silty loam, heavy
and silty loam, clayey silt 36 sands, grav els and sandy loams
of ter mi nal mo raine 22 silty and loamy sands 20
sands and grav els; very frac tured rocks - lime stones, dolomites, marls, sand stones
26 low per me able soils - loess, silts, loams 30
* – Witczak et al. (2011) as sumed that in these rocks the down ward trans port of pol lut ants mainly takes place through the frac tures, thus dif fu sion to the rock ma trix can be ig nored
T a b l e 1 The val ues of some fac tors af fect ing ground wa ter vul ner a bil ity adopted by Witczak et al. (2011) to es ti mate the MRT
are also meth ods mak ing use of nu mer i cal mod el ing to as sess vul ner a bil ity through the age of wa ter (Neukum and Azzam, 2009; Yu at al., 2014) as well as cou pling travel-time es ti ma tion and the rat ing method (Pisinaras et al., 2016).
When as sess ing ground wa ter in trin sic vul ner a bil ity to pol lu - tion, the de gree to which par tic u lar pa ram e ters of the MRT- based es ti ma tion par tic i pate in the range of ar eas of the given vul ner a bil ity de gree is not usu ally ana lysed. More over, the de - gree of this par tic i pa tion can dif fer de pend ing on the con di tions of par tic u lar aqui fer oc cur rence. The geo mor phol ogy of the area for which the ground wa ter vul ner a bil ity anal y sis was per - formed is one of the cri te ria on the ba sis of which con di tions of aqui fer oc cur rence were dif fer en ti ated. Ac cord ingly, this study is aimed at de ter min ing the par tic i pa tion of fac tors ac counted for in the as sess ment of ground wa ter vul ner a bil ity with the MRT es ti ma tion method in the ar eas of given vul ner a bil ity de gree, ac count ing for ar eas that are geomorphologically dif fer ent.
Knowl edge con cern ing the val ues and the dis tri bu tion of di ver - si fied par tic i pa tion of these pa ram e ters in ar eas of par tic u lar vul ner a bil ity de gree will help ex plain and prop erly in ter pret causes of spa tial dif fer en ti a tion of the in trin sic vul ner a bil ity of given aqui fers.
STUDY AREA
The source ma te rial used in this in ves ti ga tion was the da ta - base and GIS pro ject in which ground wa ter in trin sic vul ner a bil - ity in Po land was as sessed and mapped by Witczak et al.
(2011). Two re gions of dif fer ent geo mor phol ogy were used for the anal y ses. The first of these (de noted as S) was the up per part of the Vistula River ba sin of up land and piedmont char ac ter with a sur face of 43,109 km2 (Fig. 1). The other re gion (N) cov - ered 32,301 km2 and is a part of the mainly ag ri cul tural Wielkopolska low land. Both the up land-piedmont and low land plain re gions have geomorphological con di tions typ i cal of most land ar eas. The re gion S is also geomorphologically di verse with fore-moun tain high land (H), up land (U) and plain (P).
The geomorphological di ver sity of the re gion S re sults from its geo log i cal struc ture. In the south of the area, there are the Carpathians built of folded sand stones and shales (Cre ta - ceous/Paleogene). A flat plain (P) oc curs in the area of the Carpathian Foredeep filled with Mio cene de pos its over lain by po rous Qua ter nary de pos its. The north ern part is an up land (U), which con sists of old and par tially eroded moun tains built of fis sured quartz ite, lime stones and dolomites (Cam brian/De vo - nian), as well as lime stones and dolomites (Tri as sic/Ju ras sic).
Within the area of S, there is lo cally loess. The re gion N is cov - ered by Pleis to cene de pos its - these are usu ally al ter nat ing lay - ers of var i ous types of flu vio gla cial sands and loams.
The spa tial vari abil ity and ranges of val ues of ana lysed fac - tors used by Witczak et al. (2011) in the MRT es ti ma tion, i.e.
depth to shal low ground wa ter, av er age an nual pre cip i ta tion, ef - fec tive pre cip i ta tion in fil tra tion co ef fi cient, ter rain in cli na tion, vol - u met ric wa ter con tent of the top soil and wa ter con tent of the soils and rocks in the vadose zone in re gions S and N have been il lus trated in Fig ures 2 and 3, re spec tively, and in Ta ble 1.
The val ues of these fac tors were not de ter mined as part of the study de scribed herein. The val ues of fac tors af fect ing the in - trin sic vul ner a bil ity of ground wa ter given in Ta ble 1 were adopted by Witczak et al. (2011) as typ i cal of the val ues re - ported in the lit er a ture. The ef fec tive pre cip i ta tion in fil tra tion co - ef fi cient for soils and rocks is usu ally de ter mined on the ba sis of em pir i cal mea sure ment in a lysimeter or in an ap prox i mate method of wa ter bal ance in a catch ment (Staœko et al., 2012).
The vol u met ric wa ter con tent of the top soil, soils and rocks is usu ally de ter mined by lab o ra tory test ing of soil or rock sam ples, or in the field by geo phys i cal meth ods (Evett et al., 2008).
The val ues of fac tors af fect ing ground wa ter vul ner a bil ity which pre vail in in di vid ual re gions and sub-re gions and the pre - dom i nant de gree of vul ner a bil ity are sum ma rized in Ta ble 2.
METHOD
In the GIS-based anal y ses, ranges of ar eas with par tic u lar ground wa ter vul ner a bil ity lev els (very high, high, mod er ate, low) were de ter mined for both study re gions. Then the to tal sur face area cov ered by par tic u lar vul ner a bil ity de grees was es tab - lished for each of the ar eas. Sub ar eas dif fer ing in val ues of par - tic u lar fac tors ac counted for in the MRT es ti ma tion were found in each area of given vul ner a bil ity level. The anal y ses cov ered such fac tors as depth to shal low ground wa ter, av er age an nual pre cip i ta tion, ef fec tive pre cip i ta tion in fil tra tion co ef fi cient, ter rain in cli na tion, vol u met ric wa ter con tent of the soils and rocks in the un sat u rated zone and vol u met ric wa ter con tent of the top soil.
Land cover was omit ted in the anal y sis due to the large gen er al - iza tion of this fac tor’s val ues es tab lished in the GIS pro ject da - ta base, which was the source ma te rial for these stud ies (only three types of land cover were adopted).
There was then de ter mined the to tal sur face cov er ing the sub ar eas with par tic u lar val ues of a given pa ram e ter, in clud ing all ar eas where a given vul ner a bil ity level ap peared. Fi nally, the per cent cover of sur face ar eas with par tic u lar val ues of ana - lysed fac tors rep re sent ing suc ces sive vul ner a bil ity lev els was spec i fied. This study was per formed not on the ba sis of data from the test ing (sam pling) net work but in GIS on the ba sis of area (spa tial) data us ing ArcGIS and Quan tum-GIS soft ware.
Fac tor Re gion S
Re gion N sub-re gion H sub-re gion P sub-re gion U
depth to ground wa ter [m] >20 2–5 10–20 2–5
ef fec tive pre cip i ta tion in fil tra tion co ef fi cient [%] 13 17 13 13
ter rain in cli na tion [de grees] >10° 0.5–2° 4–6° 2–4°
vol u met ric wa ter con tent of the top soil [%] 24 24 24 12
wa ter con tent of the soils and rocks in the vadose zone [%] 5 10 5 10
vul ner a bil ity high very high very high high
T a b l e 2 The val ues of fac tors af fect ing the ground wa ter vul ner a bil ity and the as sessed de gree of vul ner a bil ity
which pre vail in the re gions ana lysed (based on Witczak et al., 2011)
RESULTS AND DISCUSSION
The spa tial dis tri bu tions of par tic u lar sub ar eas with par tic u - lar val ues of pa ram e ters af fect ing vul ner a bil ity in re gions S and N are shown, for ex am ple, in spa tial range maps of bor ders of very high ground wa ter vul ner a bil ity ar eas (Figs. 4 and 5). In par - tic u lar maps, ar eas are vis i ble with val ues which most im por - tantly af fect this de gree of vul ner a bil ity. This re fers to a small depth to the ground wa ter ta ble, higher val ues of ef fec tive pre - cip i ta tion in fil tra tion co ef fi cient and lower val ues of ter rain in cli -
na tion, vol u met ric wa ter con tent of the top soil, and soils and rocks in the un sat u rated zone.
The par tic i pa tion of sub ar eas with val ues of pa ram e ters af - fect ing the vul ner a bil ity of ar eas of par tic u lar vul ner a bil ity de - gree in re gions S and N is il lus trated in Fig ures 6 and 7. Axis X1 rep re sents de grees of in trin sic vul ner a bil ity, axis X2 rep re sents pa ram e ter val ues, and axis Y rep re sents the per cent age of sur - face ar eas with par tic u lar val ues of pa ram e ters in re la tion to the to tal sur face area char ac ter ized by a given vul ner a bil ity de gree.
As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 129
Fig. 1. Spa tial vari abil ity of ground wa ter vul ner a bil ity to pol lu tion in the re gions N and S (based on Witczak et al., 2011) and geo log i cal sketch of these re gions (based on Rühle et al., 1954, sim pli fied)
Ap prox i mate bound aries of geomorphologically di verse sub-re gions are marked in the re gion S
Fig. 2A–F – spa tial dis tri bu tion of fac tors af fect ing the ground wa ter vul ner a bil ity in re gion S (based on Witczak et al., 2011) For val ues in pan els C, E and F, see ad di tional ex pla na tions in Ta ble 1; bound aries of ar eas dif fer ing in geo mor phol ogy are ap prox i mate
As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 131
Fig. 3A–F – spa tial dis tri bu tion of fac tors af fect ing the ground wa ter vul ner a bil ity in re gion N – the low land plain (based on Witczak et al., 2011)
For val ues in pan els C, E and F, see ad di tional ex pla na tions in Ta ble 1
Fig. 4A–F – spa tial range of sub ar eas with par tic u lar val ues of fac tors af fect ing the vul ner a bil ity within ar eas of very high vul ner a bil ity – re gion S (based on Witczak et al., 2011)
White colour – ar eas where the vul ner a bil ity is other than very high; bound aries of ar eas of dif fer ent geo mor phol ogy are ap prox i mate
As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 133
Fig. 5A–F – spa tial range of sub ar eas with par tic u lar val ues of fac tors af fect ing the vul ner a bil ity within ar eas of very high vul ner a bil ity – re gion N – low land plain (based on Witczak et al., 2011)
White color – ar eas where the vul ner a bil ity is other than very high
Fig. 6A–F – par tic i pa tion of sub ar eas with par tic u lar val ues of fac tors in flu enc ing vul ner a bil ity in ar eas char ac ter ized by suc ces sive vul ner a bil ity de grees – re gion S
As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 135
Fig. 7A–F – par tic i pa tion of sub ar eas with par tic u lar val ues of fac tors in flu enc ing vul ner a bil ity in ar eas char ac ter ized by suc ces sive vul ner a bil ity de grees – re gion N
The per cent age of sur face ar eas with par tic u lar depth in ter - vals to the ground wa ter ta ble in sur face ar eas of par tic u lar vul - ner a bil ity de gree in re gions S and N is shown in Figures 6A and 7A. In ar eas with very high vul ner a bil ity, sub ar eas where the depth is <2 m and 2–5 m con sti tute 40 and 49% in re gion N, and 44 and 50% in re gion S, re spec tively.
In ar eas of high vul ner a bil ity, the sub ar eas with depths to ground wa ter of 5–10 m and 10–20 m con sti tute 24 and 25% in re gion S, re spec tively. In re gion N, depths of 2–5 m and 5–10 m were ob served in 43 and 37% of sur face ar eas with high vul ner - a bil ity, re spec tively.
In ar eas of mod er ate vul ner a bil ity, sub ar eas with depth to ground wa ter ta ble of 10–20 m were noted in 60% of re gion S. In re search re gion N, sub ar eas with depths of 2–5 m and 5–10 m cov ered 44 and 33% of the sur face, re spec tively. In re gion S, sub ar eas with depth >20 m con sti tuted 86% of the sur face of ar eas with low vul ner a bil ity. How ever, in re gion N, the depth in - ter val of 2–5 m con sti tuted 62% of the sur face of ar eas with low vul ner a bil ity.
In re gion S, there was ob served a re la tion be tween the par - tic i pa tion of ar eas of par tic u lar depth in ter vals to the ground wa - ter ta ble and in trin sic vul ner a bil ity. The greater the par tic i pa tion of sur faces with large depths, the lower is the vul ner a bil ity.
How ever, this de pend ence was not clearly ob served in re gion N, which may be the re sult of the spe cific char ac ter of the un sat - u rated zone in that area. This lo cally shows thick interbeds of low-per me able gla cial loam.
The pro por tion of sub ar eas of par tic u lar ef fec tive pre cip i ta - tion in fil tra tion co ef fi cient val ues in ar eas of suc ces sive vul ner a - bil ity de grees in re gions S and N is shown in Fig ures 6C and 7C, re spec tively. In ar eas of very high vul ner a bil ity, the sub ar - eas hav ing near-sur face sed i ments with an ef fec tive in fil tra tion co ef fi cient of 17 and 22% con sti tute 55 and 12% re spec tively of the sur face area within re gion S, and 31 and 53% re spec tively within re gion N. In ar eas of high vul ner a bil ity, sub ar eas with an in fil tra tion co ef fi cient of 13 and 17% con sti tute 56 and 20% re - spec tively of the sur face area in re gion S, and 50 and 20% re - spec tively in re gion N. In ar eas of low vul ner a bil ity, where the near-sur face sed i ments have this co ef fi cient equal to 9 and 13%, they con sti tute 21 and 73% within re gion S, and 1 and 96% within re gion N, re spec tively.
A re la tion be tween ar eas with par tic u lar val ues of ef fec tive in fil tra tion co ef fi cient with their vul ner a bil ity was ob served in both re gions. The big ger is the pro por tion of area of low co ef fi - cient value, the lower is the ground wa ter vul ner a bil ity in that area.
The pro por tion of sub ar eas with par tic u lar val ues of ter rain in cli na tion in ar eas char ac ter ized by suc ces sive vul ner a bil ity de grees in re gions S and N is shown in Fig ures 6D and 7D. In very high vul ner a bil ity ar eas, sub ar eas with ter rain in cli na tion
<2° and 2–4° con sti tute 84 and 10% of the sur face area in re - gion S, and 90% and 10% in re gion N, re spec tively. In high vul - ner a bil ity ar eas, sub ar eas with in cli na tion an gle <2° and 2–4°
con sti tute 44 and 17% sur face area within re gion S, and 92 and 8% in re gion N, re spec tively. In re gion S, in the low vul ner a bil ity area, sub ar eas with in cli na tion an gle 2–4° and 4–6° con sti tute 23 and 18% of the sur face area, re spec tively.
In re gion S, there was ob served a clear re la tion be tween the pro por tion of sur faces of sub ar eas with a par tic u lar range of ter - rain in cli na tion an gle and their vul ner a bil ity de gree. The smaller is the ter rain in cli na tion an gle, the higher is the in trin sic vul ner a - bil ity. In the low land plain re gion N, the vul ner a bil ity does not de pend on the ter rain in cli na tion be cause the ter rain in cli na tion an gles do not ex ceed 4°.
The pro por tion of sub ar eas with par tic u lar vol u met ric wa ter con tents of soils and rocks in the un sat u rated zone in ar eas char ac ter ized by suc ces sive vul ner a bil ity de grees in re gions S
and N is shown in Fig ures 6E and 7E, re spec tively. In re gion S, sub ar eas with soils hav ing 20% vol u met ric wa ter con tent (silty and loamy sands) con sti tute 90% of the sur face area with mod - er ate vul ner a bil ity and 95% of the sur face area with low vul ner a - bil ity. The sub ar eas with soils hav ing 10% vol u met ric wa ter con - tent (sands with gravel) con sti tute 51% of the sur face of very highly vul ner a ble ar eas. The sub ar eas with rocks of 5% vol u - met ric wa ter con tent (frac tured sand stones, marls and chalk) con sti tute 64% of the sur face area char ac ter ized by mod er ate vul ner a bil ity and 31% of the sur face area with very high vul ner a - bil ity.
In re gion N, sub ar eas with soils of 30% vol u met ric wa ter con tent (loess, silts, loams) con sti tute 62% sur face of ar eas of mod er ate vul ner a bil ity and 95% sur face of ar eas of low vul ner a - bil ity. The sub ar eas with ground of 10% vol u met ric wa ter con - tent (sands with gravel) con sti tute 72% of the sur face area of very high vul ner a bil ity. In both re gions, with the in creas ing par - tic i pa tion of sur face sub ar eas with higher vol u met ric wa ter con - tent val ues of soils and rocks in the un sat u rated zone, the ground wa ter in trin sic vul ner a bil ity de creases.
The pro por tion of sub ar eas with par tic u lar val ues of vol u - met ric wa ter con tent in top soil in ar eas of suc ces sive vul ner a bil - ity de grees in re gions S and N is shown in Fig ures 6F and 7F. In re gion S, the sub ar eas with top soil char ac ter ized by 24% vol u - met ric wa ter con tent con sti tute 40% of the sur face area with very high vul ner a bil ity, 83% sur face area with high vul ner a bil ity and 96% sur face area of mod er ate and low vul ner a bil ity. The sub ar eas with top soil of vol u met ric wa ter con tent of 12% con sti - tute 42% of the sur face area with very high vul ner a bil ity and only 2% sur face of ar eas of mod er ate and low vul ner a bil ity.
In re gion N, the sub ar eas with top soil char ac ter ized by 24%
vol u met ric wa ter con tent con sti tute only 3% of the sur face area with very high vul ner a bil ity, 26% sur face of ar eas with high vul - ner a bil ity, 34% sur face of area with mod er ate and 28% sur face of area with low vul ner a bil ity. The sub ar eas with top soil of 12%
vol u met ric wa ter con tent con sti tute 77% of the sur face area with very high vul ner a bil ity and only 16% of the sur face area with low vul ner a bil ity. In both re gions, there is a re la tion be - tween the in creas ing pro por tion of sur face sub ar eas with higher vol u met ric wa ter con tent of top soil and de creas ing ground wa ter vul ner a bil ity to pol lu tion.
The re sults of this pre lim i nary study should be treated as ap - prox i mate be cause the cred i bil ity of the data used is lim ited. This is mainly due to the high de gree of gen er al iza tion of in put data used by Witczak et al. (2011) to de velop the GIS pro ject of the vul ner a bil ity map at 1:500,000 scale, i.e. at a re con nais sance scale. Con se quently, as sess ment made of the spa tial ex tents of par tic u lar val ues of fac tors af fect ing ground wa ter vul ner a bil ity but de lin eated with greater ac cu racy, e.g. at 1:50,000 or greater, may dif fer some what from the pic ture ob tained in this study.
CONCLUSIONS
This in ves ti ga tion helped to ex plain the fac tors be hind spe - cific spa tial ranges of given in trin sic ground wa ter vul ner a bil ity de grees as sessed in both re gions ana lysed by the MRT es ti ma - tion method. A re la tion is shown be tween in trin sic ground wa ter vul ner a bil ity and some fac tors af fect ing this vul ner a bil ity. In the high land-piedmont re gion S, as depth to ground wa ter in - creases, the vul ner a bil ity of the ground wa ter de creases, and that as the ter rain in cli na tion de creases, the vul ner a bil ity in - creases. In the re gion N, these re la tion ships were not ob served, which may be due to it be ing a plain re gion where in the un sat u - rated zone there are interbeds of low-per me able loam and where there is no ter rain in cli na tion an gle >4°.
In both re gions of study, the in trin sic vul ner a bil ity of ground - wa ter de creases with a de creas ing value of ef fec tive pre cip i ta - tion in fil tra tion co ef fi cient. And, the ground wa ter vul ner a bil ity de creases with an in crease in the vol u met ric wa ter con tent of the soils and rocks in the vadose zone, as well as with in - creased wa ter con tent of the top soil.
In the high land-piedmont re gion S there was no clear cor re - la tion be tween the av er age an nual rain fall and the sus cep ti bil ity of ground wa ter to pol lu tion. The rea son may be that vul ner a bil - ity de pends on the ground wa ter re charge rate, and this in ad di - tion to the amount of rain fall also de pends on the li thol ogy of near-sur face sed i ments and on the ter rain in cli na tion. A re la - tion, al beit im pre cise, be tween the vul ner a bil ity de gree and the ef fec tive in fil tra tion rate, i.e. re charge, was found in area N: with a de crease of the in fil tra tion rate, the ground wa ter vul ner a bil ity de creases.
The geomorphological di ver sity of both re gions, i.e. low land plain and high land-piedmont, partly af fects the vul ner a bil ity vari abil ity, but this fac tor is not cru cial. The am bi gu ity of some of the re sults ob tained re flects the fact that both spec i fied vul ner a - bil ity de grees and their spa tial vari abil ity are a re sult of in ter ac -
tion of all fac tors af fect ing the in trin sic vul ner a bil ity as sess ment in a given area. Such an in ter ac tion of val ues of in di vid ual fac - tors is a re sult of com pen sa tion of the sum of the ef fects gen er - ated by pa ram e ters in creas ing the vul ner a bil ity and low er ing it.
The 3D spa tial dis tri bu tion of in di vid ual fac tors usu ally strongly var ies, and the de gree of this vari abil ity usu ally in creases as the sur face of the ana lysed area in creases, there fore it is dif fi cult to clearly state which fac tor was de ci sive in the re sult of any given ground wa ter vul ner a bil ity as sess ment. More over, the re sults ob tained will be af fected by the fact that most of the MRT ranges used in the clas si fi ca tion of ground wa ter in trin sic vul ner - a bil ity were broad, i.e. 20 years (high vul ner a bil ity) and 25 years (mod er ate vul ner a bil ity).
Ac knowl edge ments. This re search was fi nan cially sup - ported by the AGH Uni ver sity of Sci ence and Tech nol ogy un der grant No. 11.11.140.797. The au thors wish to thank all those in - volved in the re view of this pa per, es pe cially Prof. J. Ró¿kowski, for valu able re marks that sig nif i cantly im proved the value of this ar ti cle.
REFERENCES
Bachmat, Y., Collin, M., 1987. Map ping to as sess ground wa ter vul - ner a bil ity to pol lu tion. In: Vul ner a bil ity of Soil and Ground wa ter to Pol lut ants (eds. W. van Duijvenbooden and H.G. van Waegeningh), TNO Com mit tee on Hy dro log i cal Re search, Hague, Pro ceed ings and In for ma tion, 38: 297–307.
Bonfanti, M., Ducci, D., Masetti, M., Sellerino, M., Stevenazzi, S., 2016. Us ing sta tis ti cal anal y ses for im prov ing rat ing meth ods for ground wa ter vul ner a bil ity in con tam i na tion maps. En vi ron men - tal Earth Sci ences, 75: 1003.
Civita, M.V., 2010. The com bined ap proach when as sess ing and map ping ground wa ter vul ner a bil ity to con tam i na tion. Jour nal of Wa ter Re source and Pro tec tion, 2: 14–28.
Daly, D., Dassargues, A., Drew, D., Dunne, S., Goldscheider, N., Neale, S., Popescu, I.C., Zwhalen, F., 2002. Main con cepts of the “Eu ro pean ap proach” to karst-ground wa ter-vul ner a bil ity as - sess ment and map ping. Hydrogeology Jour nal, 10: 340–345.
Doerfliger, N., Jeannin, P.-Y., Zwahlen, F., 1999. Wa ter vul ner a bil - ity as sess ment in karst en vi ron ments: a new method of defining pro tec tion ar eas us ing a multi-at trib ute ap proach and GIS tools (EPIK method). En vi ron men tal Ge ol ogy, 39: 165–176.
Evett, S., Cepuder, P., Heng, L.K., Hignett, C., Laurent, J.P., Ruelle, P., 2008. Field es ti ma tion of soil wa ter con tent: a prac ti - cal guide to meth ods, in stru men ta tion and sen sor tech nol ogy.
Train ing Course Se ries 30, In ter na tional Atomic En ergy Agency, Vi enna.
Fos ter, S., Hirata, R., Gomes, D., D’Elia, M., Paris, M., 2002.
Ground wa ter qual ity pro tec tion: a guide for wa ter util i ties, mu - nic i pal au thor i ties and en vi ron ment agen cies. The World Bank, Wash ing ton, DC.
Gemitzi, A., Petalas, Ch., Tsihrintzis, V.A., Pisinaras, V., 2006.
As sess ment of ground wa ter vul ner a bil ity to pol lu tion: a com bi - na tion of GIS, fuzzy logic and de ci sion mak ing tech niques. En vi - ron men tal Ge ol ogy, 49: 653–673.
Gumu³a-Kawêcka, A., Szymkiewicz, A., Jaworska-Szulc, B., Pruszkowska-Caceres, M., Gorczewska-Langner, W., 2018.
Pre lim i nary es ti ma tion of ground wa ter re charge on Brda river outwash plain. 10th Con fer ence EKO-DOK 2018, E3S Web of Con fer ences, 44: 00050.
Hagedorn, B., Clarke, N., Ruane, M., Faulk ner, K., 2018. As sess - ing aqui fer vul ner a bil ity from lumped pa ram e ter mod el ing of mod ern wa ter pro por tions in ground wa ter mix tures: ap pli ca tion
to Cal i for nia’s South Coast Range. Sci ence of the To tal En vi ron - ment, 15: 1550–1560.
He, H., Li X.-G., Li, X., Cui, J., Zhang, W., Xu, W., 2018. Op ti miz ing the DRASTIC method for ni trate pol lu tion in ground wa ter vul ner - a bil ity as sess ments: a case study in China. Pol ish Jour nal of En vi ron men tal Stud ies, 27: 95–107.
Hennings, V., 2000. Methodendokumentation Bodenkunde:
Auswertungsmethoden zur Beurteilung der Empfindlichkeit und Belastbarkeit von Böden. Geologisches Jahrbuch, Reihe G, Heft SG 1. Schweizerbartsche Verlagsbuchhandlung, Stuttgart.
Hermanowski, P., Ignaszak, T., 2017. Ground wa ter vul ner a bil ity based on four dif fer ent as sess ment meth ods and their quan ti ta - tive com par i son in a typ i cal North Eu ro pean Low land river catch ment (the Pliszka River catch ment, west ern Po land). Geo - log i cal Quar terly, 61 (1): 166–176.
Hernández-Espriú, A., Reyna-Gutiérrez, A., Sánchez-León, E., Cabral-Cano, E., Carrera-Hernández, J., Martínez-Santos, P., Macías-Medrano, S., Falorni, G., Co lombo, D., 2014. The DRAS TIC-Sg model: an ex ten sion to the DRASTIC ap proach for map ping ground wa ter vul ner a bil ity in aqui fers sub ject to dif - fer en tial land sub si dence, with ap pli ca tion to Mex ico City.
Hydrogeology Jour nal, 22: 1469–1485.
Herrmann, F., Berthold, G., Fritsche, J.G., Kunkel, R., Voigt, H.J., Wendland, F., 2012. De vel op ment of a con cep tual hydrogeolo gical model for the eval u a tion of res i dence times of wa ter in soil and ground wa ter: the state of Hesse case study, Ger many. En vi ron men tal Earth Sci ences, 8: 2239–2250.
Jiménez-Ma drid, A., Carrasco, F., Martínez, C., Gogu, R.C., 2013.
DRISTPI, a new ground wa ter vul ner a bil ity map ping method for use in karstic and non-karstic aqui fers. Quar terly Jour nal of En gi - neer ing Ge ol ogy and Hydrogeology, 46: 245–255.
Kazakis, N., Voudouris, K.S., 2015. Ground wa ter vul ner a bil ity and pol lu tion risk as sess ment of po rous aqui fers to ni trate: mod i fy - ing the DRASTIC method us ing quan ti ta tive pa ram e ters. Jour - nal of Hy drol ogy, 525: 13–25.
Khemiri, S., Khnissi, A., Ben Alaya, M., Saidi, S., Zargouni, F., 2013. Us ing GIS for the com par i son of in trin sic para met ric meth ods as sess ment of ground wa ter vul ner a bil ity to pol lu tion in sce nar ios of semi arid cli mate. The case of Foussana ground - wa ter in the cen tral of Tu ni sia. Jour nal of Wa ter Re source and Pro tec tion, 5: 835–845.
As sess ment of fac tors af fect ing the in trin sic vul ner a bil ity of ground wa ter rated by the mean res i dence time es ti ma tion method 137
Krogulec, E., 2006. Meth ods and re sults of ground wa ter vul ner a bil - ity eval u a tion to con tam i na tion in the Kampinoski Na tional Park, cen tral Po land. Acta Geologica Polonica, 56: 349–359.
Krogulec, E., Trzeciak, J., 2016. DRASTIC as sess ment of ground - wa ter vul ner a bil ity to pol lu tion in the Vistula floodplain in cen tral Po land. Hy drol ogy Re search, 48: 1088–1099.
Maxe, L., Johansson, P.-O., 1998. As sess ing ground wa ter vul ner - a bil ity us ing travel time and spe cific sur face area as in di ca tors.
Hydrogeology Jour nal, 6: 441–449.
Neukum, C., Azzam, R., 2009. Quan ti ta tive as sess ment of in trin sic ground wa ter vul ner a bil ity to con tam i na tion us ing nu mer i cal sim - u la tions. Sci ence of the To tal En vi ron ment, 408: 245–254.
Newman, B.D., Osenbruck, K., Aeschbach-Hertig, W., Sol o mon, D.K., Cook, P., Rozanski, K., Kipfer, R., 2010. Dat ing of
young ground wa ter us ing en vi ron men tal trac ers: ad van tages, ap pli ca tions, and re search needs. Iso topes in En vi ron men tal and Health Stud ies, 46: 259–278.
Pisinaras, V., Polychronis, Ch., Gemitzi, A., 2016. In trin sic ground wa ter vul ner a bil ity de ter mi na tion at the aqui fer scale: a meth od ol ogy cou pling travel time es ti ma tion and rat ing meth - ods. En vi ron men tal Earth Sci ences, 75: 85.
Potrykus, D., Gumu³a-Kawêcka, A., Jaworska-Szulc, B., Prusz - kow ska-Caceres, M., Szymkiewicz, A., 2018. As sess ing ground wa ter vul ner a bil ity to pol lu tion in the Puck re gion (de nu - da tion mo raine up land) us ing ver ti cal seep age method. 10th Con fer ence EKO-DOK 2018, E3S Web of Con fer ences, 44:
00147.
Ravbar, N., Goldscheider, N., 2009. Com par a tive ap pli ca tion of four meth ods of ground wa ter vul ner a bil ity map ping in a Slo vene karst catch ment. Hydrogeology Jour nal, 17: 725–733.
Ró¿kowski, J., 2007. Eval u a tion of in trin sic vul ner a bil ity of an Up - per Ju ras sic karst-fis sured aqui fer in the Jura Krakowska (south ern Po land) to anthropogenic pol lu tion us ing the DRASTIC method. Geo log i cal Quar terly, 51 (1): 17–26.
Rühle, E., Soko³owski, S., Tyska, M., 1954. Mapa geologiczna Polski 1:1 000 000. Instytut Geologiczny, Warszawa.
Saidi, S., Bouri, S., Ben Dhia, H., Anselme, B., 2011. As sess ment of ground wa ter risk us ing in trin sic vul ner a bil ity and haz ard map - ping: ap pli ca tion to Souassi aqui fer, Tu ni sian Sahel. Ag ri cul tural Wa ter Man age ment, 98: 1671–1682.
Saidi, S., Hosni, S., Mannai, H., Jelassi, F., Bouri, S., Anselme, B., 2017. GIS-based multi-cri te ria anal y sis and vul ner a bil ity method for the po ten tial ground wa ter re charge de lin ea tion, case study of Manouba phreatic aqui fer, NE Tu ni sia. En vi ron - men tal Earth Sci ences, 76: 511.
Staœko, S., Tarka, R., Olichwer, T., 2012. Ground wa ter re charge eval u a tion based on the in fil tra tion method. In ter na tional As so - ci a tion of Hydrogeologists, Se lected Pa pers, 17: 189–197.
Tarka, R., Olichwer, T., Staœko, S., 2017. Eval u a tion of ground wa - ter re charge in Po land us ing the in fil tra tion co ef fi cient method.
Geo log i cal Quar terly, 61 (2): 384–395.
Vias, J.M., Andreo, B., Perles, M.J., Carrasco, F., Vadillo, I., Jimenez, P., 2006. Pro posed method for ground wa ter vul ner a - bil ity map ping in car bon ate (karstic) aqui fers: the COP method.
Ap pli ca tion in two pi lot sites in South ern Spain. Hydrogeology Jour nal, 14: 912–925.
Wachniew, P., 2015. En vi ron men tal trac ers as a tool in ground wa ter vul ner a bil ity as sess ment. Acque Sotterranee – Ital ian Jour nal of Ground wa ter, AS 13059: 19–25.
Wachniew, P., Zurek, A.J., Stumpp, Ch., Gemitzi, A., Gargini, A., Filippini, M., Rozanski, K., Meeks, J., Kv³rner, J., Witczak, S., 2016. To wards op er a tional meth ods for the as sess ment of in trin sic ground wa ter vul ner a bil ity: a re view. Crit i cal Re views in En vi ron men tal Sci ence and Tech nol ogy, 46: 827–884.
Witczak, S., Duda, R., ¯urek, A., 2007. The Pol ish con cept of ground wa ter vul ner a bil ity map ping. In ter na tional As so ci a tion of Hydrogeologists, Se lected Pa pers on Hydrogeology, 11: 45–59.
Witczak, S., Duda, R., ¯urek, A., Górski, J., Dragon, K. et al., 2011. Ground wa ter vul ner a bil ity map of Po land, 1:500 000 (in Pol ish with Eng lish trans la tion). Min is try of En vi ron ment, Warszawa. Pub. by AGH Uni ver sity of Sci ence and Tech nol ogy, Krakow. https://www.mos.gov.pl/fileadmin/user_up load/mos/
srodowisko/geologia/publikacje/hydrogeologia/Mapa_wrazliwo sci_wod_podz_na_zaniecz-Podatnosc_Ipoziomwod.pdf Witkowski, A.J., Ru bin, K., Kowalczyk, A., Ró¿kowski, A.,
Wróbel, J., 2003. Ground wa ter vul ner a bil ity map of the Chrzanów karst-fis sured Tri as sic aqui fer (Po land). En vi ron men - tal Ge ol ogy, 44: 59–67.
Yu, C., Zhang, B.X., Yao, Y.Y., Meng, F.H., Zheng, C.M., 2012. A field dem on stra tion of the en tropy-weighted fuzzy DRASTIC method for ground wa ter vul ner a bil ity as sess ment. Hy dro log i cal Sci ences Jour nal, 57: 1420–1432.
Yu, C., Yao, Y., Cao, G., Zheng, Ch., 2014. A field dem on stra tion of ground wa ter vul ner a bil ity as sess ment us ing trans port mod el ing and ground wa ter age mod el ing, Beijing Plain, China. En vi - ronmental Earth Sci ences, 73: 5245–5253.
Zuber, A., Witczak, S., Ró¿añski, K., Œliwka, I., Opoka, M., Mochalski, P., Kuc, T., Karlikowska, J., Kania, J., Korczyñski-Jackowicz, M., Duliñski, M., 2005. Ground wa ter dat ing with 3H and SF6 in re la tion to mix ing pat terns, trans port mod el ling and hydrochemistry. Hy dro log i cal Pro cesses, 19:
2247–2275.
Zuber, A., Ró¿añski, K., Kania, J., Purtschert, R., 2011. On some meth od olog i cal prob lems in the use of en vi ron men tal trac ers to es ti mate hydrogeologic pa ram e ters and to cal i brate flow and trans port mod els. Hydrogeology Jour nal, 19: 53–69.
