Hillslope hydrologic connectivity controls riparian groundwater turnover: Implications of catchment structure for riparian buffering and stream water sources

Hillslope hydrologic connectivity controls riparian groundwater turnover: Implications of catchment structure for riparian buffering and stream water sources
of 18
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
  Hillslope hydrologic connectivity controls riparian groundwaterturnover: Implications of catchment structure for riparianbuffering and stream water sources Kelsey G. Jencso, 1 Brian L. McGlynn, 1 Michael N. Gooseff, 2 Kenneth E. Bencala, 3 and Steven M. Wondzell 4 Received 27 October 2009; revised 7 March 2010; accepted 18 May 2010; published 15 October 2010. [ 1 ]  Hydrologic connectivity between catchment upland and near stream areas is essentialfor the transmission of water, solutes, and nutrients to streams. However, our current understanding of the role of riparian zones in mediating landscape hydrologic connectivityand the catchment scale export of water and solutes is limited. We tested therelationship between the duration of hillslope  riparian  stream (HRS) hydrologicconnectivity and the rate and degree of riparian shallow groundwater turnover along four HRS well transects within a set of nested mountain catchments (Tenderfoot Creek ExperimentalForest,MT).TransectHRSwatertableconnectivityrangedfrom9to123daysduring the annual snowmelt hydrograph. Hillslope water was always characterized bylow specific conductance ( ∼ 27  m S cm − 1 ). In transects with transient hillslope water tables, riparian groundwater specific conductance was elevated during base flowconditions ( ∼ 127  m S cm − 1 ) but shifted toward hillslope signatures once a HRSgroundwater connection was established. The degree of riparian groundwater turnover was proportional to the duration of HRS connectivity and inversely related to theriparian: hillslope area ratios (buffer ratio;  r  2 = 0.95). We applied this relationship to thestream network in seven subcatchments within the Tenderfoot Creek ExperimentalForest and compared their turnover distributions to source water contributions measuredat each catchment outlet. The amount of riparian groundwater exiting each of the sevencatchments was linearly related ( r  2 = 0.92) to their median riparian turnover time. Our observations suggest that the size and spatial arrangement of hillslope and riparian zonesalong a stream network and the timing and duration of groundwater connectivity between them is a first   order control on the magnitude and timing of water and solutesobserved at the catchment outlet. Citation:  Jencso, K. G., B. L. McGlynn, M. N. Gooseff, K. E. Bencala, and S. M. Wondzell (2010), Hillslope hydrologicconnectivity controls riparian groundwater turnover: Implications of catchment structure for riparian buffering and stream water sources,  Water Resour. Res. ,  46  , W10524, doi:10.1029/2009WR008818. 1. Introduction [ 2 ] Hydrologic investigations have been conducted acrossa wide array of research catchments and have identifiednumerouscontrolsonrunoffgeneration,includingtopography[  AndersonandBurt  ,1978;  Beven ,1978;  McGuireetal. ,2005],soil distributions [  Buttle et  al. , 2004;  Soulsby et al. , 2004; Soulsby et al. , 2006], and geology [ S haman et al. , 2004; U chida et al. , 2005]. Landscape struct ure (topography andtopology) can be particularly important for spatial patternsof water and solute movement in catchments with shallowsoils. However, the relationship between variability in catch-ment structure and the timing, magnitude, and distribution of runoffandsolutesourcesremainsunclear.Thislackofclarityis partially due to poor understanding of the role of riparianzones in mediating/buffering the upslope delivery of water and solutes across stream networks. We suggest that our understanding of catchment hydrology and biogeochemistrycan be advanced through assessment of the dominant con-trols on hydrological connectivity among hillslope  ripariansource areas and quantification of riparian buffering.[ 3 ] Hydrologic connectivity between hillslope and ripar-ian zones is typically transient but can occur when saturationdevelops across their interfaces [  Jencso et al. , 2009]. Hill-slope hydrologic connections to riparian zones may belargely controlled by topography in catchments with shallowsoil and poorly permeable bedrock. Especially important isthe convergence and divergence of catchment topographywhich controls the size of upslope accumulated area (UAA) 1 Department of Land Resources and Environmental Sciences, Montana State University, Bozeman, Montana, USA. 2 CivilEnvironmentalEngineeringDepartment, PennStateUniversity,University Park, Pennsylvania, USA. 3 U.S. Geological Survey,  Menlo Park, California, USA. 4 U.S. Forest Service,  Pacific Northwest Research Station Olympia Forestry Sciences Lab, Olympia, Washington, USA.Copyright 2010 by the American Geophysical Union.0043  1397/10/2009WR008818 WATER RESOURCES RESEARCH, VOL. 46, W10524, doi:10.1029/2009WR008818, 2010 W10524  1 of   18  [  Anderson and Burt  , 1977;  Beven , 1978]. Because of vari-ability in topography within ca tchments, hillslope UAA sizesand therefore transient groundwater inputs to riparian zonescan be spatially variable throughout the stream network [ W eyman , 1970].[ 4 ]  Jencso et al.  [2009] recently compared the duration of hillslope  riparian  stream (HRS) water table connectivity tohillslope UAA size. They found that the size of hillslopeUAA was a first   order control on the duration of HRSshallow groundwater connectivity across 24 HRS landscapetransitions ( r  2 = 0.91). Larger hillslope sizes exhibited sus-tained connections to their riparian and stream zones,whereas more transient connections occurred across HRSsequences with smaller hillslope sizes. They applied thisrelationship to the entire stream network to quantify catch-ment scale hydrologic connectivity through time and foundthat the amount of the stream network  ’ s riparian zones that were connected to the uplands varied from 4% to 67%during the year.[ 5 ] Because of their location between hillslopes andstreams, riparian zones can modulate or buffer the deliveryof water and solutes when hillslope connectivity is estab-lished across the stream network [  Hill  , 2000]. Research inheadwater catchments has emphasized the importance of theriparian zone as a relatively restricted part of the catchment which can exert a disproportionately large influence onstream hydrologic and chemical response [  M ulholland  ,1992;  Br inson , 1993;  Cirmo and McDonnell  , 1997]. Thedefinitions of riparian buffering are diverse and often depend on the water quality or hydrologic process of interest. Oneuse of the term refers to biogeochemical transformations[ Cirmo and McDonnell  ,1997] that often occurin near streamzones (e.g., redox reactions and denitrification). Another common use of the term refers to the volumetric bufferingof upslope runoff by resident near stream groundwater [  McGlynn and McDonnell  , 2003b]. In the context of thisstudy we focus on the volume buffering and source water mixing aspects of riparian function.[ 6 ] Identifying spatial and temporal hydrologic connec-tivity among HRS zones can be an important step inunderstanding the evolution of stream solute and sourcewater signatures during storm events. When a HRS con-nection is established, hillslope groundwater moves from theslope down through the adjacent riparian zone. Plot scaleinvestigations have suggested that the mixing and dis- placement of riparian groundwater (turnover) by hillsloperunoff is a first   order control on hillslope water [  McGlynnet al. , 1999], solute [  McGlynn and McDonnell  , 2003b],and nutrient [  Burt et al. , 1999;  Hill  , 2000;  Carlyle and  Hill  , 2001;  McGlynn and McDonnell  , 2003a ;  Ocampoet  al. , 2006;  P acific et al. , 2010] signatures expressed instr eamflow. Source water separations at the catchment outlet  [  H ooper et al. , 1997;  Burns et al. , 2001;  M cGlynn and  M cDonnell  , 2003b] and t heoretical exercises [ Chanat and  H ornberger  , 2003;  M cGlynn and Seibert  , 2003] have alsosuggested that the ra t e at which turnover occurs may be  proportional to the size of the riparian zone and the timing,duration, or magnitude of hillslope hydrologic connectivityto the riparian zone.[ 7 ] Information gleaned from individual plot or catchment scale tracer investigations have suggested hydrologic con-nectivity to the riparian zone as a factor in the timing of water and solute delivery to the stream. Despite these pre-vious investigations a general conceptualization of how a stream ’ s spatial sources of water change through an event remains elusive. Little field research to date has exploredhow HRS hydrologic connectivity frequency and durationrelates to the turnover of water and solutes in the riparianzone, how riparian zones  “  buffer  ”  hillslope connectivity,and how these dynamics are distributed across entire streamnetworks. This limits our ability to move forward and assessriparian buffering of hillslope groundwater connections in a whole catchment context.[ 8 ] In this paper we combine landscape analysis of HRSconnectivity [  Jencso et al. , 2009] and riparian buffering[  McGlynn and Seibert  , 2003] with high  frequency, spatiallydist ributed observations of HR S shallow groundwater con-nectivity (24 well transects; 146 wells) and solute dynamics(4 hillslope  riparian  stream transitions). We extrapolate theseobservations across seven stream networks with contrastingcatchment structure and compare them with catchment   scalehillslope and riparian spatial source water separations duringthe annual snowmelt hydrograph to address the followingquestions:[ 9 ] 1. What is the effect of HRS connectivity duration onthe degree of turnover of water and solutes in riparianzones?[ 10 ] 2. How does landscape structure influence streamnetwork hydrologic dynamics and the timing and amount of source waters detected at the catchment outlet?[ 11 ] We utilize a landscape analysis   based framework tolink landscape  scale hydrologic and solute dynamics withtheir topographic/geomorphic controls and present a way totransfer these dynamics across stream networks and catch-ments of differing structure. 2. Site Description [ 12 ] The Tenderfoot Creek Experimental Forest (TCEF)(latitude, 46°55 ′  N, longitude, 110°52 ′ W) is located in theLittle Belt Mountains of the Lewis and Clark NationalForest in Central Montana, USA (Figure 1). Tenderfoot Creek forms the headwaters of the Smith River, a tributaryof the Missouri. The TCEF is an ideal site for ascertainingrelationships between variability in landscape structure andcatchment hydrochemical response because it is composedof seven gauged catchments with a range of topographiccomplexity, watershed shapes, and hillslope and ripariansizes.[ 13 ] The seven TCEF subcatchments range in size from3 to 22.8 km 2 . In general,the catchment headwaters possessmoderately sloping (average slope  ∼ 8°) extensive (up to1200 m long) hillslopes and variable width riparian zones[  Jencso et al. , 2009]. Flathead Sandstone and Wolsey Shalecomprise the parent material in the upper portions of eachcatchment [  F arnes et al. , 1995]. Approaching the main stemof Tenderfoot Creek the strea ms become more incised,hillslopes become shorter (<500 m) and steeper (averageslope  ∼ 20°), and riparian areas are narrower than in thecatchment headwaters [  Jencso et al. , 2009]. Basement rocksof granite gneiss occur at lower elevations [  F arnes et al. ,1995], and they are visible as exposed clif fs and talus slopes. All three rock strata in the TCEF are relativelyimpermeable with potential for deeper groundwater trans-mission along geologic contacts and fractures within theWolsey shale [  Reynolds , 1995]. JENCSO ET AL.: GEOMORPHIC CONTROLS ON STREAM SOURCE WATER   W10524W10524 2 of 18  [ 14 ] Soil depths are relatively consistent across hillslope(0.5  –  1.0 m) and riparian (1  –  2.0 m) zones with localizedupland areas of deeper soils. The most extensive soil typesin the TCEF are loamy skeletal, mixed Typic Cryochreptslocated along hillslope positions and clayey, mixed AquicCryoboralfs in riparian zones and parks [  Holdorf   , 1981].Riparian soils have high organic matter.[ 15 ] The TCEF is a snowmelt dominated catchment. The1961  –  1990 average annual precipitation is 840 mm [  Farneset al. , 1995]. Monthly precipitation generally pea ks inDecember or Ja nuary (100  –  120 mm per month) and declinesto a late July through October dry period (45  –  55 mm per month). Approximately 75% of the annual precipitation fallsduring November through May, primarily as snow. Snow-melt and peak runoff typically occur in late May or earlyJune.Lowestrunoffoccursinthelatesummer throughwinter months. 3. Methods 3.1. Terrain Analysis [ 16 ] The TCEF stream network, riparian area, hillslopearea, and their buffer ratios were delineated using a 1 mAirborne Laser Swath Mapping digital elevation model(DEM) resampled to a 10 m grid cell size. Elevation mea-surements were achieved at a horizontal sampling interval of the order <1 m, with vertical accuracies of ±0.05 to ±0.15 m.We used the 10 m DEM to quantify each catchment  ’ shillslope and riparian UAA sizes following DEM landscapeanalysis methods outlined by  McGlynn and Seibert   [2003].[ 17 ] The area required for perennial stream flow (creek threshold initiation area) was estimated as 40 ha for Lower Tenderfoot Creek (LTC), Upper Tenderfoot Creek (UTC),Sun Creek (SUN), Spring Park Creek (SPC), Lower Stringer Creek (LSC), and Middle Stringer Creek (MSC) and 120 ha for Bubbling Creek (BUB). Creek threshold initiation areaswere based on field surveys of channel initiation points inTCEF [  Jencso et al. , 2009]. Accumulated area entering thestream network was calculated using a triangular multipleflow  direction algorithm [ Seibert and McGlynn , 2007].Once the accumulated area exceeded the creek thresholdvalue, it was routed downslope as stream area using a singledirection algorithm. To avoid instances where parallelstreams were computed, we used the iterative proceduresuggested by  McGlynn and Seibert   [2003]. Any stream pixelwhere we derived more than one adjacent stream pixel in a downslope direction was in the next iteration forced to drainto the downslope stream pixel with the largest accumulatedarea. We repeated this procedure until a stream network without parallel streams was obtained.[ 18 ] The TCEF riparian areas were mapped based on thefield relationship described in the study by  Jencso et al. [2009]. Landscape analysis  derived riparian area was delin-eated as all areas less than 2 m in elevation above the stream Figure 1.  Site location and instrumentation of the TCEF catchment. (a) Catchment location in the RockyMountains, MT. (b) Catchment flumes, well transects, and SNOTEL instrumentation locations. Specifictransects highlighted in this study are filled in black and labeled T1  –  T4. Transect extents are not drawn toscale. JENCSO ET AL.: GEOMORPHIC CONTROLS ON STREAM SOURCE WATER   W10524W10524 3 of 18  network pixel into which they flow. To compare the land-scape analysis  derived riparian widths to actual riparianwidths at TCEF,  Jencso et al.  [2009] surveyed 90 ripariancross sections in Stringer Creek, Spring Park Creek, andTenderfoot Creek. A regression relationship ( r  2 = 0.97)corroborated their terrain   based riparian extent mapping[  Jencso et al. , 2009].[ 19 ] The local area entering the stream network is theincremental increase in catchment area for each stream pixel(not counting upstream contributions) and is a combinationof hillslope and riparian area on either side of the streamnetwork. We separated local hillslope UAA and riparianarea into contributions from each side of the stream fol-lowing methods developed by  Grabs et al.  [2010]. TheUAA measurements for each transect  ’ s hillslope were cal-culated at the toe  slope well position. The riparian buffer ratio was computed as the ratio of local riparian area divided by the local inflows of hillslope area associated with eachstream pixel (separately for each side of the stream). The “  buffer ratio ”  represents the capacity of each riparian zoneto modulate its adjacent hillslope water inputs. Riparian buffer ratio values were measured at the riparian well position.[ 20 ]  Jencso et al.  [2009] determined the HRS connectivityfor the catchments stream network based on a relationship between UAA size and HRS connectivity duration across24 transects of HRS groundwater recording wells: %Time Connected  ¼  0 : 00002*UAA    0 : 0216 ð Þ *100 :  ð 1 Þ They found that the duration of a shallow groundwater table connection from hillslopes to the riparian and streamzones was linearly related ( r  2 = 0.92) to UAA size. For the purposes of this study, we refer to UAA size as a surrogatefor the duration of groundwater table connectivity betweenHRS zones, based on the relationship observed by  Jencsoet al.  [2009]. Larger UAA sizes indicate longer periods of connectivity duration while smaller UAA sizes are indic-ative of transient connections that only occur during thelargest snowmelt events. We applied this relationship tothe hillslope UAA values along each stream network in theseven TCEF subcatchments to determine the connectivityto riparian zones through time. 3.2. Hydrometric Monitoring [ 21 ]  Jencso et al.  [2009] instrumented 24 sites in TCEFwith transects of shallow recording groundwater wells and piezometers (146 total). At a minimum, groundwater wellswere installed across each transect  ’ s hillslope (1  –  5 m abovethe break in slope), toe slope (the break in slope betweenriparian and hillslope positions), and riparian position (1  –  2 mfrom the stream). All wells were completed to bedrock, andthey were screened from 10 cm below the ground surface totheir completion depths. Groundwater levels in each wellwere recorded with Tru Track Inc. capacitance rods (±1 mmresolution) at hourly intervals for the 2007 water year.Hydrologic connectivity between HRS zones was inferredfrom the presence of saturation measured in well transectsspanning the hillslope, toe slope, and riparian positions. Fol-lowing  Jencso et al.  [2009], we define a hillslope  riparian  stream connection as a time interval during which streamflow occurred, and the riparian, toe slope, and adjacent hillslope wells recorded water levels above bedrock.[ 22 ] Runoff was recorded in each of the seven nestedcatchments using a combination of Parshall and H  Flumesinstalled by the USFS (Figure 1). Stage in each flume wasmeasured at hourly intervals with Tru Track Inc. water level recorders and every 15 min by USFS float potenti-ometers. Manual measurements of both the well groundwater levels (electric tape) and flume stage (visual stage readings)were conducted biweekly during the summer months andmonthly during the winter to corroborate capacitance rodmeasurements. 3.3. Chemical Monitoring [ 23 ] We collected snowmelt, shallow groundwater, andstream samples once a month during the winter, every1  –  3 days during snowmelt according to runoff magnitude,and biweekly during the subsequent recession period of thehydrograph. In this paper we highlight the hydrochemicalresponse of four well transects sampled from the 24 transectswhere physical hydrology was measured. These transectswere selected to cover a range of hillslope and riparian area size and the ratio of their areas (riparian buffer ratios). High  frequency solute and SC monitoring was limited to four transects due to the time constraints associated with foot travel across the TCEF catchment during isothermal condi-tions in a 2 m snowpack. The four transects in this study arenamed in order of increasing riparian buffer ratios sequen-tially from one through four (T1  –  4). Wells were purged toensure a composite sample along the screened interval beforesample collection. Samples for solute analysis were collectedin 250 mL high  density polyethylene bottles and filteredthrough a 0.45  m m polytetrafluorethylene membrane filter.They were stored at 4°C before analyses of major cationswith a Metrohm  Peak (Herisau, Switzerland) compact ionchromatograph at Montana State University. Sodium (Na),ammonium (NH 4 ), potassium (K), calcium (Ca), and mag-nesium (Mg) were measured on a Metrosep C  2  250 cationcolumn. Detection limits for major cations were 5  –  10 m g L − 1 and accuracy was within 5% of standards. Groundwater specific conductance (SC) was measured with a handheldYSI EC300 meter (±0.1  m S cm − 1 resolution and accuracywithin ±1% of reading). We also monitored groundwater chemistry and SC in each of the 24 transects installed by  Jencso et al.  [2009] at a bimonthly interval. This corrobo-rated the range of SC dynamics observed at the four transectsused in this study and helped to determine base flow SCacross the range of riparian zone sizes in TCEF. Streamspecificconductanceandtemperatureineachsubcatchment  ’ sflume was also measured at hourly intervals with CampbellScientific CS547A conductivity probes (±0.1  m S cm − 1 reso-lution and accuracy within ±1% of reading). 3.4. Specific Conductance as a Tracer of Water Sources [ 24 ] Hillslope shallow groundwater specific conductancewas  ∼ 80% less than the SC observed in riparian wells during base flow periods of the hydrograph. We used specificconductance to distinguish between hillslope and riparianshallow groundwater and riparian saturation overland flow.PreviousstudieshaveusedSCtodistinguishthespatialsourcesof water within catchments [  K obayashi , 1986;  M cDonnell et  al. , 1991;  H asnain and Thayyen , 1994;  Cais sie et al. ,1996;  Laudon and Slaymaker  , 1997;  K obayashi et al. ,1999;  Ahmad and Hasnain , 2002;  Covino and McGlynn , JENCSO ET AL.: GEOMORPHIC CONTROLS ON STREAM SOURCE WATER   W10524W10524 4 of 18  2007;  St ewart et al. , 2007], but validation of SC with itsconstituent solutes is recommended [  Laudon and Slaymaker  ,1997;  C ovino and McGlynn , 2007]. We compared SC mea-surements with a composite ( n  = 126) of major cation con-centrations in hillslope, riparian, stream, and snowmelt grabsamples determined through IC analysis. A strong linear relationship existed between SC and Ca ( r  2 = 0.92) and SCand Mg ( r  2 = 0.89) for each spatial source supporting the useof SC as a surrogate tracer for calcium and magnesiumconcentrations in solution. Hydrochemical tracers, such asCa  + , Mg 2 are commonly used in comparable studies andwhen related to Specific conductance, recording SC probes provide high  resolution measurements for source water separations. We restrict the use of SC and solutes as tracersto the snowmelt portion of the hydrograph (1 May 2007 to1 July 2007) to minimize the potential impacts of weatheringand nonconservative behavior. 3.5. Modeling Riparian Groundwater Turnover [ 25 ] We applied a simple continuously stirred tank reactor (CSTR) [  Ramaswami et al. , 2005]) mixing model to eachriparian SC time series to quantify the turnover rate of riparian groundwater in response to hillslope water tabledevelopment and HRS connectivity. This basic exponentialmodel has been previously used to estimate [  Boyer et al. ,1997] a nd model [ S canlon et al. , 2001] flushing time con-st ants of dissolved organic carbon and silica from riparian areas and whole catchments.[ 26 ] We fit an exponential decay regression relationshipto the riparian well water SC time series at each transect.The time period analyzed for each riparian SC time serieswas the highest observed SC before snowmelt initiation andHRS connectivity until the time of lowest SC observations.Similar to  Boyer et al.  [1997], we selected sequential data  points over this period to determine the linear fit to therelationship between ln(SC) and time. The slopes of theseregressions are the turnover rate constants ( l ) or how fast the solutes that comprise SC in the riparian reservoir areturned over or mixed with more dilute hillslope inputs. Theinverse of this slope represents the  “ turnover constant  ”  of each site ( t  ), the time in days it took for the SC in theriparian zone to decrease to 37% of its initial value (Table 1)and for one volume to be flushed [  Ramaswami et al. , 2005]:    ¼  1  :  ð 2 Þ We believe a more intuitive way of describing exponentialdecay is the time required for the decaying mixture todecline to 50% of its initial concentration. This is commonlycalled the half   life and in the context of this paper is referredto as the turnover half   life ( t  50 ): t  50  ¼  ln0 : 5   ¼     ln ð 0 : 5 Þ ð 3 Þ Similarly, we calculated the time it would take to fully turnover all of the srcinal riparian SC in each transect ( t  95 ).While an exponential model can never fully reach a baselineconcentration, we chose 95% as an acceptable limit at whichthe riparian zone water SC is deemed similar to water coming from the adjacent hillslope. Thus, 5% of the srcinalriparian SC was considered the baseline at which all initialriparian water was considered turned over from a riparianzone: t  95  ¼     ln ð 0 : 05 Þ :  ð 4 Þ Wealsoestimatedhowmanyriparianvolumesmovedthroughthe riparian zone at each transect during its correspondingtime of HRS connectivity. Riparian volume turnover wascalculated by dividing the HRS water table connectionduration by the calculated turnover constant: Riparian volumes  ¼  HRS water table connection duration    :  ð 5 Þ Here we incorporate the duration of the HRS connection; themagnitude of hillslope throughflow associated with eachHRS connection is incorporated within the exponential rela-tionshipdevelopedfromthedecayrateoftheriparianSCtimeseries. 3.6. Hydrograph Separations for Hillslope, Riparian,and Saturated Area Overland Flow [ 27 ] Hydrograph separations are commonly used tools for separating the spatial and temporal sources of water exiting a catchment. They can provide an integrated measure of sourcearea contributions and their overall effect on hydrologicdynamics observed at the catchment outlet. We implemented3 component hydrograph separations to determine the spatialcontributions to stream runoff from hillslope, riparian, andsaturated overland flow sources during the annual snowmelt hydrograph (1 May 2007 to 1 July 07).  “ Real  time ”  separa-tions were developed for each subcatchment in TCEF usingcontinuousmeasurementsofriparian  saturatedoverlandflow[  Dewalle et al. , 1988] and specific conductance.[ 28 ] Saturation overland flow is limited to the near streamriparian areas in TCEF due to upland soils with high infil-tration rates. We determined the runoff contributions fromriparian overland flow using continuous measurements of  Table 1.  Transect Attributes Transect Riparian SoilDepths (m)RiparianUAA(m 2 )HillslopeUAA(m2)HRSConnection(days)RiparianBuffer RatioTurnover TimeConstant (days) t  50%Turnover (days) t  95%Turnover (days)Riparian VolumesTurned Over T1 0.7  –  1.80 783 46112 123 0.017 4 3 13 27T2 0.7  –  1.20 163 7070 46 0.023 8 6 25 6T3 0.6  –  1.10 1148 10165 29 0.113 29 20 86 1.0T4 0.7  –  0.85 700 1527 9 0.458 39 27 115 0.2 JENCSO ET AL.: GEOMORPHIC CONTROLS ON STREAM SOURCE WATER   W10524W10524 5 of 18

Model Space

Apr 28, 2018
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks