Eu

Ja

b

c

a

ARRA

KRFW2R

1

tbrPirftrih

s

0d

Ecological Engineering 36 (2010) 1251–1259

Contents lists available at ScienceDirect

Ecological Engineering

journa l homepage: www.e lsev ier .com/ locate /eco leng

valuation of physical fish habitat quality enhancement designs inrban streams using a 2D hydrodynamic model

oo Heon Leea,∗, Jun Teak Kilb,1, Sangman Jeongc,2

Department of Civil Engineering, JoongBu University, Kumsan, Chungnam-Do 312-702, Republic of KoreaDivision of Water Resources, Korea Engineering Consultants Corps, Gueui-dong, Gwangjin-gu, Seoul 462-805, Republic of KoreaDepartment of Civil & Environmental Engineering, Kongju National University, Kongju, Chungnam-Do 330-717, Republic of Korea

r t i c l e i n f o

rticle history:eceived 5 September 2009eceived in revised form 17 March 2010ccepted 31 May 2010

eywords:iver restorationish habitatUA

D hydrodynamic simulationiver2D

a b s t r a c t

Creating a habitat for a variety of forms of life, such as riparian plants and various fish, is a necessityfor stressed river ecosystems. In this study, the hydraulic characteristics of a fish habitat in an urbanchannel were analyzed using River2D, which is a two-dimensional (2D) depth-averaged finite elementhydrodynamic model, to improve the habitat of two target fish in the Daejeon Stream, Korea. These speciesare Pseudopungtungia nigra, which is an endangered species in the Daejeon Stream, and Zacco platypus,which is a dominant species. In addition, changes in the weighted usable area (WUA) were compared andreviewed as boulders were placed in the stream. The best method for improving the P. nigra’s habitat isproposed. A simulation analysis was performed on urban rivers for fish habitats. As a result, a straightand monotonous urban river flow was found to be an appropriate habitat environment for Z. platypus.The WUA for Z. platypus was about 20 times greater than that for P. nigra. Three different fish habitatenhancement methods were evaluated by calculating the WUA for the target fish in the study channel. By

calculating the WUA to create fish habitats, the V-type riffle method was found to increase the usable areaof the habitat environment for P. nigra by 360%, and the step stone method and single boulder methoddid so by 60% and 8%, respectively. For the single boulder method, boulders were placed in the channelbed at 3.5-m intervals, which significantly increased habitat availability. Moreover, centralizing the flowpattern in the channel among several types of boulder placements greatly expanded the habitat for P.

e plach spe

natdteac

u

nigra. Thus, an appropriatthe riverbed and target fis

. Introduction

Rivers have an important role in sustaining riparian ecosys-ems and human communities. However, globally, rivers haveeen severely impacted by a variety of human activities that haveesulted in the loss of many of their original ecosystem’s functions.reserving an ecological habitat is a key requirement for ripar-an species and fish species, both of which are greatly affected byivers (Lee et al., 2006). Recreating and restoring the river habitator riparian assets and endangered fish species is now an impor-

ant aspect of ecohydrological research. A significant amount ofesearch has been conducted on the habitat requirements of ripar-an and instream plants and animals; most studies are related toabitat modeling and generally focus on stable and natural chan-

∗ Corresponding author. Tel.: +82 41 750 6744; fax: +82 41 750 6391.E-mail addresses: leejh@joongbu.ac.kr (J.H. Lee), june79ms@paran.com (J.T. Kil),

mjeong@kongju.ac.kr (S. Jeong).1 Tel.: +82 2 2049 5252; fax: +82 2 2049 5111.2 Tel.: +82 41 850 8628; fax: +82 41 856 7818.

p2shtp(aSa

925-8574/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.ecoleng.2010.05.004

ement interval and boulder location that considers the characteristics ofcies should be researched and implemented.

© 2010 Elsevier B.V. All rights reserved.

els (Hauer et al., 2007; Hauer et al., 2008). For example, Crowdernd Diplas (2000) provided results that demonstrate the extento which a boulder, or a series of boulders, can influence the pre-icted flow conditions and fish habitats in a natural stream using awo-dimensional (2D) hydraulic model. García and Gortazar (2006)valuated the effectiveness of habitat enhancements, such as rifflesnd pools, with a 2D hydraulic model under different instream flowonditions in natural streams located in Spain.

Many researchers have demonstrated that improving the nat-ral habitat of fish requires knowledge of the natural recoveryrocess of river ecosystems (Cairns et al., 1977; Gore, 1985; Konrad,009; Nagaya et al., 2008; Shih et al., 2008). In particular, under-tanding the interaction between the physical habitat and theydraulic habitat is essential to rehabilitate impaired river ecosys-ems. One approach to stream restoration that focuses on thehysical fish habitat is the Instream Flow Incremental Methodology

IFIM), which is an example of how ecological characteristics can bepplied to ecosystem restoration (Bovee, 1982; National Biologicalervice, 1995; Spence and Hickley, 2000; Lopes et al., 2004). In thispproach, the Weighted Usable Area (WUA) for a fish habitat in

1252 J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259

F (a) Phor

ts

smaiitaochwmmmdv

2

2

K

tcoaDsofiids

sbuao

2

TN

ig. 1. Photo of the study reach and topographic reach map of the Daejeon Stream.each looking downstream. (c) Topographic reach map.

he stream is applied to assess the influences of different instreamtructures to provide a hydraulic habitat.

In urban streams and rivers, the amount of fish habitats iseverely limited because of the use of traditional flood controlethods for river management. Urban rivers have been man-

ged and exploited with an emphasis on both flood control andnstream water use, leading to tremendous damage to the ripar-an ecosystem and river environment. In this study, we assessedhe effectiveness of different obstacles, such as boulders, to createfish habitat using the River2D model. Several traditional meth-

ds that have been used to artificially create fish habitats wereompared and analyzed. As an alternative to eco-friendly methods,ydraulic and ecologic characteristics were analyzed after bouldersere placed on the river bed, and the WUA was estimated usingicro-habitats (depth, velocity, riverbed materials) to protect andaintain fish habitats. Instead of using traditional standardizedethods to maintain rivers, an alternative, forward-looking and

etailed method is proposed to create an environment that pro-ides an eco-friendly habitat for fish.

. Methods

.1. Study site

The study reach is situated in Daejeon Metropolitan City inorea. The basin area of the Daejeon Stream is 89.38 km2, and the

2

hnt

able 1atural flow regime of the Daejeon Stream (Daejeon Metropolitan City, 2008).

Basin area (km2) Averaged-wet flow (Q95) (m3/s) Normal flow (Q18

89.38 1.73 1.15

to of straight trapezoidal reach looking upstream. (b) Photo of straight trapezoidal

otal length of the stream is 23.08 km. The Daejeon Stream is a typi-al urban stream that flows through the heart of Daejeon City. Mostf the reaches of the Daejeon Stream that pass through the urbanrea have unchanging, straight flow patterns. Urbanization of theaejeon Stream’s basin has not only changed the morphology of the

tream but also has had a significant impact on the ichthyofaunaf the river’s ecosystem. Pseudopungtungia nigra, which is a nativesh, became endangered due to the lack of a proper habitat and

nstream flow (Hur and Kim, 2009). As a result of habitat degra-ation and the decline of P. nigra, Zacco platypus is the dominantpecies in the basin (CGRBM, 2005).

In this study, typical straight sections of urban streams that con-ist of very monotonous flow patterns were selected and used asasic modeling to analyze the current issues for fish habitats inrban rivers. The selected study reach is 100-m long, while theverage width of the channel is 17 m; it is located mid-downstreamf the Daejeon Stream (Fig. 1).

.2. Model employed in this study

.2.1. River2DA two-dimensional numerical model was selected for the

ydraulic simulation with low flow in the study reach. The 2Dumerical model was found to be very effective in comparison tohe 1D model, especially for spatially distributed phenomena, such

5) (m3/s) Low flow (Q275) (m3/s) Drought flow (Q355) (m3/s)

0.75 0.45

J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259 1253

urve o

ad

mtrot(

C

C ) + g

2

C ) + g

2

q

wvtssh

2

ATsh1toTpiacta

W

wccc

2

2

sSuflpddomeCs

Fig. 2. Flow duration c

s patterns of sediment erosion and fish habitat quality under lowischarge (Clark et al., 2006; Brown and Pasternack, 2008).

In this study, the hydrodynamic component of the River2Dodel uses the two-dimensional, depth-averaged St. Venant Equa-

ions, which are expressed conservatively. These three equationsepresent the conservation of water mass and the two componentsf the momentum vector. The depth and discharge intensities in thewo respective coordinate directions are the dependent variablesSteffler and Blackburn, 2002).

onservation of mass :∂H

∂t+ ∂qx

∂x+ ∂qy

∂u= 0 (1)

onservation of x-direction momentum :∂qx

∂t+ ∂

∂x(Uqx) + ∂

∂y(Vqx

onservation of y-direction momentum :∂qx

∂t+ ∂

∂x(Uqy) + ∂

∂y(Vqy

x = HU, qy = HV (4)

here H is the average depth of the flow; U and V are the averagedelocities of the x and y coordinates, respectively; g is the accelera-ion due to gravity; � is the density of water; S0x and S0y are the bedlopes of the x- and y-axes, respectively; Sfx and Sfy are the frictionlopes; and �xx, �xy, �yx and �yy represent the components of theorizontal turbulent stress tensor.

.2.2. WUA (Weighted Usable Area)The main goal of this study is to apply the Weighted Usable

rea (WUA) to evaluate available fish habitats in an urban stream.he WUA is the amount of physical habitat that is available for fishpecies at a given flow (Golder Kingett Mitchell, 2007). The fishabitat component of River2D is also based on the WUA (Bovee,982), which is a concept used in the PHABSIM family of fish habi-at models. The WUA is calculated as an aggregate of the productf a composite suitability index (CSI, which has a range of 0.0–1.0).he habitat suitability index that is defined in River2D includes thereferences of the target species with regards to the flow veloc-

ty, depth, and channel properties. The criteria for the flow velocitynd the depth reflect the assumption that lotic biota distributions,ertain distribution phases, and certain life cycle phases are con-rolled by the hydraulic conditions within the water column (Gorend Hamilton, 1996).

leiflfl

f the Daejeon Stream.

∂

∂xH2 = gH(S0x − Sfx) + 1

�

(∂

∂x(H�xx)

)+ 1

�

(∂

∂y(H�xy)

)(2)

∂

∂yH2 = gH(S0x − Sfx) + 1

�

(∂

∂x(H�yx)

)+ 1

�

(∂

∂y(H�yy)

)(3)

The WUA in a stream reach was calculated by Eq. (5).

UA =n∑

i=1

[f (Vi, Di, Ci) · Ai] (5)

here Ai is the stream area of the ith cell, and f(Vi, Di, Ci) is theomposite suitability index for Ai, which is usually a product of theorresponding suitability weights for the flow velocity, depth andhannel properties (Milhous et al., 1989).

.3. Monitoring activities

.3.1. Flow regime of the study reachTo analyze the flow regime of the Daejeon Stream, historical

tream flow monitoring results from the Master Plan of Daejeontream Management (2008) were used. The Daejeon Stream’s nat-ral flow regime that is presented in Table 1 was estimated from theow duration curve of the Daejeon Stream basin (Fig. 2). For exam-le, the drought flow (Q355) is an average value of the recordedischarge over a ten-year period that exceeded 97% of the totaluration (i.e., 355 days of the year). The minimum instream flowf the Daejeon Stream was legally set to 0.5 m3/s. To ensure theinimum instream flow that is required for the Daejeon Stream’s

cosystem (even during drought periods), Daejeon Metropolitanity built a diversion pipeline from the neighboring watershed’stream and a dam 9 km upstream from the Daejeon Stream’s out-

et. Thus, 0.5 m3/s was set as the basic flow rate to implement thentire habitat modeling process, to reflect the legally ensured min-mum flow rate of the study reach and to simulate the critical lowow conditions, which can be considered as the ecosystem’s baseow.

1254 J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259

ility in

2

(wi

haZscton

sbsa

2

fic

Fig. 3. Habitat suitab

.3.2. Target fishes and habitat suitability index (HSI)In this study, P. nigra (an endangered fish species) and Z. platypus

the dominant species that inhabits most of the Daejeon Stream)ere selected as the target fish to ensure fish diversity and an

ncreased number of suitable fish habitats.The habitat suitability index (HSI) indicates the suitability of

abitats based on a single parameter, such as the velocity, depthnd substrate. The preferred habitat environment for P. nigra and. platypus was monitored by Hur and Kim (2009) based on a field

urvey and fish collection. The HSI for the flow velocity, depth andhannel index (substrate) were obtained from previous research onhe same basin by Hur and Kim (2009). Fig. 3 shows the HSI curvesf the velocity and depth for P. nigra and Z. platypus. The chan-el substrate’s HSI, which is shown in Fig. 3, indicates a favorable

usmat

Fig. 4. Measuring points and hydraulic simulation results of River2D. (a) 65

dex for target fishes.

ubstrate for each target fish, which increases linearly in a rangeetween Index 1 and Index 4. Channel Indices 1, 2, 3 and 4 repre-ent silt (>0.1 mm), sand (0.1–1.0 mm), fine gravel (1.0–50.0 mm)nd coarse gravel (50.0–100.0 mm), respectively.

.3.3. Velocity and depthThe flow velocity and the depth of the reach were monitored by a

eld survey during base flow conditions. Using the observed data, aalibration process was conducted to accurately simulate the nat-

ral flow before performing a fish habitat simulation. The entiretudy reach was modeled using River 2D with 1-m finite elementeshes; 65 measuring points were used to compare the observed

nd simulated hydraulic properties, such as the flow velocity andhe flow depth in the channel.

Measuring points of study reach. (b) Result of hydraulic simulation.

J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259 1255

Fi

mhag4e

tri

Table 22D hydrodynamic simulation results for the study reach.

RMSE R R2

Velocity (m/s) 0.021 0.987 0.974Depth (m) 0.017 0.989 0.978

Table 3Calculated WUA from the River2D model.

dg1

dcb0(c

3

3

amospte

ig. 5. Correlation between observed and simulated data for 15 measuring pointsn the study reach. (a) Velocity, (b) depth.

The boundary conditions that were measured during the 0.5-3/s discharge were used in all of the model simulations presented

erein. Therefore, the upstream boundary condition was given asflow rate of 0.5 m3/s; the downstream boundary condition was

iven in the form of the water surface elevation, which is equal to9.6 EL.m; and Manning’s coefficient of 0.035 was applied to the

ntire study reach.

Fig. 4 shows the location of the 65 measuring points along withhe hydraulic simulation results, without any boulders in the studyeach and with an upstream boundary condition of 0.5 m3/s. Fig. 5llustrates the correlation between the simulated and observed

tlrh

Fig. 6. WUA estimation result in the study reach. (a) 65 Measurin

Target fish Pseudopungtungia nigra Zacco platypus

WUA of study reach (m2) 51.82 1097.88

ata at 15 measuring points in the study reach. The model did aood job of simulating the velocity’s magnitude and depth at the5 measuring points.

Table 2 shows the 2D hydraulic simulation results without boul-ers placed in the channel. The correlation coefficient (R) and theoefficient of determination (R2) of the velocity and the depthetween the model and the observation are higher than 0.98 and.97, respectively. In addition, from the root mean square errorRMSE), we can see that the model can simulate the hydraulicharacteristics of the study reach very well.

. Results and discussion

.1. Habitat modeling results for the current conditions

The physical habitats for two target fish were simulated usingRiver2D model. Because WUA was estimated using the River2Dodel without boulders, the WUA of P. nigra was lower than that

f Z. platypus by 95.28%, as shown in Table 3. The simulation resultshow that, under minimum instream flow conditions, the WUArovided a habitat environment that was much more advantageouso Z. platypus than P. nigra in the Daejeon Stream. Fig. 6 shows thestimated results of the WUA for each target fish.

Because the study reach is relatively poor for P. nigra’s habita-

ion due to urbanization, the diversity of fish in the study reach isow. This finding suggests that P. nigra has not returned to urbanivers, which are not appropriate habitats; however, a relativelyigh number of P. nigra are found in natural rivers. Thus, a variety

g points of study reach. (b) Result of hydraulic simulation.

1256 J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259

Table 4Traditional fish habitat enhancement methods (MOE, 2002).

Methods Boulder type Boulder diameter (mm) Arrangement

oft

3

(etafociesHwR

mraTi

dhV

TtptbtPnw

TW

TS

Table 7WUA simulations that represent the effects of regularly spaced boulders across thestudy reach. The spatial WUA simulations are found in Fig. 8.

Pseudopungtungia nigra

Scenario Number ofboulder

Boulderdistance (m)

WUA (m2) Difference (%)

– None – 51.82 –A1 1 15.5 53.20 2.66A2 2 14 56.01 5.28A3 3 6 67.84 21.12

aaae

3

pm

wnpa

po

sfp

sTbri

Step stone Circular 700–800 Straight lineV-type riffle Circular 700–800 110◦

Single boulder Circular 700–800 Straight line

f habitat environments for fish should be provided, and habitatsor endangered as well as many other species must be created ifhey are to survive.

.2. Evaluation of habitat enhancement design

Typical techniques for enhancing fish habitats in Korean riversMOE, 2002) were analyzed based on the results of habitat mod-ling in the study reach. Examples of fish habitat-enhancingechniques include the step stone method, the V-type riffle method,nd the single boulder method (MOE, 2002). A simulation was per-ormed using the River2D model to analyze the hydraulic effectn fish habitats, especially for both the target fishes. The typi-al methods for boulder placement and configuration are shownn Table 4 and Fig. 7. Because the three different fish habitat-nhancing methods were hypothetically modeled in River2D, theimulated results could not be validated with the observed data.owever, the model’s ability to capture variables in the channelas validated by the observed data; thus, the application of theiver2D model in this study is appropriate.

The evaluation of the traditional fish habitat improvementethods that are based on the WUA are shown in Table 5. In V-type

iffles, the flow in the channel is centralized, and the appropri-te average flow velocity is estimated for the habitats of P. nigra.his improvement extended downstream, and habitats for P. nigrancreased.

As for the single boulder method and the step stone method,iverse flow patterns occurred near boulders; however, the fishabitats for P. nigra did not improve as much as they did under the-type riffle method.

The results of the simulation are presented in Table 6 and Fig. 8.he WUA of P. nigra increased in comparison to its level beforehe boulders were placed, while that of Z. platypus decreased. Thishenomenon occurred due to the increase in the flow velocity andhe generation of an appropriate flow depth for P. nigra by placing

oulders on the riverbeds. Moreover, the V-type riffle method washe most effective method for improving the physical habitats of. nigra. However, in comparison to the increase in the WUA of P.igra, the decrease in the habitation environment for Z. platypusas low.

able 5UA results for three different habitat enhancement methods.

Methods Pseudopungtungia nigra Zacco platypus

WUA (m2) Difference (%) WUA (m2) Difference (%)

None 51.82 1097.88Step stone method 83.16 60.5 1068.82 −2.6V-type riffle method 240.60 364.3 922.87 −15.9Single boulder method 56.05 8.2 1091.95 −0.5

able 6imulation scenario that uses boulders.

Scenario Bouldertype

Boulderarrangement

Distance betweenboulders

Number ofboulder

Case A Circular Straight Regularly spaced 1–9Case B Circular Straight Irregularly spaced 4

rbh

tblhs

TWa

A4 4 3.5 85.04 25.35A5 5 2.25 89.57 5.33A6 9 1.5 92.38 4.25

The traditional method for improving fish habitats positivelyffected the habitation environment for P. nigra. The methods cre-ted an environment in which Z. platypus was also able to live. Evens the flow on the riverbeds diversified, ecosystem diversity wasnsured.

.3. Analysis of the effect of boulder placement

To capture the hydraulic and ecological effect of boulders on thehysical fish habitats, different numbers of boulders and placementethods were compared, as shown in Table 6.In Case A, the simulation results of the WUA were analyzed

hen boulders of the same size were placed in the middle of a chan-el at regular intervals. As illustrated in Table 7, additional bouldersroduce greater WUA increases. When four boulders were placedt 3.5-m intervals, the WUA increased at its highest rate.

In Case B, the WUA of P. nigra was analyzed contingent upon thelacement type of the four boulders, which is the optimal numberf boulders for the 17-m channel width at the studied reach.

Habitat simulation with four boulders in different placementshows that B5 has the highest WUA (112.10 m2). The WUA increaseor B1 to B4 and B6 was low, even though two of the boulders werelaced similarly to B5 (Table 8 and Fig. 9).

B5’s boulder placement resembled two boulders of the sameize that were closely attached together, as shown in Fig. 9(B5).his boulder placement causes an effect similar to that producedy a larger boulder; large boulders create centralized flow, whichesults in good habitats for P. nigra. This creates a pool where flows concentrated, and the water is deep, which is similar to V-typeiffles. Moreover, vortexes are created in the space between theoulders and the banks of the riverbed to provide refuges andabitats for the fish.

Even when the intervals between the boulders, with the excep-ion of two attached boulders, were different and when four

oulders were placed in any form, the results of the habitat simu-

ation did not show a significant increase in the WUA for P. nigra’sabitat environment. Table 8 and Fig. 9 show the results of the WUAimulation with four boulders that were spaced irregularly.

able 8UA simulations that represent the effects of four irregularly spaced boulders

cross the study reach.

Pseudopungtungia nigra

4 Boulders placement scenario WUA (m2) Difference (%)

B1 85.31 –B2 87.74 2.85B3 86.91 1.88B4 86.39 1.27B5 112.10 31.40B6 87.24 2.26

J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259 1257

Fig. 7. Design of traditional fish habitat enhancement methods (MOE, 2002). (a) Step stone. (b) V-type riffle. (c) Single boulder.

Fig. 8. Simulated WUA of Pseudopungtungia nigra (upper) and Zacco platypus (lower) based on 3 scenarios that represent different river habitat enhancement techniques,i.e., (a) and (b) step-stone technique, (c) and (d) V-type riffle method and (e) and (f) single boulder method.

1258 J.H. Lee et al. / Ecological Engineering 36 (2010) 1251–1259

ing fo

4

iTrompn

m8bd

ettitbt

i

tp

ddwe

rtc

R

B

B

C

Fig. 9. WUA simulations for Pseudopungtungia nigra us

. Conclusion

This study applies the WUA concept to analyze the habitat qual-ty enhancement design in an ecologically damaged urban river.he WUA of P. nigra and Z. platypus was 51.82 m2 and 1098 m2,espectively, in natural conditions in an urban stream. The WUAf Z. platypus was about 20 times higher than that of P. nigra. Theonotonous flow pattern of urban rivers was deemed as appro-

riate for the habitation environment of Z. platypus but not for P.igra.

As a result of evaluating the traditional fish habitat enhance-ent design, the WUA of P. nigra increased by 364.3%, 60.5% and

.2% with the V-type riffle method, step stone method, and singleoulder method, respectively. However, the WUA for Z. platypusecreased slightly after the boulders were placed.

An interval of 3.5 m between the boulders proved to be the mostfficient design for single boulder placement to improve fish habi-ats in the studied reach. Additionally, the results that implementedhe V-type riffle method and B5 suggest that habitats for P. nigrancreased when the flow was centralized among several placement

ypes. Thus, a boulder placement design to centralize flow shoulde reviewed by considering the characteristics of the riverbed andarget species.

Revitalizing stressed and endangered fish species in severelympacted urban streams by providing a sufficient physical habi-

C

ur irregularly spaced boulders across the study reach.

at is a fundamental step in river rehabilitation and restorationrojects.

As an alternative to traditional fish habitat enhancementesigns, such as riffles and step stones, a method that uses boul-ers and simulates physical habitats for endangered fish speciesas proposed to enhance the biodiversity of creatures in a riparian

cosystem.To create suitable physical habitats for fish in an urban river

estoration project, target species that are found to be in danger inhe target river should be accurately reviewed, and their ecologicalharacteristics should be carefully examined first.

eferences

ovee, K.D., 1982. A Guide to Stream Habitat Analysis Using the Instream Flow Incre-mental Methodology. Instream Flow Information Paper No. 12. U.S. Fish andWildlife Service. FWS/OBS-82/26.

rown, R.A., Pasternack, G.B., 2008. Comparison of methods for analyzingsalmon habitat rehabilitation designs for regulated rivers. River Res. Appl.doi:10.1002/rra.1189.

airns Jr., J., Dickson, K.L., Herricks, E.E., 1977. Recovery and Restoration of Damaged

Ecosystems, vol. 531. University of Virginia Press, Charlottesville, Virginia, USA.

lark, J.S., Hession, W.C., Rizzo, D.M., Laible, J.P., Watzin, M.C., 2006. Two-dimensional hydraulic modeling approach to linking stream morphology andaquatic habitat quality. In: Proceedings of the iEMSs Third Biennial Meeting:“Summit on Environmental Modelling and Software”. International Environ-mental Modelling and Software Society, Burlington, USA, CD ROM.

ginee

C

G

C

D

G

G

G

H

H

H

K

L

L

M

M

N

N

S

Spence, R., Hickley, P., 2000. The use of PHABSIM in the management of waterresources and fisheries in England and Wales. Ecol. Eng. 16, 153–158.

J.H. Lee et al. / Ecological En

ommittee on Geum River Basin Management, 2005. Investigation of Flora & Faunain Fresh Water Ecosystem and Biological Community Successions. NationalInstitute of Environmental Research of Korea (NIER), pp. 1–436 (in Korean).

older Kingett Mitchell, 2007. Waimakariri River Flow, Habitat Availability andAngling Suitability: Two-Dimensional Modeling Results. URS New Zealand Ltd.,URSNZ-CHC-009.

rowder, D.W., Diplas, P., 2000. Using two-dimensional hydrodynamic models atscales of ecological importance. J. Hydrol. 230, 172–191.

aejeon Metropolitan City, 2008. Mater Plan of Daejeon Stream Management. Dae-jeon Metropolitan City Report, pp. 1–294 (in Korean).

arcía de Jalon, D., Gortazar, J., 2006. Evaluation of instream habitat enhance-ment options using fish habitat simulations: case-studies in the river\as (Spain).Aquat. Ecol. 41, 461–474.

ore, J.A., 1985. The Restoration of River and Streams: Theories and Experience, vol.280. Butterworth Publishers, Boston, USA.

ore, J.A., Hamilton, S.W., 1996. Comparison of flow-related habitat evaluationsdownstream of low-head weirs on small and large fluvial ecosystem. Regul.Rivers: Res. Manage. 12, 459–469.

auer, C., Unfer, G., Schmutz, S., Habersack, H., 2007. The importance of morphody-namic processes at riffles used as spawning grounds during the incubation timeof nase (Chondrostoma nasus). Hydrobiologia 570, 15–27.

auer, C., Unfer, G., Schmutz, S., Habersack, H., 2008. Morphodynamic effects onthe habitat of Juvenile Cyprinids (Chondrostoma nasus) in a restored Austrianlowland river. Environ. Manage. 42 (2), 279–296.

ur, J.W., Kim, J.K., 2009. Assessment of riverine health condition and estimation ofoptimal ecological flowrate considering fish habitat in the lower Yongdam Dam.J. Korea Water Resour. Assoc. (KWRA), 42 (6) 481–491 (in Korean).

onrad, C.P., 2009. Simulating the recovery of suspended sediment transport andriver-bed stability in response to dam removal on the Elwha River, Washington.Ecol. Eng. 35, 1104–1115.

S

ring 36 (2010) 1251–1259 1259

ee, J.H., Jeong, S.M., Lee, M.H., Lee, Y.S., 2006. Estimation of instream flow for fishhabitat using instream flow incremental methodology (IFIM) for major tribu-taries in Han River Basin. J. Korean Society Civil Eng. (KSCE) 26 (2B), 153–160(in Korean).

opes, L.F.G., Carmo, J.S.A.D., Cortes, R.M.V., Oliveira, D., 2004. Hydrodynamicsand water quality modelling in a regulated river segment: application on theinstream flow definition. Ecol. Eng. 173, 197–218.

ilhous, R.T., Updike, M.A., Schneider, D.M., 1989. Physical habitat simulationsystem reference manual – ver. II. Instream Flow Information Paper No. 26.Biological Report 89(16). US Fish and Wildlife Service.

OE, Ministry of Environment, 2002. Stream Restoration Guideline. Ministry ofEnvironment, Republic of Korea, pp. 1–255.

agaya, T., Shiraishi, Y., Onitsuka, K., Higashino, M., Takami, T., Otsuka, N., Akiyama,J., Ozeki, H., 2008. Evaluation of suitable hydraulic conditions for spawning ofayu with horizontal 2D numerical simulation and PHABSIM. Ecol. Model. 215,133–143.

ational Biological Service, 1995. The Instream Flow Incremental Methodology. APrimer for IFIM, Biological Report 29. US Department of the Interior.

hih, S.-S., Lee, H.-Y., Chen, C.-C., 2008. Model-based evaluations of spur dikes for fishhabitat improvement: a case study of endemic species Varicorhinus barbatulus(Cyprinidae) and Hemimyzon formosanum (Homalopteridae) in Lanyang River,Taiwan. Ecol. Eng. 34, 127–136.

teffler, P., Blackburn, J., 2002. River2D Two Dimensional Depth Averaged Modelof River Hydrodynamics and Fish Habitat: Introduction to Depth AveragedModeling and User’s Manual. University of Alberta. http://www.river2D.ulberta.ca.