Simulation of Dualistic Hydrological Processes Affectedby Intensive Human Activities Based on Distributed

Hydrological ModelZuhao Zhou, Ph.D.1; Yangwen Jia, Ph.D.2; Yaqin Qiu, Ph.D.3; Jiajia Liu, Ph.D.4;Hao Wang, Ph.D.5; Chong-Yu Xu, Ph.D.6; Jia Li, Ph.D.7; and Lin Liu, Ph.D.8

Abstract: Affected by intensive human activities, basin hydrologic systems show characteristics of a dualistic structure. Thus, simulation ofhydrological processes needs to consider the social water cycle system to obtain a more accurate result. This study uses a dualistic water cyclesimulation system to simulate the hydrological processes affected by intensive human activities, which comprises the natural watercycle system and social water cycle system. The social water cycle system includes the agricultural water cycle system, industrial watercycle system, domestic water cycle system, and cross-regional allocation system. As part of the dualistic water cycle simulation system, anintegrated dualistic hydrological model is developed which couples a distributed hydrological simulation module with a water resourceallocation module. The integrated modeling approach is applied to the Haihe River basin. The results show that the model performancecan be improved when considering coupled simulation of natural and social water cycle systems. This model can also be used toinvestigate the evolutionary relationships among the natural and social water cycle systems in the study region, and to assess the regionalwater resources in other basins affected by intensive human activities. DOI: 10.1061/(ASCE)WR.1943-5452.0000990. © 2018 AmericanSociety of Civil Engineers.

Author keywords: Dualistic water cycle; Distributed hydrological model; Intensive human activities; WEP-L model; Social water cycle;Haihe River basin.

Introduction

The scientific research of water resources, hydrology, andhydroecology is based on water cycle processes. With growth inpopulation and development of the economy, intense human activ-ity has disturbed natural water cycles, resulting in severe changes inthe driving forces, structure, and parameters of these cycles. Thewater cycle system has gradually evolved from a single natural fea-ture into a nature–society dualistic water cycle system (Wang andYang 2010; Wang et al. 2013; Qin et al. 2014; Schmidt 2014;Liu et al. 2014, 2010). Different from the natural water cycle

(Horton 1931), the social water cycle is a new research field neces-sitated through the development of urbanization and water shortagethat focuses on the impacts of human activities on the water cycle(Ridolfi 2014; Merrett 1997; Linton 2014; Schellekens et al. 2017).Thus, the new Scientific Decade (2013–2022) of the InternationalAssociation of Hydrological Sciences (IAHS) takes hydrological-social modeling as the fourth science question: “How can we useimproved knowledge of coupled hydrological-social systems toimprove model predictions, including estimation of predictiveuncertainty and assessment of predictability?” (Montanari et al.2013).

1Professor, State Key Laboratory of Simulation and Regulation of WaterCycle in River Basin, China Institute of Water Resources and HydropowerResearch, A-1 Fuxing Rd., Haidian District, Beijing 100038, China;Professor, Engineering and Technology Research Center for WaterResources and Hydroecology, Ministry of Water Resources, No. 1,Yuyuantan Rd., Haidian District, Beijing 100038, China.

2Professor, State Key Laboratory of Simulation and Regulation ofWater Cycle in River Basin, China Institute of Water Resources and Hydro-power Research, A-1 Fuxing Rd., Haidian District, Beijing 100038, China;Professor, Engineering and Technology Research Center for Water Re-sources and Hydroecology, Ministry of Water Resources, No. 1, YuyuantanRd., Haidian District, Beijing 100038, China.

3Professor, State Key Laboratory of Simulation and Regulation ofWater Cycle in River Basin, China Institute of Water Resources and Hydro-power Research, A-1 Fuxing Rd., Haidian District, Beijing 100038, China;Professor, Engineering and Technology Research Center for WaterResources and Hydroecology, Ministry of Water Resources, No. 1,Yuyuantan Rd., Haidian District, Beijing 100038, China.

Note. This manuscript was submitted on August 29, 2017; approvedon May 13, 2018; published online on September 27, 2018. Discussionperiod open until February 27, 2019; separate discussions must be sub-mitted for individual papers. This paper is part of the Journal of WaterResources Planning and Management, © ASCE, ISSN 0733-9496.

4Senior Engineer, State Key Laboratory of Simulation and Regula-tion of Water Cycle in River Basin, China Institute of Water Resourcesand Hydropower Research, A-1 Fuxing Rd., Haidian District, Beijing100038, China; Senior Engineer, Engineering and TechnologyResearch Center for Water Resources and Hydroecology, Ministry ofWater Resources, No. 1, Yuyuantan Rd., Haidian District, Beijing100038, China (corresponding author). ORCID: https://orcid.org/0000-0002-5730-2590. Email: vaver@foxmail.com; wereworld@163.com

5Professor, State Key Laboratory of Simulation and Regulation ofWater Cycle in River Basin, China Institute of Water Resources andHydropower Research, A-1 Fuxing Rd., Haidian District, Beijing100038, China; Professor, Engineering and Technology Research Centerfor Water Resources and Hydroecology, Ministry of Water Resources,No. 1, Yuyuantan Rd., Haidian District, Beijing 100038, China.

6Professor, Dept. of Geosciences Hydrology, Univ. of Oslo, SemSaelands vei 1, P.O. Box 1047 Blindern, N-0316 Oslo, Norway.

7Engineer, Bureau of South to North Water Transfer of Planning, De-signing and Management, Ministry of Water Resources, No. 3, YuyuantanRd., Haidian District, Beijing 100038, China.

8Engineer, China Construction Water Affairs Environmental ProtectionCo., Ltd., Chongxin Bldg., No. 13 Hangfeng Rd., Fengtai District, Beijing100070, China.

© ASCE 04018077-1 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

The natural water cycle is driven only by natural forces such asthe sun and wind. The social water cycle is driven by mechanical,electrical, and other forces having obvious artificial features,which is another water circulation in addition to the naturalwater cycle and could contain all the water-use processes ofhuman activities, such as agricultural water use, industrialwater use, domestic water use, cross-regional allocation, andso on. The dualistic water cycle system combines the naturalwater cycle system, which includes the natural transformationprocesses of the soil–vegetation–atmosphere continuum, withthe social water cycle system, which includes the processes ofwater storage, water intake, water conveyance, water use, drain-age, and water reuse (Wang et al. 2006, 2013; Liu et al. 2010). Thesocial and natural water cycle systems are dependent on andaffected by each other. The social water cycle system takes theavailable water resources from the natural water system, whereasthe latter is affected by the former because of its water intake anddrainage process.

Because the impact of human activity varies among regions, dis-tributed hydrological models are widely used in water cycle sim-ulation. The most recent distributed hydrological models focusmainly on the evolution mechanism of the natural water cycle,which simulate the natural processes of precipitation, evaporation,infiltration, runoff yield, and confluence. However, these models,such as the Soil and Water Assessment Tool (SWAT) (Arnold et al.1998; Piniewski et al. 2017; Liang et al. 2017; Shrestha et al. 2017),system hydrological european (SHE) (Abbott et al. 1986), distrib-uted basin simulator (DBSIM) (Garrote and Bras 1995), distributedhydrology-vegetation model (DHVM) (Wigmosta et al. 1994),geomorphology-based hydrological model (GBHM) (Yang et al.1998, 2002), variable infiltration capacity model (VIC) (Liangand Xie 2001; Luo et al. 2013), soil and water integrated model(SWIM) (Krysanova et al. 1998), and TOPgraphy based hydrolog-ical MODEL (TOPMODEL) (Beven and Kirkby 1979; Famigliettiet al. 1992), simplify or neglect the influence of the social watercycle process. Many works have emphasized the influences of so-cial factors but with a simplified natural water cycle (using a con-ceptual hydrological model or using observation data) (Döll et al.1999, 2003; Alcamo et al. 2003; Carcy et al. 2014; DeFries andEshleman 2004; Harou et al. 2009; Kyle et al. 2013; Blancet al. 2014; Hejazi et al. 2014; Liu et al. 2015; Kim et al.2016). Thus, many researchers developed or improved the hydro-logical models to evaluate the impact of human activities (such aswater use, irrigation schemes, groundwater pumping, and reservoiroperations) on hydrology (Biemans et al. 2011; Condon andMaxwell 2014, 2015; De Graaf et al. 2014, 2015; Sacks et al.2009; Leng et al. 2013, 2014, 2015; Van Der Knijff et al. 2010;Döll et al. 2009, 2012; Pokhrel et al. 2015; Voisin et al. 2013,2017; Wada et al. 2014). Biemans et al. (2011) studied the impactof reservoirs on discharge and irrigation water supply at differentspatial scales and found that reservoirs have a significant effect onirrigation water supply. Condon and Maxwell (2014) studiedthe impact of irrigation on groundwater depth and found thatgroundwater-fed irrigation amplifies the annual streamflow cyclein climatological dry periods which have a higher irrigation de-mand. De Graaf et al. (2014) explored the attribution of water de-mand to simulated water availability and found that the dynamicrepresentation of abstractions and feedbacks have an important ef-fect on the process of assessing the spatial and temporal impacts ofglobal water demand change. Pokhrel et al. (2015) developed anintegrated hydrologic model which is capable of studying the dy-namic relationship between human water use, groundwater storage,and the entire hydrologic cycle. Voisin et al. (2017) integrated aspatially distributed allocation system in the regional Earth system

to detect the influence of water withdrawals and consumptive de-mands and their allocation on surface and groundwater sources.They found that groundwater use has a pronounced effect on reduc-ing water supply deficit and that the sectoral return flow is a majordrive for spatial redistribution of water resources and water deficits.Most studies are global and continental in scale, and do not focuson all aspects of the social water cycle.

The aims of this study are to develop a watershed-scale inte-grated dualistic water cycle model to simulate the natural–socialhydrological processes, and to analyze the impact of social wateruse on the natural hydrological processes, such as streamflow,evapotranspiration, and so on. The water and energy transfer pro-cess in large river basins (WEP-L) model (Jia et al. 2006) is used asthe basis of the hydrological module, and is embedded in the socialwater cycle system module, which includes irrigation water use,domestic water use, industrial water use, reservoir water allocation,groundwater pumping, and so on. This study introduces the mecha-nism and principles of the dualistic water cycle process, and usesa case study of the Haihe River basin to investigate the modelperformance over traditional hydrological models.

Study Area and Data

Study Area

The Haihe River basin is located in North China (112°–120° E, 35°–43° N) across eight provinces, municipalities, and autonomous re-gions (Fig. 1). The total drainage area is 320,041 km2. The hillyregion has an area of 170,460 km2, accounting for 53% of the totalarea, and the rest is plains. The climate of this basin is semiarid andsemihumid. The mean annual rainfall across the study area for theperiod 1956–2005 was 535 mm, and the multiyear average annualtotal water resources (1956–2005) were 37.0 × 109 m3, of whichthe surface water resources were 21.6 × 109 m3, groundwater re-sources were 15.4 × 109 m3, and the average utilization was 23.5 ×109 m3 (MWR 2006).

With the rapid social and economic development, the intensityof human activities in the Haihe River basin has increased dra-matically since the 1970s. Until 2005, about 1.07 × 107 hm2,or 33% of the study area, was covered by agriculture, of whichabout 7.54 × 106 hm2 constituted effective irrigation area. Thelarge-scale irrigation areas in the Haihe River basin are mainlylocated in the piedmont plain and the Yellow River irrigation area.With a significant role in national economic and social develop-ment, the Haihe River basin has become an important industrialand high-tech base, which includes the large cities of Beijing,Tianjin, Shijiazhuang, and Tangshan. In order to increase thewater supply capacity and fulfill water demands, a large numberof water storage projects were constructed during recent decades.In 2005, the number of reservoirs reached 1,865, with a totalcapacity of 3.13 × 1010 m3. Of these, 33 large reservoirs had atotal capacity of 26.2 × 109 m3 (Fig. 1). In addition, a volumeof 3.72 × 109 m3 of water still needs to be diverted from theYellow River (MWR 2005).

In 2005, the total water consumption of the Haihe River basinwas 38.0 × 109 m3, of which the consumptions of domestic water,industrial water, agricultural water, and ecoenvironmental water were5.6 × 109 m3 (14.6%), 5.7 × 109 m3 (14.9%), 26.4 × 109 m3

(69.5%), and 0.4 × 109 m3 (1.0%), respectively. In other words,the water consumption is larger than the average yearly availablewater resources. Thus, human activities have significant affectedthe formation and transformation of water resources. Therefore,

© ASCE 04018077-2 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

a dualistic water cycle system modeling approach that integrates thenatural and social cycle in this basin needs to be developed.

Data

The data used in this case included the following types: (1) dailymeteorological data from 57 national weather stations (1956–2005),

including air temperature, wind speed, sunshine duration, vaporpressure, and relative humidity, which were provided by NationalMeteorological Information Center of China (2012); (2) dailyprecipitation data from 1,560 stations (1956–2005) obtained fromVolume 3 of Annual Hydrological Report of China (Ministryof Water Resources of China 2005); (3) monthly flow data of

Fig. 1. Study area of Haihe River basin, China.

© ASCE 04018077-3 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

22 runoff station (1956–2000), also obtained from the Volume 3 ofAnnual Hydrological Report of China; (4) the Haihe River basinremote sensing evapotranspiration data comes fromMODIS Evapo-transpiration data set (NASA 2010); (5) surface elevation data ob-tained from the United States Geological Survey (USGS) LandProcesses Distributed Active Archive Center (LP DAAC) (USGS1998); (6) a river distribution map (spatial resolution 1∶25 × 106)supplied by Ren (2007); (7) land-use maps of year 1996, 2000,and 2004 are obtained from National Earth System Science DataSharing Infrastructure, National Science and Technology Infra-structure of China (National Earth System Science Data SharingInfrastructure n.d.); (8) spatial distribution data of soil type (spatialresolution 1∶10 × 107) obtained from the second national soil surveyand a related study by Ma (1994); (9) data of plains aquifer hydro-geological unit divisions and permeability coefficients reported byRen (2007); (10) data of large reservoir characteristics and sluiceparameters obtained from Haihe RiverWater Conservancy Commis-sion (HRWCC); (11) monthly storage changes of 33 reservoirsduring 1956–2005 obtained from HRWCC; (12) data of the area,plant structure, and irrigation patterns of 77 large irrigation zonesobtained from HRWCC; and (13) the water use data of agriculture,industry, and domestic use for 1980–2005 obtained from HRWCC.

Model Development

Natural Water Cycle Part

Natural water cycle processes and the model simulation results areaffected by evaporation and other factors driven by precipitation.Watershed subdivision and codification are critical for the watercycle simulation process. The simulated drainage network was firstextracted from a digital elevation model (DEM) modified to include

the surveyed river network. Then the simulated drainage networkwas used for the watershed subdivision and codification based onthe coding rule (Luo et al. 2006; Liu et al. 2014). The model sub-divided the entire basin into 3,067 subwatersheds to reflect the up-stream and downstream relationship, and the subwatersheds werefurther subdivided into 11,752 contour bands to reflect the heighteffect and flow movements on the overland route. The verticalstructure of the contour bands included the interception layer,depression layer, upper soil layer, transition layer, unconfined aqui-fer, and confined aquifer (Fig. 2). The mosaic method (Avissar andPielke 1989) was used to consider the heterogeneity of land use,which was divided into five groups within each contour band: waterbody, soil–vegetation, impervious area, irrigated farmland, and non-irrigated farmland. Essentially, the authors classified the land useunits and estimated the water and heat fluxes of each land type,and used the area’s mean value as the water and heat fluxes ofthe unit.

The natural water cycle was simulated based on the WEP-Lmodel, which is a distributed hydrologic model that combinesthe merits of physically based, spatially distributed (PBSD) andsoil–vegetation–atmosphere transfer scheme (SVAT) models. Themodel is capable of dynamically assessing water resources in vari-ous forms, such as the surface water, groundwater, and rainfallwater consumed by vegetation. It is also capable of modeling in-filtration excess, saturation excess, and mixed runoff generationmechanisms. Jia et al. (2001, 2006) provided more details.

Social Water Cycle Part

Social Water Cycle SystemThe social water cycle system was subdivided into three subsys-tems according to the water supply targets: agricultural, industrial,

Fig. 2. Vertical structure of natural water cycle simulation.

© ASCE 04018077-4 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

and domestic water cycle systems. In addition, there is a cross-regional water supply system for the water transfer from reservoirsor another river basin to the study area. Fig. 3 shows a generaliza-tion of the social water cycle system. In this model, the social watercycle has the same basic calculation unit with natural water cycle,i.e., contour bands within the subwatershed, and the same dailytime step for the coupled calculation.

The social water cycle focuses on the social water circulation,which includes six aspects: water intake, water conveyance, waterstorage, water use, water drainage, and water reuse (Wang et al.2013). Mostly, water intake and water drainage are the two criticalprocesses related to the natural water cycle. The coupling mecha-nism occurs mainly during these two processes. Rivers, reservoirs,and groundwater can be considered as water intake sources. Thewater intake processes decrease the water in the river and ground-water, whereas the drainage processes increase the water in the riverand underground. In addition, the model considered waste waterreuse. Fig. 4 shows the coupling principals.

To simulate the social water cycle, every water cycle subsystemmust first be generalized by determining the distribution of demandnodes in each subsystem and examining the hydraulic connectionbetween the demand nodes and the water intake position. This stepassumes that the groundwater intake position is located in the con-tour centroid and its water supply is used only for the social watersystem in the contour band. For surface water, the intake position isthe subwatershed outlet if no diversion work or water intake proj-ects exist. The reservoir intake position is located in the outlet of itsown subwatershed dam site. Once all of the subsystem structureswere generalized, the social water cycle was simulated in termsof water storage, intake, conveyance, use, drainage, and reuse.

The water intake process was omitted in the case of water divisionengineering without dams and water lifting engineering.

Unit of Social Water Cycle SystemIn the agricultural water cycle system, the demand nodes are theirrigated farmland, forestland with irrigation facilities, and fish-ponds. The spatial distribution of demand nodes can be determinedby using statistical data of the cultivated land (or irrigated forest-land or fishponds) in addition to a spatial distribution map of theirrigated farmland (or artificial forestland, reservoirs, and ponds).Each contour band included two types of irrigation areas: large-to-medium-sized areas, and small area. Each large-to-medium-sizedarea were viewed as a combined demand node, whereas the otherareas were generalized as one demand node. For large-to-medium-sized areas, the hydraulic connection can be determined by exam-ining the relationship between irrigation area and water intake. Interms of small areas, it is assumed that surface water and ground-water water intakes are located in sub-watersheds and contourbands where the small-sized irrigation areas are. Each irrigated for-estland (or fishpond) can be generalized to one demand node. It wasassumed that surface water and groundwater intakes were locatedin the same subwatersheds and contour bands as the irrigatedforestlands (or fishponds).

In the industrial water cycle system, the demand nodes are in-dustrial and mining enterprises, whose spatial distribution could bedetermined by the map of industrial and mining enterprises distri-butions. The industrial and mining enterprises were generalizedinto two categories: those that use local water were generalizedto one demand node, and others that take water from water transferprojects were generalized to another demand node. It was assumed

Fig. 3. Generalizations of social water cycle systems.

© ASCE 04018077-5 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

that surface water and groundwater intakes were located in thesame subwatersheds and contour bands as the industrial and miningenterprises. In terms of the conditions in which water is taken froman interbasin water diversion project, the hydraulic connection be-tween the demand nodes and the water intakes can be determinedby examining the conditions of the water diversion project.

In the domestic water cycle system, the demand nodes are urbanresidents and rural residents. The spatial distribution can be deter-mined by using the statistical data of urban population (or ruralpopulations) in addition to a spatial distribution map of urban land(or rural residential) areas. Two types of urban residents were con-sidered during the generalization process: those who use localwater were generalized to one demand node, and those who takewater from an interbasin water transfer project were generalized toa different demand node. For the first type of residents, it was as-sumed that the surface water and groundwater intakes ware locatedin the subwatersheds and contour bands where they reside. In termsof the residents who take water from an interbasin water diversionproject, the hydraulic connection between the demand nodes andthe water intakes can be determined by examining the conditions ofthe water diversion project. Rural residents in each contour bandcan be generalized to one demand node. It was assumed that surfacewater and groundwater intakes were located in the subwatershedand contour bands where they reside.

To simplify the model structure, the historical statistical wateruse data were used directly in the dualistic water cycle model.Usually, the available historical statistical water use data are ona yearly scale and at the administrative regional level, which islarger than the model scale. Therefore, it was necessary to down-scale these data to obtain daily water use in each calculation unit inboth spatial and temporal aspects. The yearly water use of eachsubsystem is

Wa ¼XY

t¼1

XN

u¼1

fðWu;t;Pt;R;MÞ ð1Þ

Wi ¼XY

t¼1

XN

u¼1

Gu;t ×Wu ð2Þ

Wl ¼XY

t¼1

XN

u¼1

Wt × ρu × Au ð3Þ

where Wa (m3) = total annual water consumption for agriculturalirrigation; Wu;t (m3) = water demand of various calculation unitsfor the number of days t, calculated by the Penman-Monteith equa-tion (Monteith 1973); Pt (m3) = amount of daily rainfall, calculatedby the precipitation (mm) multiplied by the area of agricultural ir-rigation; R (m3) = amount of available water resources in the area;M = agriculture irrigation management factors such as irrigationwater management level and irrigation patterns; f = functionalrelationship between practical water consumption and water re-quirement under certain conditions; Wi (m3) = total annual waterconsumption for industrial use; Gu;t (million yuan) = daily grossdomestic product (GDP) value in each calculation unit; Wu(m3=million yuan) = GDP water consumption in every calculationunit; Wl (m3) = total annual water consumption for domestic use;Wt (m3=day=per person) = daily water consumption per capita in acalculation unit; ρu (person=km2) = population density in a calcu-lation unit; Au (km2) = area of calculation unit; u = calculationunit number; t = daily serial number; N = total number of calcu-lation units within the region; and Y = total days in a year(365 or 366).

In model applications, the daily irrigation water consumption inthe calculation units was downscaled from yearly irrigation wateruse data considering precipitation, evapotranspiration, and irriga-tion management factors. The daily industrial water consumptionin the calculation units was downscaled from yearly industrialwater use and used GDPs as weight. The daily domestic waterconsumption in unit calculation was downscaled from yearlydomestic water use data and used population as weight. The de-tailed method follows that reported by Cui et al. (2010) andCao et al. (2010).

Water Resources and Water Allocation SystemThe water resources of the social water cycle system include fourtypes: local surface water, cross-regional surface water, cross-basinsurface water, and groundwater. Water is supplied by the four typesof sources consequently. The local surface water comes from the

Fig. 4. Simulation principals of social water cycle.

© ASCE 04018077-6 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

streamflow of the local river, the storage of the local reservoir orpond, the retained water, and so on. The cross-regional surfacewater comes from the storage of large reservoirs upstream ofthe water-use unit, and the water is transferred over long distance.The cross-basin surface water comes from the streamflow out of thelocal basin or water transferred from larger water diversion proj-ects. In the Haihe River basin, it could be water from Yellow Rivebasin or the South-to-North Water Diversion Project of China.

Fig. 5 shows the generalized cross-regional water distributionsystem. The demand nodes of the cross-regional water distributionsystems are for industrial, agricultural, and domestic water use.Every subwatershed in the water supply area can be generalizedto one demand node. The hydraulic connection between demandnodes and the reservoir can be determined according to the watersupply range.

Some reservoirs supply water to the region near it. However,some reservoirs supply water to regions far from them. For thesereservoirs, the entire water supply area can be divided into severalwater-use zones, and the amount of reservoir water supply is equalto the sum of water consumption of each zone. Usually, the waterdemand is always supplied if the reservoirs have adequate water. Ifthe reservoirs cannot meet the demand during drought conditions,river water will be used to meet the shortfall. If demand still cannotbe met, a caution will be written out for warning. The simulation ofthe reservoir is estimated as

Sr ¼XNz

z¼1

Wz ¼XNz

z¼1

ðWa;z þWi;z þWl;z − SlocalÞ ð4Þ

ΔS ¼ Win − Sr − E −D ð5Þ

where Sr (m3) = reservoir water supply;Wz (m3) = water consump-tion in water-use zone z; z = water-use zone; Nz = sum of all thezones supplied water by the reservoir; Wa;z (m3) = amount of agri-cultural irrigation water consumption in water-use area z, whichcan be calculated by Eq. (1);Wi;z (m3) = industrial water consump-tion in water-use zone z, which can be calculated by Eq. (2);Wl;z (m3) = domestic water consumption in water-use zone z,which can be obtained by Eq. (3); Slocal (m3) = available amountof water supply by local water resources, including surface water,groundwater, and reclaimed water; ΔS = water storage change ofthe reservoir; Win = income water of the reservoir from upstream

streamflow; E = evaporation of the reservoir; and D = drainage ofthe reservoir when the storage is over the storage capacity of thereservoir.

In the model, both the surface water and groundwater use dataare introduced into the model separately and downscaled using theaforementioned method. The natural water cycle is simulated usingWEP-L model, whereas the social water cycle is simulated basedon the statistical data, so an allocation scheme is needed for thecoupling. First, the local surface water and groundwater withinthe subbasin in which the computation unit is located is used tomeet the demand of the social water cycle system; second, thecross-regional surface water is used for the remainder demand;third, the cross-basin surface water, if it exists, is used for theremainder demand; and finally, if demand still cannot be met, acaution is recorded in the output file.

In the social water cycle system, the irrigation water is used likeadditional precipitation, which falls on the irrigated farmland andtakes part in the infiltration and evapotranspiration processes. Mostof the industrial water drains back to the river, and the rest isclassified as evaporation, which contains the evaporation fromwater-use processes and the water solidified in the commodity.The domestic water is subdivided into two groups, back to the riverand evaporated.

Distribution of Social Water Use

The areas of land-use categories in the contour bands were used asweights for downscaling the administrative regional-level water-use data to the computation units. Then the annual water-use datawere further downscaled to each day according to the characteris-tics of different social water cycle subsystems. Fig. S1 in the Sup-plemental Data shows the spatial distribution of the diversity socialwater cycle subsystems in 2005, including the distributions ofirrigated water use, industrial water use, residential water use, andso on, for illustration purpose.

Results

Impact on Stream Flow

The data series of 1956–2000 were used for long-term daily sim-ulation of the dualistic water cycle system structure of the HaiheRiver basin, of which the calibration period was from 1956 to1979, and the validation period was from 1980 to 2000. The maincalibration parameters were the high-sensitivity parameters desig-nated by Jia et al. (2006). Because large-scale hydrological mod-eling normally is concerned about the hydrological stations in themain streams of the rivers with longer data records, and many of theother stations do not have long-term observation data, the hydro-logical stations of Chengde, Luanxian, Daiying, Guantai, Guant-ing, and Huangbizhuang were selected to show the impact onvalidation period (Table 1 and Fig. S2). In addition, box-plotsof Nash–Sutcliffe coefficient of validation periods of all 22 stationsare shown to compare the performance of coupled and noncoupledmodel (Fig. S3). The results indicate that the performance of thecoupled model matched the observation data better than did theperformance of the noncoupled model.

Fig. 6 uses the Guantai station to show the details of the impactof coupled and noncoupled model in a typical year. The observationdata were matched better by the coupled model than by the non-coupled model. Moreover, the coupled simulation had a significantinfluence on normal and low flow years. However, it had a lessereffect on high flow years due mainly to the relatively large amount

Fig. 5. Generalized cross-regional water allocation system.

© ASCE 04018077-7 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

of water in these years. Moreover, the impact on the flood seasonwas less than that on the nonflood season.

Impact of Social Water Cycle on Evapotranspiration

Fig. 7 compares the results of the coupled and noncoupled modelsimulations of watershed evapotranspiration processes. The remotesensing image indicates that evapotranspiration in plain areas washigher than that in the hilly areas. Considering the two differentsimulation scenarios, the authors determined that the coupledmodel matched the observation data better than the noncoupledmodel in the piedmont and Yellow River diversion irrigated areas,because the coupled model considers the impact of social watercycle on the basin evapotranspiration.

Fig. 8 compares the situation with and without cross-regionalallocation. Without cross-regional allocation means that the wateronly comes from the local surface water. With cross-regionalallocation means that the water comes from the local, cross-regional, and cross-basin surface water. The figure shows theshortfall of the local river streamflow in meeting the socialwater demand. It indicates that considering the cross-regional al-location scheme could make the social water cycle more accurate.Because the water allocation of the coupled model was closer tothe actual water use, the simulated basin evapotranspiration of thecoupled model was 520 mm, which was larger than that of thenoncoupled model (455 mm), especially in the piedmont andYellow River diversion irrigated areas. Meanwhile, the simulatedstreamflow of the noncoupled model was larger than that of thecoupled model.

Impact on Reservoir Storage Change

Four typical large reservoirs were selected for studying the impacton reservoir storage change of with and without reservoir operation.The selected reservoirs are typical in the Haihe River basin, and are

located on the upstream mountain area of the main river and havecomplicated water supply connections (Fig. 1). They also had along-term storage change in the observation data. The observationsof the Wangkuai reservoir and Huangbizhuang reservoir were from1980 to 2005. The observations of the Miyun reservoir were from1980 to 2000. The observations of the Panjiakou reservoir werefrom 1984 to 2000.

Two situations were considered for testing the impacts with orwithout cross-regional reservoir allocation. With cross-regional res-ervoir allocation, the reservoir storage variable is equal to precipi-tation plus runoff minus evaporation and cross-regional watertransfer. Without cross-regional reservoir allocation, the storagevariable of the reservoir is equal to the precipitation and runoff mi-nus evaporation. The monthly reservoir storage change was stablewhen the cross-regional reservoir allocation was neglected. Thereservoir change in the situation with cross-regional reservoir allo-cation was closer to the observations, although there were someunderestimations and overestimations (Fig. 9). Because the watertransfer of the reservoir is affected by the water supply range andthe water supply amount, deviation in the calculation of water de-mand may cause a deviation in the water supply. Generally speak-ing, the storage was reduced in the preflood period and increased inthe flood recession period, which is consistent with regional wateroperation laws.

Discussions and Conclusions

Simulation of dualistic hydrological processes is of vital impor-tance for dynamic water resources management of regions withintensive human influence, but such studies are rarely seen inthe literature. This study developed an integrated dualistic watercycle modeling approach to analyze the impact of social water useon the natural hydrological processes, which mainly contain thenatural water cycle and the social water cycle. The natural water

Table 1. Model validation results, 1980–2000

Hydrological stations

Coupled model Noncoupled model

Relativeerror (%)

Nash–Sutcliffecoefficient

Correlationcoefficient

Relativeerror (%)

Nash–Sutcliffecoefficient

Correlationcoefficient

Chengde −5.8 0.72 0.85 2.8 0.72 0.72Luanxian −1.3 0.60 0.86 −2.7 0.40 0.76Daiying −4.0 0.65 0.81 7.1 0.50 0.54Guantai 3.6 0.81 0.93 3.1 0.71 0.79Huangbizhuang −9.6 0.68 0.83 −17.6 0.72 0.73

0

100

200

300

400

500

Jan.

Feb.

Mar

.A

pr.

May

.Ju

n.Ju

l.A

ug.

Sep.

Oct

.N

ov.

Dec

.

Jan.

Feb.

Mar

.A

pr.

May

.Ju

n.Ju

l.A

ug.

Sep.

Oct

.N

ov.

Dec

.

Jan.

Feb.

Mar

.A

pr.

May

.Ju

n.Ju

l.A

ug.

Sep.

Oct

.N

ov.

Dec

.Ave

rage

wolf ylhtnom (m

3 /s)

Coupled model Non-coupled model Observed flow

high flow year (1996) normal flow year(1993) low flow year(1999) Guantai

Fig. 6. Comparison of streamflow in three typical years at Guantai station.

© ASCE 04018077-8 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

cycle was simulated using the WEP-L model, so the social watercycle is a key point in this study. Normally, the social water cyclesystem can be divided into agricultural water cycle system,industrial water cycle system, domestic water cycle system,

and cross-regional allocation system. Two hydrological models,a natural–social dualistic water cycle model and a natural watercycle model, were used to simulate hydrological processes ofthe Haihe River basin. This paper comparatively analyzed the

Fig. 7. Comparison of evapotranspiration simulation in 2005: (a) coupled model scenario; (b) noncoupled model scenario; and (c) remote sensingdata.

© ASCE 04018077-9 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

streamflow, evapotranspiration, changes of reservoir storage, andcross-regional surface water allocation. In addition, the couplingprincipals between natural and social water cycles were discussed.

The study concluded that (1) the model simulation efficiencycan be much improved when considering coupled simulation of

natural and social water cycle systems; (2) influenced by humanactivities, the streamflow of the basin is significantly less thanthe natural runoff, which is more prominent in normal and dryyears; (3) as a result of the substantial increase in social water con-sumption, although the river streamflow was greatly reduced, the

Fig. 7. (Continued.)

© ASCE 04018077-10 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

amount of watershed evapotranspiration increased significantly,and, based on the modelling results, the evapotranspiration inthe piedmont and Yellow River diversion irrigated areas is higherthan that in other areas; (4) this paper developed the natural–socialdualistic water cycle model, and the modeled water balance process

of the reservoir and the processes of storage are in good agreementwith the measured processes; and (5) the investigation of the impactof the cross-regional surface water allocation indicated that consid-ering the cross-regional allocation scheme could make the socialwater cycle more accurate.

Fig. 7. (Continued.)

© ASCE 04018077-11 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

This paper proposed an integrated dualistic water cycle model,and applied it to Haihe River basin, and the findings of the studyhave the following implications. The dualistic model can evalu-ate the future trends of water resources, ecology, and the water

environment by creating scenarios based on the predictions ofa climate model and water control conditions; this work couldrefer to Wang et al. (2013). Because the dualistic model cansimulate different hydrological process and human activities

Fig. 8. Impact of cross-regional water allocation: (a) without cross-regional allocation noncoupled model; and (b) with cross-regional allocationcouple model.

© ASCE 04018077-12 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

(corresponding to the subsystem of social water cycle), it couldbe used for the water resources attribution analysis; this workcould refer to Jia et al. (2012). The integrated model still hassome shortcomings that need to be addressed in future studies,such as strengthening the close coupling of the social water cycle

rather than using the statistical data as external input, improvingthe function of simulating agricultural practices with rule-basedbehavior, operating the reservoir using dynamic schedulingrule, strengthening the coupled simulation of surface water andgroundwater, and so on.

Fig. 8. (Continued.)

© ASCE 04018077-13 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Acknowledgments

This research was financially supported by the NationalKey Research and Development Program of China(2016YFC0402405), the National Science and Technology MajorProject of Water Pollution Control and Prevention of China(2008ZX07207-006, 2012ZX07201-006), the major state basicresearch development program of China (2015CB452701), andthe National Natural Science Foundation of China (41601032).The authors also acknowledge the help of the Natural-SocialDualistic Water Cycle Model Research Group and the helpfulsuggestions of external reviewers.

Supplemental Data

Figs. S1–S3 are available online in the ASCE Library (www.ascelibrary.org).

References

Abbott, M., J. Bathurst, J. A. Cunge, P. E. O’Connell, and J. Rasmussen.1986. “An introduction to the European Hydrological System—SystemeHydrologique Europeen, ‘SHE’. 2: Structure of a physically-based, dis-tributed modeling system.” J. Hydrol. 87 (1–2): 61–77. https://doi.org/10.1016/0022-1694(86)90115-0.

(a)

(b)

(c)

(d)

-4

-2

0

2

4

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006

Res

ervo

ir s

tora

ge c

hang

e(1

08m

3 )R

eser

voir

sto

rage

cha

nge

(108

m3 )

Res

ervo

ir s

tora

ge c

hang

e(1

08m

3 )R

eser

voir

sto

rage

cha

nge

(108

m3 )

year

Observed data With reservoir allocation Without reservoir allocation

year

Observed data With reservoir allocation Without reservoir allocation

year

Observed data With reservoir allocation Without reservoir allocation

year

Observed data With reservoir allocation Without reservoir allocation

Fig. 9. Comparison of reservoir storage change: (a) Wangkuai reservoir; (b) Huangbizhuang reservoir; (c) Miyun reservoir; and (d) Panjiakoureservoir.

© ASCE 04018077-14 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Alcamo, J., P. Döll, T. Henrichs, F. Kaspar, B. Lehner, T. Rösch, andS. Siebert. 2003. “Development and testing of the Water GAP 2 globalmodel of water use and availability.” Hydrol. Sci. J. 48 (3): 317–337.https://doi.org/10.1623/hysj.48.3.317.45290.

Arnold, J. G., R. Srinivasan, R. Muttiah, and J. R. Williams. 1998. “Largearea hydrologic modeling and assessment. Part I: Model development.”J. Am. Water Resour. Assoc. 34 (1): 73–89. https://doi.org/10.1111/j.1752-1688.1998.tb05961.x.

Avissar, R., and R. A. Pielke. 1989. “A parameterization of heterogeneousland-surface for atmospheric numerical models and its impact onregional meteorology.” Mon. Wea. Rev. 117 (10): 2113–2136. https://doi.org/10.1175/1520-0493(1989)117<2113:APOHLS>2.0.CO;2.

Beven, K. J., and M. J. Kirkby. 1979. “A physically based, variable con-tributing area model of basin hydrology.” Hydrol. Sci. Bull. 24 (1):43–69. https://doi.org/10.1080/02626667909491834.

Biemans, H., I. Haddeland, P. Kabat, F. Ludwig, R. W. A. Hutjes, J. Heinke,W. von Bloh, and D. Gerten. 2011. “Impact of reservoirs on riverdischarge and irrigation water supply during the 20th century.” WaterResour. Res. 47 (3): W03509. https://doi.org/10.1029/2009WR008929.

Blanc, E., K. Strzepek, A. Schlosser, H. Jacoby, A. Gueneau, C. Fant, S.Rausch, and J. Reilly. 2014. “Modeling U.S. water resources under cli-mate change.” Earths Future 2 (4): 197–224. https://doi.org/10.1002/2013EF000214.

Cao, Z., Y. Qiu, and Z. Zhou. 2010. “Water distribution of Songliao Basinin time and space.” In Vol. 2 of Proc., 9th Int. Conf. on Hydroinfor-matics 2010, 1663–1670. Beijing: Chemical Industry.

Carcy, M., M. Baraer, B. G. Mark, A. French, J. Bury, K. R. Young, andJ. M. McKenzie. 2014. “Toward hydro-social modeling: Merging hu-man variables and the social sciences with climate-glacier runoff models(Santa River, Peru).” J. Hydrol. 518 (Part A): 60–70. https://doi.org/10.1016/j.jhydrol.2013.11.006.

Condon, L. E., and R. M. Maxwell. 2014. “Feedbacks between managedirrigation and water availability: Diagnosing temporal and spatialpatterns using an integrated hydrologic model.” Water Resour. Res.50 (3): 2600–2616. https://doi.org/10.1002/2013WR014868.

Condon, L. E., and R. M. Maxwell. 2015. “Evaluating the relationshipbetween topography and groundwater using outputs from a continental-scale integrated hydrology model.” Water Resour. Res. 51 (8): 6602–6621. https://doi.org/10.1002/2014WR016774.

Cui, X., Z. Zhou, and D. Qin. 2010. “Simulation and analysis of agricul-tural irrigation in Songliao Basin.” In Vol. 2 of Proc., 9th Int. Conf. onHydroinformatics 2010, 1655–1662. Beijing: Chemical Industry.

Defries, R., and K. N. Eshleman. 2004. “Land-use change and hydrologicprocesses: A major focus for the future.” Hydrol. Processes 18 (11):2183–2186. https://doi.org/10.1002/hyp.5584.

De Graaf, I., E. Sutanudjaja, L. van Beek, and M. Bierkens. 2015. “A high-resolution global-scale groundwater model.” Hydrol. Earth Syst. Sci.19 (2): 823–837. https://doi.org/10.5194/hess-19-823-2015.

De Graaf, I., L. van Beek, Y. Wada, and M. Bierkens. 2014. “Dynamicattribution of global water demand to surface water and groundwaterresources: Effects of abstractions and return flows on river discharges.”Adv. Water Resour. 64: 21–33. https://doi.org/10.1016/j.advwatres.2013.12.002.

Döll, P., K. Fiedler, and J. Zhang. 2009. “Global-scale analysis of river flowalterations due to water withdrawals and reservoirs.”Hydrol. Earth Syst.Sci. 13 (12): 2413–2432. https://doi.org/10.5194/hess-13-2413-2009.

Döll, P., H. Hoffmann-Dobrev, F. T. Portmann, S. Siebert, A. Eicker, M.Rodell, G. Strassberg, and B. Scanlon. 2012. “Impact of water with-drawals from groundwater and surface water on continental water stor-age variations.” J. Geodyn. 59–60: 143–156. https://doi.org/10.1016/j.jog.2011.05.001.

Döll, P., F. Kaspar, and J. Alcamo. 1999. “Computation of global wateravailability and water use at the scale of large drainage basins.” Math-ematische Geologie 4: 111–118.

Döll, P., F. Kaspar, and B. Lehner. 2003. “A global hydrological modelfor deriving water availability indicators: Model tuning and validation.”J. Hydrol. 270 (1–2): 105–134. https://doi.org/10.1016/S0022-1694(02)00283-4.

Famiglietti, J. S., E. F. Wood, M. Sivapalan, and D. J. Thongs. 1992. “Acatchment scale water balance model for FIFE.” J. Geophys. Res.97 (D17): 18997–19007. https://doi.org/10.1029/92JD01049.

Garrote, L., and R. L. Bras. 1995. “A distributed model for real-timeflood forecasting using digital elevation models.” J. Hydrol. 167 (1):279–306. https://doi.org/10.1016/0022-1694(94)02592-Y.

Harou, J. J., M. Pulido-Velazquez, D. E. Rosenberg, J. Medellin-Azuara,J. R. Lund, and R. E. Howitt. 2009. “Hydro-economic models:Concepts, design, applications, and future prospects.” J. Hydrol.375 (3): 627–643. https://doi.org/10.1016/j.jhydrol.2009.06.037.

Hejazi, M., et al. 2014. “Long-term global water projections using sixsocioeconomic scenarios in an integrated assessment modeling frame-work.” Technol. Forecasting Soc. Change 81: 205–226. https://doi.org/10.1016/j.techfore.2013.05.006.

Horton, R. E. 1931. “The field, scope, and status of the science of hydrol-ogy.” Eos, Trans. Am. Geophys. Union 12 (1): 189–202. https://doi.org/10.1029/TR012i001p00189-2.

Jia, Y., X. Ding, H. Wang, Z. Zhou, Y. Qiu, and C. Niu. 2012. “Attributionof water resources evolution in the highly water-stressed Hai RiverBasin of China.”Water Resour. Res. 48 (2): W02513. https://doi.org/10.1029/2010WR009275.

Jia, Y., G. Ni, Y. Kawahara, and T. Suetsugi. 2001. “Development of WEPmodel and its application to an urban watershed.” Hydrol. Processes15 (11): 2175–2194. https://doi.org/10.1002/hyp.275.

Jia, Y., H. Wang, Z. Zhou, Y. Qiu, X. Luo, J. Wang, D. Yan, and D. Qin.2006. “Development of the WEP-L distributed hydrological modeland dynamic assessment of water resources in the Yellow River basin.”J. Hydrol. 331 (3–4): 606–629. https://doi.org/10.1016/j.jhydrol.2006.06.006.

Kim, S. H., M. I. Hejazi, L. Liu, K. V. Calvin, L. E. Clarke, J. Edmonds,P. Kyle, P. L. Patel, and M. A. Wise. 2016. “Balancing global wateravailability and use at basin scale in an integrated assessment model.”Clim. Change 136 (2): 217–231. https://doi.org/10.1007/s10584-016-1604-6.

Krysanova, V., D. I. Müller-Wohlfeil, and A. Becker. 1998. “Developmentand test of a spatially distributed hydrological/water quality model formesoscale watersheds.” Ecol. Modell. 106 (2–3): 261–289. https://doi.org/10.1016/S0304-3800(97)00204-4.

Kyle, P., E. Davies, J. J. Dooley, S. J. Smith, L. Clarke, J. Edmonds, andM. I. Hejazi. 2013. “Influence of climate change mitigation technologyon global demands of water for electricity generation.” Int. J. Green-house Gas Control 13: 112–123. https://doi.org/10.1016/j.ijggc.2012.12.006.

Leng, G., M. Huang, Q. Tang, H. Gao, and L. Y. R. Leung. 2014.“Modeling the effects of groundwater-fed irrigation on terrestrial hy-drology over the conterminous United States.” J. Hydrometeorol.15 (3): 957–972. https://doi.org/10.1175/JHM-D-13-049.1.

Leng, G., M. Huang, Q. Tang, and L. Y. R. Leung. 2015. “Amodeling studyof irrigation effects on global surface water and groundwater resourcesunder a changing climate.” J. Adv. Model. Earth Syst. 7 (3): 1285–1304.https://doi.org/10.1002/2015MS000437.

Leng, G., M. Huang, Q. Tang, W. J. Sacks, H. Lei, and L. Y. R. Leung.2013. “Modeling the effects of irrigation on land surface fluxes andstates over the conterminous United States: Sensitivity to input dataand model parameters.” J. Geophys. Res. 118 (17): 9789–9803.https://doi.org/10.1002/jgrd.50792.

Liang, X., and Z. Xie. 2001. “A new surface runoff parameterization withsub grid-scale soil heterogeneity for land surface models.” Adv. WaterResour. 24 (9): 1173–1193. https://doi.org/10.1016/S0309-1708(01)00032-X.

Liang, Z. M., T. T. Tang, B. Q. Li, T. Liu, J. Wang, and Y. M. Hu. 2017.“Long-term streamflow forecasting using SWAT through the integrationof the random forests precipitation generator: Case study of DanjiangkouReservoir.” Hydrol. Res. 49 (5): 1513–1527. https://doi.org/10.2166/nh.2017.085.

Linton, J. 2014. “Modern water and its discontents: A history of hydrosocial renewal.” Wiley Interdiscip. Rev. Water 1 (1): 111–120.https://doi.org/10.1002/wat2.1009.

© ASCE 04018077-15 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.

Liu, J., D. Qin, H. Wang, M. Wang, and Z. Yang. 2010. “Dualistic watercycle pattern and its evolution in Haihe River basin.” Chin. Sci. Bull.55 (16): 1688–1697. https://doi.org/10.1007/s11434-010-3043-5.

Liu, J., Z. Zhou, Y. Jia, H. Wang, and X. Chen. 2014. “A stem-branch topo-logical codification for watershed subdivision and identification to sup-port distributed hydrological modelling at large river basins.” Hydrol.Processes 28 (4): 2074–2081. https://doi.org/10.1002/hyp.9746.

Liu, L., M. Hejazi, P. Patel, P. Kyle, E. Davies, Y. Zhou, L. Clarke, andJ. Edmonds. 2015. “Water demands for electricity generation in theUS: Modeling different scenarios for the water-energy nexus.” Technol.Forecasting Soc. Change 94: 318–334. https://doi.org/10.1016/j.techfore.2014.11.004.

Luo, X., Y. Jia, J. Wang, H. Wang, D. Qin, and Z. Zhou. 2006. “Method fordelineation and codification of a large basin based on DEM and sur-veyed river network.” [In Chinese.] Adv. Water Sci. 17 (2): 259–264.

Luo, X., X. Liang, and H. R. Mccarthy. 2013. “VIC+ for water-limitedconditions: A study of biological and hydrological processes and theirinteractions in soil-plant-atmosphere continuum.” Water Resour. Res.49 (11): 7711–7732. https://doi.org/10.1002/2012WR012851.

Ma, T. 1994. Chinese soils (third version). [In Chinese.] Beijing: ChinaAgriculture.

Merrett, S. 1997. Introduction to the economics of water resources: Aninternational perspective. Lanham, MD: Rowman & Littlefield.

Ministry of Water Resources of China. 2005. Vol. 3 of annual hydrologicalreport of China (1964–2005): Hydrological data of Haihe River basin.Weifang: Shandong Hydrographic Printing Co., Ltd.

Montanari, A., et al. 2013. “Panta Rhei—Everything flows: Change inhydrology and society—The IAHS Scientific Decade 2013–2022.”Hydrol. Sci. J. 58 (6): 1256–1275. https://doi.org/10.1080/02626667.2013.809088.

Monteith, J. L. 1973. Principles of environmental physics. London: EdwardArnold.

MWR (Ministry of Water Resources of China). 2005. China waterstatistical yearbook 2005. Beijing: China Water Power.

MWR (Ministry of Water Resources of China). 2006. China water resour-ces bulletin. Beijing: China Water Power.

NASA. 2010. “The MOD16 global MODIS evapotranspiration data pro-ducts.”Accessed September 4, 2018. ftp://ftp.ntsg.umt.edu/pub/MODIS/NTSG_Products/MOD16/.

National Earth System Science Data Sharing Infrastructure. n.d. “Chinaland use/land cover data set.”Accessed September 4, 2018. http://www.geodata.cn.

National Meteorological Information Center of China. 2012. “China’s sur-face climate data daily value data set (V3.0).”Accessed September 4,2018. http://data.cma.cn/data/cdcdetail/dataCode/SURF_CLI_CHN_MUL_DAY_V3.0.html.

Piniewski, M., M. Szcześniak, S. C. Huang, and Z. W. Kundzewicz. 2017.“Projections of runoff in the Vistula and the Odra River basins with thehelp of the SWAT model.” Hydrol. Res. 49 (2): 303–317. https://doi.org/10.2166/nh.2017.280.

Pokhrel, Y. N., S. Koirala, P. Yeh, N. Hanasaki, L. Longuevergne, S. Kanae,and T. Oki. 2015. “Incorporation of groundwater pumping in a globalland surface model with the representation of human impacts.” WaterResour. Res. 51 (1): 78–96. https://doi.org/10.1002/2014WR015602.

Qin, D., C. Lu, J. Liu, H. Wang, J. Wang, H. Li, J. Chu, and G. Chen.2014. “Theoretical framework of dualistic nature-social water cycle.”Chin. Sci. Bull. 59 (8): 810–820. https://doi.org/10.1007/s11434-013-0096-2.

Ren, X. 2007. Haihe river basin water resources evaluation. [In Chinese.]Beijing: China Water Power.

Ridolfi, E. 2014. “Exploring the urban hydro social cycle in tourist envi-ronments.” Investigaciones Geograficas 61: 17–38. https://doi.org/10.14198/INGEO2014.61.02.

Sacks, W. J., B. I. Cook, N. Buenning, S. Levis, and J. H. Helkowski. 2009.“Effects of global irrigation on the near-surface climate.” Clim. Dyn.33 (2–3): 159–175. https://doi.org/10.1007/s00382-008-0445-z.

Schellekens, J., et al. 2017. “A global water resources ensemble of hydro-logical models: The eartH2Observe Tier-1 dataset.” Earth Syst. Sci.Data 9 (2): 389–413. https://doi.org/10.5194/essd-9-389-2017.

Schmidt, J. J. 2014. “Historicising the hydrosocial cycle.” Water Altern.7 (1): 220–234.

Shrestha, S., M. Shrestha, and P. K. Shrestha. 2017. “Evaluation of theSWAT model performance for simulating river discharge in the Hima-layan and tropical basins of Asia.” Hydrol. Res. 49 (3): 846. https://doi.org/10.2166/nh.2017.189.

USGS (US Geological Survey). 1998. “GTOPO30: Global 30 arc-secondsdigital elevation model.”Accessed September 4, 2018. http://edcdaac.usgs.gov/gtopo30/gtopo30.asp.

Van Der Knijff, J. M., J. Younis, and A. P. J. De Roo. 2010. “LISFLOOD: AGIS-based distributed model for river basin scale water balance andflood simulation.” Int. J. Geogr. Inf. Sci. 24 (2): 189–212. https://doi.org/10.1080/13658810802549154.

Voisin, N., M. I. Hejazi, L. R. Leung, L. Liu, M. Huang, H.-Y. Li, andT. Tesfa. 2017. “Effects of spatially distributed sectoral water manage-ment on the redistribution of water resources in an integrated watermodel.” Water Resour. Res. 53 (5): 4253–4270. https://doi.org/10.1002/2016WR019767.

Voisin, N., H. Li, D. Ward, M. Huang, M. Wigmosta, and L. R. Leung.2013. “On an improved sub-regional water resources managementrepresentation for integration into earth system models.” Hydrol. EarthSyst. Sci. 17 (9): 3605–3622. https://doi.org/10.5194/hess-17-3605-2013.

Wada, Y., D. Wisser, and M. F. P. Bierkens. 2014. “Global modeling ofwithdrawal, allocation and consumptive use of surface water andgroundwater resources.” Earth Syst. Dyn. 5 (1): 15–40. https://doi.org/10.5194/esd-5-15-2014.

Wang, H., Y. Jia, G. Yang, Z. Zhou, Y. Qiu, C. Niu, and H. Peng. 2013.“Integrated simulation of the dualistic water cycle and its associatedprocesses in the Haihe River Basin.” Chin. Sci. Bull. 58 (27):3297–3311. https://doi.org/10.1007/s11434-012-5371-0.

Wang, H., J. Wang, D. Qin, and Y. Jia. 2006. “Theory and methodologyof water resources assessment based on dualistic water cycle model.”[In Chinese.] ShuiliXuebao 37 (12): 1496–1502.

Wang, H., and G. Yang. 2010. “Preliminary study on new concept ofwater resources management under dualistic water cycle condition.”[In Chinese.] Chin. J. Nat. 32 (3): 130–133.

Wigmosta, M. S., L. W. Vail, and D. P. Lettenmaier. 1994. “A distributedhydrology-vegetation model for complex terrain.” Water Resour. Res.30 (6): 1665–1679. https://doi.org/10.1029/94WR00436.

Yang, D., S. Herath, and K. Musiake. 1998. “Development of ageomorphology-based hydrological model for large catchments.” Annu.J. Hydraul. Eng. JSCE 42: 169–174. https://doi.org/10.2208/prohe.42.169.

Yang, D., S. Herath, and K. Musiake. 2002. “A hill slope-based hydrolog-ical model using catchment area and width functions.” Hydrol. Sci. J.47 (1): 49–65. https://doi.org/10.1080/02626660209492907.

© ASCE 04018077-16 J. Water Resour. Plann. Manage.

J. Water Resour. Plann. Manage., 2018, 144(12): 04018077

Dow

nloa

ded

from

asc

elib

rary

.org

by

Nor

thw

est A

&F

Uni

vers

ity o

n 09

/28/

18. C

opyr

ight

ASC

E. F

or p

erso

nal u

se o

nly;

all

righ

ts r

eser

ved.