Research Article Evolution of Hydrological Drought in Human …downloads.hindawi.com/journals/amete/2016/5102568.pdf · 2019. 7. 30. · Research Article Evolution of Hydrological

Research ArticleEvolution of Hydrological Drought in Human Disturbed AreasA Case Study in the Laohahe Catchment Northern China

Yi Liu12 Liliang Ren12 Ye Zhu12 Xiaoli Yang2 Fei Yuan2

Shanhu Jiang2 and Mingwei Ma2

1State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering Hohai University Nanjing 210098 China2College of Hydrology and Water Resources Hohai University Nanjing 210098 China

Correspondence should be addressed to Liliang Ren njrll9999126com

Received 29 May 2015 Accepted 25 November 2015

Academic Editor Maurits W Ertsen

Copyright copy 2016 Yi Liu et al This is an open access article distributed under the Creative Commons Attribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

A case study on the evolution of hydrological drought in nonstationary environments is conducted over the Laohahe catchment innorthern China Using hydrometeorological observations during 1964ndash2009 meteorological and hydrological droughts are firstlyanalyzed with the threshold level method Then a comprehensive analysis on the changes within the catchment is conducted onthe basis of hydrological variables and socioeconomic indices and the whole period is divided into two parts the undisturbedperiod (1964ndash1979) and the disturbed period (1980ndash2009) A separating framework is further introduced to distinguish droughtsinduced by different causes that is the naturalized drought and human-induced drought Results showed that human activitiesare more inclined to play a negative role in aggravating droughts Drought duration and deficit volume in naturalized conditionsare amplified two to four times and three to eight times respectively when human activities are involved For the two dry decades1980s and 2000s human activities have caused several consecutive drought events with rather long durations (up to 29 months)These results reflect the considerable impacts of human activities on hydrological drought which could provide some theoreticalsupport for local drought mitigation and water resources management

1 Introduction

Drought is a recurring natural phenomenon primarilyinduced by successive reduction of precipitation over a periodof time [1ndash4] Different from aridity which is a permanentfeature of climate condition drought is a consequence ofclimate variability reflecting temporal water deficiencies andcan be observed in many regions rather than constrained todry climate zones [5] In view of the widespread impacts ofdroughts including economic agricultural hydrological andecological losses and costsmuch concern is given to droughtsin recent years Climate changes associatedwith global warm-ing and other climate extremes have aggravated the damageof droughts even in some areas such as southern Europe andWest Africa droughts with higher peaks and severity levelshave been frequently observed during recent decades [6]

Drought can be classified into four types according to thevariable that is used that is meteorological (precipitation)agricultural (soil moisture) hydrological (streamflow water

level reservoir storage and groundwater discharge) andsocioeconomic droughts [7] Among them the hydrologicaldrought is mostly related to our social activities whichcan be defined as inadequate surface and subsurface waterresources for establishedwater uses of a givenwater resourcesmanagement system [4]

The development scheme of hydrological drought israther complex and is subject to the effects of climate andcatchment control or a combination of the two factorsAmong various climatic variables precipitation and tem-perature are two key variables which largely determine theweather pattern and antecedent condition for the occurrenceof hydrological droughtMeanwhile since a sustained changein precipitation or temperature is commonly induced by vari-ation in large-scale circulations analyzing the relationshipbetween hydrological drought and atmospheric circulationindices (eg the El Nino Southern Oscillation (ENSO) andhigh-pressure areas) is also a hot issue in recent droughtresearches [8] Catchment control also plays an important

Hindawi Publishing CorporationAdvances in MeteorologyVolume 2016 Article ID 5102568 12 pageshttpdxdoiorg10115520165102568

2 Advances in Meteorology

role in drought propagation and influences processes likepooling attenuation lag and lengthening [9] Some studiesfurther investigated the individual roles of climate andcatchment control in governing hydrological drought char-acteristics and the conclusion is highly dependent on spatialscales Globally hydrological drought duration and deficitvolume aremore related to climate than to catchment controlHowever for regional scale where climate is assumed tobe relatively uniform catchment characteristics like geologyarea slope and groundwater system play an important role ingoverning hydrological drought duration and deficit volume[10]

In fact the impacts on hydrological drought are notlimited to the above-mentioned natural factors (ie climaticand catchment factors) and influences from human activitiesin forms of land cover change reservoir regulation agricul-tural irrigation and water withdrawal from streams or riverchannels should not be ignored From the perspective ofhydrology science human activities influence the process ofhydrological cycle for example infiltration and evapotran-spiration processes and further alter flow regimes and theirspatiotemporal distributions [11] Accordingly the processof hydrological drought would also be influenced Withthis in mind the traditional understanding of hydrologicaldroughts should never be limited to the naturalized drysituation drought caused by human activities has become anew challenging issue that needs to be noticed and discussed

The focus of our study is to investigate the impact ofhuman activities on hydrological drought In this paperwe call drought induced by climate variability the ldquonatu-ralized droughtrdquo and drought caused by human activitiesthe ldquohuman-induced droughtrdquo As aforementioned humanactivities potentially influence underlying surfaces whichfurther leads to a nonstationary response in hydrologicalvariables (eg runoff) Here comes the question will thesechanges influence the characteristics of hydrological droughtsand what is the difference between naturalized and human-induced droughts Based on the above questions a com-prehensive investigation on the evolution of hydrologicaldroughts is conducted in the Laohahe catchment in northernChina (this catchment is selected due to its representativerunoff change pattern influenced by combined effects ofclimate change and human activities which is quite commonin China water-stressed regions) combined with a separatingframework for distinguishing the naturalized and human-induced droughts

2 Study Area and Data

The Laohahe catchment (41∘Nsim4275∘N 11725∘Esim120∘E)located in northern China is selected as an example con-sidering its drastic runoff change pattern which is a com-mon and typical phenomenon for water-stressed regionsin northern China The drainage area of the Laohahecatchment is 18112 km2 and its elevation ranges between427m and 2054m with a general increasing trend fromnortheast to southwest The climate belongs to the semiaridzone according to the observation during the period 1964ndash2009 mean annual temperature precipitation and runoff

2010 0 20 40 60

(km)

N

DEM (m)

Hig

h2054

Low

427

42∘30

998400N

42∘0998400N

41∘30

998400N

41∘0998400N

118∘30

998400E 119∘30

998400E117∘30

998400E 118∘0998400E 119

∘0998400E

Chifeng cityStreamflow stationMeteorological stationRain gauge

ReservoirRiverBasin boundary

Figure 1 Location of the study area and the geographic distributionof hydrometeorological stations and reservoirs

are 758∘C 4183mm and 287mm respectively Due to thetemporally unevenly distributed precipitation (80 of annualprecipitation concentrated between May and September)and temperature the Laohahe catchment presents a strongseasonality in runoff Normally winter is cold and dry withrelatively low streamflow observed while in summer it is hotand wet with most peak flow occurring during this season

The data used in this study includes materials for hydro-logical drought identification and input forcing for hydro-logical modeling Daily precipitation of 52 rain gauges andstreamflow records of the Xinglongpo hydrological stationsituated at the outlet of the Laohahe basin during 1964ndash2009 is provided by the Water Resources Department ofthe Inner Mongolia Autonomous Region Streamflow datais further converted to catchment runoff by averaging therunoff amounts over the catchment area Daily meteoro-logical forcing (1964ndash2009) including maximum and min-imum air temperature wind speed relative humidity andsunshine duration of 3 national standard meteorologicalstations in and around the Laohahe catchment is downloadedfrom the China Meteorological Data Sharing Service System(httpdatacmagovcn) Their geographic distributions areshown in Figure 1 Besides locations of three large reservoirs(Erdaohezi Dahushi and Sanzuodian) within the Laohahecatchment are also given in Figure 1 considering their poten-tial impacts on runoff

Geographic information is obtained as follows 3-arcsecond (about 90m) digital elevation from the shuttleradar topography mission (SRTM) digital elevation modelsoil types from the 5-minute FAO dataset [12] and land

Advances in Meteorology 3

d1 d2 d3 d4 d5

1 2 3 4 5i1 i2 i3

i4

t1 t2 t3 t4 t9

Qz

Stre

amflo

w (m

3s

)

Figure 2 Illustration of threshold level method time of occurrence119905

119894

duration 119889119894

deficit volume or severity V119894

drought interval 119894119896

andthe threshold level 119876

119911

cover data provided by the Chinese Academy of ScienceWe also collect socioeconomic statistics of the Chifengcity from the local statistical bureau including populationagricultural production gross industrial product (GIP) andgross domestic product (GDP) In addition the baseflowand baseflow index (BFI) are calculated by the HydrologicalUtility Package (httpwwwcnhupcom) using the recursivedigital filtering method (RDF [13]) BFI itself is not acatchment characteristic but it integrates storage and releaseproperties of a catchment [14] and has been shown to havea close relationship with hydrological drought duration [1516] Therefore we use this index to reflect varied catchmentcharacteristics

3 Methodology

31 Hydrological Drought Analysis The threshold levelmethod is so far the most commonly used approach in viewof recent hydrological drought researches Originating fromthe statistical theory of runs introduced by [17] this methodis commonly used to analyze a sequential time series with atime resolution of one month or longer Figure 2 gives thegeneral illustration of this method A sequence of droughtevents can be derived from a streamflow hydrograph whenthe flow falls below a certain threshold level119876

119911 Accordingly

drought characteristics including the time of occurrence 119905119896

drought duration119889119896 deficit volume V

119896 and drought interval

119894

119896(the time period between two consecutive drought events)

can be obtainedThe selection of threshold value 119876

119911is subjective but

essential since it influences the number of events droughtduration and deficit volume For perennial streams likeour studied catchment threshold levels between the 70-percentile flow and the 95-percentile flow from the flowduration curve (FDC) are recommended [18] In this studythe 70-percentile flow is used to identify meteorologicaland hydrological droughts fromprecipitation and streamflowseries respectively

32 The Framework for Separating Naturalized and Human-Induced Drought Distinguishing droughts induced by dif-ferent factors is essential for objective understanding of theunderlying causes of varied regional drought which alsomakes sense for water planning and management In this

section a framework developed by van Loon and van Lanen[19] is applied to separate naturalized and human-induceddroughts Part of this framework is similar to the commonmethod used in hydrology of assessing the impacts ofclimate change and human activities on hydrological systemwhile the remaining part focuses on quantitative analysis onhydrological drought

Specifically this framework can be divided into threesections In the first step the time series of observed hydrom-eteorological variables are tested so as to find possible changepoints The whole period can then be divided into two partsby the change point that is the baseline period (ldquoundis-turbedrdquo) and changed period (ldquodisturbedrdquo) For change pointdetection a number of methods are available such as thePettitt test [20] Kendall test [21] and the precipitation-runoff double cumulative curves methodThe second sectionfocuses on hydrological model simulation and reconstruct-ing runoff series of the changed period For this purposethe hydrological model should be firstly calibrated usingmeteorological forcing of the baseline period Then keepingoptimized parameters unchanged the simulation driven bymeteorological forcing of the changed period is the recon-structed (simulated) series for which no human disturbancesare involved For hydrological model simulation varioushydrological models can be chosen as long as they are capableof reproducing the natural situation especially during lowflow and drought The Variable Infiltration Capacity (VIC)model [22] is adopted in this study and its specific descriptionis given in the following section The third step is thecore of this framework Hydrological droughts induced bydifferent causes during the changed period can be derivedby identifying observed and simulated runoff series with thethreshold level method respectively The former representsa combination of naturalized and human-induced droughtswhile the latter refers to droughts that develop in naturalconditions Therefore their difference is the human-induceddroughts The threshold values are calculated based on theundisturbed period and applied to the entire time seriesMeanwhile to reduce the impact of modeling errors ondrought separation different threshold values are calculatedfrom observed and simulated runoff series respectively

33 A Brief Description of the VIC Model The semidis-tributed three-layer Variable Infiltration Capacity (VIC)model [22] is chosen for hydrological modeling It has beenused in many drought related researches such as recon-structing and analyzing drought events [23ndash25] and droughtprediction [26] The VIC model considers the dynamicvariation in both water and energy balance according to thesubgrid heterogeneity represented by soil moisture storageevaporation and runoff generation Partitioning of rainfallinto infiltration and surface runoff is controlled by a VariableInfiltration Capacity curve [27] Vertical water movementoccurs within discrete soil layers through diffusion andbaseflow is modeled from a lower soil moisture zone asa nonlinear recession [28] A separate routing model isemployed to transport grid cell surface runoff and baseflowto the outlet then into the river system including a lineartransfer function model used to calculate the within-cell


Table 1 Physical meanings and ranges of the sensitive parameters in the VIC model

Parameter Physical meaning Unit Range119894 Infiltration curve parameter NA 0-1119889

1

Thickness of top thin soil moisture layer m 005ndash01119889

2

Thickness of middle soil moisture layer m 0ndash2119889

3

Thickness of lower soil moisture layer m 0ndash2119863

119904

Fraction of119863119904max where nonlinear baseflow begins Fraction 0-1

119863

119904max Maximum velocity of baseflow mmday 0ndash30119882

119904

Fraction of maximum soil moisture where nonlinear baseflow occurs Fraction 0-1

routing and the linearized Saint-Venant equation serving forchannel routing [29]

The description of land surface characteristics for eachgrid cell is mainly implemented through numerous param-eters of two categories vegetation and soil parameters Vege-tation parameters comprise vegetation leaf area index rough-ness length displacement height architectural resistance andminimum stomatal resistance with relevant information thatcan be obtained from the University of Marylandrsquos (UMD)land cover classification [30] and estimated according to theLand Data Assimilation Systems developed by the NationalAeronautics and Space Administration (NASA) As for soilparameters some can be determined from empirical valuesand need not be adjusted such as porosity saturated soilpotential saturated hydraulic conductivity and the exponent119861 for unsaturated flow (available from the 5-minute Foodand Agriculture Organization data set [12]) while otherparameters are subject to calibration based on the agreementbetween simulated and observed hydrographs Commonlycalibration is conducted by optimizing the seven sensitiveparameters (119894 119889

1 1198892 1198893 119863119904 119863119904max and 119882119904) with two

criteria Nash-Sutcliffe Coefficient of Efficiency (NSCE) andBIAS The specific meanings and ranges of these sensitiveparameters are listed in Table 1

4 Results and Discussion

41 Evolution of Meteorological and Hydrological DroughtsMeteorological drought extracted from monthly precipita-tion series shows that there is no significant variation trendduring the past 46 years (Figure 3(a)) The drought intervalis generally stable with most of drought events occurringin the dry season In contrast hydrological drought (Fig-ure 3(b)) derived from monthly runoff series shows differentpattern with obvious decadal fluctuations observed Amongthe five decades the 1980s (1980ndash1989) and 2000s (2000ndash2009) are the two most severe dry decades which sufferconsecutive droughts with large deficit volume whereas forthe 1990s (1990ndash1999) almost no hydrological droughts occurA comparison of drought characteristics between meteo-rological and hydrological droughts further demonstratestheir difference Figures 3(c) and 3(d) are the scatterplotsof drought duration and deficit volume Obviously meteo-rological droughts occur every decade but for hydrologicaldroughts they are mostly concentrated in the 1980s and2000s Comparing to meteorological drought the duration

of hydrological drought is lengthened (up to 17 months)and deficit volume is attenuated (not more than 6mm)Meanwhile the relationship between drought duration anddeficit volume is modified and tends to be linear for thehydrological drought

The above-mentioned differences between the twodrought categories on the one hand reflect a propagationscheme from meteorological drought to hydrologicaldrought Due to the effects of climate and catchmentcontrol meteorological droughts are combined into a longerhydrological drought Meanwhile the drought signal isweakened in catchment stores during this process [31] Onthe other hand the striking contrast among decades interms of hydrological drought seems quite abnormal andis far beyond the scope of naturalized drought propagationprocess To further explore the reasons for the unusualbehavior of hydrological drought during the 1990s theflow duration curves of precipitation and runoff for eachdecade are compared shown in Figure 4 For precipitationminor decadal difference of the flow duration curves isfound especially in the low flow section and using the70-percentile threshold of whole periods would ensure 25ndash35 of droughts (accordingly 65ndash75 of nondrought)recognized for different decades However for runoff ratherlarge decadal difference is observed Using the 70-percentilethreshold of whole periods would recognize approximately50ndash60 of droughts for the 1980s and 2000s but no morethan 10 for the 1990s This shows the inconsistent behaviorbetween meteorological and hydrological droughts Fordrought propagation catchment control plays an importantrole in determining process features like duration and deficitThe discrepant performances between these two droughtcategories suggest that the effect of catchment control isnot constant but changes over time Besides the naturalhydrology system hydrological drought at this catchmentscale is potentially disturbed by human activities

42 Analysis on the Changing Environments To furtherexplore the causes of the abnormal pattern in hydrologi-cal drought a comprehensive analysis on drought relatedvariables (including precipitation runoff runoff coefficientbaseflow and BFI) is conducted Among them precipitationprovides the antecedent condition for the occurrence ofhydrological drought and governs drought deficit volumeRunoff coefficient represents the proportion of runoff inprecipitation and can be used to reflect temporally varied


1965

1969

1973

1977

1981

1985

1989

1993

1997

2001

2005

2009

PrecipitationDeficit volume

minus2

0

2

4

6

8

Prec

ipita

tion

(mm

)

(a) Meteorological drought

RunoffDeficit volume

2001

2005

2009

1981

1985

1989

1993

1997

1973

1969

1977

1965

00

02

04

06

08

10

Runo

ff (m

m)

(b) Hydrological drought

1960s1970s1980s

1990s2000s

1 2 3 4 5 6 70

Duration (month)

0

minus2

minus4

minus6

minus8

minus10

minus12

minus14

minus16

Defi

cit v

olum

e (m

m)

(c) Meteorological drought

1960s1970s1980s

1990s2000s

2 4 6 80 12 14 16 1810

Duration (month)

0

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Defi

cit v

olum

e (m

m)

(d) Hydrological drought

Figure 3 Temporal evolutions of identified (a) meteorological drought (derived from precipitation) and (b) hydrological drought (derivedfrom runoff) and the scatterplots of drought duration and deficit volume for (c) meteorological and (d) hydrological droughts

80 857565 70 90 9560 100

Frequency ()

0

2

4

6

8

10

Mon

thly

pre

cipi

tatio

n (m

m)

0 8020 40 60 1000

50

100

150

200

250

300

Whole period1960s1970s

1980s1990s2000s

(a)

0

1

2

3

4

Mon

thly

runo

ff (m

m)

1006030 40 50 70 908020100

Frequency ()

1006040 802000

10

20

30

40

50

60


1980s1990s2000s

(b)

Figure 4 Monthly flow duration curves of precipitation (a) and runoff (b) for each decade Dull yellow and blue lines represent curves fordry and wet decades respectively


Table 2 Mean annual precipitation runoff runoff coefficient baseflow and BFI of every decadal-year over the Laohahe catchment duringthe past half-century

Precipitation (mm) Streamflow (mm) Runoff coefficient Baseflow (mm) BFI1964ndash1969 41460 3834 009 1098 0291970ndash1979 45501 3587 008 1006 0291980ndash1989 39631 1595 004 489 0351990ndash1999 47338 4298 009 1292 0322000ndash2009 37906 973 003 370 044

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)

PrecipitationLine trend of hydrologic variables

(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent

Line trend of hydrologic variablesRunoffRunoff coefficientLine trend of runoff coefficient and BFI

(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)

Line trend of hydrologic variablesBaseflowBFILine trend of runoff coefficient and BFI

(c)

Figure 5 Variations of annual precipitation runoff runoff coefficient baseflow and baseflow index (BFI) over the Laohahe catchment since1964 The red dashed line represents linear trend for hydrological variables (precipitation runoff and baseflow) and the blue dashed line forrunoff coefficient and BFI

rainfall-runoff relationshipsThe BFI itself is not a catchmentcharacteristic but it integrates the effect of catchment storageand release properties and is found to have strong positivecorrelation with hydrological drought duration

As shown in Figure 5 and Table 2 precipitation (Fig-ure 5(a)) presents a slightly decreasing trend during the past50 years accompaniedwith regularly decadal fluctuation (the1980s and 2000s are the two driest decades whereas the1990s is the wettest decade) Runoff (Figure 5(b)) also showsa descending trend but in a more rapid rate This incon-sistent pattern between precipitation and runoff is furtherillustrated by the varied runoff coefficient Normally runoffcoefficient at this catchment scale varies around 009 but inthe two driest decades 1980s and 2000s a rather low value(no more than 005) is observed implying the significantly

reduced proportion of runoff in precipitation Meanwhilerunoff components (Figure 5(c)) have also changed and theproportion of baseflow generally increases with an upwardtrend detected in BFI This indicates that catchment controllike storage and release properties has changed over timewhich indirectly influences the development of hydrologicaldrought

Based on the above analysis we further investigate therunoff changes (ie varied response to precipitation andrunoff components) with the double cumulative curvesmethod Figure 6(a) is the double cumulative curve ofprecipitation and runoff From this graph we can find twoprominent inflection points where the gradient for the cumu-lative curve of runoff significantly changes compared withthat of precipitation that is 1979 and 1998 This deviation is


RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)

Figure 6 Double cumulative curves of (a) annual precipitation and runoff and (b) annual runoff and baseflow

quite related to human activities which lead to sharp declinein runoff [32] In response to the 1978 land reform policymost regions in northern China (grain production base)switched their focus to the development of agriculture andirrigation has become a routine agronomic practice [33] TheLaohahe catchment is one of these cases as shown in Figure 7indices like agricultural production (Figure 7(a)) and grossdomestic product (GDP Figure 7(d)) have increased in a stag-gering speed since 1979 The price paid for the fast-growingagriculture is that more water is consumed and evaporated tomeet irrigation demands and livelihood purpose (indicatedby a continuous growth in population)The second inflectionpoint which emerged in 1998 implies that reduction in runoffhas been aggravated since then Figure 6(b) shows that after1998 runoff decreases more rapidly than baseflow meaningthat consumption from surface water has further increasedDuring this period the gross industrial product (GIP) hasexperienced fast growth (Figure 7(c)) which tremendouslycontributes to water depletion for the sake of second andthird industry development and industrial structure adjust-ment [11] In addition increased reservoir storage capacitiesduring the 2000s might be another reason for surface waterreduction Before 2000s the Laohahe catchment is mainlyregulated by two large reservoirs namely the Erdaohezi andDahushi with a total regulation capacity of 20 times 108m3(a sum of the two reservoirsrsquo storage capacity) In 2003this regulation capacity has been increased to 569 times 108m3with the construction of the San Zuodian reservoir whichpotentially brings some pressure to the surface runoff

As a matter of fact human activities carried out atthis catchment are not limited to the above-mentionedaspects Other forms such as abstraction from groundwaterland cover and land use change and livestock breedingall influence runoff patterns in a direct or indirect wayAccordingly the naturalized process of hydrological droughtis disturbed and develops in a more complicated mode

From the perspective of water resources management ourtraditional drought mitigation solutions should be adjustedso as to adapt to droughts driven by different causes Thisonce again shows the significance of making the distinctionbetween naturalized drought and human-induced drought

43 Separating Naturalized and Human-Induced HydrologicalDrought According to the above analysis on the changes ofcatchment characteristics the temporal span is divided intotwo parts that is the baseline (undisturbed) period from1964 to 1979 duringwhich catchment conditions are relativelynatural and minor human activities are negligible and thechanged (disturbed) period from 1980 to 2010 Following theseparating framework described in Section 32 we first cali-brate the hydrological VICmodel using hydrometeorologicalforcing during the baseline period Specifically the VICmodel is run at a daily temporal and 00625∘times 00625∘ spatialresolution Calibration contains a relative long time 1964ndash1974 which aims to ensure an appropriate sample size withboth wet and dry years involved and further improves therepresentativeness of parameters Accordingly the remainingfive years (1975ndash1979) belong to the validation period Fig-ure 8(a) is the monthly simulation results during the baselineperiod Overall a satisfying model performance is found forboth calibration and validation periods Values of NSCE andBIAS are 085 and 42 for the calibration period and 080and 18 for the validation period respectively suggestingthat VIC is capable of capturing natural variability of theobserved hydrograph (eg water amounts peak magnitudeand recession)

With the calibrated model we reconstructed the stream-flow series using hydrometeorological forcing during thechanged period (1980ndash2009) As shown in Figure 8(b) thedifference between observed and simulated (reconstructed)series mainly reflects the anthropogenic effects on runoffcombined with minor simulation errors


Table 3 Drought characteristics (using the 70-percentile monthly threshold) for the observed and simulated runoff during 1980ndash2009

Changed period Number of droughts Duration (month) Deficit (mm)Mean Max Mean Max

Observed 27 75 29 minus312 minus1547

Simulated 32 4 6 minus112 minus186

Baseline periodChanged period

1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

(a) Agricultural production

Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)

(c) Gross industrial product (GIP)


1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)

(d) Gross domestic product (GDP)

Figure 7 Time series of (a) agricultural products (b) population (c) gross industrial product (GIP) and (d) gross domestic product (GDP)in Chifeng the main city within the Laohahe catchment from 1949 to 2005

Figure 9 presents the evolution of hydrological droughtsseparately identified from observed and simulated stream-flow series Obviously the upper panel (Figure 9(a)) rep-resenting droughts caused by combined effects of naturalcondition and human disturbances shows a more severedry pattern than the naturalized condition displayed inthe middle panel (Figure 9(b)) Table 3 lists their specific

differences in statistics of drought characteristics Althoughin the naturalized circumstance five more drought events areobserved drought duration and deficit volume are amplifiedtwo to four times and three to eight times respectively whenhuman disturbances are involved This means that the sever-ity of human-induced drought far exceeds that of droughts innaturalized situations Figure 9(c) is the temporal difference


NSCE = 083BIAS = 25 NSCE = 080

BIAS = 18

Observed runoffSimulated runoffPrecipitation

1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period

Figure 8 VIC simulated monthly runoff at the Xinglongpo hydrologic station for (a) the baseline period 1964ndash1979 and (b) changed period1980ndash2009

Observed runoffDeficit volume of observed runoff

minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)

Simulation runoffDeficit volume of simulated runoff

1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)

Drought aggravationDrought mitigation

minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)

Figure 9 Identified hydrological droughts (with 70-percentile threshold derived from the baseline period) during 1980ndash2009 from (a)observed runoff and (b) simulated runoff series and (c) is the difference between observed and simulated runoff series representing thenet effect of human activities on drought

between observed and simulated droughts which reflects thenet impact of human activities The positive values (lightgrey area) represent a negative role of aggravating droughtwhereas the negative values (light blue area) imply a positiveeffect of drought mitigation We can find that during thewhole changed period human activities are more inclined toplay a negative role In particular in the two dry decades 1980sand 2000s human activities have induced several consecutivedrought events with rather long durations such as droughtsfrom November 1980 to April 1982 (18 months) and fromAugust 2007 to December 2009 (29 months) In contrast theinfluence is rather moderate during the wet decade 1990seven in 1994 and 1995 a positive effect is observed Theabnormal human activity behaviors in these two years maybe related to the interannual regulation roles of reservoirs

Figure 10 shows the monthly variations of precipitation andrunoff during the past 46 years (only low flow seasons aredisplayed from the recession month (September) in this yearto May of next year) We can see that precipitation in 1994and 1995 is generally the lowest however correspondingrunoff is extremely high We speculate that the drainagefrom reservoirs could be one contributing factor Due to lackof reservoir records this unusual phenomenon needs to befurther analyzed in the future

In addition since the whole separation framework ofhydrological droughts largely depends on hydrologicalmodelsimulations accurate model performance in capturing thecharacteristics of hydrograph is essential for the separationresults between naturalized and human-induced droughtsAsmany other hydrological simulations minor biases (errors


1994

1995

JanSep FebOct AprDec MarNov MayMonth

0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff

Figure 10 Monthly variations of (a) precipitation and (b) runoff during the past 46 years The red and blue lines represent variables in 1994and 1995 respectively and the grey lines represent variables in other 44 years

not more than 25 equivalent to approximately 09mmin streamflow) in produced water amounts are observedduring the baseline period with overestimations in peakstreamflow contributing to most of the errors (Figure 8)Since droughts recognized by the fixed threshold are morelikely to happen in low flow seasons this systematic positivedeviation in flood peak would not influence our separatedresults in any significant way as long as the errors for thelow flow section are effectively controlled In a more directway we compared the differences between observations andhydrological model simulations identified streamflow deficitvolume during the baseline period to analyze the propagatedbias in droughts Results show that among the 73 monthswhich experience hydrological droughts during the baselineperiod the difference between the two datasets on average is001mmmonth withmaximumabsolute value nomore than005mmmonth This magnitude in general is rather smallcompared with the streamflow deficit volume of human-induced droughts where an order of 02ndash08mmmonth isfound for most of the drought events (Figure 9(c)) In otherwords minor errors derived frommodel simulations thoughexist they will not influence the dominant contribution ofhuman activities substantially especially for the two driestdecades 1980s and 2000s According to the above findingswe could conclude that with the participation of humanactivities hydrological drought in the Laohahe catchment hasbeen further aggravated and become more severe

5 Summary and Conclusions

In this paper a case study over the Laohahe catchmentis conducted which serves as a representative example ofthe evolution of hydrological drought in the context ofchanging environments The common understanding of thehydrological drought development is mainly governed by

climate and catchment control However for catchments likeour studied one which is intensively disturbed by humanactivities the causes for hydrological drought become rathercomplicated and the effects from human activities should beconsidered

The comparison between meteorological droughts andhydrological droughts occurring in the Laohahe catchmentduring 1964ndash2009 shows an interesting pattern that ismeteorological droughts occur regularly in every decadebut for hydrological droughts significant decadal differencesare observed the 1980s and 2000s suffer extremely severehydrological droughts whereas for the 1990s almost nodroughts occurThis striking contrast is far beyond the scopeof naturalized drought process dominated by climate andcatchment control indicating the potential impact of humandisturbances on hydrological drought

The long-term series of runoff related variables are furtheranalyzed and changes in runoff are confirmed according toseveral indices that is inconsistent trend between precipi-tation and runoff varied runoff coefficient and BFI amongdecades which commonly reflect the varied responses ofrunoff to precipitation and runoff components Then thedouble cumulative curves method is employed to detect thechanged points which further divides the whole period intotwo parts that is the undisturbed period (1964ndash1979) anddisturbed period (1980ndash2009) Meanwhile four socioeco-nomic indices are also analyzed which imply increased waterconsumption in the changed period

Based on the divided periods a separating framework isintroduced to distinguish between naturalized drought andhuman-induced drought during the changed period Fromthe comparison between their drought characteristics it canbe found that the drought duration and deficit volume of nat-uralized drought are amplified two to four times and three toeight times respectively when human activities are involved


This reflects the considerable impact of human activities onhydrological drought Generally human activities are moreinclined to play a negative role which aggravates droughtsIn particular in the two dry decades 1980s and 2000shuman activities have induced several consecutive droughtevents with rather long durations (up to 29 months) Witha comprehensive analysis on the individual roles of naturalcondition and human activities on hydrological drought thisstudy is promising to provide some theoretical support forfuture drought mitigation and water resources management

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This work was supported by the Special Basic Research Fundfor Methodology in Hydrology (Grant no 2011IM011000)from the Ministry of Sciences and Technology China the111 Project (Grant no B08048) from the Ministry of Edu-cation and State Administration of Foreign Experts AffairsChina the China Scholarship Council (CSC) the NationalKey Technology RampD Program by Ministry of Sciencesand Technology China (2013BAC10B02) the FundamentalResearch Funds for the Central Universities (2015B14514)the National Natural Science Foundation of China (nos51579066 41501017 and 41201031) the project sponsoredby SRF for ROCS SEM (515025512) and Natural ScienceFoundation of Jiangsu Province (BK20150815)

References

[1] D A Wilhite ldquoDrought as a natural hazard concepts anddefinitionsrdquo inDrought AGlobal Assessment D AWilhite Edvol 1 pp 1ndash18 Routledge New York NY USA 2000

[2] L M Tallaksen and H A J van LanenHydrological DroughtmdashProcesses and Estimation Methods for Streamflow and Ground-water vol 48 of Developments in Water Sciences ElsevierAmsterdam The Netherlands 2004

[3] A K Mishra and V P Singh ldquoA review of drought conceptsrdquoJournal of Hydrology vol 391 no 1-2 pp 202ndash216 2010

[4] A Dai ldquoIncreasing drought under global warming in observa-tions and modelsrdquo Nature Climate Change vol 3 pp 52ndash582010

[5] K Stahl Hydrological droughtmdasha study across Europe [PhDthesis] Albert-Ludwigs-Universitat Freiburg FreiburgerSchriften zur Hydrologie Freiburg Germany no 15 2001httpwwwfreidokuni-freiburgdevolltexte202

[6] S I Seneviratne N Nicholls D Easterling et al ldquoChangesin climate extremes and their impacts on the natural physicalenvironmentrdquo in A Special Report of Working Groups I and II ofthe Intergovernmental Panel on Climate Change (IPCC) chapter3 pp 109ndash230 Cambridge University Press Cambridge UK2012

[7] D A Wilhite and M H Glantz ldquoUnderstanding the droughtphenomenon the role of definitionsrdquo Water International vol10 no 3 pp 111ndash120 1985

[8] D G Kingston A K Fleig L M Tallaksen and DM HannahldquoNorth Atlantic sea surface temperature atmospheric circula-tion and summer drought inGreat Britainrdquo in Proceedings of the6thWorld FRIENDConference ldquoGlobal Change Facing Risks andThreats to Water Resourcesrdquo E Servat S Demuth A Dezetteret al Eds vol 340 pp 598ndash604 IAHS-AISH Publication FezMorocco October 2010

[9] H A J van Lanen L Kasparek O Novicky et al HydrologicalDroughtmdashProcesses and EstimationMethods for Streamflow andGroundwater vol 48 of Developments in Water Science BVElsevier Science 2004

[10] E Peters H A J van Lanen P J J F Torfs and G BierldquoDrought in groundwatermdashdrought distribution and perfor-mance indicatorsrdquo Journal of Hydrology vol 306 no 1ndash4 pp302ndash317 2005

[11] L Ren F Yuan B Yong et al ldquoWhere does blue water go inthe semi-arid area of northern China under changing environ-mentsrdquo in Evolving Water Resources Systems UnderstandingPredicting andManagingWater-Society Interactions Proceedingsof ICWRS 2014 Bologna Italy vol 364 IAHS Publications2014

[12] R G Allen L S Pereira D Raes and M Smith ldquoCrop evap-otranspiration guidelines for computing crop requirementsrdquoIrrigation and Drainage Paper 56 FAO Rome Italy 1998

[13] V D Lyne and M Hollick ldquoStochastic time-variable rainfallrunoff modellingrdquo in Proceedings of the Hydrology and WaterResources Symposium vol 32 pp 89ndash92 Institution of Engi-neers Perth Australia September 1979

[14] J L Salinas G Laaha M Rogger et al ldquoComparative assess-ment of predictions in ungauged basins-part 2 flood and lowflow studiesrdquoHydrology and Earth System Sciences vol 17 no 7pp 2637ndash2652 2013

[15] M D Zaidman H G Rees and A R Young ldquoSpatio-temporaldevelopment of streamflow droughts in northwest EuroperdquoHydrology and Earth System Sciences vol 6 no 4 pp 733ndash7512002

[16] M Fendekova and M Fendek ldquoGroundwater drought in thenitra river basinmdashidentification and classificationrdquo Journal ofHydrology and Hydromechanics vol 60 no 3 pp 185ndash193 2012

[17] V Yevjevich ldquoAn objective approach to definitions and investi-gations of continental hydrologic droughtsrdquo Hydrology Papers23 Colorado State University Fort Collins Colo USA 1967

[18] A K Fleig LM Tallaksen H Hisdal and S Demuth ldquoA globalevaluation of streamflow drought characteristicsrdquo Hydrologyand Earth System Sciences vol 10 no 4 pp 535ndash552 2006

[19] A F van Loon and H A J van Lanen ldquoMaking the distinctionbetween water scarcity and drought using an observation-modeling frameworkrdquoWater Resources Research vol 49 no 3pp 1483ndash1502 2013

[20] A N Pettitt ldquoA non-parametric approach to the change-pointproblemrdquo Journal of the Royal Statistical Society Series C AppliedStatistics vol 28 no 2 pp 126ndash135 1979

[21] M G Kendall Rank Correlation Methods Charles GriffinLondon UK 1975

[22] X Liang D P Lettenmaier E F Wood and S J Burges ldquoAsimple hydrologically based model of land surface water andenergy fluxes for GCMsrdquo Journal of Geophysical Research D vol99 no 7 pp 14415ndash14428 1994

[23] J Sheffield G Goteti F Wen and E F Wood ldquoA simulated soilmoisture based drought analysis for the United Statesrdquo Journalof Geophysical ResearchD Atmospheres vol 109 no 24 pp 1ndash192004


[24] K M Andreadis and D P Lettenmaier ldquoTrends in 20th cen-tury drought over the continental United Statesrdquo GeophysicalResearch Letters vol 33 no 10 2006

[25] A Wang T J Bohn S P Mahanama R D Koster and D PLettenmaier ldquoMultimodel ensemble reconstruction of droughtover the continental United Statesrdquo Journal of Climate vol 22no 10 pp 2694ndash2712 2009

[26] L F Luo and E F Wood ldquoMonitoring and predicting the 2007US droughtrdquoGeophysical Research Letters vol 34 no 22 2007

[27] R J Zhao Y L Zhang L R Fang et al ldquoThe Xinanjiangmodelhydrological forecastingrdquo in Proceedings of the 477th OxfordSymposium vol 129 pp 351ndash356 IAHS Publisher 1980

[28] L Dumenil and E Todini ldquoA rainfall-runoff scheme for usein the Hamburg climate modelrdquo in Advances in TheoreticalHydrology P OrsquoKane Ed vol 1 of European GeophysicalSociety Series on Hydrological Sciences pp 129ndash157 ElsevierAmsterdam The Netherlands 1992

[29] D Lohmann E Raschke B Nijssen and D P LettenmaierldquoRegional scale hydrology I Formulation of the VIC-2L modelcoupled to a routing modelrdquo Hydrological Sciences Journal vol43 no 1 pp 131ndash141 1998

[30] M C Hansen R S Defries J R G Townshend and RSohlberg ldquoGlobal land cover classification at 1 km spatialresolution using a classification tree approachrdquo InternationalJournal of Remote Sensing vol 21 no 6-7 pp 1331ndash1364 2000

[31] A F van LoonMH J vanHuijgevoort andH A J van LanenldquoEvaluation of drought propagation in an ensemble mean oflarge-scale hydrological modelsrdquo Hydrology and Earth SystemSciences vol 16 no 11 pp 4057ndash4078 2012

[32] L Ren M Wang C Li and W Zhang ldquoImpacts of humanactivity on river runoff in the northern area of Chinardquo Journalof Hydrology vol 261 no 1ndash4 pp 204ndash217 2002

[33] Z Zhang X Chen C-Y Xu L Yuan B Yong and S YanldquoEvaluating the non-stationary relationship between precipita-tion and streamflow in nine major basins of China during thepast 50 yearsrdquo Journal of Hydrology vol 409 no 1-2 pp 81ndash932011

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Geological ResearchJournal of


Geology Advances in


role in drought propagation and influences processes likepooling attenuation lag and lengthening [9] Some studiesfurther investigated the individual roles of climate andcatchment control in governing hydrological drought char-acteristics and the conclusion is highly dependent on spatialscales Globally hydrological drought duration and deficitvolume aremore related to climate than to catchment controlHowever for regional scale where climate is assumed tobe relatively uniform catchment characteristics like geologyarea slope and groundwater system play an important role ingoverning hydrological drought duration and deficit volume[10]

In fact the impacts on hydrological drought are notlimited to the above-mentioned natural factors (ie climaticand catchment factors) and influences from human activitiesin forms of land cover change reservoir regulation agricul-tural irrigation and water withdrawal from streams or riverchannels should not be ignored From the perspective ofhydrology science human activities influence the process ofhydrological cycle for example infiltration and evapotran-spiration processes and further alter flow regimes and theirspatiotemporal distributions [11] Accordingly the processof hydrological drought would also be influenced Withthis in mind the traditional understanding of hydrologicaldroughts should never be limited to the naturalized drysituation drought caused by human activities has become anew challenging issue that needs to be noticed and discussed

The focus of our study is to investigate the impact ofhuman activities on hydrological drought In this paperwe call drought induced by climate variability the ldquonatu-ralized droughtrdquo and drought caused by human activitiesthe ldquohuman-induced droughtrdquo As aforementioned humanactivities potentially influence underlying surfaces whichfurther leads to a nonstationary response in hydrologicalvariables (eg runoff) Here comes the question will thesechanges influence the characteristics of hydrological droughtsand what is the difference between naturalized and human-induced droughts Based on the above questions a com-prehensive investigation on the evolution of hydrologicaldroughts is conducted in the Laohahe catchment in northernChina (this catchment is selected due to its representativerunoff change pattern influenced by combined effects ofclimate change and human activities which is quite commonin China water-stressed regions) combined with a separatingframework for distinguishing the naturalized and human-induced droughts

2 Study Area and Data

The Laohahe catchment (41∘Nsim4275∘N 11725∘Esim120∘E)located in northern China is selected as an example con-sidering its drastic runoff change pattern which is a com-mon and typical phenomenon for water-stressed regionsin northern China The drainage area of the Laohahecatchment is 18112 km2 and its elevation ranges between427m and 2054m with a general increasing trend fromnortheast to southwest The climate belongs to the semiaridzone according to the observation during the period 1964ndash2009 mean annual temperature precipitation and runoff

2010 0 20 40 60

(km)

N

DEM (m)

Hig

h2054

Low

427

42∘30

998400N

42∘0998400N

41∘30

998400N

41∘0998400N

118∘30

998400E 119∘30

998400E117∘30

998400E 118∘0998400E 119

∘0998400E

Chifeng cityStreamflow stationMeteorological stationRain gauge

ReservoirRiverBasin boundary

Figure 1 Location of the study area and the geographic distributionof hydrometeorological stations and reservoirs

are 758∘C 4183mm and 287mm respectively Due to thetemporally unevenly distributed precipitation (80 of annualprecipitation concentrated between May and September)and temperature the Laohahe catchment presents a strongseasonality in runoff Normally winter is cold and dry withrelatively low streamflow observed while in summer it is hotand wet with most peak flow occurring during this season

The data used in this study includes materials for hydro-logical drought identification and input forcing for hydro-logical modeling Daily precipitation of 52 rain gauges andstreamflow records of the Xinglongpo hydrological stationsituated at the outlet of the Laohahe basin during 1964ndash2009 is provided by the Water Resources Department ofthe Inner Mongolia Autonomous Region Streamflow datais further converted to catchment runoff by averaging therunoff amounts over the catchment area Daily meteoro-logical forcing (1964ndash2009) including maximum and min-imum air temperature wind speed relative humidity andsunshine duration of 3 national standard meteorologicalstations in and around the Laohahe catchment is downloadedfrom the China Meteorological Data Sharing Service System(httpdatacmagovcn) Their geographic distributions areshown in Figure 1 Besides locations of three large reservoirs(Erdaohezi Dahushi and Sanzuodian) within the Laohahecatchment are also given in Figure 1 considering their poten-tial impacts on runoff

Geographic information is obtained as follows 3-arcsecond (about 90m) digital elevation from the shuttleradar topography mission (SRTM) digital elevation modelsoil types from the 5-minute FAO dataset [12] and land


d1 d2 d3 d4 d5

1 2 3 4 5i1 i2 i3

i4

t1 t2 t3 t4 t9

Qz

Stre

amflo

w (m

3s

)


119894

duration 119889119894




119911


3 Methodology


119911 Accordingly




119894












1


2


3


119904


119863


119904




1 1198892 1198893 119863119904 119863119904max and 119882119904) with two








1965

1969

1973

1977

1981

1985

1989

1993

1997

2001

2005

2009


minus2

0

2

4

6

8

Prec

ipita

tion

(mm

)



2001

2005

2009

1981

1985

1989

1993

1997

1973

1969

1977

1965

00

02

04

06

08

10

Runo

ff (m

m)


1960s1970s1980s

1990s2000s

1 2 3 4 5 6 70

Duration (month)

0

minus2

minus4

minus6

minus8

minus10

minus12

minus14

minus16

Defi

cit v

olum

e (m

m)


1960s1970s1980s

1990s2000s

2 4 6 80 12 14 16 1810

Duration (month)

0

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Defi

cit v

olum

e (m

m)



80 857565 70 90 9560 100

Frequency ()

0

2

4

6

8

10

Mon

thly

pre

cipi

tatio

n (m

m)

0 8020 40 60 1000

50

100

150

200

250

300


1980s1990s2000s

(a)

0

1

2

3

4

Mon

thly

runo

ff (m

m)

1006030 40 50 70 908020100

Frequency ()

1006040 802000

10

20

30

40

50

60


1980s1990s2000s

(b)





2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)


(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent


(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)


(c)







RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


d1 d2 d3 d4 d5

1 2 3 4 5i1 i2 i3

i4

t1 t2 t3 t4 t9

Qz

Stre

amflo

w (m

3s

)


119894

duration 119889119894




119911


3 Methodology


119911 Accordingly




119894












1


2


3


119904


119863


119904




1 1198892 1198893 119863119904 119863119904max and 119882119904) with two








1965

1969

1973

1977

1981

1985

1989

1993

1997

2001

2005

2009


minus2

0

2

4

6

8

Prec

ipita

tion

(mm

)



2001

2005

2009

1981

1985

1989

1993

1997

1973

1969

1977

1965

00

02

04

06

08

10

Runo

ff (m

m)


1960s1970s1980s

1990s2000s

1 2 3 4 5 6 70

Duration (month)

0

minus2

minus4

minus6

minus8

minus10

minus12

minus14

minus16

Defi

cit v

olum

e (m

m)


1960s1970s1980s

1990s2000s

2 4 6 80 12 14 16 1810

Duration (month)

0

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Defi

cit v

olum

e (m

m)



80 857565 70 90 9560 100

Frequency ()

0

2

4

6

8

10

Mon

thly

pre

cipi

tatio

n (m

m)

0 8020 40 60 1000

50

100

150

200

250

300


1980s1990s2000s

(a)

0

1

2

3

4

Mon

thly

runo

ff (m

m)

1006030 40 50 70 908020100

Frequency ()

1006040 802000

10

20

30

40

50

60


1980s1990s2000s

(b)





2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)


(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent


(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)


(c)







RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




1


2


3


119904


119863


119904




1 1198892 1198893 119863119904 119863119904max and 119882119904) with two








1965

1969

1973

1977

1981

1985

1989

1993

1997

2001

2005

2009


minus2

0

2

4

6

8

Prec

ipita

tion

(mm

)



2001

2005

2009

1981

1985

1989

1993

1997

1973

1969

1977

1965

00

02

04

06

08

10

Runo

ff (m

m)


1960s1970s1980s

1990s2000s

1 2 3 4 5 6 70

Duration (month)

0

minus2

minus4

minus6

minus8

minus10

minus12

minus14

minus16

Defi

cit v

olum

e (m

m)


1960s1970s1980s

1990s2000s

2 4 6 80 12 14 16 1810

Duration (month)

0

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Defi

cit v

olum

e (m

m)



80 857565 70 90 9560 100

Frequency ()

0

2

4

6

8

10

Mon

thly

pre

cipi

tatio

n (m

m)

0 8020 40 60 1000

50

100

150

200

250

300


1980s1990s2000s

(a)

0

1

2

3

4

Mon

thly

runo

ff (m

m)

1006030 40 50 70 908020100

Frequency ()

1006040 802000

10

20

30

40

50

60


1980s1990s2000s

(b)





2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)


(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent


(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)


(c)







RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


1965

1969

1973

1977

1981

1985

1989

1993

1997

2001

2005

2009


minus2

0

2

4

6

8

Prec

ipita

tion

(mm

)



2001

2005

2009

1981

1985

1989

1993

1997

1973

1969

1977

1965

00

02

04

06

08

10

Runo

ff (m

m)


1960s1970s1980s

1990s2000s

1 2 3 4 5 6 70

Duration (month)

0

minus2

minus4

minus6

minus8

minus10

minus12

minus14

minus16

Defi

cit v

olum

e (m

m)


1960s1970s1980s

1990s2000s

2 4 6 80 12 14 16 1810

Duration (month)

0

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Defi

cit v

olum

e (m

m)



80 857565 70 90 9560 100

Frequency ()

0

2

4

6

8

10

Mon

thly

pre

cipi

tatio

n (m

m)

0 8020 40 60 1000

50

100

150

200

250

300


1980s1990s2000s

(a)

0

1

2

3

4

Mon

thly

runo

ff (m

m)

1006030 40 50 70 908020100

Frequency ()

1006040 802000

10

20

30

40

50

60


1980s1990s2000s

(b)





2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)


(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent


(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)


(c)







RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in




2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

200

300

400

500

600

Prec

ipita

tion

(mm

)


(a)

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

0

20

40

60

80

100

Runo

ff (m

m)

000

005

010

015

020

025

030

Runo

ff co

effici

ent


(b)10

08

00

02

04

06

2004

2000

2008

1984

1980

1988

1992

1972

1968

1976

1964

1996

Base

flow

inde

x (B

FI)

0

5

10

15

20

25

Base

flow

(mm

)


(c)







RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


RunoffPrecipitation

2000 2005 201019851980 199519751970 199019650

25

50

75

100

125

150

175

200

Accu

mul

ated

pre

cipi

tatio

n (m

m)

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1998

1979

(a)

RunoffBaseflow

1970 1975 1990 19951965 1980 2000 2005 20101985

times102

times102

0

2

4

6

8

10

12

14

16

Accu

mul

ated

runo

ff (m

m)

1

2

3

4

5

Accu

mul

ated

bas

eflow

(mm

)

1998

(b)













1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in







1979

0

5

10

15

20

25

30

35

40

10

(5

ton)

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year


Baseline period

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

10

15

20

25

30

35

40

45

50

(mill

ion)

(b) Population


1998

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

(bill

ion

yuan

)



1979

2000

2004

1980

1984

1988

1976

1968

1964

1972

1992

1996

1956

1960

1952

Year

0

50

100

150

200

250

300

350

(bill

ion

yuan

)







BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in



BIAS = 18


1966

1968

1970

1972

1974

1976

1978

1964

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

0

10

20

30

40

50

Runo

ff (m

m)

(a) Baseline period


2002

2004

2006

2000

2008

1986

1984

1988

1982

1980

1996

1994

1992

1990

1998

0

20

40

60

Runo

ff (m

m)

500

400

300

200

100

0

Prec

ipita

tion

(mm

)

(b) Changed period



minus1

0123

Obs

erve

d ru

noff

(mm

)

1985 1989 1993 1997 2001 2005 2009 1981

(a)


1985 1989 1993 1997 2001 2005 20091981minus05

00

05

10

15

Sim

ulat

edru

noff

(mm

)

(b)


minus04minus02000204060810

Diff

eren

ce o

f d

efici

t (m

m)

1985 1989 1993 1997 2001 2005 2009 1981

(c)






1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


1994

1995


0

20

40

60

80

100Pr

ecip

itatio

n (m

m)

(a) Precipitation

1994

1995


0

2

4

6

8

10

Runo

ff (m

m)

(b) Runoff













Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in





Acknowledgments


References












































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in





















Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in










Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in

Documents

Research Article Evolution of Hydrological Drought in Human …downloads.hindawi.com/journals/amete/2016/5102568.pdf · 2019. 7. 30. · Research Article Evolution of Hydrological