Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Pre-event water contributions and streamwater residence times in different land use settings of the transboundary mesoscale Lužická Nisa catchment

Pre-event water contributions and streamwater residence times in different land use settings of... The objective of the study was to evaluate the spatial distribution of peakflow pre-event water contributions and streamwater residence times with emphasis on land use patterns in 38 subcatchments within the 687 km 2 large mesoscale transboundary catchment Luzická Nisa. Mean residence times between 8 and 27 months and portions of preevent water between 10 and 97% on a storm event peakflow were determined, using 18O data in precipitation and streamwater from a weekly monitoring of nearly two years. Only a small tracer variation buffering effect of the lowland tributaries on the main stem was observed, indicating the dominant impact on the mountainous headwaters on the runoff generation. Longest mean streamwater residence times of 27 months were identified in the nearly natural headwaters of the Jizera Mountains, revealing no ambiguous correlation between the catchment area and altitude and the mean residence time of streamwater. Land use control on the pre-event water portions were determined for three land use categories with percentage of urban areas from 0 to 10%, 10 to 20% and more than 20%. The fraction of pre-event water in the first category decreases from 97% to 65% with the increasing percentage of forest from 76% to 100%, revealing that forests may provide only a limited infiltration of precipitation due to leaf interception and soil water use for transpiration. Fractions of pre-event water of 39­87% in the second (agricultural catchments) and of 10­35% in the third (urbanized catchments) category increase with percentage of non-urban areas. Keywords: 18O isotope; Stormflow event; Peak pre-event contributions; Mean residence time; Land use. INTRODUCTION Tracing of hydrological processes in nested mesoscale catchments at scales between 100 and 1000 km2 (Uhlenbrook et al., 2004) is often hampered by differences and inconsistency of data between small and large scales. Upscaling of abundant process knowledge from small headwater catchments has been more widely performed to mesoscale catchments around 100 km2 of area, for example in Switzerland (Köplin et al., 2014) and UK (Tetzlaff et al., 2007). The role of groundwater contributions to streams and streamwater residence times in catchment hydrology has been widely acknowledged over the past decades. Several studies have addressed the role of soil cover (Capell et al., 2012), topography and geometry (McGuire et al., 2005, Soulsby et al., 2010) of nested smaller subcatchments in water pathways and residence times, often adopting tracer approaches. Hydrological processes at the mesoscale are also often related to land use, which becomes an important parameter in runoff modelling (Isik et al., 2013; Montzka et al., 2008; Niehoff et al., 2002). However, recent reviews of the assessment of subsurface water contributions to streams and streamwater residence times (Klaus and McDonnell, 2013; McDonnell et al., 2010) revealed a principal need for a better understanding of the spatial distribution of their patterns across catchment scales. Linkages between pre-event water contributions, residence times and landscape settings often show opposite results, explained by specific catchment conditions (Klaus and McDonnell, 2013). Despite of abundant knowledge of mixing processes at headwater scales, little is known about the transfer of the typical approaches (two-component mixing, sine-wave estimation of residence times) from small experimental catchments to more complex scales. Although new tracer analytical techniques facilitate a better in-situ temporal monitoring at smaller scales (e.g. Holko et al., 2011), the spatial distribution of the runoff generation parameters and their interpretation remain a challenge. More work on the understanding of hydrological processes is particularly needed in catchments of size towards 1000 km2. Successful examples of a consistent isotopic assessment of nested streamwaters of mesoscale and large-scale catchments in the contiguous US (Dutton et al., 2005; Kendall and Coplen, 2001) reveal important hydrological patterns and links to climatic and geological settings that are often hidden at headwater or large scale. Tracing of hydrological processes in the mesoscale catchments between 500 and 1000 km2 has been often performed through surveys of isotopic or chemical tracers along the stream courses. It has been typically linked to hydrological (Popescu et al., 2008), geochemical or pollution patterns (Markovics et al., 2010; Pardo et al., 2004), whereas the largescale surveys have addressed geochemical, climatic or anthropogenic patterns (Pawellek et al., 2002), often resulting from channel deviations and water abstractions (Herczeg and Leaney, 2011). This paper summarizes the nearly 2-years long period of hydrometeorological and isotopic monitoring in the Luzická Nisa catchment and its 38 subcatchments with the objective to evaluate magnitude and distribution of pre-stormflow-event water portion and streamwater residence times linked to the altitude, catchment area and land use patterns. Although several studies have addressed the increasing occurrence and magnitude of floods in the Luzická Nisa catchment in pan-European (Alfieri et al., 2014), and regional (Bissolli et al., 2011) context, little is known on the hydrological processes that cause the floods and pollution fluxes. These topics have been addressed dominantly at point or small headwater catchment scale (Kändler and Seidler, 2009; Sanda et al. 2014). Hydrological evaluation of the catchment has been largely missing. A synthesis over the transboundary mesoscale catchment was hampered by data inconsistency. Except for the small headwater Uhlíská in the Jizera Mountains (Sanda et al., 2014), this is the first study to employ environmental isotopic data in the larger Luzická Nisa mesoscale catchment. Study site The mesoscale catchment of the Luzická Nisa (Lausitzer Neisse in German) river from the Jizera Mountains in the Czech Republic to the German lowlands in the vicinity of Zittau is characterized by many land use types and a large variety of hydrological and hydrochemical patterns. The differences are mostly related to the interaction of the forested-mountainous Jizera Mountains in the Luzická Nisa subcatchment, the agricultural-lowland Zittau basin in the subcatchment Mandava (Mandau in German) and the major cities of Liberec and Jablonec n.N. Names of the rivers in this paper match the language of the country of their spring. The Luzická Nisa catchment (Fig. 1) covers parts of Czech Republic, Germany and Poland and reaches from the headwaters region in the Jizera, Luzice and Zittau Mountains down to the gauge station on the Luzická Nisa in Zittau (German/Polish border). The tributaries above Zittau include small rivers originating in the Jizera Mountains (e.g. Bílá Nisa, Harcovský potok, Cerná Nisa, Jeice) and along the Jestd Ridge (Frantiskovský potok). The most prominent tributary, the Mandava River, originates in the north of Bohemian-Saxonian Swiss sandstone area, flows southeast-east across the region of Upper Lusatia north of Zittau Mountains, and joins the Luzická Nisa in Zittau. Important tributaries of the Mandava are Luznicka (Lausur in German) and Landwasser in Germany. The catchment (687 km2) divides into Czech (476 km²), German (205 km²) and Polish (6 km²) parts, with altitudes between 886 m a.s.l. in the Jizera Mountains (Olivetská hora summit) and 229 m a.s.l. at the gauge station Zittau. Water is transfered from outside of the Luzická Nisa catchment for the drinking water supply of the metropolitan area of Liberec and Jablonec n. N. and related settlements forming agglomeration of about 200 thousand inhabitants. Both water reservoirs of Josefv Dl on the Kamenice River and Sous on the Cerná Desná River are located in the Labe river regional catchment. They collect water from the headwaters of the highest locations of the Czech part of the Jizera Mountains reaching above 1000 m a.s.l. This results in additional volumes and hydrochemical signatures in the Luzická Nisa catchment (Farský, 1992). The Luzická Nisa catchment belongs to the temperate climate zone in the transient region between maritime and continental conditions, with precipitation maxima in July and August (Pohle et al., 2015). Heavy rains occur frequently during spring and summer. Long term annual precipitation amount varies between approximately 1400 mm·a­1 in higher altitudes of the mountains and 640 mm·a­1 in the lowlands. The Jizera Mountains are characterized by long lasting snow cover of up to 160 days/year. Mean annual temperature varies between 8°C in the lowlands and 5°C in the mountains (Sanda et al., 2014). The complex geological structure of the catchment is dominated by granites and granodiorites in the Jizera Mountains (Klomínský, 1969), sandstones in the Zittau Mountains and granite with basaltic and phonolitic hilltops in the western part. Fig. 1. Map of the study area (main stem of the Luzická Nisa is between CZ1 and D1). The lowlands are dominated by loess. The soils have different properties, particularly with respect to runoff formation, erosion, solute transport and water storage capacity. While soils on mountainous hillslopes are mostly shallow, often skeleton rich dystric cambisols, podzols (Nikodem et al., 2013) and leptosols with high infiltration rate and low storage capacities, the valleys are filled by organosols that release humic and fulvic acids. Agricultural soils are mostly luvisols and stagnic luvisols with high silt content and low infiltration rate. They are partly affected by poorly drained horizons. Luvisols are easily erodible and contribute to suspended load in the rivers. The lowland flood plains are typically covered by gleysols (Kändler and Seidler, 2009). Land use differs according to the soil types. While the mountainous areas are dominated by spruce forests and pastures (Pavl et al., 2007), the fertile soils in the lowlands are agriculturally exploited. Towns and villages in this region are located directly upon the streams. The rivers are linking towns and villages, river beds are often artificially changed and the banks are under revetment. Particularly the agriculturally used loess soils in the lower catchment parts are endangered by surface runoff and erosion, leading to high suspended particle loads and high nutrient concentrations in the rivers (Kändler and Seidler, 2009). DATA AND METHODS Monitoring The long-term hydrometeorological monitoring network in the Luzická Nisa catchment includes five precipitation (Table 1a) monitoring stations (four in the Czech Republic and one in Saxony) and seven runoff (Table 2) gauging stations (four in the Czech Republic and three in Saxony). Precipitation gauges Mlýnice, Chibská, are operated by the River Labe and River Ohe Authorities respectively. The Liberec station was a temporal station for the purposes of this project and station Bedichov is operated by the Czech Hydrometeorological Institute. The Czech runoff gauges are operated by the River Labe Authority (the Uhlíská gauge is operated by the Czech Technical University in Prague), and the German gauges are operated by the Saxonian Evironmental Authority. Weekly water samples for determination of 18O/2H content in rainwater and streamwater were collected in the period from September 2012 until April 2014 typically in weekly mode. They included five (Table 1b) precipitation stations (four in the Czech Republic and one in Saxony), and 38 streamwater (Table 3) profiles (24 in the Czech Republic and 14 in Saxony). Streamwater discrete samples were manually collected into 20 ml HDPE bottles and stored frozen until the analysis. The streamwater sampling campaign during each week was performed within one day along both the main Luzická Nisa river stem and its tributaries in the Czech Republic and the Mandava River and its tributaries and two stations on the Luzická Nisa River in Germany. The whole catchment was therefore sampled within a few hours. Precipitation was collected in a simple funnel-container device with oil protection of the collected water. In case of snow precipitation, samples were manually collected from the funnel into a closed plastic container and melted. The isotope monitoring in the Uhlíská catchment has been gradually established since 2006, delivering data for studies at catchment (Sanda et al., 2014), hillslope (Dohnal et al., 2006; Dusek et al., 2012) and point (Snhota et al., 2008) scales. Laboratory analyses The analyses of 18O and 2H were performed at the Czech Technical University in Prague, Faculty of Civil Engineering, using the Liquid Water Isotope Analyzer, LGR Inc. device (Penna et al., 2010). The values are expressed as 18O, 2H in of V-SMOW with typical precision of 18O ±0.15 and 2H ±0.7 V-SMOW (Vienna Standard Mean Ocean Water). Contents of 18O and 2H at CZ17, CZ24, D15 are also analyzed in monthly samples within the framework of the IAEA isotope hydrology databases GNIP and GNIR (Vitvar et al., 2007). Table 1. Precipitation amount (a-left) and precipitation sampling sites for the water isotopes (b-right). Precipitation XI/12-X/13 (mm) 1613 1119* 1222 1066 965 Precipitation 27-Aug-13 to 4-Sep-13 (mm) 104.0 37.3 51.1 31.6 14.0 27-Aug-13 to 4-Sep-13 (d18O ()) ­4.76 ­3.76 ­4.01 ­4.37 ­4.66 Amplitude X/12-IV/14 (d18O ()) 3.01 4.16 3.88 3.78 3.23 Provider CHMI project P. Labe P. Ohre ZÖF Zittau Location Bedichov Liberec Mlýnice Chibská Zittau Altitude (a.s.l.) 770 367 390 440 235 Profile CZ24 CZ25 CZ26 CZ50 D15 Location Uhlíská Liberec Oldichov Lucany Zittau Altitude (a.s.l.) 825 367 415 565 235 * correlated with Mlýnice on VII/13-IV/14 data Table 2. Outflow characteristics of the streams in the catchment. Runoff gauge - river / ID Uhlíská - Cerná Nisa / CZ17 Prosec - Luzická Nisa / CZ4 Mnísek - Jeice / CZ9 Varnsdorf - Mandava / above D4 Zittau - Mandava / D3 Hartau - Luzická Nisa / (CZ16-D2) Zittau - Luzická Nisa / D1 Altitude (a.s.l.) 792 401 375 314 239 241 235 Catchment area (km2) 1.2 53.7 32.2 88.9 293.9 377.5 694.0 Mean outflow XI/12-X/13 (m3/s) 0.05 1.42 0.35 1.53 3.68 7.12 11.85 Specific outflow XI/12-X/13 (m3/s/km2) 41.0 26.4 10.9 17.2 12.5 18.9 17.1 Table 3. List of streamwater sampling sites, land use characteristics and results of the analyses utilizing stable isotopes in streams in the Luzická Nisa catchment ,,wtp" abbreviates water treatment plant in Liberec. "Lake" denotes the catchment of the artificial lake Olbersdorfer See. "Urban" denotes small catchments entirely located in developed zones and drained through artificially designed channels or pipes. Landuse categories: Forest-more than 90% of non-urban areas, with a dominant forest component greater than the sum of urban land and arable land and grassland; Agriculture-more than 80% of non-urban areas, with a prevailing arable land and grassland component; Urbanbelow 80% of non-urban areas; Mix-mixed category of more than 80% of non-urban land with prevailing forest, but not dominant (greater than the sum of other components). Gauge altitude (a.s.l.) 235 237 239 302 340 325 353 275 291 247 243 240 383 387 565 529 472 401 381 354 339 349 351 333 333 333 333 333 375 375 394 387 290 286 265 245 792 407 Catchment average altitude (a.s.l.) 440 479 402 434 416 487 403 355 369 344 410 240 433 503 639 628 620 603 584 543 530 urban urban 528 urban urban 528 528 602 576 620 596 486 526 503 490 828 668 Catchment area (km2) 694.0 377.5 293.9 162.4 75.1 48.4 21.8 49.7 28.4 9.8 15.9 lake 18.3 16.8 4.2 7.4 48.1 56.4 63.5 120.7 137.3 urban urban 138.8 urban urban 138.8 138.8 7.1 32.8 15.2 4.2 77.9 291.0 320.7 349.5 1.2 1.0 Non urban land (%) 83 83 84 84 83 93 80 83 80 87 80 91 91 94 83 86 74 76 76 74 69 60 54 69 ­ ­ ­ ­ 89 94 98 99 93 76 82 83 100 99 Urban land (%) 17 17 16 16 17 7 20 17 20 13 20 9 9 6 17 14 26 24 24 26 31 40 46 31 ­ ­ ­ ­ 11 6 2 1 7 24 18 17 0 1 Forest (%) 39 49 28 33 24 49 15 9 9 6 15 45 25 52 34 45 47 50 48 46 43 56 50 43 ­ ­ ­ ­ 75 74 89 79 49 51 51 49 100 95 Arable and grassland (%) 42 32 53 48 57 41 62 72 68 80 63 39 65 39 48 39 25 24 26 26 25 3 3 25 ­ ­ ­ ­ 13 20 9 20 42 23 29 32 0 4 Preevent/total water ratio (­) 0.61 0.71 0.76 0.74 0.54 0.63 0.46 0.57 0.39 0.82 0.71 0.87 0.72 0.59 0.44 0.47 0.35 0.29 0.32 0.33 0.28 0.10 0.30 0.28 0.49 0.10 0.31 0.30 0.97 0.78 0.69 0.72 0.74 0.36 0.48 0.51 0.47 0.65 Mean residence time (months) 18.3 21.1 14.4 11.8 12.6 9.4 11.9 13.2 16.4 16.5 27.4 8.8 14.8 8.2 21.3 19.6 14.5 16.5 15.5 15.4 15.8 ­ ­ 17.1 ­ ­ ­ ­ 15.5 22.8 25.6 25.3 21.3 18.4 19.0 20.6 14.5 20.7 Profile D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 D14 CZ1 CZ2 CZ3 CZ4 CZ5 CZ6 CZ7 CZ7-1 CZ7-2 CZ8 CZ8-1 CZ8-1a CZ8-2 CZ8-3 CZ9 CZ10 CZ11 CZ12 CZ13 CZ14 CZ15 CZ16 CZ17 CZ44 River Luzická Nisa Luzická Nisa Mandava Mandava Mandava Luznicka Leutersdorfer Bach Landwasser Landwasser Grundbach Goldbach Olbersdorfer See Mandava Luznicka Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Frantiskovský potok Jizerský potok Luzická Nisa city of Liberec city of Liberec Luzická Nisa Luzická Nisa Fojtecký potok Jeice Malá Jeice Jeice Jeice Luzická Nisa Luzická Nisa Luzická Nisa Cerná Nisa Malá Jeice Location Zittau Zittau Zittau above Grossschönau Seifhennersdorf Neuschönau Seifhennersdorf Mittelherwigsdorf Oderwitz Zittau Zittau Zittau Rumburk Dolní Podluzí Lucany n. N. Jablonecké Paseky down Jablonec n. N. Prosec Vratislavice centre of Liberec above wtp Liberec Liberec 1 m below wtp discharge of wtp spillway of wtp 20 m below wtp 50 m below wtp Mnísek Mnísek with Fojtecký brook Oldichov v Hájích Oldichov v Hájích Chrastava Chrastava Chotyn Hrádek nad Nisou Uhlíská Betlém Landuse Agric. Mix Agric. Agric. Agric. Forest Agric. Agric. Agric. Agric. Agric. Agric. Agric. Forest Agric. Mix Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Forest Forest Forest Forest Forest Urban Mix Mix Forest Forest Data interpretation Isotopic separation of event and pre-event water was based on the mixing equation the period August 27­ September 3, 2013 (see Table 1 for details). The cs values are equal to the 18O content in streamwater at each of 38 profiles on August 24, 2013 ­ the last sampling, considered as baseflow before the event. Water residence times Streamflow residence times were estimated using the seasonal sine-wave variations of 18O in monthly step in the streamflow and cumulative precipitation (Kralik, 2015). The seasonal variations in 18O are expressed (Eq. 2) Rs = Qs ct - cn = Qt cs - cn (1) where Qt ­ total (event & pre-event) flow (m3/s), Qs ­ pre-event flow fraction (m3/s) ( V-SMOW), ct ­ 18O content in the streamwater, cs ­ 18O content in the pre-event baseflow ( VSMOW), cn ­ 18O content in the rainwater during the event ( V-SMOW), and Rs ­ volumetric ratio of the pre-event water in the total outflow. The calculation was performed for the 38 studied gauges at the peakflow September 3, 2013, using the average and standard deviation of 18O content (cn) in precipitation of five sampled stations (Uhlíská, Liberec, Oldichov, Lucany, Zittau) during 18O mean 18O Asin 2t / b c (2) where mean(18O) ( V-SMOW) is the mean value of 18O, A ( V-SMOW) is the seasonal amplitude of 18O, b is the period of a seasonal cycle (one-year period is 2), t is time (months), and c (rad) is the phase shift. Parameters mean(18O), A and c were obtained by fitting Eq. (2) on the experimental data via least squares optimization. Mean residence time (MRT) is reflected in the decrease of the input amplitude in precipitation (Ap) relative to the output amplitude (e.g. in streamflow A) in the linear reservoir, according to Eq. (3). 2 1 Ap MRT 1 b A 0.5 (3) Luzické Mountains is observed in the Mandava tributary Luznicka (D6, D14) in more depleted 18O values compared to profiles in their vicinity (D8, D9). The lowland tributary Landwasser coming from northwest (D9, D8) has no impact of mountains. The Mandava tributaries Grundbach (D10) and Goldbach (D11) drain the area of the Lusatian fault where possible impact of groundwater circulation with particular isotopic composition and smaller isotopic variations may be hypothesized. Finally, D12 represents the outlet of the Olbersdorfer See (lake) with associated evaporative effects on the 18O content. Isotopic separation of pre-event water and calculation of water residence times Isotopic separation of event and pre-event water (Fig. 4a) was performed for the peakflow on September 3rd, 2013 with causal pre-event rainfall during August 27­September 4, 2013. This event has occured after three weeks with no precipitation and caused a discharge rise at all studied profiles. It is hypothesized (Fig. 4b) that the minor precipitation amounts on August 28 and September 1 contributed to the increase of the antecedent moisture, causing a major streamflow peak on September 3. This event was characterized by particularly enriched 18O values in the causal pre-event rainfall (Table 1b), reaching from ­3.76 V-SMOW at Liberec to ­4.76 V-SMOW at Uhlíská, with average ­4.13 V-SMOW and standard deviation 0.41 V-SMOW. 18O values in streamwater at the peakflow (September 3rd, 2013) reached up to ­5.8 V-SMOW from the background of ­10.0 to ­8.5 V-SMOW (Figs. 4cd). Fig. 5 and Table 3 display the calculated ratio of pre-event water (Eq. 1). The highest portions of the pre-event water were identified in two catchments CZ9 and D12, affected by the lakes Fojtka and Olbersdorfer, respectively. It is hypothesized that older water was released from the lakes during the stormflow event, causing a higher portion of the pre-event water. High portions of the pre-event water appeared also in the nearly natural Jeice (CZ10­CZ13, C44) subcatchment, covered dominantly by forests and meadows, and with little or no impact of settlements. The source of elevated pre-event water portions at D10 and D11 may include groundwater inflow along the Lusatian fault. More than 70% of the total runoff has pre-event origin in the Mandava catchment (D13, D11, D4 and D3). The main stem of the Luzická Nisa River (CZ1­D1) ranges from 69% to 79% of pre-event water, with lower values in the agglomerations (CZ4, CZ7). The Leutersdorfer brook (D7) and the Landwasser (D8, D9) were drained by 46%, 57% and 39% of pre-event water, respectively. Lowest values of Rs were identified at the profile CZ8-1a (10%) discharging from the water treatment plant. It can be hypothesized that the dominant rainfall amount prior to the event at Bedichov (Table 1a) resulted in significant portions of event water along the Luzická Nisa main stem, whereas the smaller causal rainfall amounts in the western lowland part of the catchment (Zittau) resulted in a less pronounced rapid runoff in the Mandava catchment. Five calculations of mean residence times using sine-wave amplitudes of five rainfall stations were carried out for each streamflow gauge (Tab. 1b). The average amplitude of 18O values in precipitation at the five sampling stations in the period October 2012­April 2014 was 3.61 V-SMOW with a standard deviation of 0.47 V-SMOW. The highest mean residence time of streamwater (25-27 months) were estimated in the nearly natural Jeice (CZ11, CZ12) and in the Goldbach (D11) catchments (Table 3). where (1/b) = 12/2 is the factor for the residence time in months. This fitting was performed for the 18O values at all five precipitation sampling stations Uhlíská, Liberec, Oldichov, Lucany, Zittau (cumulative weekly samples) and the stream (grab weekly samples) at the studied 38 profiles for the period October 2012­April 2014. The variability of Ap (expressed as average and standard deviation of each of the five precipitation stations) is therefore caused by the different amplitudes of 18O values at the respective five rainfall stations (see Table 1). The coefficient of determination R2 expresses the linear least square determination. RESULTS AND DISCUSSION Monitoring Fig. 2 shows selected hydroclimatical and isotopic characteristics of precipitation and streamwater for the period September 2012­April 2014. It displays daily temperature and precipitation amount at the wettest station Bedichov, daily discharge at headwaters of the Luzická Nisa (CZ17), catchment outlet (D1) and main tributary (D3), weekly 18O content in precipitation at the five sampling stations (see also Tab. 1b), and weekly 18O content in streamwater at seven selected profiles. No significant gradients in isotopic composition of precipitation were observed, indicating a homogeneous origin of precipitation over the study area. Very similar 18O patterns in streams are observed during winter and early spring. The 18O depleted streamwater of the mountain headwaters occurs only along the main Luzická Nisa stem and its tributaries in the Czech Republic and sustains during summer and autumn in particular. In contrast, streamwater in the the Mandava and Luznicka (D3 and D6) follow patterns of elevated 18O, due to non-existing impact of mountainous headwaters. These overall patterns were biased by the isotopically distinct rainfall-runoff event in September 2013. The 18O content in streamwaters at all 38 monitoring profiles is shown in Fig. 3 (top panel). The main stream of the Luzická Nisa (15 profiles from CZ1 to D1) shows a gradual increase of 18O values and decrease of their variations along with the progressing mixing towards the catchment outlet. The median 18O values along the main stem of Luzická Nisa reach from ­10 to ­9.5 V-SMOW, which corresponds to the median 18O values in precipitation (Fig. 2). The profiles CZ17 (Cerná Nisa) and CZ9-13, CZ44 (Jeice) represent mountainous headwaters from the Jizera Mountains with depleted 18O values. Streams at the profiles CZ7-1 and CZ7-2 (Frantiskovský and Jizerský brooks, resp.) drain entirely developed city areas, and the profiles CZ8-1 and CZ8-1a are outlet and spillway of the city water treatment plant. The 18O values at these four profiles are therefore highly affected by human impact. Profiles D13, D5, D4 and D3 represent the Mandava River with no major impact of mountainous headwaters. Impact of the precipitation (mm/day) 30 20 10 0 -10 -20 -30 -40 3 3 CZ17 Uhlíská - Cerná Nisa D3 Zittau - Mandava D1 Zittau - Luzická Nisa 3 3 -6 -9 -12 -15 -18 -21 -24 -5 CZ24 - Uhlíská CZ50 - Lucany CZ26 - Oldichov CZ25 - Liberec D15 - Zittau 3 D1 Luzická Nisa Zittau D3 Mandava Zittau -6 -7 D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn -8 -9 -10 -11 3 3 Fig. 2a, b, c, d (top to bottom). a) Air temperature and daily precipitation at the Bedichov station, b) streamflow and c) 18O content in precipitation and d) in streamwater at selected profiles, for the period September 2012 ­ April 2014. The latter may be associated with deeper groundwater contribution along the Lusatian fault. The main stream of the Luzická Nisa shows streamwater residence times between 15 and 20 months (with lowest values in the agglomeration CZ3-CZ7) and the Mandava (D13, D5, D4, D3), Grundbach (D10) and Landwasser (D8, D9) between 10 and 16 months. The relatively shorter streamwater residence times in the lowland Mandava catchment and its tributaries may be attributed to river terraces and alluvia. Therefore, the bulk of the river water probably comes from areas which are relatively close to the river, resulting 3 0.01 -3 daily average temp. Bedrichov (°C) -4 -5 -6 Jeice and Mandava are major L. Nisa tributaries with their stream networks urban denotes minor streams originating in Liberec (CZ7-1 -2) and outlet/overspill of the wtp - water treatment plant (CZ8-1, 1-a) water reservoir Olbersdorfer See Cerná Nisa 792 m a.s.l 565 m a.s.l. 235 m a.s.l. 18O ( SMOW) -7 -8 -9 -10 -11 -12 -13 forest Jeice forest urban 407-290 m a.s.l Mandava agricultural 383-247 m a.s.l Luzická Nisa stem stream agricultural urban Mandava tributaries forest and agricultural mix -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 low baseflow 30.10.12 iso-light rain 16.4.13 baseflow 27.8.13 iso-heavy storm 3.9.13 CZ8 CZ1 CZ2 CZ3 CZ4 CZ5 CZ6 CZ7 CZ9 D13 D14 D11 CZ8-1a CZ10 CZ17 CZ14 CZ15 CZ16 CZ44 CZ11 CZ12 CZ8-2 CZ8-3 CZ7-1 CZ7-2 CZ8-1 CZ13 Fig. 3a, b (top to bottom). 18O in streamwaters along the principal water courses and their tributaries (number of collected samples in brackets). a) Box-Whisker-Plots for the period September 2012 - April 2014. b) Evolution of 18O along the river network in selected hydrological situations (light isotopic rain, pre-event baseflow, heavy isotopic storm, and typical baseflow). in shorter mean residence times (Dósa et al., 2011). The shortest residence times were identified in the Luznicka catchment (D9 and D14) with very shallow aquifer. Residence time was not calculated for the channelled streams CZ7-1and CZ7-2 and the overflow and spillway from the water treatment plant CZ81 and CZ8-1a. Similarly to several previous studies on mesoscale nested catchments (Soulsby et al., 2010; Tetzlaff et al., 2007), the mean streamwater residence times in the Luzická Nisa catchment reveal no significant correlation with the catchment altitude and area (Figs. 6a, b). In contrast, the pre-event water portion Rs increases with the catchment area (Fig. 6c). This supports the hypothesis that larger rivers have larger portion of pre-event water from river banks and alluvium. This portion, however, do not always imply longer streamwater residence time in the lowland subcatchments, because the groundwater discharge from lowland river banks is often younger than contributions from deeper fractures in the granitic headwaters. For example, Dósa et al. (2011) used 18O in the sine-wave method and reported a shorter mean residence time in streamwater of the Váh River below the Slovak Tatra Mountains (13 months) than in the headwaters Jalovecký potok Creek (19 months). Similarly, Soulsby et al. (2010) reported a shorter mean residence time of the Scottish streamwater Upper Dee (600 days) than in its headwaters (more than 2 years). In contrast, Capell et al. (2011, 2012) have found a largely muted response of stable isotopes in lowland streamwater in the 749km2 large mesoscale catchment North Eck in Scotland (very similar catchment size of 687 km2 to the Luzická Nisa). They have ascribed this result to the strong differences in geological and soil conditions between the lowland catchments and headwaters, which causes a strong dumping of the tracer input. According to Fig. 2d, however, this phenomemon is not observed in the Luzická Nisa catchment. All profiles along the main Luzická Nisa stem and the lowland Mandava tributary reveal similar annual variations of 18O of around 3 VSMOW (excluding the event from September 2013). During autumn baseflow periods, the entire course of the Luzická Nisa carries isotopically depleted waters, originating in the mountainous wetlands with dominantly snowmelt-induced recharge (Sanda et al., 2014). We argue that the pronounced isotopic variability of streamwaters throughout the entire Luzická Nisa catchment is caused by relatively shallow aquifers in the dominantly Neogene Zittau basin, which precludes development of deeper aquifers. D12 D10 D2 D1 D5 D4 D3 D7 D6 D9 D8 CZ17 (153) . CZ1 (83) CZ2 (83) CZ3 (82) CZ4 (82) CZ5 (82) CZ6 (83) CZ7 (82) CZ8 (83) CZ8-2 (67) CZ8-3 (67) CZ14 (83) CZ15 (82) CZ16 (82) D2 (85) D1 (85) . CZ7-1 (75) CZ7-2 (74) . CZ8-1 (67) CZ8-1A (58) . CZ44 (73) . CZ9 (82) CZ10 (82) CZ11 (83) CZ12 (82) CZ13 (82) . D13 (98) D5 (85) D4 (85) D3 (85) . D7 (83) . D14 (81) D6 (84) . D9 (88) D8 (61) . D10 (84) . D11 (87) . D12 (58) precipitation Bedrichov (mm/day) CZ17 Uhlíská - Cerná Nisa D3 Zittau - Mandava D1 Zittau - Luzická Nisa precipitation Bedichov (mm/hour) 8 CZ17 Uhlíská - Cerná Nisa 6 D3 Zittau - Mandava D1 Zittau - Luzická Nisa 4 22-Aug-13 29-Aug-13 10-Oct-13 5-Sep-13 3-Oct-13 19-Sep-13 12-Sep-13 26-Sep-13 28-Aug-13 4-Sep-13 2-Sep-13 3-Sep-13 5-Sep-13 6-Sep-13 7-Sep-13 8-Sep-13 27-Aug-13 29-Aug-13 30-Aug-13 3 9-Sep-13 -5.5 -6.0 -6.5 -7.0 D1 Luzická Nisa Zittau D3 Mandava Zittau D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn CZ17 - Cerná Nisa Uhlíská CZ17 - Cerná Nisa, automated -5.5 -6.0 -6.5 -7.0 D1 Luzická Nisa Zittau D3 Mandava Zittau D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn CZ17 - Cerná Nisa Uhlíská CZ17 - Cerná Nisa, automated -7.5 -8.0 -8.5 -9.0 -9.5 -10.0 -10.5 22-Aug-13 03-Oct-13 -7.5 -8.0 -8.5 -9.0 -9.5 -10.0 -10.5 19-Sep-13 10-Oct-13 27-Aug-13 28-Aug-13 29-Aug-13 30-Aug-13 29-Aug-13 3 05-Sep-13 12-Sep-13 26-Sep-13 0 02-Sep-13 03-Sep-13 04-Sep-13 05-Sep-13 06-Sep-13 07-Sep-13 08-Sep-13 Fig. 4a, b, c, d (top left to bottom right). Rainfall-runoff episode used for isotopic separation of the peakflow of September 3, 2013. a) Precipitation and discharge during the entire episode August 22­ October 10, b) Detailed record during the peakflow period August 27­ September 9, c) 18O in streamflow during the entire episode August 22­ October 10, and d) 18O in streamflow during the peakflow period August 27­September 9, showing the response of the catchments on the storm event with significantly different isotopic content . Bold lines for CZ17 headwater catchment Uhlíská indicate manual weekly and automated (daily or 4x daily sampling at the only station of the network), showing relatively good capture of the event by weekly sampling, including the peakflow. Cerná Nisa 792 m a.s.l Rs - Rs 565 m a.s.l. 235 m a.s.l. forest Jeice forest urban 407-290 m a.s.l Mandava agricultural 383-247 m a.s.l Mandava tributaries forest and agricultural -0.1 -0.2 Luzická Nisa stem stream agricultural urban mix Fig. 5. Volumetric ratio of pre-event water in the peakflow during the event on September 3 2013 based on 18O isotopic separation and mean residence time (MRT) of the streamwaters. The error bars show the variability of the results calculated using different 18O contents in the causal rainfall at the five sampling stations. MRT - mean residence time (month) water reservoir Olbersdorfer See MRT 09-Sep-13 all streams all streams Luzická Nisa main stem D11 Luzická Nisa main stem R² = 0.171 mean residence time (months) mean residence time (months) R² = 0.453 15 CZ17 10 R² = 0.256 average catchment altitude (m a.s.l.) catchment area (km2) catchment area (km2) Fig. 6a, b, c. Three examples of relationships between altitude, area, mean residence time and fraction of pre-event water (for Luzická Nisa ­ main stem, when applicable). The error bars for the pre-event water fraction show the variability of the results calculated using different 18 O contents in the causal rainfall at the five sampling stations. The error bars for the mean residence times show the variability of residence time calculations using different amplitudes of 18O contents in precipitation at the respective five rainfall stations. 1.00 0.95 0.90 0.85 0.80 0.75 R² = 0.531 0.70 0.65 0.60 fraction of non-urban area in the catchment (-) fraction of forest in the catchment (-) 0.90 R² = 0.739 0.85 0.92 0.42 0.40 0.90 fraction of the non-urban in the catchment (-) fraction of urban area in the catchment (-) 0.30 0.28 R² = 0.831 0.82 R² = 0.632 0.80 0.20 0.0 0.1 0.2 0.3 0.4 0.5 Fig. 7a, b, c, d. Relationship of fraction of pre-event water and landuse. All nested catchments (a-top left), small forested Czech catchments, category I (b-top right), agricultural catchments, mostly German, category II (c-bottom left), most urbanized Czech catchments Jablonec n.N, Liberec, Chrastava , category III (d-bottom right). The landuse categories I, II and III are defined in Table 3. The error bars for the pre-event water fraction show the variability of the results calculated using different 18O contents in the causal rainfall at the five sampling stations. Fig. 7 shows the relationship between landuse (Table 3) and the pre-event water fraction. Fig. 7a shows a weak overall increase of the pre-event water fraction with increasing portion of non-urban landuse in the respective catchment, where nonurban is understood as sum of forest, arable land, grassland, orchards and open water. Three distinct categories evolve upon distribution of catchments according to the degree of non-urban landuse (Table 3). The category "Forest" of 90­100% nonurban landuse contains all Czech mountainous headwaters CZ10-CZ13, CZ17 and CZ44 (Fig 7b), with a dominant portion of forest which is greater than the sum of urban land, arable land and grassland. The fraction of pre-event water in this group decreases with the percentage of forest, revealing that forests may provide only a limited infiltration of precipitation due to leaf interception and soil water use for transpiration (Nadezhdina et al., 2010). Not included in this assessment are the Luznicka stream profiles D6 and D14. Despite the low degree of urban landuse, these catchments have only a shallow aquifer that does not allow for storage of and mixing with preevent water. The category "Agriculture" of 80-90% of nonurban landuse dominantly contains the Mandava, Landwasser and Leutersdorfer Bach subcatchments, with prevailing arable and grassland. In this group, the fraction of pre-event water is directly related to the non-urban landuse degree (Fig. 7c). The category "Urban" contains most developed catchments below 80% of non-urban land. In this group the increasing fraction of pre-event water is accompanied by a decrease of the percentage of urban landuse. Other catchments ("Mix") that contain more than 80% of non-urban land, but no dominant forest component such as in category "Forest", are not included in Fig. 7. They are considered as mixed-landuse catchments. CONCLUSIONS The study presents a readily available approach of an assessment how individual subcatchments contribute to the hydrological response of a larger mesoscale catchment with highly heterogeneous land use and landscape characteristics. It improves the understanding of the role of topography, geology and land use with respect to origin, mixing and residence time of water in the subcatchments. The study also shows the variability of calculated residence times and pre-event water fractions, caused by the uncertainty and heterogeneity of the isotopic rainfall input to the catchment. Although the approach is based on the standard two-component mixing techniques, it provides a comparison of heterogeneous settings that cannot be assessed using techniques at hillslopes or headwater scales. Unlike several previous studies conducted typically in less developed catchments, the Luzická Nisa catchment includes a large variety of landscapes to explore phenomena such as the impact of urbanization on baseflow and its contributions to peak event flow or impact of the lowland alluvial riparian zone on the streamwater residence times. This assessment may be used in both gauged and ungauged mesoscale basins. Acknowledgement. The research was funded by the Czech Science Foundation, project No. 14-15201J and by EU programme Ziel3/Cíl3, No. 100114993 and student grant of CTU SGS16/143/OHK1/2T/11. Authors thank Labe River Authority and Ohe River Authority for hydrological data provision. REFERENCES Alfieri, L., Salamon, P., Bianchi, A., Neal, J., Bates, P., Feyen, L., 2014. Advances in pan-European flood hazard mapping. Hydrol. Proc., 28, 4067­4077. Bissolli, P., Friedrich, K., Rapp, J., Ziese, M., 2011. Flooding in eastern central Europe in May 2010 - reasons, evolution and climatological assessment. Weather, 66, 147­153. Capell, R., Tetzlaff, D., Malcolm, I.A., Hartley, A.J., Soulsby, C., 2011. Using hydrochemical tracers to conceptualise hydrological function in a large scale catchment draining contrasting geologic provinces. J. Hydrol., 408, 164­177. Capell, R., Tetzlaff, D., Soulsby, C., 2012. Can time domain and source area tracers reduce uncertainty in rainfall runoff models in larger heterogeneous catchments? Water Resour. Res., 48, W09544. Dohnal, M., Dusek, J., Vogel, T., 2006. The impact of the retention curve hysteresis on prediction of soil water dynamics. J. Hydrol. Hydromech., 54, 258­268. Dósa, M., Holko, L., Kostka, Z., 2011. Estimation of the mean transit times using isotopes and hydrograph recessions. Die Bodenkultur, 62, 47­52. Dusek, J., Vogel, T., Dohnal, M., Gerke, H.H., 2012. Combining dual-continuum approach with diffusion wave model to include a preferential flow component in hillslope scale modeling of shallow subsurface runoff. Adv. Water Res., 44, 113­125. Dutton, A., Wilkinson, B.H., Welker, J.M., Bowen, G.J., Lohmann, K.C., 2005. Spatial distribution and seasonal variation in 18O/16O of modern precipitation and river water across the conterminous United States. Hydrol. Process., 19, 4121­4146. Farský, I., 1992. A comment on the anthropogenic impacts at the upper flow of the river Luzická Nisa. Sborník ces. geogr. spol., 97, 26­32. Herczeg, A.L., Leaney, F.W., 2011. Review: environmental tracers in arid-zone hydrology. Hydrogeol. J., 19, 17­29. Holko, L., Kostka, Z., Sanda, M., 2011. Assessment of Frequency and Areal Extent of Overland Flow Generation in a Forested Mountain Catchment. Soil & Water Res., 6, 1, 43­53. Isik, S., Kalin, L., Schoonover, J.E., Srivastava, P., Lockaby, G.B., 2013. Modeling effects of changing land use/cover on daily streamflow: An artificial neural network and curve number based hybrid approach. J. Hydrol., 485, 103­112. Kändler, M., Seidler, C., 2009. Hydrochemical Load in a Small River Following Heavy Rain Events. Folia Geographica, Series Geographica-Physica, 40, 27­32. Kendall, C., Coplen, T.B., 2001. Distribution of oxygen-18 and deuterium in river waters across the United States. Hydrol. Process., 15, 1363­1393. Klaus, J., McDonnell, J.J., 2013. Hydrograph separation using stable isotopes: review and evaluation. J. Hydrol., 505, 47­64. Klomínský, J., 1969. The Krkonose-Jizera granitoid massif. Sb. geol. vd, 15, 1­134. Köplin, N., Viviroli, D., Schädler, B., Weingartner, R., 2014. How does climate change affect mesoscale catchments in Switzerland? - A framework for a comprehensive assessment. Adv. Geosci., 27, 111­119. Kralik, M., 2015. How to estimate mean residence times of groundwater. Proc. Earth Planet. Sci., 13, 301­306. Markovics, R., Kanduc, T., Szramek, K., Golobocanin, D., Milacic, R., Ogrinc, N., 2010. Chemical dynamics of the Sava riverine system. J. Eviron. Monit., 12, 2165­2176. McDonnell, J.J., McGuire, K., Aggarwal, P., Beven, K.J., Biondi, D.,Destouni, G., Dunn, S., James, A., Kirchner, J., Kraft, P., Lyon, S., Maloszewski, P., Newman, B., Pfister, L., Rinaldo, A., Rodhe, A., Sayama, T., Seibert, J., Solomon, K., Soulsby, C., Stewart, M., Tetzlaff, D., Tobin, C., Troch, P., Weiler, M., Western, A., Wörman, A., Wrede, S., 2010. How old is streamwater? Open questions in catchment transit time conceptualization, modelling and analysis. Hydrol. Process., 24, 1745­1754. McGuire, K.J., McDonnell, J.J., Weiler, M., Kendall, C., McGlynn, B.L., Welker, J.M., Seibert, J., 2005. The role of topography on catchment scale water residence time. Water Resour. Res., 41, 1­14. Montzka, C., Kanty, M., Kunkel, R., Menz, G., Vereecken, H.,Wendland, F., 2008. Modelling the water balance of a mesoscale catchment basin using remotely sensed land cover data. J. Hydrol., 353, 322­334. Nadezhdina, N., David, T.S., David, J.S., Ferreira, M.I., Dohnal, M., Tesa, M., Gartner, K., Leitgeb, E., Nadezhdin, V., Cermák, J., Jimenez, M.S., Morales, D., 2010. Trees never rest: the multiple facets of hydraulic redistribution. Ecohydrol., 3, 431­444. Niehoff, D., Fritsch, U., Bronstert, A., 2002. Land-use impacts on storm-runoff generation: Scenarios of land-use change and simulation of hydrological response in a meso-scale catchment in SW-Germany. J. Hydrol., 267, 80­93. Nikodem, A., Kodesová, R., Bubenícková, L., 2013. Simulation of the influence of rainfall redistribution in spruce and beech forest on the leaching of Al and SO42­ from forest soils. J. Hydrol. Hydromech., 61, 39­49. Pardo, L.H., Kendall, C., PettRidge, J., Chang, C.C.Y., 2004. Evaluating the source of streamwater nitrate using d15N and d18O in nitrate in two watersheds in New Hampshire, USA. Hydrol. Process., 18, 2, 699­712. Pavl, V., Hejcman, M., Pavl, L., Gaisler, J., 2007. Restoration of grazing management and its effect on vegetation in an upland grassland. Appl. Veget. Sci., 10, 375­382. Pawellek, F., Frauenstein, F., Veizer, J., 2002. Hydrochemistry and isotope geochemistry of the upper Danube River. J. Hydrol., 66, 3839­3853. Penna, D., Stenni, B., Sanda, M., Wrede, S., Bogaard, T.A., Gobbi, A., Borga, M., Fischer, B.M.C., Bonazza, M., Chá- rová, Z., 2010. On the reproducibility and repeatability of laser absorption spectroscopy measurements for 2H and 18O isotopic analysis. Hydrol. Earth Syst. Sci., 14, 1551­1566. Pohle, I., Koch, H., Conradt, T., Gädeke, A., Grünewald, U., 2015. Potential impacts of climate change and regional anthropogenic activities in Central European mesoscale catchments. Hydr. Sci. J., 60, 912­928. Popescu, R., Costinel, D., Ionete, R.E., Axente, D., 2014. Isotopic fingerprint of the middle Olt River basin, Romania. Isotopes Eviron. Health Stud., 50, 461­474. Sanda, M., Vitvar, T., Kulasová, A., Jankovec, J., Císlerová, M., 2014. Runoff formation in a humid, temperate headwater catchment using a combined hydrological, hydrochemical and isotopic approach Jizera Mountains, Czech Republic. Hydrol. Process., 28, 3217­3229. Snhota, M., Sobotková, M., Císlerová, M., 2008. Impact of the entrapped air on water flow and solute transport in heterogeneous soil: experimental set-up. J. Hydrol. Hydromech., 56, 247­256. Soulsby, C., Tetzlaff, D., Hrachowitz, M., 2010. Spatial distribution of transit times in montane catchments: conceptualization tools for management. Hydrol. Process., 24, 3283­ 3288. Tetzlaff, D., Waldron, S., Brewer, D.J., Soulsby, C., 2007. Assessing nested hydrological and hydrochemical behaviour of a mesoscale catchment using continuous tracer data. J. Hydrol., 336, 430­443. Uhlenbrook, S., Roser, S., Tilch, N., 2004. Hydrological process representation at the meso-scale: the potential of a distributed, conceptual catchment model. J. Hydrol., 291, 278­296. Vitvar, T., Aggarwal, P.K., Herczeg, A.L., 2007. Global network is launched to monitor isotopes in rivers. EOS, Trans. AGU, 88, 325­326. Received 5 May 2016 Accepted 14 October 2016 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Hydrology and Hydromechanics de Gruyter

Pre-event water contributions and streamwater residence times in different land use settings of the transboundary mesoscale Lužická Nisa catchment

Loading next page...
 
/lp/de-gruyter/pre-event-water-contributions-and-streamwater-residence-times-in-6YfAbDZGQg
Publisher
de Gruyter
Copyright
Copyright © 2017 by the
ISSN
0042-790X
eISSN
0042-790X
DOI
10.1515/johh-2017-0003
Publisher site
See Article on Publisher Site

Abstract

The objective of the study was to evaluate the spatial distribution of peakflow pre-event water contributions and streamwater residence times with emphasis on land use patterns in 38 subcatchments within the 687 km 2 large mesoscale transboundary catchment Luzická Nisa. Mean residence times between 8 and 27 months and portions of preevent water between 10 and 97% on a storm event peakflow were determined, using 18O data in precipitation and streamwater from a weekly monitoring of nearly two years. Only a small tracer variation buffering effect of the lowland tributaries on the main stem was observed, indicating the dominant impact on the mountainous headwaters on the runoff generation. Longest mean streamwater residence times of 27 months were identified in the nearly natural headwaters of the Jizera Mountains, revealing no ambiguous correlation between the catchment area and altitude and the mean residence time of streamwater. Land use control on the pre-event water portions were determined for three land use categories with percentage of urban areas from 0 to 10%, 10 to 20% and more than 20%. The fraction of pre-event water in the first category decreases from 97% to 65% with the increasing percentage of forest from 76% to 100%, revealing that forests may provide only a limited infiltration of precipitation due to leaf interception and soil water use for transpiration. Fractions of pre-event water of 39­87% in the second (agricultural catchments) and of 10­35% in the third (urbanized catchments) category increase with percentage of non-urban areas. Keywords: 18O isotope; Stormflow event; Peak pre-event contributions; Mean residence time; Land use. INTRODUCTION Tracing of hydrological processes in nested mesoscale catchments at scales between 100 and 1000 km2 (Uhlenbrook et al., 2004) is often hampered by differences and inconsistency of data between small and large scales. Upscaling of abundant process knowledge from small headwater catchments has been more widely performed to mesoscale catchments around 100 km2 of area, for example in Switzerland (Köplin et al., 2014) and UK (Tetzlaff et al., 2007). The role of groundwater contributions to streams and streamwater residence times in catchment hydrology has been widely acknowledged over the past decades. Several studies have addressed the role of soil cover (Capell et al., 2012), topography and geometry (McGuire et al., 2005, Soulsby et al., 2010) of nested smaller subcatchments in water pathways and residence times, often adopting tracer approaches. Hydrological processes at the mesoscale are also often related to land use, which becomes an important parameter in runoff modelling (Isik et al., 2013; Montzka et al., 2008; Niehoff et al., 2002). However, recent reviews of the assessment of subsurface water contributions to streams and streamwater residence times (Klaus and McDonnell, 2013; McDonnell et al., 2010) revealed a principal need for a better understanding of the spatial distribution of their patterns across catchment scales. Linkages between pre-event water contributions, residence times and landscape settings often show opposite results, explained by specific catchment conditions (Klaus and McDonnell, 2013). Despite of abundant knowledge of mixing processes at headwater scales, little is known about the transfer of the typical approaches (two-component mixing, sine-wave estimation of residence times) from small experimental catchments to more complex scales. Although new tracer analytical techniques facilitate a better in-situ temporal monitoring at smaller scales (e.g. Holko et al., 2011), the spatial distribution of the runoff generation parameters and their interpretation remain a challenge. More work on the understanding of hydrological processes is particularly needed in catchments of size towards 1000 km2. Successful examples of a consistent isotopic assessment of nested streamwaters of mesoscale and large-scale catchments in the contiguous US (Dutton et al., 2005; Kendall and Coplen, 2001) reveal important hydrological patterns and links to climatic and geological settings that are often hidden at headwater or large scale. Tracing of hydrological processes in the mesoscale catchments between 500 and 1000 km2 has been often performed through surveys of isotopic or chemical tracers along the stream courses. It has been typically linked to hydrological (Popescu et al., 2008), geochemical or pollution patterns (Markovics et al., 2010; Pardo et al., 2004), whereas the largescale surveys have addressed geochemical, climatic or anthropogenic patterns (Pawellek et al., 2002), often resulting from channel deviations and water abstractions (Herczeg and Leaney, 2011). This paper summarizes the nearly 2-years long period of hydrometeorological and isotopic monitoring in the Luzická Nisa catchment and its 38 subcatchments with the objective to evaluate magnitude and distribution of pre-stormflow-event water portion and streamwater residence times linked to the altitude, catchment area and land use patterns. Although several studies have addressed the increasing occurrence and magnitude of floods in the Luzická Nisa catchment in pan-European (Alfieri et al., 2014), and regional (Bissolli et al., 2011) context, little is known on the hydrological processes that cause the floods and pollution fluxes. These topics have been addressed dominantly at point or small headwater catchment scale (Kändler and Seidler, 2009; Sanda et al. 2014). Hydrological evaluation of the catchment has been largely missing. A synthesis over the transboundary mesoscale catchment was hampered by data inconsistency. Except for the small headwater Uhlíská in the Jizera Mountains (Sanda et al., 2014), this is the first study to employ environmental isotopic data in the larger Luzická Nisa mesoscale catchment. Study site The mesoscale catchment of the Luzická Nisa (Lausitzer Neisse in German) river from the Jizera Mountains in the Czech Republic to the German lowlands in the vicinity of Zittau is characterized by many land use types and a large variety of hydrological and hydrochemical patterns. The differences are mostly related to the interaction of the forested-mountainous Jizera Mountains in the Luzická Nisa subcatchment, the agricultural-lowland Zittau basin in the subcatchment Mandava (Mandau in German) and the major cities of Liberec and Jablonec n.N. Names of the rivers in this paper match the language of the country of their spring. The Luzická Nisa catchment (Fig. 1) covers parts of Czech Republic, Germany and Poland and reaches from the headwaters region in the Jizera, Luzice and Zittau Mountains down to the gauge station on the Luzická Nisa in Zittau (German/Polish border). The tributaries above Zittau include small rivers originating in the Jizera Mountains (e.g. Bílá Nisa, Harcovský potok, Cerná Nisa, Jeice) and along the Jestd Ridge (Frantiskovský potok). The most prominent tributary, the Mandava River, originates in the north of Bohemian-Saxonian Swiss sandstone area, flows southeast-east across the region of Upper Lusatia north of Zittau Mountains, and joins the Luzická Nisa in Zittau. Important tributaries of the Mandava are Luznicka (Lausur in German) and Landwasser in Germany. The catchment (687 km2) divides into Czech (476 km²), German (205 km²) and Polish (6 km²) parts, with altitudes between 886 m a.s.l. in the Jizera Mountains (Olivetská hora summit) and 229 m a.s.l. at the gauge station Zittau. Water is transfered from outside of the Luzická Nisa catchment for the drinking water supply of the metropolitan area of Liberec and Jablonec n. N. and related settlements forming agglomeration of about 200 thousand inhabitants. Both water reservoirs of Josefv Dl on the Kamenice River and Sous on the Cerná Desná River are located in the Labe river regional catchment. They collect water from the headwaters of the highest locations of the Czech part of the Jizera Mountains reaching above 1000 m a.s.l. This results in additional volumes and hydrochemical signatures in the Luzická Nisa catchment (Farský, 1992). The Luzická Nisa catchment belongs to the temperate climate zone in the transient region between maritime and continental conditions, with precipitation maxima in July and August (Pohle et al., 2015). Heavy rains occur frequently during spring and summer. Long term annual precipitation amount varies between approximately 1400 mm·a­1 in higher altitudes of the mountains and 640 mm·a­1 in the lowlands. The Jizera Mountains are characterized by long lasting snow cover of up to 160 days/year. Mean annual temperature varies between 8°C in the lowlands and 5°C in the mountains (Sanda et al., 2014). The complex geological structure of the catchment is dominated by granites and granodiorites in the Jizera Mountains (Klomínský, 1969), sandstones in the Zittau Mountains and granite with basaltic and phonolitic hilltops in the western part. Fig. 1. Map of the study area (main stem of the Luzická Nisa is between CZ1 and D1). The lowlands are dominated by loess. The soils have different properties, particularly with respect to runoff formation, erosion, solute transport and water storage capacity. While soils on mountainous hillslopes are mostly shallow, often skeleton rich dystric cambisols, podzols (Nikodem et al., 2013) and leptosols with high infiltration rate and low storage capacities, the valleys are filled by organosols that release humic and fulvic acids. Agricultural soils are mostly luvisols and stagnic luvisols with high silt content and low infiltration rate. They are partly affected by poorly drained horizons. Luvisols are easily erodible and contribute to suspended load in the rivers. The lowland flood plains are typically covered by gleysols (Kändler and Seidler, 2009). Land use differs according to the soil types. While the mountainous areas are dominated by spruce forests and pastures (Pavl et al., 2007), the fertile soils in the lowlands are agriculturally exploited. Towns and villages in this region are located directly upon the streams. The rivers are linking towns and villages, river beds are often artificially changed and the banks are under revetment. Particularly the agriculturally used loess soils in the lower catchment parts are endangered by surface runoff and erosion, leading to high suspended particle loads and high nutrient concentrations in the rivers (Kändler and Seidler, 2009). DATA AND METHODS Monitoring The long-term hydrometeorological monitoring network in the Luzická Nisa catchment includes five precipitation (Table 1a) monitoring stations (four in the Czech Republic and one in Saxony) and seven runoff (Table 2) gauging stations (four in the Czech Republic and three in Saxony). Precipitation gauges Mlýnice, Chibská, are operated by the River Labe and River Ohe Authorities respectively. The Liberec station was a temporal station for the purposes of this project and station Bedichov is operated by the Czech Hydrometeorological Institute. The Czech runoff gauges are operated by the River Labe Authority (the Uhlíská gauge is operated by the Czech Technical University in Prague), and the German gauges are operated by the Saxonian Evironmental Authority. Weekly water samples for determination of 18O/2H content in rainwater and streamwater were collected in the period from September 2012 until April 2014 typically in weekly mode. They included five (Table 1b) precipitation stations (four in the Czech Republic and one in Saxony), and 38 streamwater (Table 3) profiles (24 in the Czech Republic and 14 in Saxony). Streamwater discrete samples were manually collected into 20 ml HDPE bottles and stored frozen until the analysis. The streamwater sampling campaign during each week was performed within one day along both the main Luzická Nisa river stem and its tributaries in the Czech Republic and the Mandava River and its tributaries and two stations on the Luzická Nisa River in Germany. The whole catchment was therefore sampled within a few hours. Precipitation was collected in a simple funnel-container device with oil protection of the collected water. In case of snow precipitation, samples were manually collected from the funnel into a closed plastic container and melted. The isotope monitoring in the Uhlíská catchment has been gradually established since 2006, delivering data for studies at catchment (Sanda et al., 2014), hillslope (Dohnal et al., 2006; Dusek et al., 2012) and point (Snhota et al., 2008) scales. Laboratory analyses The analyses of 18O and 2H were performed at the Czech Technical University in Prague, Faculty of Civil Engineering, using the Liquid Water Isotope Analyzer, LGR Inc. device (Penna et al., 2010). The values are expressed as 18O, 2H in of V-SMOW with typical precision of 18O ±0.15 and 2H ±0.7 V-SMOW (Vienna Standard Mean Ocean Water). Contents of 18O and 2H at CZ17, CZ24, D15 are also analyzed in monthly samples within the framework of the IAEA isotope hydrology databases GNIP and GNIR (Vitvar et al., 2007). Table 1. Precipitation amount (a-left) and precipitation sampling sites for the water isotopes (b-right). Precipitation XI/12-X/13 (mm) 1613 1119* 1222 1066 965 Precipitation 27-Aug-13 to 4-Sep-13 (mm) 104.0 37.3 51.1 31.6 14.0 27-Aug-13 to 4-Sep-13 (d18O ()) ­4.76 ­3.76 ­4.01 ­4.37 ­4.66 Amplitude X/12-IV/14 (d18O ()) 3.01 4.16 3.88 3.78 3.23 Provider CHMI project P. Labe P. Ohre ZÖF Zittau Location Bedichov Liberec Mlýnice Chibská Zittau Altitude (a.s.l.) 770 367 390 440 235 Profile CZ24 CZ25 CZ26 CZ50 D15 Location Uhlíská Liberec Oldichov Lucany Zittau Altitude (a.s.l.) 825 367 415 565 235 * correlated with Mlýnice on VII/13-IV/14 data Table 2. Outflow characteristics of the streams in the catchment. Runoff gauge - river / ID Uhlíská - Cerná Nisa / CZ17 Prosec - Luzická Nisa / CZ4 Mnísek - Jeice / CZ9 Varnsdorf - Mandava / above D4 Zittau - Mandava / D3 Hartau - Luzická Nisa / (CZ16-D2) Zittau - Luzická Nisa / D1 Altitude (a.s.l.) 792 401 375 314 239 241 235 Catchment area (km2) 1.2 53.7 32.2 88.9 293.9 377.5 694.0 Mean outflow XI/12-X/13 (m3/s) 0.05 1.42 0.35 1.53 3.68 7.12 11.85 Specific outflow XI/12-X/13 (m3/s/km2) 41.0 26.4 10.9 17.2 12.5 18.9 17.1 Table 3. List of streamwater sampling sites, land use characteristics and results of the analyses utilizing stable isotopes in streams in the Luzická Nisa catchment ,,wtp" abbreviates water treatment plant in Liberec. "Lake" denotes the catchment of the artificial lake Olbersdorfer See. "Urban" denotes small catchments entirely located in developed zones and drained through artificially designed channels or pipes. Landuse categories: Forest-more than 90% of non-urban areas, with a dominant forest component greater than the sum of urban land and arable land and grassland; Agriculture-more than 80% of non-urban areas, with a prevailing arable land and grassland component; Urbanbelow 80% of non-urban areas; Mix-mixed category of more than 80% of non-urban land with prevailing forest, but not dominant (greater than the sum of other components). Gauge altitude (a.s.l.) 235 237 239 302 340 325 353 275 291 247 243 240 383 387 565 529 472 401 381 354 339 349 351 333 333 333 333 333 375 375 394 387 290 286 265 245 792 407 Catchment average altitude (a.s.l.) 440 479 402 434 416 487 403 355 369 344 410 240 433 503 639 628 620 603 584 543 530 urban urban 528 urban urban 528 528 602 576 620 596 486 526 503 490 828 668 Catchment area (km2) 694.0 377.5 293.9 162.4 75.1 48.4 21.8 49.7 28.4 9.8 15.9 lake 18.3 16.8 4.2 7.4 48.1 56.4 63.5 120.7 137.3 urban urban 138.8 urban urban 138.8 138.8 7.1 32.8 15.2 4.2 77.9 291.0 320.7 349.5 1.2 1.0 Non urban land (%) 83 83 84 84 83 93 80 83 80 87 80 91 91 94 83 86 74 76 76 74 69 60 54 69 ­ ­ ­ ­ 89 94 98 99 93 76 82 83 100 99 Urban land (%) 17 17 16 16 17 7 20 17 20 13 20 9 9 6 17 14 26 24 24 26 31 40 46 31 ­ ­ ­ ­ 11 6 2 1 7 24 18 17 0 1 Forest (%) 39 49 28 33 24 49 15 9 9 6 15 45 25 52 34 45 47 50 48 46 43 56 50 43 ­ ­ ­ ­ 75 74 89 79 49 51 51 49 100 95 Arable and grassland (%) 42 32 53 48 57 41 62 72 68 80 63 39 65 39 48 39 25 24 26 26 25 3 3 25 ­ ­ ­ ­ 13 20 9 20 42 23 29 32 0 4 Preevent/total water ratio (­) 0.61 0.71 0.76 0.74 0.54 0.63 0.46 0.57 0.39 0.82 0.71 0.87 0.72 0.59 0.44 0.47 0.35 0.29 0.32 0.33 0.28 0.10 0.30 0.28 0.49 0.10 0.31 0.30 0.97 0.78 0.69 0.72 0.74 0.36 0.48 0.51 0.47 0.65 Mean residence time (months) 18.3 21.1 14.4 11.8 12.6 9.4 11.9 13.2 16.4 16.5 27.4 8.8 14.8 8.2 21.3 19.6 14.5 16.5 15.5 15.4 15.8 ­ ­ 17.1 ­ ­ ­ ­ 15.5 22.8 25.6 25.3 21.3 18.4 19.0 20.6 14.5 20.7 Profile D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D11 D12 D13 D14 CZ1 CZ2 CZ3 CZ4 CZ5 CZ6 CZ7 CZ7-1 CZ7-2 CZ8 CZ8-1 CZ8-1a CZ8-2 CZ8-3 CZ9 CZ10 CZ11 CZ12 CZ13 CZ14 CZ15 CZ16 CZ17 CZ44 River Luzická Nisa Luzická Nisa Mandava Mandava Mandava Luznicka Leutersdorfer Bach Landwasser Landwasser Grundbach Goldbach Olbersdorfer See Mandava Luznicka Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Luzická Nisa Frantiskovský potok Jizerský potok Luzická Nisa city of Liberec city of Liberec Luzická Nisa Luzická Nisa Fojtecký potok Jeice Malá Jeice Jeice Jeice Luzická Nisa Luzická Nisa Luzická Nisa Cerná Nisa Malá Jeice Location Zittau Zittau Zittau above Grossschönau Seifhennersdorf Neuschönau Seifhennersdorf Mittelherwigsdorf Oderwitz Zittau Zittau Zittau Rumburk Dolní Podluzí Lucany n. N. Jablonecké Paseky down Jablonec n. N. Prosec Vratislavice centre of Liberec above wtp Liberec Liberec 1 m below wtp discharge of wtp spillway of wtp 20 m below wtp 50 m below wtp Mnísek Mnísek with Fojtecký brook Oldichov v Hájích Oldichov v Hájích Chrastava Chrastava Chotyn Hrádek nad Nisou Uhlíská Betlém Landuse Agric. Mix Agric. Agric. Agric. Forest Agric. Agric. Agric. Agric. Agric. Agric. Agric. Forest Agric. Mix Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Urban Forest Forest Forest Forest Forest Urban Mix Mix Forest Forest Data interpretation Isotopic separation of event and pre-event water was based on the mixing equation the period August 27­ September 3, 2013 (see Table 1 for details). The cs values are equal to the 18O content in streamwater at each of 38 profiles on August 24, 2013 ­ the last sampling, considered as baseflow before the event. Water residence times Streamflow residence times were estimated using the seasonal sine-wave variations of 18O in monthly step in the streamflow and cumulative precipitation (Kralik, 2015). The seasonal variations in 18O are expressed (Eq. 2) Rs = Qs ct - cn = Qt cs - cn (1) where Qt ­ total (event & pre-event) flow (m3/s), Qs ­ pre-event flow fraction (m3/s) ( V-SMOW), ct ­ 18O content in the streamwater, cs ­ 18O content in the pre-event baseflow ( VSMOW), cn ­ 18O content in the rainwater during the event ( V-SMOW), and Rs ­ volumetric ratio of the pre-event water in the total outflow. The calculation was performed for the 38 studied gauges at the peakflow September 3, 2013, using the average and standard deviation of 18O content (cn) in precipitation of five sampled stations (Uhlíská, Liberec, Oldichov, Lucany, Zittau) during 18O mean 18O Asin 2t / b c (2) where mean(18O) ( V-SMOW) is the mean value of 18O, A ( V-SMOW) is the seasonal amplitude of 18O, b is the period of a seasonal cycle (one-year period is 2), t is time (months), and c (rad) is the phase shift. Parameters mean(18O), A and c were obtained by fitting Eq. (2) on the experimental data via least squares optimization. Mean residence time (MRT) is reflected in the decrease of the input amplitude in precipitation (Ap) relative to the output amplitude (e.g. in streamflow A) in the linear reservoir, according to Eq. (3). 2 1 Ap MRT 1 b A 0.5 (3) Luzické Mountains is observed in the Mandava tributary Luznicka (D6, D14) in more depleted 18O values compared to profiles in their vicinity (D8, D9). The lowland tributary Landwasser coming from northwest (D9, D8) has no impact of mountains. The Mandava tributaries Grundbach (D10) and Goldbach (D11) drain the area of the Lusatian fault where possible impact of groundwater circulation with particular isotopic composition and smaller isotopic variations may be hypothesized. Finally, D12 represents the outlet of the Olbersdorfer See (lake) with associated evaporative effects on the 18O content. Isotopic separation of pre-event water and calculation of water residence times Isotopic separation of event and pre-event water (Fig. 4a) was performed for the peakflow on September 3rd, 2013 with causal pre-event rainfall during August 27­September 4, 2013. This event has occured after three weeks with no precipitation and caused a discharge rise at all studied profiles. It is hypothesized (Fig. 4b) that the minor precipitation amounts on August 28 and September 1 contributed to the increase of the antecedent moisture, causing a major streamflow peak on September 3. This event was characterized by particularly enriched 18O values in the causal pre-event rainfall (Table 1b), reaching from ­3.76 V-SMOW at Liberec to ­4.76 V-SMOW at Uhlíská, with average ­4.13 V-SMOW and standard deviation 0.41 V-SMOW. 18O values in streamwater at the peakflow (September 3rd, 2013) reached up to ­5.8 V-SMOW from the background of ­10.0 to ­8.5 V-SMOW (Figs. 4cd). Fig. 5 and Table 3 display the calculated ratio of pre-event water (Eq. 1). The highest portions of the pre-event water were identified in two catchments CZ9 and D12, affected by the lakes Fojtka and Olbersdorfer, respectively. It is hypothesized that older water was released from the lakes during the stormflow event, causing a higher portion of the pre-event water. High portions of the pre-event water appeared also in the nearly natural Jeice (CZ10­CZ13, C44) subcatchment, covered dominantly by forests and meadows, and with little or no impact of settlements. The source of elevated pre-event water portions at D10 and D11 may include groundwater inflow along the Lusatian fault. More than 70% of the total runoff has pre-event origin in the Mandava catchment (D13, D11, D4 and D3). The main stem of the Luzická Nisa River (CZ1­D1) ranges from 69% to 79% of pre-event water, with lower values in the agglomerations (CZ4, CZ7). The Leutersdorfer brook (D7) and the Landwasser (D8, D9) were drained by 46%, 57% and 39% of pre-event water, respectively. Lowest values of Rs were identified at the profile CZ8-1a (10%) discharging from the water treatment plant. It can be hypothesized that the dominant rainfall amount prior to the event at Bedichov (Table 1a) resulted in significant portions of event water along the Luzická Nisa main stem, whereas the smaller causal rainfall amounts in the western lowland part of the catchment (Zittau) resulted in a less pronounced rapid runoff in the Mandava catchment. Five calculations of mean residence times using sine-wave amplitudes of five rainfall stations were carried out for each streamflow gauge (Tab. 1b). The average amplitude of 18O values in precipitation at the five sampling stations in the period October 2012­April 2014 was 3.61 V-SMOW with a standard deviation of 0.47 V-SMOW. The highest mean residence time of streamwater (25-27 months) were estimated in the nearly natural Jeice (CZ11, CZ12) and in the Goldbach (D11) catchments (Table 3). where (1/b) = 12/2 is the factor for the residence time in months. This fitting was performed for the 18O values at all five precipitation sampling stations Uhlíská, Liberec, Oldichov, Lucany, Zittau (cumulative weekly samples) and the stream (grab weekly samples) at the studied 38 profiles for the period October 2012­April 2014. The variability of Ap (expressed as average and standard deviation of each of the five precipitation stations) is therefore caused by the different amplitudes of 18O values at the respective five rainfall stations (see Table 1). The coefficient of determination R2 expresses the linear least square determination. RESULTS AND DISCUSSION Monitoring Fig. 2 shows selected hydroclimatical and isotopic characteristics of precipitation and streamwater for the period September 2012­April 2014. It displays daily temperature and precipitation amount at the wettest station Bedichov, daily discharge at headwaters of the Luzická Nisa (CZ17), catchment outlet (D1) and main tributary (D3), weekly 18O content in precipitation at the five sampling stations (see also Tab. 1b), and weekly 18O content in streamwater at seven selected profiles. No significant gradients in isotopic composition of precipitation were observed, indicating a homogeneous origin of precipitation over the study area. Very similar 18O patterns in streams are observed during winter and early spring. The 18O depleted streamwater of the mountain headwaters occurs only along the main Luzická Nisa stem and its tributaries in the Czech Republic and sustains during summer and autumn in particular. In contrast, streamwater in the the Mandava and Luznicka (D3 and D6) follow patterns of elevated 18O, due to non-existing impact of mountainous headwaters. These overall patterns were biased by the isotopically distinct rainfall-runoff event in September 2013. The 18O content in streamwaters at all 38 monitoring profiles is shown in Fig. 3 (top panel). The main stream of the Luzická Nisa (15 profiles from CZ1 to D1) shows a gradual increase of 18O values and decrease of their variations along with the progressing mixing towards the catchment outlet. The median 18O values along the main stem of Luzická Nisa reach from ­10 to ­9.5 V-SMOW, which corresponds to the median 18O values in precipitation (Fig. 2). The profiles CZ17 (Cerná Nisa) and CZ9-13, CZ44 (Jeice) represent mountainous headwaters from the Jizera Mountains with depleted 18O values. Streams at the profiles CZ7-1 and CZ7-2 (Frantiskovský and Jizerský brooks, resp.) drain entirely developed city areas, and the profiles CZ8-1 and CZ8-1a are outlet and spillway of the city water treatment plant. The 18O values at these four profiles are therefore highly affected by human impact. Profiles D13, D5, D4 and D3 represent the Mandava River with no major impact of mountainous headwaters. Impact of the precipitation (mm/day) 30 20 10 0 -10 -20 -30 -40 3 3 CZ17 Uhlíská - Cerná Nisa D3 Zittau - Mandava D1 Zittau - Luzická Nisa 3 3 -6 -9 -12 -15 -18 -21 -24 -5 CZ24 - Uhlíská CZ50 - Lucany CZ26 - Oldichov CZ25 - Liberec D15 - Zittau 3 D1 Luzická Nisa Zittau D3 Mandava Zittau -6 -7 D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn -8 -9 -10 -11 3 3 Fig. 2a, b, c, d (top to bottom). a) Air temperature and daily precipitation at the Bedichov station, b) streamflow and c) 18O content in precipitation and d) in streamwater at selected profiles, for the period September 2012 ­ April 2014. The latter may be associated with deeper groundwater contribution along the Lusatian fault. The main stream of the Luzická Nisa shows streamwater residence times between 15 and 20 months (with lowest values in the agglomeration CZ3-CZ7) and the Mandava (D13, D5, D4, D3), Grundbach (D10) and Landwasser (D8, D9) between 10 and 16 months. The relatively shorter streamwater residence times in the lowland Mandava catchment and its tributaries may be attributed to river terraces and alluvia. Therefore, the bulk of the river water probably comes from areas which are relatively close to the river, resulting 3 0.01 -3 daily average temp. Bedrichov (°C) -4 -5 -6 Jeice and Mandava are major L. Nisa tributaries with their stream networks urban denotes minor streams originating in Liberec (CZ7-1 -2) and outlet/overspill of the wtp - water treatment plant (CZ8-1, 1-a) water reservoir Olbersdorfer See Cerná Nisa 792 m a.s.l 565 m a.s.l. 235 m a.s.l. 18O ( SMOW) -7 -8 -9 -10 -11 -12 -13 forest Jeice forest urban 407-290 m a.s.l Mandava agricultural 383-247 m a.s.l Luzická Nisa stem stream agricultural urban Mandava tributaries forest and agricultural mix -4 -5 -6 -7 -8 -9 -10 -11 -12 -13 low baseflow 30.10.12 iso-light rain 16.4.13 baseflow 27.8.13 iso-heavy storm 3.9.13 CZ8 CZ1 CZ2 CZ3 CZ4 CZ5 CZ6 CZ7 CZ9 D13 D14 D11 CZ8-1a CZ10 CZ17 CZ14 CZ15 CZ16 CZ44 CZ11 CZ12 CZ8-2 CZ8-3 CZ7-1 CZ7-2 CZ8-1 CZ13 Fig. 3a, b (top to bottom). 18O in streamwaters along the principal water courses and their tributaries (number of collected samples in brackets). a) Box-Whisker-Plots for the period September 2012 - April 2014. b) Evolution of 18O along the river network in selected hydrological situations (light isotopic rain, pre-event baseflow, heavy isotopic storm, and typical baseflow). in shorter mean residence times (Dósa et al., 2011). The shortest residence times were identified in the Luznicka catchment (D9 and D14) with very shallow aquifer. Residence time was not calculated for the channelled streams CZ7-1and CZ7-2 and the overflow and spillway from the water treatment plant CZ81 and CZ8-1a. Similarly to several previous studies on mesoscale nested catchments (Soulsby et al., 2010; Tetzlaff et al., 2007), the mean streamwater residence times in the Luzická Nisa catchment reveal no significant correlation with the catchment altitude and area (Figs. 6a, b). In contrast, the pre-event water portion Rs increases with the catchment area (Fig. 6c). This supports the hypothesis that larger rivers have larger portion of pre-event water from river banks and alluvium. This portion, however, do not always imply longer streamwater residence time in the lowland subcatchments, because the groundwater discharge from lowland river banks is often younger than contributions from deeper fractures in the granitic headwaters. For example, Dósa et al. (2011) used 18O in the sine-wave method and reported a shorter mean residence time in streamwater of the Váh River below the Slovak Tatra Mountains (13 months) than in the headwaters Jalovecký potok Creek (19 months). Similarly, Soulsby et al. (2010) reported a shorter mean residence time of the Scottish streamwater Upper Dee (600 days) than in its headwaters (more than 2 years). In contrast, Capell et al. (2011, 2012) have found a largely muted response of stable isotopes in lowland streamwater in the 749km2 large mesoscale catchment North Eck in Scotland (very similar catchment size of 687 km2 to the Luzická Nisa). They have ascribed this result to the strong differences in geological and soil conditions between the lowland catchments and headwaters, which causes a strong dumping of the tracer input. According to Fig. 2d, however, this phenomemon is not observed in the Luzická Nisa catchment. All profiles along the main Luzická Nisa stem and the lowland Mandava tributary reveal similar annual variations of 18O of around 3 VSMOW (excluding the event from September 2013). During autumn baseflow periods, the entire course of the Luzická Nisa carries isotopically depleted waters, originating in the mountainous wetlands with dominantly snowmelt-induced recharge (Sanda et al., 2014). We argue that the pronounced isotopic variability of streamwaters throughout the entire Luzická Nisa catchment is caused by relatively shallow aquifers in the dominantly Neogene Zittau basin, which precludes development of deeper aquifers. D12 D10 D2 D1 D5 D4 D3 D7 D6 D9 D8 CZ17 (153) . CZ1 (83) CZ2 (83) CZ3 (82) CZ4 (82) CZ5 (82) CZ6 (83) CZ7 (82) CZ8 (83) CZ8-2 (67) CZ8-3 (67) CZ14 (83) CZ15 (82) CZ16 (82) D2 (85) D1 (85) . CZ7-1 (75) CZ7-2 (74) . CZ8-1 (67) CZ8-1A (58) . CZ44 (73) . CZ9 (82) CZ10 (82) CZ11 (83) CZ12 (82) CZ13 (82) . D13 (98) D5 (85) D4 (85) D3 (85) . D7 (83) . D14 (81) D6 (84) . D9 (88) D8 (61) . D10 (84) . D11 (87) . D12 (58) precipitation Bedrichov (mm/day) CZ17 Uhlíská - Cerná Nisa D3 Zittau - Mandava D1 Zittau - Luzická Nisa precipitation Bedichov (mm/hour) 8 CZ17 Uhlíská - Cerná Nisa 6 D3 Zittau - Mandava D1 Zittau - Luzická Nisa 4 22-Aug-13 29-Aug-13 10-Oct-13 5-Sep-13 3-Oct-13 19-Sep-13 12-Sep-13 26-Sep-13 28-Aug-13 4-Sep-13 2-Sep-13 3-Sep-13 5-Sep-13 6-Sep-13 7-Sep-13 8-Sep-13 27-Aug-13 29-Aug-13 30-Aug-13 3 9-Sep-13 -5.5 -6.0 -6.5 -7.0 D1 Luzická Nisa Zittau D3 Mandava Zittau D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn CZ17 - Cerná Nisa Uhlíská CZ17 - Cerná Nisa, automated -5.5 -6.0 -6.5 -7.0 D1 Luzická Nisa Zittau D3 Mandava Zittau D6 Luznicka Grossschönau CZ4 Luzická Nisa Prosec CZ8-3 Luzická Nisa Liberec CZ13 Velká Jeice Chrastava CZ15 Luzická Nisa Chotyn CZ17 - Cerná Nisa Uhlíská CZ17 - Cerná Nisa, automated -7.5 -8.0 -8.5 -9.0 -9.5 -10.0 -10.5 22-Aug-13 03-Oct-13 -7.5 -8.0 -8.5 -9.0 -9.5 -10.0 -10.5 19-Sep-13 10-Oct-13 27-Aug-13 28-Aug-13 29-Aug-13 30-Aug-13 29-Aug-13 3 05-Sep-13 12-Sep-13 26-Sep-13 0 02-Sep-13 03-Sep-13 04-Sep-13 05-Sep-13 06-Sep-13 07-Sep-13 08-Sep-13 Fig. 4a, b, c, d (top left to bottom right). Rainfall-runoff episode used for isotopic separation of the peakflow of September 3, 2013. a) Precipitation and discharge during the entire episode August 22­ October 10, b) Detailed record during the peakflow period August 27­ September 9, c) 18O in streamflow during the entire episode August 22­ October 10, and d) 18O in streamflow during the peakflow period August 27­September 9, showing the response of the catchments on the storm event with significantly different isotopic content . Bold lines for CZ17 headwater catchment Uhlíská indicate manual weekly and automated (daily or 4x daily sampling at the only station of the network), showing relatively good capture of the event by weekly sampling, including the peakflow. Cerná Nisa 792 m a.s.l Rs - Rs 565 m a.s.l. 235 m a.s.l. forest Jeice forest urban 407-290 m a.s.l Mandava agricultural 383-247 m a.s.l Mandava tributaries forest and agricultural -0.1 -0.2 Luzická Nisa stem stream agricultural urban mix Fig. 5. Volumetric ratio of pre-event water in the peakflow during the event on September 3 2013 based on 18O isotopic separation and mean residence time (MRT) of the streamwaters. The error bars show the variability of the results calculated using different 18O contents in the causal rainfall at the five sampling stations. MRT - mean residence time (month) water reservoir Olbersdorfer See MRT 09-Sep-13 all streams all streams Luzická Nisa main stem D11 Luzická Nisa main stem R² = 0.171 mean residence time (months) mean residence time (months) R² = 0.453 15 CZ17 10 R² = 0.256 average catchment altitude (m a.s.l.) catchment area (km2) catchment area (km2) Fig. 6a, b, c. Three examples of relationships between altitude, area, mean residence time and fraction of pre-event water (for Luzická Nisa ­ main stem, when applicable). The error bars for the pre-event water fraction show the variability of the results calculated using different 18 O contents in the causal rainfall at the five sampling stations. The error bars for the mean residence times show the variability of residence time calculations using different amplitudes of 18O contents in precipitation at the respective five rainfall stations. 1.00 0.95 0.90 0.85 0.80 0.75 R² = 0.531 0.70 0.65 0.60 fraction of non-urban area in the catchment (-) fraction of forest in the catchment (-) 0.90 R² = 0.739 0.85 0.92 0.42 0.40 0.90 fraction of the non-urban in the catchment (-) fraction of urban area in the catchment (-) 0.30 0.28 R² = 0.831 0.82 R² = 0.632 0.80 0.20 0.0 0.1 0.2 0.3 0.4 0.5 Fig. 7a, b, c, d. Relationship of fraction of pre-event water and landuse. All nested catchments (a-top left), small forested Czech catchments, category I (b-top right), agricultural catchments, mostly German, category II (c-bottom left), most urbanized Czech catchments Jablonec n.N, Liberec, Chrastava , category III (d-bottom right). The landuse categories I, II and III are defined in Table 3. The error bars for the pre-event water fraction show the variability of the results calculated using different 18O contents in the causal rainfall at the five sampling stations. Fig. 7 shows the relationship between landuse (Table 3) and the pre-event water fraction. Fig. 7a shows a weak overall increase of the pre-event water fraction with increasing portion of non-urban landuse in the respective catchment, where nonurban is understood as sum of forest, arable land, grassland, orchards and open water. Three distinct categories evolve upon distribution of catchments according to the degree of non-urban landuse (Table 3). The category "Forest" of 90­100% nonurban landuse contains all Czech mountainous headwaters CZ10-CZ13, CZ17 and CZ44 (Fig 7b), with a dominant portion of forest which is greater than the sum of urban land, arable land and grassland. The fraction of pre-event water in this group decreases with the percentage of forest, revealing that forests may provide only a limited infiltration of precipitation due to leaf interception and soil water use for transpiration (Nadezhdina et al., 2010). Not included in this assessment are the Luznicka stream profiles D6 and D14. Despite the low degree of urban landuse, these catchments have only a shallow aquifer that does not allow for storage of and mixing with preevent water. The category "Agriculture" of 80-90% of nonurban landuse dominantly contains the Mandava, Landwasser and Leutersdorfer Bach subcatchments, with prevailing arable and grassland. In this group, the fraction of pre-event water is directly related to the non-urban landuse degree (Fig. 7c). The category "Urban" contains most developed catchments below 80% of non-urban land. In this group the increasing fraction of pre-event water is accompanied by a decrease of the percentage of urban landuse. Other catchments ("Mix") that contain more than 80% of non-urban land, but no dominant forest component such as in category "Forest", are not included in Fig. 7. They are considered as mixed-landuse catchments. CONCLUSIONS The study presents a readily available approach of an assessment how individual subcatchments contribute to the hydrological response of a larger mesoscale catchment with highly heterogeneous land use and landscape characteristics. It improves the understanding of the role of topography, geology and land use with respect to origin, mixing and residence time of water in the subcatchments. The study also shows the variability of calculated residence times and pre-event water fractions, caused by the uncertainty and heterogeneity of the isotopic rainfall input to the catchment. Although the approach is based on the standard two-component mixing techniques, it provides a comparison of heterogeneous settings that cannot be assessed using techniques at hillslopes or headwater scales. Unlike several previous studies conducted typically in less developed catchments, the Luzická Nisa catchment includes a large variety of landscapes to explore phenomena such as the impact of urbanization on baseflow and its contributions to peak event flow or impact of the lowland alluvial riparian zone on the streamwater residence times. This assessment may be used in both gauged and ungauged mesoscale basins. Acknowledgement. The research was funded by the Czech Science Foundation, project No. 14-15201J and by EU programme Ziel3/Cíl3, No. 100114993 and student grant of CTU SGS16/143/OHK1/2T/11. Authors thank Labe River Authority and Ohe River Authority for hydrological data provision. REFERENCES Alfieri, L., Salamon, P., Bianchi, A., Neal, J., Bates, P., Feyen, L., 2014. Advances in pan-European flood hazard mapping. Hydrol. Proc., 28, 4067­4077. Bissolli, P., Friedrich, K., Rapp, J., Ziese, M., 2011. Flooding in eastern central Europe in May 2010 - reasons, evolution and climatological assessment. Weather, 66, 147­153. Capell, R., Tetzlaff, D., Malcolm, I.A., Hartley, A.J., Soulsby, C., 2011. Using hydrochemical tracers to conceptualise hydrological function in a large scale catchment draining contrasting geologic provinces. J. Hydrol., 408, 164­177. Capell, R., Tetzlaff, D., Soulsby, C., 2012. Can time domain and source area tracers reduce uncertainty in rainfall runoff models in larger heterogeneous catchments? Water Resour. Res., 48, W09544. Dohnal, M., Dusek, J., Vogel, T., 2006. The impact of the retention curve hysteresis on prediction of soil water dynamics. J. Hydrol. Hydromech., 54, 258­268. Dósa, M., Holko, L., Kostka, Z., 2011. Estimation of the mean transit times using isotopes and hydrograph recessions. Die Bodenkultur, 62, 47­52. Dusek, J., Vogel, T., Dohnal, M., Gerke, H.H., 2012. Combining dual-continuum approach with diffusion wave model to include a preferential flow component in hillslope scale modeling of shallow subsurface runoff. Adv. Water Res., 44, 113­125. Dutton, A., Wilkinson, B.H., Welker, J.M., Bowen, G.J., Lohmann, K.C., 2005. Spatial distribution and seasonal variation in 18O/16O of modern precipitation and river water across the conterminous United States. Hydrol. Process., 19, 4121­4146. Farský, I., 1992. A comment on the anthropogenic impacts at the upper flow of the river Luzická Nisa. Sborník ces. geogr. spol., 97, 26­32. Herczeg, A.L., Leaney, F.W., 2011. Review: environmental tracers in arid-zone hydrology. Hydrogeol. J., 19, 17­29. Holko, L., Kostka, Z., Sanda, M., 2011. Assessment of Frequency and Areal Extent of Overland Flow Generation in a Forested Mountain Catchment. Soil & Water Res., 6, 1, 43­53. Isik, S., Kalin, L., Schoonover, J.E., Srivastava, P., Lockaby, G.B., 2013. Modeling effects of changing land use/cover on daily streamflow: An artificial neural network and curve number based hybrid approach. J. Hydrol., 485, 103­112. Kändler, M., Seidler, C., 2009. Hydrochemical Load in a Small River Following Heavy Rain Events. Folia Geographica, Series Geographica-Physica, 40, 27­32. Kendall, C., Coplen, T.B., 2001. Distribution of oxygen-18 and deuterium in river waters across the United States. Hydrol. Process., 15, 1363­1393. Klaus, J., McDonnell, J.J., 2013. Hydrograph separation using stable isotopes: review and evaluation. J. Hydrol., 505, 47­64. Klomínský, J., 1969. The Krkonose-Jizera granitoid massif. Sb. geol. vd, 15, 1­134. Köplin, N., Viviroli, D., Schädler, B., Weingartner, R., 2014. How does climate change affect mesoscale catchments in Switzerland? - A framework for a comprehensive assessment. Adv. Geosci., 27, 111­119. Kralik, M., 2015. How to estimate mean residence times of groundwater. Proc. Earth Planet. Sci., 13, 301­306. Markovics, R., Kanduc, T., Szramek, K., Golobocanin, D., Milacic, R., Ogrinc, N., 2010. Chemical dynamics of the Sava riverine system. J. Eviron. Monit., 12, 2165­2176. McDonnell, J.J., McGuire, K., Aggarwal, P., Beven, K.J., Biondi, D.,Destouni, G., Dunn, S., James, A., Kirchner, J., Kraft, P., Lyon, S., Maloszewski, P., Newman, B., Pfister, L., Rinaldo, A., Rodhe, A., Sayama, T., Seibert, J., Solomon, K., Soulsby, C., Stewart, M., Tetzlaff, D., Tobin, C., Troch, P., Weiler, M., Western, A., Wörman, A., Wrede, S., 2010. How old is streamwater? Open questions in catchment transit time conceptualization, modelling and analysis. Hydrol. Process., 24, 1745­1754. McGuire, K.J., McDonnell, J.J., Weiler, M., Kendall, C., McGlynn, B.L., Welker, J.M., Seibert, J., 2005. The role of topography on catchment scale water residence time. Water Resour. Res., 41, 1­14. Montzka, C., Kanty, M., Kunkel, R., Menz, G., Vereecken, H.,Wendland, F., 2008. Modelling the water balance of a mesoscale catchment basin using remotely sensed land cover data. J. Hydrol., 353, 322­334. Nadezhdina, N., David, T.S., David, J.S., Ferreira, M.I., Dohnal, M., Tesa, M., Gartner, K., Leitgeb, E., Nadezhdin, V., Cermák, J., Jimenez, M.S., Morales, D., 2010. Trees never rest: the multiple facets of hydraulic redistribution. Ecohydrol., 3, 431­444. Niehoff, D., Fritsch, U., Bronstert, A., 2002. Land-use impacts on storm-runoff generation: Scenarios of land-use change and simulation of hydrological response in a meso-scale catchment in SW-Germany. J. Hydrol., 267, 80­93. Nikodem, A., Kodesová, R., Bubenícková, L., 2013. Simulation of the influence of rainfall redistribution in spruce and beech forest on the leaching of Al and SO42­ from forest soils. J. Hydrol. Hydromech., 61, 39­49. Pardo, L.H., Kendall, C., PettRidge, J., Chang, C.C.Y., 2004. Evaluating the source of streamwater nitrate using d15N and d18O in nitrate in two watersheds in New Hampshire, USA. Hydrol. Process., 18, 2, 699­712. Pavl, V., Hejcman, M., Pavl, L., Gaisler, J., 2007. Restoration of grazing management and its effect on vegetation in an upland grassland. Appl. Veget. Sci., 10, 375­382. Pawellek, F., Frauenstein, F., Veizer, J., 2002. Hydrochemistry and isotope geochemistry of the upper Danube River. J. Hydrol., 66, 3839­3853. Penna, D., Stenni, B., Sanda, M., Wrede, S., Bogaard, T.A., Gobbi, A., Borga, M., Fischer, B.M.C., Bonazza, M., Chá- rová, Z., 2010. On the reproducibility and repeatability of laser absorption spectroscopy measurements for 2H and 18O isotopic analysis. Hydrol. Earth Syst. Sci., 14, 1551­1566. Pohle, I., Koch, H., Conradt, T., Gädeke, A., Grünewald, U., 2015. Potential impacts of climate change and regional anthropogenic activities in Central European mesoscale catchments. Hydr. Sci. J., 60, 912­928. Popescu, R., Costinel, D., Ionete, R.E., Axente, D., 2014. Isotopic fingerprint of the middle Olt River basin, Romania. Isotopes Eviron. Health Stud., 50, 461­474. Sanda, M., Vitvar, T., Kulasová, A., Jankovec, J., Císlerová, M., 2014. Runoff formation in a humid, temperate headwater catchment using a combined hydrological, hydrochemical and isotopic approach Jizera Mountains, Czech Republic. Hydrol. Process., 28, 3217­3229. Snhota, M., Sobotková, M., Císlerová, M., 2008. Impact of the entrapped air on water flow and solute transport in heterogeneous soil: experimental set-up. J. Hydrol. Hydromech., 56, 247­256. Soulsby, C., Tetzlaff, D., Hrachowitz, M., 2010. Spatial distribution of transit times in montane catchments: conceptualization tools for management. Hydrol. Process., 24, 3283­ 3288. Tetzlaff, D., Waldron, S., Brewer, D.J., Soulsby, C., 2007. Assessing nested hydrological and hydrochemical behaviour of a mesoscale catchment using continuous tracer data. J. Hydrol., 336, 430­443. Uhlenbrook, S., Roser, S., Tilch, N., 2004. Hydrological process representation at the meso-scale: the potential of a distributed, conceptual catchment model. J. Hydrol., 291, 278­296. Vitvar, T., Aggarwal, P.K., Herczeg, A.L., 2007. Global network is launched to monitor isotopes in rivers. EOS, Trans. AGU, 88, 325­326. Received 5 May 2016 Accepted 14 October 2016

Journal

Journal of Hydrology and Hydromechanicsde Gruyter

Published: Jun 1, 2017

References