Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
J/~fKRASNY NG U - BU L L 43 9 , 2 0 0 2 - PAGE 7
Quantitative hardrock hydrogeology in a regional scale
JI A' KRAs NY
Krasny, J. 2002: Quant itative hardrock hydrogeology in a regiona l scale.Norgesgeologiske undersekelse Bulletin 439,7-14.
In the hydrogeological characterisation of an area, two quantitat ive aspectsare to be taken into account, namely th egeological sett ing that defines the geometry and anatomy of a hydrogeological environment, and the spatial andt ime variat ions of available natural groun dwater resources, that most ly depend on recharge possibil it ies.Methodological approaches aiming to quanti fy th ese two aspects in hardrock areas at a regional scale are presented. A hardrock environment can be defined as an intrica te hierarchical system consist ing of inhomogeneouselements of different extent.A regionally prevailing transmissivity, mostly in units m'ld with anomalies ranging from0.1 m'ld up to 100 m'ld , seems to be very similar in hardrock environments th roughout th e world.Considering thisregio nal distribu t ion of tran smissivit y, natural groundwater resources typically depend on th e prevail ing climaticconditions.On a backgro und of the Eart h's basic climatic zonation (arid, humid, temperate zones),a vertical climaticdifferenti ation (mountains,lowlands,etc.) is of imp ortance.Under favourable conditi ons in temperate climatic zonesnatural groundwater resources might reach up to 15 Li s km' in the highest parts of hardrock mountainous areas.
JiNKrosny, CharlesUniversity Prague,Faculty of Science,Albertov 6, 128 43 Praha 2, CzechRepublic.
IntroductionMost of th e present-d ay hydrogeological projects in th e
world are of local character. Issues such as cont aminant
hydrogeology, most water supp ly problems, groundwater
assessment in civil eng ineering and mining projects are of
thi s type.However, results of regional hydrogeological studies are indispensable for administrators and decision mak
ers,and will have a considerable influ ence on any land-useplanning.They also determine th e condit ions of an optimum
sustainable use of groundwater and it s protection in
extended areas. Regional results are also of importance for
hydrogeologists, in enabling them to draw genera l hydroge
ological conclusions and to compare th e cond itions in different areas. In this way, factors causing similarities or differ
ences in parti cular hydrogeological environments can be
considered and discussed at region al, nationa l or internat ionallevels.Methodologies involving hydrogeological data
regionalisation are important especially for hardrock areasas there, in contrast to hydrogeological basins occupied by
stratabound sedimentary aquife rs, regional approaches in
hydrogeolog ical stud ies have not been common.
When characterising the hydrogeology of an area, twoquant itat ive aspects should be taken into account:
• Static properties of the aquifer system: the geometryand anatomy of hydrogeological bodie s and the spatialdist ribution of their hydraul ic characteristics . No hydro
geologica l environment is homogeneous and isotropicunder natural cond itions ,and the hardrock environment
in particular is to be considered as an int ricate hierarch i
cal system of hydrogeolog ical inhomogeneit ies on different scales.
• Geological sett ing determines the character of a hydro
geological environ ment within which groundwater, the
dynamic element that is the object of our att ention,moves and is distributed. Groundwater recharge, which
varies in tim e and space,determines our sustainable natural groundwater resources. Under specific conditions,
induced and/or art ificial resources mig ht com plement
the natur al resources.In th is contribution, th e results of various hydrogeolog i
cal studies are summarised and a meth odology is suggested
which aims to help in th e assessment of reasonable ground
water development in hardrock areas at a regiona l scale.
General characteristics of hardrockenvironmentsGeometry and anatomy of hydrogeologicalbodiesThe geom etry (horizontal extent and thickness) and
anatomy (interna l character, distrib ution and type of poros
ity ) of hydrogeologica l bodies (aquifers and aquitards) are
st rongly dependent on the age and character of the last
important orogenesis and on the general geological devel
opment during the post-orogenic period . The age of therocks and th e tectonic activities control th e degree of diagenesisand the relationship between the intergranular and fis
sure porosity of rocks. Consequently, th ese features det ermine the magnitude and distribution of their hydraulic para
mete rs (t ransmissivity and permeability, storativi ty ).
Comparatively old and folded sedimentary rocks have gen
erally lost most of their primary intergranular poro sity.
Instead of this, fracture porosity prevails. This is especially
the case with bedrock, which is usually represented by crys-
NGU - BULL 439 , 200 2 - PAG E 8
talline (igneo us and metamorp hic) or by sedimenta ry, highly
cemented and/or folded rocks. The common feature of a
hardrock environment, designated as a 'hydrogeological
massif', is a vertical sequence of three zones, termed upp erweathered, middle fractu red and lower massive (Krasny
1996a).As in the caseof the intergranular environment, th e per
meabilit y of fractures and of fault zones typ ically decreases
th rough geolog ical t ime due to different geo log icalprocesses such as hydrothermal alteration, mineral precipi
tati on and mechanical clogg ing (Mazurek 2000).Therefore,
in any hardrock environment geolog ically young fractures
are the most important for groundwat er flow, but as not ed
by Banks et al. (1993), such fractures might not be recorded
by standard field geological and geophysical meth ods.
Relatively solub le rocks such as carbonates,gypsum and saltdepos it s are the only exceptions to thi s general rule as their
permeabiliti es general ly increase with t ime.
Hydraulic parameters for identifyinghardrock environmentsAquifers and aquitards are dist ingu ished both qualit atively
by th eir geomet ry and anato my, and quant itatively by the
magn itude of representat ive hydraulic paramete rs. Thecom monly used coefficient of hydraulic cond uct ivit y (permeabi lity ) was orig inally derive d to characterise an inter
granular and homogeneous environment. Subsequent ly, itwas commonly used in a mechanical senseand wi thout tak
ing into account the origi nal preconditions in interpreting
the results of aquifer tests in other hydrogeolog ical situa
tions . In hydrogeological environments with a prevailing
fracture porosity, however, hydraulic conduct ivity can beused for hydrogeological interpretat ions only very cau
t iously, taking into account scale differences in pathways ofgroundwater flow and the objectives of the hydrogeological
study (Krasny 2001).In contrast to hydraulic conductivity, the coeffic ient of
t ransmissivity is defined as a parameter expressing the
property of the entire thic kness of an aquifer. In a hardrock
environment, permeabil ity typically decreases wi thin a
depth of some tens of metres from the surface. As waterwells sited in such an environment most ly reach depths ofbetween 30 and 60 metres, the ir t ransmissivity can be con
sidered as representi ng the ent ire hardrock aquifer (Krasny1993 a). Consequent ly, transmissivity may well express th e
capab ility of the hydrogeological environment fo r ground
water abstraction. This is generall y th e main objective of
most hydrogeological stud ies regardless as to whet her the
groundwater is considered as a natu ral resource for water
supply,a nuisance factor dur ing diffe rent unde rground construct ions, or a t ransport medium for the dispersion of cont
amin ants.Results of thousands of pumping tests available world
wide in hardrock environments have enabled a quantitative
and standard ised approach for study ing transmissivity dls-
JlflfKRASNY
t ribut ion under various condi tio ns. In many cases, how ever,
only data of a less reliable character are available in archives.
Such data are typically not suitable for determining the
exact hydraulic parameters as the coefficient of transmissiv
ity.On the other hand, many of these data are good enough
for providing ideas on t ransmissivity distribution and couldbe treated stat ist ically.On this basis,an 'index of t ransmissiv
ity Y' [ = log (10· q) where q = specific capacity in Us m] was
intro duced as one of the comparat ive regional parameters(Jetel & Krasny 1968).This facilitated a statistical treatment
of th e available data from pump-tested wells,and the objec
t ive classification of transmissivity magnitude and variation
is now the principal procedure for draw ing important con
clusions regarding transmissivity spatial distribution.
Statistica l treatment and classification of transmissivity dataWhere there exist suffic ient transmissivity data for the distri
but ion to be examined by stat ist ical techn iques, sample
popul ations based on lithological units, geog raphical areas,
etc., can be characterised by a central value and a stat ist ical
parameter expressing the variation in the distribution. If one
makes no assumpt ions abou t the natu re of the distri buti on,non-paramet ric paramete rs such as 'median' and 'interquar
ti le range' are best suited for this pur pose. Several authors,however, have noted that distr ibution s of well yield, specific
capacity and transmissivity tend to approxima te to a lognormal distr ibut ion (e.g.,Jetel & Krasny 1968, Banks 1998).lf
one assumes that such distribut ions are log-no rmal, they
may be characterised by the geometric mean and the stan
dard deviation of log-t ransformed values. As the index oftransmisivity Y is already log-transfo rmed, its distribution
can be characterised by the arithmetic mean and standarddeviation.
Transmissivity data distr ibut ion of particular statis tical
samples can be graphically represented on probability
paper by cumu lative relative frequencies of t ransmissivity
values (Fig. 1).Treated values can be expressed by the index
of transmissivity Y, by the coeffic ient of t ransmissivity T,or by
the specific capacity q.The last two parameters should then
be expressed in a logarithmic form .
By using th is procedure, the range of prevailing transmissivity values x ± s ( x = arith met ic mean, s = standarddeviation of a statistical sample) represent the t ransmissivity
backgro und of a stat ist ically treated hydrogeolog ical envi
ronme nt. Transmissivity values outside the transmissivity
background are considered as out lying data - posit ive or
negative anomalies (+A, -A). The far outl iers, or ext reme
anomal ies, positiv e (++A) and negative (- -A), can be found
outside the interval x± 2s (Fig. 1).
To classify transmissivity occurring in different hydrogeological environments, the whole range of possible trans
missivity values was separated into six classes represent ing
the orders of transmissivity magnitude introd uced by
Krasny (1993a) (Table 1). A class (or more than one class) of
JlfiiKRASNY NG U -B UL L 4 3 9 , 2002 - PA G E 9
+A
x +s
'" '"'"+' +' Xrx IX
x-s
-A
z-zs
--A
CLASSES OF TRAN SMISSIVITY MAG NITUDE
VERY LO W LOW INTERMEDI ATE HIGH VERY HIGH
V IV 111 11 I
r 'J
~~Y +17
+A .
4 V
' I :~ :1'" ~- i ~~ , I~V' ,
----:.~, - ~ ,I
~-~- Itv
~/'"- - - I I~ ,1 <0
/ r . ' I V
/ /. rI:JJY ~l/
I Il l. -A ·
I
1 0 35 ' 0 '5 50 55 60 6 5 7.0 1 5 ab
g7. 7
Q5
IlO
' 5
' 0
2.3
Index Y
0001 0002 !0P';i, .9101 I Oj2 ,o p i , .9j1 0 2 I O,5! , " I' f ! ! "'10 20I 5p " WP q [I/s m], ! I , I I , , J
0 1 02 , 0,5 " ,,) f " ,jO zp I ~ I ,,:pO, s " ,!p3 1 L "10' T [m 2/d], , r , , I I ! I I
Fig.l . Prevailing t ransmissivity and it s classification expressed as fields of cumula tive relative frequencies (modified after Krasny 1999). q = specificcapacity in Lis m,T = coeffic ient of t ransmissivity in m'ld, Index Y = index of t ransmissivity Y (Index Y= log 1O'q ,q = specific capacity in Lis m),x = arith meti c mean, s = standard deviat ion ,x ± s = int erval of prevailing t ransmissivit y (hydrogeo log ical or t ransmissivity background) compr ising approximately th e centra l 68% of transmissivity values of a stat ist ical sample,++A,+A,-A, - -A =field s of positive and negative anomalies (+A,-A) and extre meanomalies (++A,- -Al, respect ively (outside th e interval x ± s of prevailing t ransmissivity).A - field comprising t ransmissivity values of the majori ty of hardrock types, B - field of t ransmissivity values in crystall ine limestones andlor in ot herhardrocks of higher prevail ing t ransmissivity, C - cumu lative relative frequency of t ransmissivity of fluvial depos its along the Labe River in the CzechRepub lic (for comparison wi th transmissivity of hardro cks expressed by fields A and B), indicating not only high er transmissivity but insig nificant variability (steep slope of the line) aswe ll (t ransmissivity class lIa).Classificat ion of t ransmissivity magnitude and variat ion after Krasny (1993a).
Table 1. Classification of t ransmissivity magnitu de (afte r Krasny 1993a).
Comparative regional parameters VeryCoefficient Classof Designation approximately corresponding to the approximate
of trans- transmis- of trans- coeff icient of transmissivity Groundwater supply expectedmissivity sivity missivity potential discharge in LIs of(m' /d) magnitude magnitude Non-logarithmic: Logarithm ic: a single well at
Specific capacity (LIsm) Index Y 5 m drawdown
Very high Withdrawals of great > 50regional importance
- - 1,000 - - - - - - - - - - - - - - - - - - - - - - - - - - --- - --- - - - - - - - - - - 10 - - - - - - - - - - - - --- ----- 7.0 --- ---- - --- --- - --- - --- - - - - - - - - - - - - - --- - ---------11 High Withd rawals of lesser
regional impor tance 5 - 50- - - - 100 - - - - - - - - - - - - - - - - - - - - - - --- - - - - - - - - - - - - - - - - --- 1- - - - - - - - - - - - - - - - - - - - 6.0 - - - - - - - --- - --- - - - - - --- - - - - - - - - - - - - - - - - - - - - - - - - - -
Withdrawals for localIII Intermediate water supply (small 0.5 - 5
communit ies,plants, etc.)- - - - - 10 - ---- --- - - - - - - - - - - - - - - --- --- ------ - - - - - --- - - 0.1- - - - - - - - - - - - - - - - --- 5.0-- - ----- - ---- - ----- ---- - - - - - - - - - - - - - --- ---------
Smaller wi thdrawals forIV Low local water supply 0.05 - 0.5
(private consum pt ion, etc.)------ 1 --- ----- ---- ------ ---- --- - ----- ------ ---- -- 0.01--- - ------------- 4.0 - - --- - - - - --- - - - - - - - - - - - - - - - - - - - - --- - - --- --------
Withd rawals for localV Very lo w water supply wit h 0.005 - 0.05
limited consumption- - - -- 0.1 - - --- ---- ---- ------ ---- ------ ----- - --- - - - - 0.001- ------ - - - ------- 3.0 - - - - - ----- - - - - - - - - - - ---- --- - ----- ------- - - - - ----
Sources for local waterVI Imperceptible supply are diff icult < 0.005
(if possible) to ensure
NGU -B U L L 4 3 9 , 2002 - PAG E 10 JII'i{KRASNY
Or logarithmic tran sformat ion of any parameter exp ressing tran smissivity
.. Usablealsofor hydraulicconductivity evaluation
Designation oftransmissivity
variatio n
Standarddeviation Class ofof transmissivity transmissivity
index y " variation
Table 2. Classification of t ransmissivit y variation (after Krasny 1993a).
Hydrogeological
environmentfrom the point of
view of itshydraulicheteroqene ity"
a Ins ignificant Homogeneous- ---- - - 0.2 --------------------------- --- -------------------
b Small Sligh tlyheter ogeneous
--- --- - 0.4 - - - --- --- ------------------------------------- ---c Moderate Fairly
heterogeneous------- 0.6 - - --- ---- - - - --- ------------------ ---- --- --- --- - --
d Large Considerablyheterogeneous
- - - - - - - 0.8 ------------------------- - - --- - - - - --- --- --- ---- --e Very large Very heterogeneous
- ---- --1 .0 - - - - - - - --- --- - - - - - --------------------- ------ - - - -f Extremely large Extremely
heterogeneous
Scale effect in permeabil ity distri bution: hierarchy of inhomogeneityelementsThe above -mentioned conclusions are based on results of
aquife r tests carried out in water wells, mostly to depths of
some tens of metres.The charac ter of a spat ial distribut ion
of inhomogeneity elements, and consequently also of
hydraulic parameters in a fractured environment, strongly
depend on the exten t of a study area and its relation to the
size of respect ive inh omogeneity elements (Rats 1967, Rats
& Chernyshov 1967, Kiraly 1975). Consequent ly, because of
this 'scale effect' values of hydraul ic parameters are typically
influenced by methods used in their determination .
In a hydrogeolog ical environment characterised by
inhomogeneity elements (fracturing) of similar size, trans
missivi ty mean values becom e closer wi th increasing testing
density wit hin the study area. Finally, t hese values are pract i
cally the same, irrespective of t he position of a tested area
wi thin th is envi ronment (Krasny 2000a). The exten t of an
area representi ng the lowe st limit above which pract ically
no changes in mean t ransmissivity values occur is desig
nated as a representative elementary vol ume (REV).
Accord ing to Bear (1994), th e main characteristi c of a REVis
that the average value of fluid and sol id propert ies taken
over it are ind ependent of its size. Occurrences of larger
inhomogeneous elements (e.g., large fractu re zones), how
ever, may cause other supp lementary differences in t rans
missivity values and the REV size might expand consider
ably.
In a fractured hardrock envi ronmen t, permeability and
transmissivity d istribution is commonly considere d dis
arranged, wi t hout any possibi lit y to be predicted. However,
stud ies on transmissivity distribu t ion in hard rocks of the
transmissivity magnitude is determined aft er th e percent
age of th e interval x ± s (transmissivity backg round ) belong
ing to part icular classes (Fig. 1) (for a detailed description of
the classification system, see Krasny 1993a). The part icular
classes mig ht indi cate the prospect of groundwater supply
in different hydrogeological envi ronments (Table 1).
Another important prope rty of a set of t ransmissivity
values (i.e.,of a stati stical sampl e) is their variabili ty.This sug
gests spat ial transmissivit y changes and, consequently, indi
cates the int ernal character (anatomy) of a hydr og eologi cal
environment and its degree of hydrau lic heterogeneity.
Similarly as wi th t ransm issivity mag nitude, t ransmissivity
variat ion is also classified by six classes, denominated a to f
(Table 2) and based on a standa rd deviat ion of transm issiv
it y of a stat istic al sample. Any transmi ssivi ty parameter
expressed in a logarithmic form, but preferably the t ransmis
sivity index Y, can be used. On probabil ity paper (Fig. 1),sam
ples wi t h low variation will plot alon g lines wi th steeper
slopes th an samples wi th a large variat ion .
This classificat ion system enables a realistic assessment
of aquifer capabili ty to wi t hd raw groundwater in different
areas,and helps in di scussions on the inf luences causing dif
ferences in t ransmissivity values. It also makes it possible to
express quant itati vely reg ional hydrogeolog ical cond it ions
in a com pact form and to compare them in tables, figu res
and in hyd rogeo logical maps.
Transmissivity data from aquifer tests have been stati sti
cally t reated du ring recent years in many hardrock areas of
th e Czech Republ ic, especially wit hin th e framework of the
count ryw ide hydrogeolog ical mapping programm e. This
has brou ght to light important informat ion on transmissiv
ity distr ibution.Stat ist ical samp les were chosen accordi ng to
hydrogeological units, rock ty pes and th e structu ral, geo
morph olog ical and hydrogeological position of water well s.
Part icul ar statisti cal samples represent areas rangi ng from
several km ' to tens or hundreds of km', w it h th e least used
data frequency around seven. Prevail ing transmissivity was
mo stly in un its of m'/d up to slightly mor e than 10 m' /d
(Krasny 1999, Fig. 1).
Examples from di fferent count ries show the possib ilit ies
fo r correlati ve transm issivity stud ies in hardrock areas based
on th is stan dardi sed approa ch. In Poland, Stasko & Tarka
(1996) analysed the transrnisslvity distribu t ion of
Precambr ian and Lower Palaeozoic gnei sses, migmatites,
horn fels and amphibol ites. In Sweden , Carlsson & Carlstedt
(1977) determined cumulative relat ive frequencies of t rans
missivity indices Y from gneisses, granites, Algo nkian and
Cambrian sandstones and Ordovician limesto nes. A similar
proc edure was used for tran smissivity data analysis from
hardrock regions in Ghana (Darko & Krasny 1998, Darko
2001) and in Korea (expressed by Krasny, having used data
from Callahan & Choi 1973). Everywh ere a similar prevailing
transmissivity of hardrocks was determi ned th at can be
expr essed by th e classes IV (V,1I1) c, d (Fig. 1).
JIf/fKRASNY
Bohemian Massif enabled th e definition of a hierarchical
system of inhomogeneity elemen ts of different scalesdesignated as local, med ium -scale (sub-regional) and regional
scale inhomogeneiti es (Krasny 2000b).On a local scale, irregular changes in transmissivity spa
tial distribution within th e 'near-surface aquifer' of hard
rocks are common, and are marked by a weathered (upper)
zone, often wi th j uxtaposed Quaternary deposits, and by a
fractured (middle) zone.These are indicated by differences
in yields and transmissivity of nearby wells dri lled in th esame rock that mig ht reach several orders of magnitude
(Fig. 1). In more extensive areas, transmissivity determ inedby aquifer tests tend s to attain considerably closer valuesboth in prevailing ranges and in arithmetic means. As men
tioned above, these prevailing values represent th e trans
missivity background as ment ioned above.
Superposed upon this background, however, significant
differences in transmissivity magnitude might occur due tothe presence of inhomogeneity elements of a higher scale
level (medium-scale or sub-regional inhomogeneitie s).
These belo ng to tectonically strongly affected zones with aconsiderably higher tran smissivity that might be of impor
tance for groundwater abstraction in hardrock environments (Krasny 1996c).
Other authors have comm ented upon the higher pre
vailing permeabili ty of rocks in valleys than on slopes and
topographic elevations (LeGrand 1954, Krasny 1974, 1998,
Henriksen 1995).Following these findings, the hydrogeolog
ical environment in fractu red rocks should not be considered as regionally homogeneous. It should rather be consid
ered as a compl ex system where belts of regionally highe rprevai ling permeability occur, usually fol lowing the valleys
and depressions.These belts can be perceived as manifesta
tions of inhomogeneity elements of a higher order superim
posed on an environme nt where local inhomoge neity ele
ments can be averaged.The differences in statist ically prevailing values between valleysand elevations may reach one
order of magn it ude or even more. The ratio of arithmeticmeans ranges from 1.6 to 38 in different types of hydrogeological environments (Krasny 1998).These differences seem
to be of genera l validi ty, even though they are of distinct
magn itu des in different hydrogeological environme nt. The
overall tectonic pred isposition of valleys or depressions may
well be the main cause of these differences,a lthough, hydro
geo logica l inf luences wo uld seem to be addi tiona l facto rs
leading to increases in these differences, as reported byKrasny (1974).
Regional-scale transmissivity changes were determined
in extended hardrock areas in South Bohemia (Krasny et al.1984). This regional transmi ssivity distribution evidently
ref lects the intensi ty of so-called neotectonic activity, lasti ng
from the Late Tertiary up to recent times . A lower prevailing
transm issivity is characterist ic of relatively flat areas with
negligible neotectonic activi ty, and higher in zones where
neotectonic deformat ion has been more pronounced (e.g.,
NGU- BULL 4 39 , 2002 - PAGE 11
the Sumava Mts.). The differences in regionally prevailing
t ransmissivit y reach more than one order of t ransmissivity
magn itude.This represents an important shift in a regional
transmissivity background upon which local changes oft ransmissivity (posit ive and negative anomalies) are super
imposed (Havlfk & Krasny 1998).This might be of practical
inter est for groundwater supply studies and the sit ing of
waste deposit s or deep repositories.
Similar gradual changes in th e regio nally prevailing
transmissivity of hardrocks, caused apparent ly by different
intensiti es of rock fractu ring, have been reported fromNorway by Rohr-Torp (1994). His study showed a wel l yie ld
correlation with th e amount of post-g lacial isostatic upl ift
following the Weichselian glaciation. Even though the rea
sons for the differen t degrees of fracturing in South
Bohemia and in Norway are dissimilar, th e comparable
effects indicate possibilit ies of regional variations in trans
missivity also in other fracture-dominated environments.The above-ment ioned hierarchical system of t ransmis
sivity d istribution due to disparate inhom ogeneity elementsis to be expected in all fractured and dou ble-porosi ty envi
ronments. Practical conclusions should be drawn for con
ceptua l model impl ementation, grou ndwater flow model
ling, safe yield assessment, well sitin g and studies on
groundwater vulnerability.
Influences of lithology on regionaltransmissivity distribution inhardrock areasResults of hundreds of pumping tests carried out in drilled
and dug wells in hard rocks within th e Czech part of theBohemian Massif have enabled a quantitative and standard
ised approach to transmissivity distribution studies.Irregular
local permeability changes, determin ed from results of par
ti cular aquifer tests, usually scatter over a wide interval of
several orders of magnitude. On the other hand, regionally
prevai ling transmissivity values (hydrogeological back
ground) resulting from a statistical t reatment, mostly belongto classes IV(V,III) c,d, i.e. very low to inte rmediate transmis
sivity with moderate to large t ransmissivity variation - seefield A in Fig. 1. Based on the differences in pet rograph ic
composi tions of rocks and their different ways of fissuring
and weathering, differences in hydrogeological propert ies
have also often been considered and expected.An objective
comparison of the transmissivity magnitude of particular
rocks in the Czech part of the Bohemian Massif, however,
indicated only small differences in the regionally prevailingt ransmissivit y of partic ular areas. Except for some types of
rocks, the influence of petrography on t ransmissivity spatialdistribution cannot be considered important. Areas com
pr ising crystalline limestones (marbles) represent the most
prominent exception of th is rule. Marble s commonly occur
as intercalations of variable th ickness in bedrock and their
hydrogeological properti es usually differ considerably from
other rock types.The prevailing transmissivity and hydraulic
N GU-BUL L 4 39 , 2 002 - PAG E 1 2
conductivity of marb les is usually a half to one order of mag
nitu de higher compared with other crystalline rocks (Krasny
1999) and might be expressedschemat ically by field B in Fig.
1 (most ly classes Ill-IV c.d).
There are some indicat ions that relativ ely high er trans
missiviti es may be expected in areas underlain by basicigneous rocks and also by some types of granit e. On the
other hand, phyllites displayed a lower region ally prevailingtransmissivity in some areas (Krasny 1993b). No sign if icant
transmissivity differences on a regiona l scale, however, have
been demo nstrated between most granites and the majo r
ity of metamorphic rock types,even though the fo rmer have
oft en been considered more permeable. Possible small dif
ferences in t ransmissivity caused by dist inct ive petrographies of hardrocks are evident ly masked by th e more impor
tant influence of fracturing. Differences in the ir geomechan
ical properties, frequency of fractu ring and th e character ofweathering are obvious, however,and may result in local dif
ferences in th eir permeab ility. At depth , massive granit ic
rocks seem to be more predisposed to form imp ortant frac
ture zones as many th ermal waters are usually associated
with them. Thermo-mineral waters in the Czech Republic,
such as those at Karlovy Vary (Carlsbad) Spa and Teplice Spa,
emanate from granite and quartz porphyry (rhyolite),respect ively (Hanzlik &Krasny 1998,Jakes &Krasny 1998).
The influence of weathering and the presence of gener
ally more highly permeable Quaterna ry deposits, regoli t hs,debris and fluvia l deposit s result in higher transmissivit ies in
wells where hardrocks are covered by these thick, unconsol
idated depos its. The transmissivity variation of hardrocks
com prising fluvial Quaternary depos its is usually lower than
th at of hardrocks wi thout a Quaternary cover.This suggestsan equalising effect of the hydraulically more homoge
neous,Quaternary depos it s (Krasny 1993b).
Natural groundwater resourcesAn assessment of groundwater runoff into rivers draining
surroundin g aquifers isgenerally considered the best way to
estimate natura l (renewable) groundwater resources of th e
tempe rate climatic zone on a regional scale. During a map
ping programme focused on th e assessment of groundwa
ter runoff in the Czech Repub lic, the to tal, long-term, mean
natu ral grou ndwate r resources of the whole coun t ry wereestimated to be ashigh as 205 m'/s.The result ing map of th e
JlfI{KRAsNY
long-term, mean groun dwater runoff in the former
Czechoslovakia at 1:1,000,000 scale w ith explanato ry notes
(Krasny et al. 1981, 1982), might serve as an examp le of a
national cartographi c representation of natura l groundwa
ter resources. Previous assessments of natural groundwa ter
resources (=recharge) in hardrock areas of Cent ral Europewere low, mostly up to a maximum of 1-2 Us km' (e.g., Hynie
1961). The results of the above-mentio ned mapping programme have indicated regionally valid values in the
hardrock environment ranging from 1-2 Us km2 to more
than 10 Us km'.
To compile such a map, a number of methods for
groundwater runoff esti matio n were tested and compared.
For a variety of reasons, mainly due to its obje ct ive data pro
cessing, th e method of Kille (1970) was chosen andemp loyed by all co-authors.The results were compared with
a number of othe r hydrolog ical methods aiming to separatea ground water component from the total runoff, e.g.,
Castany et al. (1970), Kliner & Knezek (1974) and Makarenko
(1948).An independent met hod applying regionally prevail
ing t ransmissivity values and morphometri c character istics
of the area (Krasny & Knezek 1977) was also used.
Differences between th e results ob tained by part icularmethods were not significant. In some catchments, in the
vicin ity of mountain summ its, the long -term specificgroundwater runo ff from hardrock areas reaches in excess
of 15 L/s km'.This is mainly because of favourable climat icand geomorpholo gical conditions .The latter result in a rela
t ively high hydraulic gradient of groundwater flow even in
this less transmissible hardrock aquifer (Krasny 1999). With
decreasing elevation , and mostly due to a decrease in pre
cip itation, th e rate of grou ndwater runoff generally gradually diminishes down to 1-2 Us km'.The approximate rela
t ionsh ip bet ween climat ic and hypsometric condit ions andground water runoff (= natu ral groundwater resources) dis
t ribution in hardrock areasof the Bohemian Massif is shown
in Table 3. Maximum long-term recharge may thus be esti
mated at mo re th an 300 mm /year.This accounts for more
than 20 % of the mean annual precipitat ion.Natural ground wate r resources in hardrock areas of the
Bohemian Massif were also studied in the neighbouring
regionsof Germany and Poland.Killes (1970) method, part ly
mod ified by Kopf & Rothascher (1980),was used for regiona lrenewab le ground water resource estimation in Bavaria,
Table 3. Relatio nship between climatic and hypsometric cond it ion s and groundwater runoff (natural groundwater resources) distri bution in hardrockareas of the Bohem ian Massif (aft er Krasny 1996b).
Morph ological Approximate Mean annual Mean annual Groundwater runoff(hypsometric) unit elevat ion (m a.s.l.l precipitation (mm) evapot ranspirat ion (natural ground water
(esti mat ion in mm) resour ces- LIs km ')
Mo untains 1,200 - 1,600 1,000 - 1,200 450 10 - 15Lower mountains 800 - 1,200 800 - 1,000 7 -10Piedmo nt areas 300 - 800 600- 800 3 - 7Flat areas, low lands less th an 300 500 - 600 650 1 - 3
)I(fi KRASNY
Germany, by Apel et al. (1996). Results of regional studies on
groundwater runoff distributi on in Poland were presented
by Bocheriska et al. (1997) and Kryza & Kryza (1997). The
results obtained both in Bavaria and in Poland were quite
comp arable wi th those from the Czech Repub lic.Residence t ime, based on isotope stud ies,was assessed
to between a half to one year for a shallow grou ndwater
flow and up to ten yearsfor a deep one in crystalline rocks of
the Bavarian Forest in Germany (Seiler & Muller 1996).
Previously, Mart inec (1975) had reported a gro undwater res
idence tim e in crystalline rocks of the Krkonose Mts. (Czech
Republic) of one to two years.
ConclusionsDepending on hydrogeolog ical and climat ic cond itions , the
limitin g factor for groundwater resource develop ment may
be either:
hydraulic prop erti es of rocks,orground water recharge.
A regional assessment and knowledge of the se two
aspects is indispensable for any decision -making on the sustainable develop ment and protec tion of groundwater in
more extended areas such as regions, states,or even conti
nent s.
In arid and semi-arid regions, due to th e limited precipi
tation and high evaporation, natural groundwater rechargetypically represents a clear limitation for groundwater
abstraction. Intensive groundwater withdrawals may result
in a long-term overdraft.
On th e other hand, within the Eart h's temp erate climaticzone,a relationship betwee n groundwate r abst ract ion pos
sibili t ies of different hydrogeolog ical environme nts, givenby their hydraulic parameters, and the available natura l
gro undwater resources determines what is effectively a safe
yield for an area. In hardrock environments where high nat
ural (renewable) groundwater resources reach up to 10-15
LIs km' in some catchments in the vicinity of mountain summits, recharge is suffic ient to cover abst ract ion possibilities
th at are limited only by the tran smissivit y of th e rocks.
Mountains with th eir surprisingly high natu ral groundwaterresources, in spite of th eir relat ively low t ransmissivity, com
monly represent source areas high enough to maintain flowin water courses in adjacent piedmont zones du ring dry
periods.Therefore, when considering the format ion of nat
ural groundwater resources (groundwater recharge) against
a background of th e Earth's clim atic zonation (arid, hum id,temp erate, etc.) th e vert ical climat ic zonat ion mostly arising
from hypsometric differences (mountains, lowlands, etc.),should also be taken into account.
Under the se conditions, and considering the present
day general t rends towards water manageme nt optimisa
ti on and also increases in existing wate r demand, adequately sited water wells or other water int ake systems in
hardrock areas can cover th e requirements on wate r supp ly
for small communiti es, plants or farms,and also for domestic
N G U -B U L L 439 , 2002 - PAGE 13
water consumpt ion. In areas wi th high t ransmissivity ofhardrocks, th e groundwater abst ract ion possibi lit ies might
even be high enough to supply small towns. Therefore,
within hardrock areas with suffi cient water resources, scat
tered water intake sites can effective ly cover the water sup
ply demands of a large number of consumer s. A hydrog eo
log ical basis, i.e., an adequate understanding of transm issiv
ity and permeabili ty distribu tion , is an essential factor in
attaining this goal.
AcknowledgementsThe author acknow ledges the suppo rt of the EC Project LOWRGREP(Landuse opti misation w it h respect to groundwater resources prot ect ion in the mountain hard rock areas) and of th e Ministr y of Education ofthe Czech Repub lic (Grant No. J13/98: 113100005).Thanks are also dueto th e reviewers, David Banks and Helge Henrik sen, for th eir helpful andconstructive comments on th e manuscript .
ReferencesApel, R., Klemm, A. & Rudiger, F. 1996: Grund lagen zum Wasserw irt-
schaft lichen Rahmenplan Naab-Regen Hydrogeologi e.Bayerisches Geologisches Landesamt.Munchen, 106 pp.
Banks, D. 1998:Predictin g the probabil ity distribution of yield f rom mult iple boreholes in crystall ine bedrock.Ground Water 36,2,269-2 74.
Banks, D.,Rohr-Torp,E.,& 5karphagen,H. 1993:Groundwater resources inhard rock; experiences from Hvaler stu dy, South western Norwa y. InS.Banks& D.Banks(eds.):Hydrogeology of hard rocks,Memoirsof th eXXIVt h Congress, Internat ional Association of Hydrogeologists, As(Oslo), 39-51.
Bear,J.1994: Use of models in decision making . In Dracos,T.& Stauffer, F.(eds.):Transport and react ive processes in aquifers,Proceeding s ofthe IAHRSymposium, 3-9 .
Bocheriska, T., Marszalek, H. & Stasko, S. 1997: Hard rocks of th e SudetyMts. as a groundw ater collecto r and its vulnerability . In Bocheriska,T. & Stasko, S. (eds.): Hydrogeology. The second Workshop onHardrock hydrogeology of the Bohemian Massif, Acta Universita tis
Wratis laviensis20S2, 11-19.
Callahan, J.T. & Choi, S.1. 1973: Development of water from fracturedcrystalline rocks, Repub lic of Korea. Proceedings 2"" International
symposium on Groundwater, Palermo.Carlsson, L.& Carlstedt, A. 1977: Estimation of transmissibility and per
meability in Swedish bedrock.Nordi c Hydrology 8, 103-116.Castany G, Margat J, Alb inet M. & Dellarozlere-Boui llin, O. 1970:
Evaluati on rapid e de ressources en aux dune region . AttiConvegno internaz. sulle acque sotte ranee, 462-482.
Darko, P.K. 2001: Quant itat ive aspects of hard rock aquifers: Regionalevaluatio n of groundwater resources in Ghana. PhD .Thesis,CharlesUniversity, Prague.
Darko, P. K.& Krasny,J. 1998: Compariso n of hardrock hydraulic parameters in distinct climatic zones: t he Ghana and th e Bohemian Massifareas. In: Annau, R., Bender, S. & Wohnli ch, S. (eds.): HardrockHydrogeology of the Bohemian Massif, Proceedings 3'dInt ernational
Workshop, Windi scheschenbach. MOnchner Geologische Hefte, B8,3-10.
Hanzlik, J. & Krasny, J. 1998: Brine occurrences in Czech-German bord erregio n. In: Annau, R., Bender, S. & Wohnlich, S. (eds.): HardrockHydrog eology of th e Bohemian Massif.Proceeding s 3'dInternational
Workshop, Windi scheschenbach. MOnchner Geologische Heft e, B8,71-78.
Havlfk,M.& Krasny). 1998:Transmissivity distribution in southern part oft he Bohemian Massif : regio nal t rends and local anomalies. In:Annau,R., Bender,S.& Wohnlich,S.(eds.),Hardrock Hydrogeology ofth e Bohemian Massif, Proceeding s 3'd Internatianal Workshop,
N GU- BU L L 439, 2002 - PAG E 14
Windischeschenbach.MunchnerGeologische Heite, B8, 11-18 .
Henri ksen, H. 1995:Relat io n bet ween to pography and well yield in bore
ho les in crystall ine rocks, Sogn og Fjo rda ne, No rway. Ground Water
33,635-643.
Hynie, O. 1961: Hyd roge ology of Czechoslovakia I (in Czech).
Nakladatelstviceskoslovenske Akademie ved, Praha, 562 pp.
Jakes, P. & Krasny, J. 1998: The Teplice Spa in th e Northern Bohemia: 120
years of hydr ogeological issues and possib ilities of it s fu ture devel
opme nt . In: Annau, R., Bend er , S. & Wohnl ich , 5. (eds.): Hardrock
Hyd rogeology of the Bohemian Massif. Proceedings 3'dInternational
Workshop, Windischeschenbach. Mimchner Geologische Hefte, B8,
79-86 .
Jetel, J. & Krasny, J. 1968: Approxim ati ve aquifer charact eri sti cs in
regiona l hydrog eological stu dy. Vestnik Ustiedniho Ustavu geolog
ickeho 43, 5, 459-461 .
Kille, K. 1970: Das Verfa hren MoMNQ, ein Beitrag zur Berechnung der
mittleren lanqjahriqe n Gru ndwassern eub ild ung mit Hilfe de r
mo nat lichen Niedrigwasserabflusse. Zeit schri ft Deutsch. geo l.
Gessetlschatt,89-95 .
Kiraly, L. 1975: Rapport sur l'etat actuel des connaissances dans le
domaine des caracteres physiques des roches karstiques. In
Hydr ogeology of karstic terrains.IAH & UNESCO, Paris, 53-67.
Kliner, K. & Knezek, M. 1974: Separat ion of groundw ater run off by using
groundwater level (In Czech). Vodohospodarsky casopis 22, 5.
Brati slava.
Kopf E. & Rot hascher, A. 1980: Das natOrl iche Grun dwasserdargebot in
Bayern im Vergl eich zu de n Hauptkompon enten des
Wasserkreislau fes. Schriftenreihe Bayerisches Landesamt fur
Wasserwirtschaft, 13.
Krasny, J. 1974: Les differences de la t ra nsrnissivite, statistiquement sig
n ifi cat ives, dans les zones de I'infilt rat ion et du d rainage. Mem.
Assoc. lntern. Hydrogeol . 10, I.Communica tions , 204-2 11.
Krasny,J. 1993a:Classifica ti on of t ransmissivit y mag nitude and variation,
Ground Water 31, 2, 230- 236.
Krasny.J, 1993b: Prevail ing t ransmissivity of hard rocks in the Czech part
of th e Krkonose and Jizerske hory Mts. In: Popra w ski, L. &
Bochenska, T. (eds.): Sb. Sympozia "Wsp6lczesne problemy hydro
geologii VI", 79-86.
KrasnY,J.1996a: Hydr ogeological environment in hard rocks :an attempt
at its schematizing and terminol ogical con siderat ions. In: KrasnY,J.&
Mls, J. (eds.): First Workshop on "Hard rock hydr og eolog y of th e
Boh emi an Massif" 1994,Acta Universitatis Carolinae Geologica, 40, 2,
115-122 .
Krasny, J. 1996b: Sta te-of-the-art of hydrogeological investigations in
hard rocks: the Czech Repu blic. In: Krasny, J. & Ml s, J. (ed s.): First
Wor kshop on "Hardrock hydrog eol ogy of the Bohemi an Massif"
1994 ,Acta Universitatis Carolinae Geoloqica, 40, 2, 89-1 01.
Krasny, J. 1996c: Scale effe ct in t ransmissivity data distribution. In:
Krasny, J. & Mls, J. (ed s.): First Workshop on "Hardrock hyd rogeology
of th e Boh em ian Massif" 1994, Acta Universitatis Carolinae
Geologica, 40, 2, 123-133.
Krasny,J. 1998:Groundwater di scharge zones : sensitive areas of surface
water - groundwater interaction. In: Van Brahama , J., Eckstein, Y.,
Ong ley, L.K., Schnei der, R. & Mo ore, J.E. (eds.): Gam bling w it h
groun dwate r - physical, chemical,a nd biol og ical aspec ts of aq uifer
stream relati on s.Proceedings of the 28"' IAH Congress, 111-1 16.
Krasny, J. 1999: Hard-rock hydrogeology in the Czech Republ ic.
Hydroqeoloqie, 2, 25-38. Orleans.
Krasny, J. 2000a: Geolog ic fact ors in flu en cing spatia l dist ribution of
hard rock transmi ssivit y. In: Sili lo, O. et al. (ed s.), Groundwa ter: past
achievements and fu ture challenges, Proceeding s 30 IAH Congres s,
187-191.
Krasny, J. 2000 b: Hierarchy of in ho moge nei ty elements infl uen cing
tr ansm issivit y d istribution in hard rock s. In: Knutsson, G. (ed.):
Hard rock hydrogeology of the Fennoscandi navian shield,
Proceedings Workshop on Hard rock Hydrogeology, Aspo, Sw eden,
May 26-27, 1996, Nordic Hydrological Programme, NHP Report 45, 37
47.
Krasny.J.200 1: Unde rstand ing of hyd rogeolog ic environment The deci -
JIF/{KRASNY
sive condition to predict undergrou nd technogenic changes. In:
Howard, K.& Israfil ov, R. (eds.):Current prob lems of hyd rogeolog y in
urban areas, urban agglomerates and ind ust rial cent res.
Proceedings Advanced Research NATO Worksho p, May 2001. Baku,
Kluver Academic Pub lishers. Dordrecht. 381-398.
Krasny, J. & Knezek, M. 1977: Regional estimate of groundwater run-off
from fi ssured rocks on using transmissivity coefficient and geomor
phologic charac teristics. Journal of Hydrological Sciences 4, 2, 149
159.
Krasny,J., Darikova, H., Hanzel, V.,Knezek , M.,Matuska, M.& Suba.J . 1981:
Map of g roun dwater runoff in Czechoslovakia in 1:1,000,000. Cesky
nydromete oroloqickv ustav. Praha.
Krasny, J., Knezek , M., Subova , A., Darikova. H., Matuska, M. & Hanzel, V.
1982: Groundwater runoff in th e territor y of Czechoslovakia (in
Czech).Cesky hydrometeoroloqicky ustav , 50 pp.
Krasny, J.et al. 1984: Explanatory no te to basic hydrogeological ma p of
Czecho slova kia in 1: 200,000, shee t 32 Ceske Budejovice (in Czech).
Ustiedni ustav geologicky, 128 pp .
Kryza, J. & Kryza, H. 1997: Renewal rate of groundwater in the Sude ty
Mts . - it s d isposable and exploited resources. In: Bochenska, T. &
Sta sko, S. (eds.). Hydrogeolog y.The second Workshop on Hardrock
hydrogeology of th e Bohem ian Massif, Acta Universitatis
Wrati slaviensis2052, 223-22 7.
LeGrand, H.E. 1954: Geology and g round water in the Statesv ille area,
North Caro lina: North Carolin a Dept. of Conservat ion and
Development, Div.of Min eral Resources Bull.68, 68 pp. In: Davis, S,N.,
and De Wiest, R.J.M., 1966. Hydrogeology: John Wil ey & Sons, NewYork-l ondon-Sydney, p.327.
Makarenko, FA 1948: 0 podzem nom pitanii rek (in Russian: On ground
wa ter di scharge into river s). Trudy Laboratorii Gidrogeolog itsheskikh
Problem AN SSSR, 1,67-72 .
Martinec, J. 1975: Subsurface flow fro m snowmel t tr aced by trit iu m.
Water Resources Research 3, 496-498.
Mazurek, M. 2000: Geo log ical and hyd raulic properties of water-con
ducting features in cry stall ine rock s. In: Stober, I. & Bucher, K. (ed s.):Hydrogeology of crystalline rocks, Kluwer Acad em ic Publishers, 3-26.
Rats, M.v. 1967:Neodnorodnost gornykh porod i ikh fiz ich eskikh svo jstv,Nauk a, Mos kva.
Rats, M.V. & Chernyshov, S.N. 1967: Stat ist ical aspects of the problem on
the per meabi lity of th e j o in ty rocks. International Symposium on
Hydrogeology of fra ctured rocks. Dubrovn ik.
Rohr-Torp, E. 1994: Present uplift rates and groundwater potential in
No rw eg ian hard roc ks, Norges geologiske undersokelse, Bulletin 426,
47-52.
Seiler, P.-K. & Muller, K. 1996: Discharg e and groundwater recharg e in
crystalline rocks of the Bavarian Forest, Germ any. In: Krasny, J.& Mls,
J. (eds.): First Wor kshop on "Hardrock hydrogeology of the
Bohem ian Ma ssif " 1994, Acta Universitatis Carolinae Geoloqica, 40, 2,
205-219.
Stas ko, S. & Tarka, R. 1996: Hydraul ic paramet ers of hard rocks based on
long-term field experiment in Po lish Sudet es. In: Krasny, J. & Mls, J.
(eds.): First Workshop on "Hard rock hydrogeology of the Bohe mian
Massif " 1994, Acta Universi ta tis Carolinae Geolog ica, 40, 2, 167- 178.