Groundwater fl ooding hazard in river valleys of hill regions: example of the Kapos River, Southwest-Hungary 157 Hungarian Geographical Bulletin 62 (2) (2013) 157–174. Groundwater fl ooding hazard in river valleys of hill regions: example of the Kapos River, Southwest-Hungary Dénes LÓCZY1 and József DEZSŐ1 Abstract In the riverine fl oodplains of hill regions built of sand and loess, interactions between river channels and groundwater reservoirs result from the high permeability of the riverbed and the spatial heterogeneity of fl oodplain deposits and soils. Although in dry periods, groundwater sustains the river in the form of basefl ow, and the relationship is the opposite during wet spells, the predictability of inundations from rising groundwater levels is rather low. Also the spatial and temporal development of inundation in narrow fl oodplains of hill regions (like the Kapos River fl oodplain) takes a course in several respects diff erent from that in broad lowlands. In the study areas of the Kapos fl oodplain topographic, remote sensing and soil distribution surveys are jointly applied to assess the true extent of frequent inundation hazard. Keywords: fl oodplain, groundwater monitoring, Histosols, “perirheic zone”, waterlog- ging, hill region Introduction The evaluation of fl ood hazard calls for answering a range of questions: where are inundations expected (i.e. the potential fl oodplain has to be delimited); how oft en do inundations happen; what duration do they have and in which part of the year are they expected with the highest prob- ability? In addition to geomorphological factors, local fl ood inundation also de- pends on the ecological conditions in the fl oodplain: the density of vegeta- tion, tillage and other cultivation methods applied in land utilization and soil moisture state prior to the fl ood (Lastra, J. et al. 2007). Waterlogging precludes 1 Institute of Environmental Sciences, University of Pécs, H-7624 Pécs, Ifj úság útja 6. E-mails: loczyd@gamma.tt k.pte.hu, dejozsi@gamma.tt k.pte.hu – – – – 158 certain types of land use, while ephemeral wetlands are maintained by regular temporal waterlogging. Elements of infrastructure (road and railway embank- ments, fl ood-control dykes, irrigation and drainage canals) modify the passage of fl oods. Large-scale farming has also substantially transformed the natural patt ern of inundation. The frequency and duration of inundation depends on fl ood discharge and slope of the river as well on the climate of the catchment. The fl oodplains of major rivers can be inundated for months, signifi cantly reducing their agroecological value. Since the abiotic environments of fl oodplains are generated at critical discharges above geomorphological thresholds, the spatial patt erns of fl ood- plains are also governed by the spatial and temporal patt ern of fl oods. In order to properly appreciate the impacts of fl oods, fi rst of all, their recurrence intervals has to be compared to the duration of natural fl oodplain succession. If both intervals are of similar length, it is probable that a more ’mature’ vegeta- tion type, a fl oodplain forest of limited spatial diversity, is formed (Whited, D.C. et al. 2007). If fl oods recur within shorter intervals, such as along rivers of braided channel, their disturbance character is more pronounced, and an earlier stage of succession with a more complex patt ern becomes prevalent in the fl oodplain (Arscott, D.B. et al. 2002). A medium-long recurrence interval or a more complicated history of disturbance (on a decadal scale) may create maximum spatial complexity (Ward, J.V. et al. 1999). Earlier the zones of fl ood hazard have been delimited on hydraulic, hydrological basis. More recently ecological considerations, e.g. land suitability are also included in the delimitation, in the sense of the slogan ’living with fl oods’. Land use restrictions vary with the zones. It is necessary to mention that from the aspect of fl ood and inundation hazards small watercourses also deserve att ention. The riparian zones of headwa- ters may be in natural conditions and, therefore, may be more effi cient in fl ood control than the zones along larger rivers, completely transformed by human activities (for instance, along the Rhine – Dister, E. et al. 1990). The restora- tion/rehabilitation of fl oodplain habitats – where it is still possible – is the most economical tool of fl ood control (McCartney, M.P. and Naden, P.S. 1995). Inundation hazard from excess water in the lowlands of Hungary Excess water (waterlogging) had been long associated with river fl ooding. Re- cently, the defi nitions of excess water (Pálfai, I. 2001) have been extended also to include upbursting groundwater even in total absence of any watercourse. In addition to groundwater levels raised on the fl oodplains of major rivers during fl ood stages, any waterlogging in lowland areas is included in this broad category. The presently used defi nition of excess water originating from 159 rainfall or snowmelt which covers any extensive but temporary inundation of lowland areas and fully saturates the soil. Whether soil saturation necessar- ily leads to seasonally waterlogged surfaces, remains to be an open question (Rakonczai, J. et al. 2003). Recurrence intervals of extreme waterlogging have been calculated for Hungary recently (Pálfai, I. 2009 – Table 1) and found to be rather irregular for the mid- and late 20th century. In recent decades, excess water hazard has been observed to increase dramatically. Inundation hazard from excess water is more diffi cult to delimit both temporally and spatially than river fl ood hazard. Rapid snowmelt in spring, early summer cyclonal rains as well as occasional summer cloudbursts are held to be responsible for it. Although the average depth and duration of snow cover is on the decline, extreme values of such parameters oft en occur. The excess water hazard map of Hungary (Pálfai, I. 2009) identifi es four classes: 1 – no hazard areas, where highly permeable surface deposits (sands) prevent enduring inundation; 2 – moderate hazard, where natural levees in fl oodplains and lower sections of alluvial fans are occasionally aff ected; 3 – medium hazard, where one-time fl oodways and backswamps and swamps enclosed between alluvial fans are exposed to rising groundwater and 4 – serious hazard, where inundations regularly recur in wet years. The fi rst three categories make up more than two million hectares in Hungary, i.e. one-third of the agricultural area. In wet years excess water is a source of great damage to Hungarian agriculture, public transport (washing away railway embankments) and tourism (the proliferation of mosquitoes). Waterlogging in river valleys of hill regions The spatial and temporal development of inundation in narrow fl oodplains of hill regions (like the Kapos River fl oodplain under study) takes a course Table 1. Recurrence intervals of major excess water inundations in Hungary * Probability of occurrence, per cent Average return period, years Approximate minimum inundated area, hectares Example years 50 20 10 5 2 2 5 10 20 50 60,000 170,000 270,000 360,000 480,000 1960, 1997 1963, 2010 1956, 1967 1966, 2000 1940, 1941, 1942, 1999 * Modifi ed aft er Pálfai, I. 2009. 160 diff erent from that in broad lowlands (for instance, of the Tisza River and its tributaries). In the former case concentrated cloudbursts create inundations which aff ect the fl oodplain all along the river, particularly in broader sections (embayments), while in the Great Plain extensive partial areas are fl ooded with rapid and hardly predictable dynamics. The ’fl ood pulse’ concept (Junk, W.J. et al. 1989) portrays a simple time sequence of fl oodplain fl ooding (Figure 1, I. A–C). However, during fl oods the incursion of river water across the surface of a ’convex’ fl oodplain may be strongly aff ected by fl oodplain ’wetness’ (groundwater, hyporheic water, run- off from the hillslopes surrounding the fl oodplain, direct precipitation and antecedent water from earlier fl oods) (Mertes, L.A.K. 1997). Fig. 1 Comparison of fl ood development according to the fl ood pulse concept (I.A–C, aft er Junk, W.J. and Wantzen, K.M. 2004) and according to the perirheic zone concept of fl oodplain inundation (II.A–C, aft er Mertes, L.A.K. 1997). I.A = dry fl oodplain before fl ood; I.B = fl ood inundation extending from river channel; I.C = complete inundation of the fl oodplain; II.A = fl oodplain with high groundwater table and local wetlands before fl ood; II.B = extending excess groundwater patches during rising river stage; II.C = inun- dated fl oodplain with streamwater/groundwater mixing (”perirheic zone”). 1 = fl oodplain; 2 = river channel and streamwater-inundated areas; 3 = excess groundwater-inundated areas and the perirheic zone 161 Consequently, somewhat diff erent temporal and spatial patt erns of fl ooding result: along with the hyporheic zone, a mixing zone of stream and (excess) groundwater, the ’perirheic zone’, is created (Figure 1, II.A–C, aft er Mertes, L.A.K. 1997). Study area: the Kapos River catchment The medium-sized catchment of the Kapos River covers 3,295.4 km2 in the Outer Somogy Hills region (Figure 2). The trunk river is 112.7 km long, a 5th-order stream at confl uence with the Sió Canal (the outfl ow of Lake Balaton to the Danube). The topographical fl oodplain (without that of the tributaries) extends over 104.2 km2, which makes up 3.3 per cent of the total catchment area. Fig. 2. DEM representation of the Kapos River catchment 162 High-water fl ow regulation in the early 19th and mid-20th century (Ihrig, D. 1973) did not fully eliminate fl ood and inundation hazards in the Kapos Valley. Even today all streams of the catchment show high fl ood hazards since global climate change increases the probability of non-predictable rainfall events and fl ash fl oods (Czigány, Sz. et al. 2010). Water regime shows low- water stages in August–early September and high water most oft en in March (caused by snowmelt in the hills). Most of the other extremes are due to sum- mer showers. In the embayments downstream of the town of Dombóvár rainy weather can raise groundwater levels rapidly and create extensive temporary waterlogging. Water seepage beneath dykes and impoundments at the confl u- ences of tributary streams may further aggravate the situation. The events of May and June 2010 called att ention to inundations in the perirheic zone and increased vulnerability to fl ooding also along smaller tributaries (Lóczy, D. et al. 2012). For the mapping of the spatial extension of waterlogging and estimating inundation hazard, alternative methods have been tried. Although it cannot be confi rmed yet by groundwater table moni- toring, the 2010 fl ooding in the Döbrököz embayment clearly shows a water- logged perirheic zone also signifi cantly contributed to the inundation of the fl oodplain. Inundation hazard evaluation from topographic and drainage analyses Through the detailed survey and mapping of landforms and DEM representa- tion of topography, most (and earliest) fl ood endangered tracts on the fl ood- plain can be relatively easily identifi ed (see e.g. Lastra, J. et al. 2007). Among the GIS methods the MrVBF index (Gallant, J.C. and Dowling, T.I. 2003) is of outstanding signifi cance. In order to be able to use the MrVBF approach for inundation hazard assessment in Hungary, a table to assess sensitivity to inun- dation was prepared (Table 2). It is based on inundation frequency, soil drainage and position in relief. When applied for the Kapos fl oodplain, it was supple- mented with reference sites fi eld-checked aft er the 2010 rainfall events. Inundation hazard evaluation by remote sensing The interpretation of remote sensing images taken during fl oods (particularly high-resolution Ikonos and SPOT images and aerial photographs) can also be of help in the identifi cation of areas with inundation hazard (Rakonczai, J. et al. 2003). Unfortunately, few images are suitable for this purpose. They have to be taken shortly aft er fl ooding, and the percentage of cloud cover has to remain below 10 per cent. For the fl oodplain embayments the map of possible inun- 163 Ta bl e 2 . I nu nd at io n se ns iti vi ty cl as se s o f t he fl oo dp la in * Ra nk sc or e Se ns iti vi ty c la ss D es cr ip tio n Ex am pl e fr om th e K ap os V al le y 0 no t s en si bl e (in un da - tio n no t p ro ba bl e) so ils w ith m ed iu m to g oo d w at er b ud ge t i n hi gh er po si tio n na tu ra l l ev ee (s ou th o f R eg öl y) 1 lo w so ils w ith m ed iu m w at er b ud ge t o cc as io na lly in un da t- ed in w in te r a nd s pr in g on h ill s um m its a nd s lo pe s fo ot sl op es o n th e ri gh t b an k (T ol na H ill s) (e .g . a t K es ző hi de gk út , B el ec sk a) 2 lo w to m ed iu m so ils w ith m ed iu m w at er b ud ge t p ot en tia lly in un da te d in w in te r a nd s pr in g, li m ite d cu lti va bi lit y, o n hi ll m id s- lo pe s an d fo ot sl op es fo ot sl op e zo ne o f t er ra ce le ve ls (e .g . D öb rö kö z, K ur d – ba ck g ar de ns ) 3 m ed iu m so ils w ith p oo r w at er b ud ge t a nd re du ce d cu lti va bi lit y be ca us e of s at ur at io n or in un da tio n, o n fo ot sl op es , fl a t su rf ac es , d ep re ss io ns m ar gi ns o f b ac ks w am ps in th e D om bó vá r– D öb rö kö z em ba ym en t 4 m ed iu m to h ig h un cu lti va bl e so ils w ith p oo r w at er b ud ge t, se as on al ly in un da te d, o n fo ot sl op es , fl a t s ur fa ce s an d de pr es si on s bo tt o m o f b ac ks w am p in th e Sz ak ál y em ba ym en t 5 hi gh so ils w ith p oo r w at er b ud ge t u nd er e nd ur in g in un da - tio n, c ul tiv at io n is li m ite d th ro ug ho ut th e ye ar , f ou nd on v al le y fl o or s, in d ep re ss io ns ol d m ea nd er s, in fi l le d ox bo w s (e .g . so ut he as t o f R eg öl y) * C om pi le d by L óc zy , D . f ro m v ar io us s ou rc es . dation (Figure 3) was based on the fi rst available image aft er the fl ooding and was constructed from band 6 of the Landsat-7 (ETM+) im- age for 24 September 2010. It shows the actual distribu- tion of pixels where refl ect- ance was predominantly controlled by water surface. (Refl ectance was calibrated for fi sh-ponds in the study area. The drainage net- work was superimposed on the image from the Hungarian Water Management Database. (The allocation error of drain- age lines may amount to ca 100 m.) The smoothed envelope curve embraces all ’water’ pixels and provides at least and approximation of areas potentially affected by wa- terlogging (Figure 3). (On the basis of fi eld observations, it is assumed that the patches of early summer inundations survived to a large extent well into the autumn.) Studying the fi gure, the following observations can be made. The contigu- ous inundated areas are closely associated with the elements of the drainage network (the Kapos canal, the also channelized tribu- tary streams and the drain- age canals). At the same time, minor water surfaces in the marginal zone of the 164 fl oodplain, which derive from rainwater runoff and throughfl ow generated on the neighbouring hillslopes. Some manmade features of the fl oodplain impound both kinds of fl ow. This kind of reconstruction, however, cannot show a complete picture since, for various reasons, along some sections no water surface can be ob- served at all. Here the approximate boundary of maximum possible inunda- tion is shown by a dashed envelope line. Also areas with groundwater table immediately (less than 20 cm) below the surface could have been rightfully included among those stricken by excess water (Rakonczai, J. et al. 2003). The authors of the mentioned paper on mapping excess water inundations in the Great Hungarian Plain cite several sources of incorrect identifi cation. Inundation hazard evaluation from soil distribution Land drainage measures, as corollaries to river regulation, modify or even re- verse the soil formation sequences in the former fl oodplains. As a consequence of the hydromorphic eff ect, on higher grounds of the fl oodplain meadow soil dynamics had been prevalent before river regulation. With land drainage groundwater levels dropped and chernozem dynamics became predominant. Fig. 3 Excess water inundation in the Kapos fl oodplain between Nagyberki and Kurd on 24 September 2010 (based on Landsat-7 ETM+ image). The dashed line indicates sections where only approximate width of the inundated zone can be established 165 In lower-lying spots of the fl oodplain (in the infi lling oxbows and backswamps) bog formation had been the typical pedological process, but aft er water man- agement interventions the peat bogs (Fibric Histosol) began to transform into muck (Hemic Histosol) and ’earthy’ or humifi ed peat (Sapric Histosol), where the groundwater lies at 1.5–2 m deep below the surface (Dömsödi, J. 1988). In the Kapos fl oodplain this process is of particularly great signifi cance (Lóczy, D. 2013). Bog degradation in the Kapos Valley is a process with unfavourable impact on temporal waterlogging. In the Kapos fl oodplain bog soils (Histosols) are related to the former bogs of the valley fl oor drained during the water regulations in the early 19th century (Figure 4). Peat occurs in the most extensive areas and thickest (up to 6 m thick) beds (with silt interbeddings) along the uppermost course of the Kapos River and in the valleys of tributary streams there (Gergely, E. et al. 2000). Fig. 4. Soil catenas across the Kapos Valley (edited by Dezső, J. 2012). Main soil types: 1 = Haplic Gleysol; 2 = Histic Gleysol; 3 = Cambisols; 4 = Histosols. Parent materials: 5 = clay; 6 = loess; 7 = silt; 8 = fi ne sand; 9 = coarse sand; 10 = layer with mol- lusc shells; 11 = gleyic horizon 166 Along the Upper Kapos the peat beds are typically underlain by un- consolidated silts, peat-bearing silt and calcareous silt deposited upon clay, sandy clay and fi ne sand layers. In the major fl oodplain embayments down- stream, thinner humifi ed peat and muck beds alternate with meadow clays. In the Upper Kapos Valley most of beds are composed of fi brous Sphagnum peat of fi ne fabric (Table 3). The most extensive peat area is ca 12 km long, 400–500 m wide and the peat beds are 2 m deep. The soil profi le shows evidence of organic matt er accumulation and de- composition as well as biotic action and hydromorphic infl uence. If exposed, the decomposed muck of porous structure and low bulk density is highly suscep- tible to wind erosion, particularly in spring when the surface is still barren. Meadow soils are the widest spread soil type of the fl oodplain, typical of the waterlogged bott om surfaces of backswamps and oxbows. The uppermost, ploughed horizon is of crumbly structure, but oft en degraded to porous. It is underlain by a ferric horizon of marked red colour (Table 4). Hydromorphic eff ect (mott led fabric) is observed at 0.5 to 1 m depth. Located in depressions, such soils receive surplus water from the surrounding, somewhat higher, surfaces and, therefore, are usually water- logged. In addition to the reconstruction of former channels and wetlands, which are important elements of the landscape structure, another benefi t of collecting soil survey information is the help they provide for the delimitation of areas of excess water hazard. The water budget classes of the genetic soil types occurring on the fl ood- plain are identifi ed using a table (Table 5) compiled from various literary sourc- es and also drawing information from the interpretation of the Landsat image. When detailed soil data are available, the wetness classes of the Soil Survey of England and Wales can also be applied (McRae, S.G. and Burnham, C.P. 1981 – Table 6). Classifi cation is based on depth to groundwater table and the duration of soil water saturation. The subtypes and varieties of meadow soils (Histosols and Gleysols) which are liable to be waterlogged in rainy periods can be iden- tifi ed on the soil map: boggy meadow soil, mucky meadow soil, ’earthy peat’ and peaty meadow soil. Since all the typologies presented in the three tables defi ne six classes, comparisons between them are relatively easy. Having completed the assess- ments, the embayments of the Lower Kapos fl oodplain are mostly found to belong to the inundation sensitivity classes 2–3 (low to medium sensitivity); show rank scores 2–3 (medium susceptibility to inundation; poor to medium infi ltration capacity and permeability; high water storage) and fall into the British wetness classes II or (less typically) III (moderately well drained or imperfectly drained). 167 Ta bl e 3 . F iel d de sc rip tio n of a w id es pr ea d H ist os ol v ar iet y fro m th e K ap os fl oo dp la in (b y D ez ső , J . a nd L óc zy , D .) Lo ca lit y Lo ca lit y de sc ri pt io n G en et ic s oi l t yp e W RB s oi l t yp e C od e: Re gö ly R7 F La nd u se : m ea do w /g ra zi ng la nd (a nc ie nt ro ot tr ac es in di ca te o ne - tim e fl o od pl ai n so ft w oo d fo re st ) La nd fo rm : b ott o m o f o xb ow D at e of su rv ey 21 .0 8. 20 11 G PS c oo rd in at es Ps eu do gl ey ic m ea do w so il G ley ic H ist os ol x y z Pa re nt m at er ia l: si lt G ro un dw at er ta bl e: -2 80 c m So il pi t 0– 15 0 cm 13 77 11 N 60 15 80 E 10 1 m H or iz on s, cm Pr ofi le d es cr ip tio n C ol ou r Ph ys ic al ty pe Eff e r- ve sc en ce / ca rb on at e, % K A * Sa lts , pe r ce nt 0– 15 cr um bl y in th e ro ot z on e, m uc k, n ot p lo ug he d da rk y el lo w to lig ht b ro w n, 10 YR 3 /4 cl ay ey s ilt st ro ng /> 10 66 0. 11 15 –2 5 de co m po se d pe at o f p or ou s st ru ct ur e, re d fe rr ic pr ec ip ita tio ns ye llo w is h re d, 5Y R 6/ 8 – ve ry s lig ht /< 2 66 0. 11 25 –4 5 co m pa ct , d is in te gr at in g in to s la bs , p itc h bl ac k, sh in y cl ay fi lm s on s ke le ta l p ar tic le s bl ac k, 1 0 YR 2/ 1 cl ay lo am no n- eff e r- ve s- ce nt /0 47 0. 08 45 –5 0 bi ot ur ba te d (? ), tr an si tio na l t ow ar ds g le ye d he av y cl ay m ix ed cl ay ey s ilt st ro ng / 10 –2 5 47 – 50 –9 0 ps eu do gl ey ic h or iz on w ith ru st y pr ec ip ita tio ns al on g ro ot tr ac es , r oo t a nd b io tu rb at io n ch an - ne ls gr ey co m pa ct ed si lt vi ol en t/ >2 5 47 0. 09 90 –1 20 la ye r r ic h in ru st y pr ec ip ita tio ns a lo ng ro ot ch an ne ls m ix ed w ith ol iv e gr ey m a- tr ix , 5 Y 6/ 2 fi n e si lt vi ol en t/ >2 5 32 0. 09 12 0– 15 0 ru st y fe rr ic a nd s oft c al ca re ou s pr ec ip ita tio ns al on g ro ot c ha nn el s m ix ed fi n e si lt vi ol en t/ >2 5 – – 168 Ta bl e 3 . f ol yt at ás a Lo ca lit y Lo ca lit y de sc ri pt io n G en et ic s oi l t yp e W RB s oi l t yp e C od e: Re gö ly R7 F La nd u se : m ea do w /g ra zi ng la nd (a nc ie nt ro ot tr ac es in di ca te o ne - tim e fl o od pl ai n so ft w oo d fo re st ) La nd fo rm : b ott o m o f o xb ow D at e of su rv ey 21 .0 8. 20 11 G PS c oo rd in at es Ps eu do gl ey ic m ea do w so il G ley ic H ist os ol x y z Pa re nt m at er ia l: si lt G ro un dw at er ta bl e: -2 80 c m So il pi t 0– 15 0 cm 13 77 11 N 60 15 80 E 10 1 m H or iz on s, cm Pr ofi le d es cr ip tio n C ol ou r Ph ys ic al ty pe Eff e r- ve sc en ce / ca rb on at e, pe r c en t K A * Sa lts , pe r ce nt 15 0– 21 0 fl u vi al ly re w or ke d lo es s? lig ht g re y, 5 Y 7/ 2 si lt vi ol en t/ >2 5 – – 21 0– 27 0 ho m og en eo us fl uv ia l d ep os it lig ht g re y, 5 Y 7/ 2 si lt st ro ng / 10 –2 5 – – 27 0– 31 0 re de po si te d lo es s, g ro un dw at er ta bl e at 2 80 c m lig ht g re y, 5 Y 7/ 2 si lt st ro ng / 10 –2 5 – – 31 0– 34 0 ho m og en eo us fl uv ia l d ep os it lig ht g re y, 5 Y 7/ 1 fi n e sa nd st ro ng / 10 –2 5 – – >3 40 ho m og en eo us fl uv ia l d ep os it lig ht g re y, 5 Y 7/ 1 si lt st ro ng / 10 –2 5 – – * A ra ny ’s Pl as tic ity In de x (fo r e xp la na tio n se e th e te xt ) 169 Ta bl e 4 . F iel d de sc rip tio n of a ty pi ca l G ley so l ( m ea do w so il) fr om th e K ap os fl oo dp la in (b y D ez ső , J . a nd L óc zy , D .) Lo ca lit y Lo ca lit y de sc ri pt io n G en et ic s oi l t yp e W RB s oi l t yp e C od e: Re gö ly R 3F La nd u se : a ra bl e fi e ld w ith h or se ra dd is h La nd fo rm : b ott o m o f ba ck sw am p M ea do w so il un de r c ul tiv at io n M ol lic G ley so l D at e of s ur ve y 20 .0 8. 2 01 1 G PS c oo rd in at es Pa re nt m at er ia l: fi n e sa nd G ro un dw at er ta bl e: >1 50 c m So il pi t 0– 15 0 cm x y z 13 46 44 N 59 89 25 95 m H or iz on s, c m Pr ofi le d es cr ip tio n C ol ou r Ph ys ic al ty pe Eff e r- ve sc en ce / ca rb on at e, pe r c en t 0– 25 re dd is h br ow n m at ri x w ith d ar k gr ey b io tu rb at ed el em en ts ye llo w is h re d; 5 YR 5/ 6 si lt st ro ng / 10 –2 5 25 –4 5 co m pa ct ed , h om og en eo us , s la b st ru ct ur e, d ar k gr ey s hi ny c la y fi l m s bl ac k, 10 YR 2 /1 cl ay no n- eff e r- ve s- ce nt /0 45 –5 5 cr um bl y w ith ru st y pr ec ip ita tio ns , g le yi c m at ri x w ith b la ck c la y in c ha nn el s m ott le d fi n e sa nd , c la y st ro ng / 10 –2 5 55 –1 20 gl ey ed s an d w ith p al e fe rr ic p re ci pi ta tio ns lig ht y el lo w is h br ow n, 2. 5Y 6 /4 m ed iu m s an d st ro ng / 10 –2 5 >1 20 in di st in ct s tr uc tu re lig ht g re y, 5Y 7 /2 m ed iu m s an d sl ig ht / 2– 10 170 Ta bl e 5 . E va lu at io n of g en et ic so il ty pe s o cc ur rin g on th e K ap os fl oo dp la in a cc or di ng to th eir su sc ep tib ili ty to in un da tio n * Ra nk sc or e Pr ed ic ta bl e sa tu ra tio n G en et ic s oi l ( su b) ty pe s D ra in ag e pr op er tie s Fr eq ue nc y, ye ar s D ur at io n, w ee ks In fi l tr at io n ca pa ci ty , m m d -1 Tr an sm is si on ca pa ci ty , m m d -1 St or ag e ca pa ci ty , m m m -1 0 50 –1 00 le ss th an o ne m ea do w c he rn oz em , ch er no ze m m ea do w s oi l go od : 30 0– 10 00 go od : 15 0– 50 0 go od : 10 0– 15 0 1 20 –5 0 1– 2 m ea do w s oi l, ca lc ar eo us al lu vi al m ea do w s oi l hi gh : >1 ,0 00 hi gh : 50 0– 1, 00 0 m ed iu m : 50 –1 00 2 10 –2 0 3– 4 bo gg y m ea do w s oi l m ed iu m : 10 0– 30 0 m ed iu m : 50 –1 50 hi gh : 15 0– 20 0 3 5– 10 4– 8 ea rt hy p ea t ( ’b la ck e ar th ’) po or : 50 –1 00 po or : 10 –5 0 hi gh : 15 0– 20 0 4 2– 5 se ve ra l m on th s bo g so il w ith m uc k po or : 10 –1 00 ve ry p oo r: < 10 hi gh : 15 0– 20 0 5 1 se ve ra l m on th s bo g so il w ith p ea t po or : 10 –1 00 ve ry p oo r: <1 0 ve ry h ig h: > 20 0 * C om pi le d by L óc zy , D . 171 Table 6. Wetness classes of soils according to the depth of the soil horizon saturated to water capacity Wetness class Water saturation Drainage class (approximate) Depth, cm Duration, day per year I II III >70 <70 <70 <30 30–90 90–180 well drained moderately well drained imperfectly drained IV <40 >180 poorly drained V <40 or <70 >180 or >335 very poorly drained (water- logged) VI <40 >335 very poorly drained (regularly inundated) Source: Soil Survey of England and Wales. Groundwater table monitoring At high (fl ood) stages the unconsolidated fl oodplain deposits and soils are assumed to temporarily store water before it is conveyed downstream. The overall eff ect of this water storage is the delay and att enuation of the fl ood peak in downstream areas. Flow pathways are oft en defl ected from the align- ment of the main channel and oft en run diagonal towards the channel in a downstream direction (Kelly, B.P. 2001). Over a narrow fl oodplain (such as that of the Kapos River) another major element of subsurface fl ow is the spatial continuation of throughfl ow from the neighbouring hillslopes, which is close to perpendicular to the channel. It oft en causes waterlogging during high river stages (perirheic zone). The major controls on the alignment of groundwater fl ow (’underfl ow’) paths are the hydraulic properties of fl oodplain deposits, regional slope and sinuosity (Larkin, R.G. and Sharp, J.M. Jr. 1992). Unfortunately, to realistically depict groundwater fl ow a dense network of observation wells with long time series would be necessary. The national monitoring system of groundwater levels only very sparsely covers the Kapos fl oodplain and the embayments of the lower segments are not monitored at all. In order to receive information on the position of groundwater in the fl oodplain for the period November 2011–October 2012, we installed measuring instru- ments (Dataqua DA-S-LRB 122 SMART rigid sound water level gauges, preci- sion: ±0.1 per cent; measurement range: 0–200 cm; manufactured by Dataqua Electronic Co., Balatonalmádi, Hungary) into two observations wells (at Kurd, at a short distance from the river gauge, and downstream of the constriction, at Dúzs). The fi rst year when an uninterrupted record of groundwater table fl uc- tuations could be obtained was 2012 (Figure 5). Naturally, the laws of ground- water fl ow could only be revealed aft er a much longer period of monitoring. 172 Assessing the intensity of mutual stream/groundwater interactions in the Kapos fl oodplain, Considerable recharge is observed from infi ltration (snowmelt) and early spring fl oods, when evaporation losses are not yet re- markable, and in the saturated fl oodplain deposits perirheic fl ow (Mertes, L.A.K. 1997) is regularly observed. The river stages and water levels of ob- servation well I were raised by rapid snowmelt in late February. The much lower groundwater table in well II responded with a remarkable delay. The groundwater reserves, however, are heavily depleted by evaporation caused by rising temperature in the fi rst third of the growing season. Although the highly variable amounts of (early) summer precipitation are of great ecologi- cal signifi cance, high temperatures considerably reduce their contribution to groundwater recharge. The rainfall event in July did not infl uence water level in well I, while its impact with a three-day delay (similar to that in February) was observed for well no II. Since infi ltration does not reach the groundwater table, summer show- ers are mostly ineffi cient in groundwater recharge. In lack of by-channels and Fig. 5. River stages of the Kapos River at the Kurd gauge and groundwater levels recorded in observation wells I (Kurd) and II (Dúzs) between November 2011 and October 2012 (by Dezső, J. 2012) 173 oxbows and backswamps in the perirheic zone drained, higher river water stages are unable to saturate fl oodplain soils. Where high-porosity layers are uninterrupted between the channel and more remote areas of the fl oodplain, groundwater recharge also occurs in drought periods. Conclusions Diff erent approaches have been tried to present inundation hazard in the narrow fl oodplain of a medium-sized river in a hill region of Hungary. Sup- plemented with long-term groundwater level monitoring, the joint applica- tion of the presented methods can lead to realistic identifi cation of areas with high-probability groundwater fl ooding. REFERENCES Arscott, D.B., Tockner, K. and Ward, J.V. 2002. Aquatic habitat dynamics along a braided Alpine river ecosystem (Tagliamento River, N.E. Italy). Ecosystems 5. 802–814. CZIGÁNY, SZ., PIRKHOFFER, E., FÁBIÁN, SZ.Á. and ILISICS, N. 2010. Flash fl oods as natural hazards in Hungary, with special focus on SW-Hungary. Riscuri şi catastrofe 8.1. 131–152. DISTER, E., GOMER, D., OBRDLIK, P., PETERMANN, P. and SCHNEIDER, E. 1990. Water management and ecological perspectives of the Upper Rhine’s fl oodplains. Regulated Rivers: Research and Management 5. 1–15. Dömsödi, J. 1988. Lápképződés, lápmegsemmisülés (Bog formation, bog destruction). Elmélet– Módszer–Gyakorlat 46. Budapest, Geographical Research Institute, Hungarian Academy of Sciences, 120 p. Gallant, J.C. and Dowling, T.I. 2003. A multiresolution index of valley bott om fl atness for mapping depositional areas. Water Resources Research 39. 1347–1353. Gergely, E., Géczi, Cs., Horváth, J., Jakab, A., Jónás, Gy.-né, Károlyi, Z.-né, Mattányi, Zs., Szalai, Z., Szabó, I. and Ress, S. 2000. Kapos folyóvölgy – Lehetőségterv (Kapos River valley. Feasibility plan). Budapest, Öko Zrt. 157 p. Ihrig, D. ed. 1973. A magyar vízszabályozás története (History of river regulations in Hungary). Budapest, OVH (National Water Management Offi ce), 398 p. Junk, W.J. and Wantzen, K.M. 2004. The fl ood pulse concept: new aspects, approaches and applications - an update. In Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries, Vol. II. Sustaining Livelihoods and Biodiversity in the New Millennium. 11–14 February 2003, Eds. Welcomme, R.L. and Petr, T Phnom Penh (RAP Publication 2004. 7). 117–140. Junk, W.J., Bayley, P.B. and Sparks, R.E. 1989. The fl ood-pulse concept in river-fl oodplain systems. Canadian Journal of Fisheries and Aquatic Sciences 106. 110–127. Kelly, B.P. 2001. Relations among river stage, rainfall, ground-water levels, and stage at two Missouri River fl ood-plain wetlands. US Geological Service Water-Resources Investigation Report 01–4123. 18 p. htt p://www.usgs.WRIR014123.pdf 174 LARKIN, R.G. and SHARP, J.M. JR. 1992. On the relationship between river basin geomorphology, aquifer hydraulics, and ground-water fl ow direction in alluvial aquifers. Geological Society of America Bulletin 104. 1608–1620. LASTRA, J., FERNÁNDEZ, E., DÍEZ, A. and MARQUÍNEZ, J. 2007. Flood hazard delineation combining geomorphological and hydrological methods: an example in the Northern Iberian Peninsula. Natural Hazards 45. 277–293. Lóczy, D. 2013. Hydromorphological and geoecological foundations of fl oodplain rehabilitation: Case study from Hungary. Saarbrücken, Lambert Academic Publishing, (in press) Lóczy, D., Czigány, Sz. and Pirkhoffer, E. 2012. Flash fl ood hazards. In Studies on Water Management Issues. Ed. Kumarasamy, M. Rij eka, InTech, 27–52. McCartney, M.P. and Naden, P.S. 1995. A semi-empirical investigation of the infl uence of fl oodplain storage on fl ood fl ow. Journal of CIWEM 9. 236–246. McRae, S.G. and Burnham, C.P. 1981. Land Evaluation. Monographs of Soil Survey 7. Oxford, Clarendon Press, 239 p. Mertes, L.A.K. 1997. Documentation and signifi cance of the perirheic zone on inundated fl oodplains. Water Resources Research 33. 1749–1762. Pálfai, I. 2001. A belvíz defi níciói (Defi nitions of excess water). Vízügyi Közlemények 83. (3): 376–392. Pálfai, I. 2009. Inland fl ooding in Hungary. Riscuri şi catastrofe 8. (7): 193–201. RAKONCZAI, J., CSATÓ, SZ., MUCSI, L., KOVÁCS, F. and SZATMÁRI, J. 2003. Az 1999. és 2000. évi alföldi belvíz-elöntések kiértékelésének gyakorlati tapasztalatai (Experience from the evaluation of the 1999 and 2000 excess water inundations in the Great Hungarian Plain). Vízügyi Közlemények Special Issue 4. 317–336. Ward, J.V., Tockner, K. and Schiemer, F. 1999. Biodiversity of fl oodplain river ecosystems: ecotones and connectivity. Regulated Rivers: Research & Management 15. 126–139. Whited, D.C., Lorang, M.S., Harner, M.J., Hauer, F.R., Kimball, J.S. and Stanford, J.S. 2007. Climate, hydrologic disturbance, and succession: drivers of fl oodplain patt ern. Ecology 88. (4): 940–953. << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.3 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.0000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 100 /Optimize false /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments true /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 300 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 300 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /CreateJDFFile false /Description << /ARA /BGR /CHS /CHT /CZE /DAN /DEU /ESP /ETI /FRA /GRE /HEB /HRV (Za stvaranje Adobe PDF dokumenata najpogodnijih za visokokvalitetni ispis prije tiskanja koristite ove postavke. Stvoreni PDF dokumenti mogu se otvoriti Acrobat i Adobe Reader 5.0 i kasnijim verzijama.) /ITA /JPN /KOR /LTH /LVI /NLD (Gebruik deze instellingen om Adobe PDF-documenten te maken die zijn geoptimaliseerd voor prepress-afdrukken van hoge kwaliteit. De gemaakte PDF-documenten kunnen worden geopend met Acrobat en Adobe Reader 5.0 en hoger.) /NOR /POL /PTB /RUM /RUS /SKY /SLV /SUO /SVE /TUR /UKR /ENU (Use these settings to create Adobe PDF documents best suited for high-quality prepress printing. Created PDF documents can be opened with Acrobat and Adobe Reader 5.0 and later.) /HUN >> /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ << /AsReaderSpreads false /CropImagesToFrames true /ErrorControl /WarnAndContinue /FlattenerIgnoreSpreadOverrides false /IncludeGuidesGrids false /IncludeNonPrinting false /IncludeSlug false /Namespace [ (Adobe) (InDesign) (4.0) ] /OmitPlacedBitmaps false /OmitPlacedEPS false /OmitPlacedPDF false /SimulateOverprint /Legacy >> << /AddBleedMarks false /AddColorBars false /AddCropMarks false /AddPageInfo false /AddRegMarks false /ConvertColors /ConvertToCMYK /DestinationProfileName () /DestinationProfileSelector /DocumentCMYK /Downsample16BitImages true /FlattenerPreset << /PresetSelector /MediumResolution >> /FormElements false /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles false /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ] >> setdistillerparams << /HWResolution [2400 2400] /PageSize [612.000 792.000] >> setpagedevice