Changing human landscape interactions after development of tourism in the northern Vietnamese Highlands

Changing human landscape interactions after development of tourism in the northern Vietnamese Highlands tài liệu, giáo á...

Changing human–landscape interactions after development of

Huong Thi Thu Hoanga,b,d,* , Veerle Vanackera, Anton Van Rompaeyb,

a

Earth and Life Institute, Georges Lemaıˆtre Center for Earth and Climate Research, Universite´ Catholique de Louvain, Place L Pasteur 3, Bte L4.03.08,

Louvain-la-Neuve, Belgium

b

Division of Geography, Department of Earth and Environmental Sciences, K.U Leuven, Celestijnenlaan 200 E, bus 2409, B-3001 Heverlee, Belgium

c Institute of Vietnamese studies and Development sciences, VNU, 336 Nguyen Trai street, Thanh Xuan district, Hanoi, Viet Nam

d

Faculty of Geography, Hanoi University of Sciences, VNU, 334 Nguyen Trai street, Thanh Xuan district, Hanoi, Viet Nam

Introduction

Manytropicalareasworldwidearecharacterizedbyhighrates

ofdeforestation.AccordingtoLambinandGeist(2003),one-third

ofthehumidforestinSoutheast Asiawasclearedbetween the

beginningofthetwentiethcenturyandWorldWarII.Thedecline

inforestcovercontinuedafterthe1950s.Presently,forestscover

46–48%ofthelandsurfaceinSoutheastAsia,butlessthan10%of

theprimarytropicalrainforestispreserved(FAO,2010;Dongetal.,

2012) Recentdeforestationratesforthis regionareassessedat

1.5%per year(FAO, 2006; Grainger,2008) Nevertheless,forest

dynamicsarediverse.Insomecountries,suchasthePhilippinesor Cambodia, the deforestation rate is much higher than the Southeast Asian average; while in other countries, such as Vietnam,thestart ofa foresttransitionisreported (FAO,2006; MeyfroidtandLambin,2008b) Foresttransitionisdefinedbya reversal of the trend of deforestation so that net reforestation occurs (Meyfroidt and Lambin, 2008b) Causes of tropical deforestation and forest transition are still poorly understood, andarethescopeofongoingresearchprogrammes.Deforestation andreforestationpatternsarelinkedtomultiplebiophysicaland socio-economicvariablessuchasethnicity(Castellaetal.,2005;Vu

etal.,2013),landtenure(Mottetetal.,2006),increasingdemand forfoodproduction(Zhang,2000;GeistandLambin,2001;Casse

etal.,2004;MeyfroidtandLambin,2008a),poverty(Hobbs,2001; Adamsetal.,2004;Dasguptaetal.,2005;Robinson,2006;Zwane,

2007),soilfertility(Szillassietal.,2010;Vanackeretal.,2014),and accessibility(Koning,2000;Castellaetal.,2005;Etteretal.,2006; VanDesseletal.,2008)

A R T I C L E I N F O

Article history:

Received 16 January 2014

Received in revised form 7 August 2014

Accepted 27 August 2014

Available online 6 September 2014

Keywords:

Coupled human–environmental changes

Tourism development

Ethnicity

Land use pressure

Forest transition

Northern Vietnam

A B S T R A C T

Indevelopingcountriesintropicalregions,thepoorestsegmentsoftheruralpopulationoftenrelyon forestsforsurvival.Thecreationofoff-farmjobsinthetourismsector,constructionormanufacturing hasbeensuggestedasapotentialwaytoalleviatepressureontropicalforests.UsingSaPadistrictasa casestudy,weevaluated thecouplingofhumanandforestdynamics.Thedistrictwasopenedfor internationaltourismin1993,whichhadalargeimpactondailylifeinSaPatownanditssurrounding communities.Analysisoflandcoverchangefortheperiod1993–2014,usinghigh-resolutionsatellite imagesfromthreetimeperiodsandananalysisofcovariance,detectedpossibleassociationsbetween forestcoverchangeandsocio-economic,culturalandbiophysicalvariablesatthevillagelevel.Between

1993and2006,SaPadistrictexperiencedanetdecreaseofforestinfavourofarableland,whilethis trend was reversed in the period 2006–2014 However, trends at district level mask substantial heterogeneityatvillagelevel.Resultsshowthatdeforestationisconsiderablylowerinvillagesthatare stronglyinvolvedintourismactivities Marginal agricultural fieldswithlow productivityare also preferentiallyabandoned.Becauseofdiversificationinalternativeeconomicactivities,ruralhouseholds maybecomelessdependentonnaturalresourcesandagriculturalproductsfortheirsurvival.These resultssuggestthatthecreationofoff-farmincomesourcesactivitiescanbeadriverofshiftsinhuman– environmentinteractions,asnewlivelihoodstrategiescanoffsetthepressureonforestedland

ß2014ElsevierLtd.Allrightsreserved

* Corresponding author at: Earth and Life Institute, Georges Lemaıˆtre Center for

Earth and Climate Research, Universite´ Catholique de Louvain, Place L Pasteur 3, Bte

L4.03.08, Louvain-la-Neuve, Belgium Tel.: +32 494694385; fax: +32 16322980.

E-mail addresses: thi.th.hoang@student.uclouvain.be ,

huonghoangbg@yahoo.com (H.T.T Hoang).

ContentslistsavailableatScienceDirect

Anthropocene

j ou rna l h ome p a ge : w ww e l se v i e r co m/ l oc a te / a nce ne

http://dx.doi.org/10.1016/j.ancene.2014.08.003

2213-3054/ß 2014 Elsevier Ltd All rights reserved.

thepopulationoftenrelyonforestsforsurvival(Tugault-Lafleur,

2007;Coulibaly-Linganiet al.,2009) Theuseofforestproducts

allowslivelihooddiversification,butmayleadtoforest

degrada-tionand/ornetdeforestationwhenitisnotcontrolled(Jadinetal.,

2013).TheVietnamesemountainareasaredominantlypopulated

byethnicminoritiesthatspeaklocallanguagesandstilladopta

traditionallivelihoodbasedonself-subsistencefarming(Foxetal.,

relativelyisolatedconditionsanddonot fullyparticipatein the

majoreconomictransformationofVietnamthatistakingplacein

thelowlands.Thelivelihoodoftheselocalpeoplestronglydepends

ontheavailablenaturalresourcesbecauseofalackof

infrastruc-ture and education that would allow them to participated in

market-orientedactivities(FrontierVietnam,1997).Duringpast

decades, the scarcity of arable land coupled with population

growthhas led to increasingpressure on forests (Burgess and

Barbier,2001).Exploitationofforestresourcesbyethnic

minori-ties,respondingtosocio-economicpressures,isgenerallythought

tobethecauseofrapidforestdegradationand/ordeforestationin

SoutheastAsia(Foxetal.,2000;GeistandLambin,2001).Studies

byJodha(1998),Ravnborg(2003),Scherr(2000)andJadinetal

(2013), however, showed that deforestation is not necessarily

associatedwithpoverty

Thecreationofoff-farmjobshasbeensuggestedasapotential

waytoalleviatepressuresontropicalforests(Mather,1992;Rudel

etal.,2005;Getahunetal.,2013;Teka-Belayetal.,2013).Off-farm

jobsareoftencreatedbyeconomicdevelopmentofurbanareas

resultingin rural–urbanmigrations (Vanegas and Henry,2012;

Vermeirenetal.,2012).Developmentoftourismactivitiesinrural

areashasalsobeensuggestedasaviablemeanstooffsetpressures

on forests (Garcı´a-Martı´nez et al.,2011; Nyaupane andPoudel,

2011) When rural households can generate additional income

fromtourismactivities,abandonmentoflow-productivefarmland

and spontaneous establishment of secondary forest on former

agriculturalplots mayresult.Dong etal (2008)supported this

hypothesisbasedon a case-studyin Lugulake(China).Joband

Paesler(2013)alsodescribedhowtheintensificationoftourismin

Wasini(Kenya)hasledtolessintensivelanduseforagricultural

purposes,eventually resultingin adecreaseoffarmlandandan

increaseoftheforestarea.Nevertheless,thishypothesishasbeen

challenged by other studies suggesting that tourism activities

stimulate deforestation and forest degradation Research by

thetourismsectordidnotdecreaseagriculturalpressureonforests

andsoilresourcesbecausehouseholdsinvestedtheirincomefrom

tourismintheexpansionofarablefieldsandincreasingfrequency

of cultivation by hiring external labour Additionally, Gaughan

etal.(2009)showedthattheincreasednumberofvisitorstothe

archaeological sites of Angkor Kwat in Cambodia accelerated

deforestationintheAngkorbasin.Thedeforestationoccurreddue

toincreasedcharcoalproductionfornewrestaurantsandhotels,

whichrequiredwoodproductsfromforests.Inthecoastalareasof

HainanIsland(SouthernChina)andthe Mediterranean(Turkey),

WangandLiu(2013)andAtiketal.(2010)respectivelyindicated

thattourismdevelopmentledtoarapidincreaseofthebuilt-uparea

Theseactivitiesresultedinadecreaseofagriculturallandandcoastal

forest,causinglandscapefragmentationandcoastalerosion

In this study, we evaluate possiblechanges in the human–

environment interactions after the development of tourism

activities Using Sa Pa district in the northern Vietnamese

Highlandsasatestcase,weaddressedthefollowingquestions:

First,howhasforestcoverchangedintheperiodbetween1993and

2014?Second,how does forest cover changerelate totourism

development?Third,whatarethelikelyimpactsofthechanging

human–landscaperelationshipsonlocallivelihoods?

Studyarea

SaPadistrictislocatedinNorthernVietnam(Fig.1)andcovers

anareaofca.680km2.Ithasatotalof55,900inhabitants(GSO,

2010)livingin17communesanditsadministrativecentre,SaPa town The district is considered as a gateway to the northern VietnameseHighlands.Thetopographyisrough,withanelevation

of180mintheMuongHoavalleyandupto3143mattheFansipan peak (highest elevation in Vietnam,locatedwithin Hoang Lien NationalPark).ThemajorriversaretheMuongHoaandTaTrung

HoRiverthatflowintheRedRivernearbyLaoCai.Theregionis characterized by a sub-tropical and temperate climate withan annualrainfallof2763mm(FrontierVietnam,1999)

SaPadistrictishometo6majorethnicgroups:theHmong,the Yao,theTa`y,theGia´y,theXaPhoandtheKinh.TheTa`yoccupied thefertilevalleysandmiddlealtitudes.Theotherethnicgroups suchastheHmongandYaoenteredNorthernVietnamonlyinthe 19thcentury(Michaudand Turner, 2006), andsettled onsteep forestedslopesgenerallyabove800m.Before1960s,therewere onlyafewKinhlowlanderslivinginSaPatownasthesurveillance andmaintenancestaffsofFrenchmilitary(MichaudandTurner,

2006).From1960sonwards,KinhmigratedtoSaPadistrictasthis wasstimulatedbytheNewEconomicZonePolicyofthenational government (Michaud and Turner, 2000, 2006) The Kinhwere mainly involved in administration, tourism, and education and settled in thedistrict’s capital, while most of theother ethnic groupspracticeddifferenttypesofsubsistenceagriculturemostly

intheformofshiftingcultivation(Tugault-Lafleur,2007).Apart from the shifting cultivation, ethnic minorities also used to cultivate opium and collect forest products for their survival

whichcouldhavecontributedtopastforestclearance.Today,the ethnicgroupscultivatewaterriceonpermanentterracedpaddy fields;maizeandothercropsonuplandfields(Leiszetal.,2004; Turner,2011).Terracedpaddyfieldswerefirstintroducedbythe HmongandYaowhomigratedfromsouthernChinatonorthern Vietnamduringthelate19thandearly20thcenturies(Michaud,

1997) Additionally, many households cultivate cardamom (Amomum aromaticum) underforest cover as a substitutecash crop,afterthebanonopiumin1992(Tugault-LafleurandTurner, 2009;Turner,2011)

Because of its scenic landscape and presence of five ethnic groupswiththeirtraditionalwayofliving,SaPaisconsideredas oneofthemostattractivetourismareasinVietnam.TheHoang Lien Mountainscomprise probably thelast remnantsof native forestofthenorthernVietnamesehighlands.Itbecameoneofthe firstareasrecognizedasa‘specialuseforest’inVietnam,anditwas convertedintotheHoangLienNationalPark(HLNP)inJuly2002 followingthePrimeMinister’sDecision90/2002/QD-TTgtoprotect biodiversity bypreservingthesubtropical and temperateforest ecosystems(Le,2004;Jadinetal.,2013).AlreadyundertheFrench Regime(1887–1940),SaPadistrictwasawell-knownholidayand relaxationresort(MichaudandTurner,2006).NorthernVietnam sufferedalotunderthefirstIndochinawar(1945–1954).Thetown sunkintooblivion,asalargepartofthepopulationofSaPatown fledawayfromthehostilities.Intheearly1960s,intheframework

of the New Economic Zones Policy, migration schemes were designed by the new socialist regime that stimulated the Vietnamese Kinh from the lowlands to populate the northern VietnameseHighlands(Hardy,2005).Thedecisionofthenational governmenttoopenSaPadistrictforinternationaltourismin1993 hadalargeimpactondailylifeinSaPatownanditssurrounding communities.Thenumberofdomesticandinternationalvisitors increasedexponentiallyfrom16,100in1995to405,000in2009 (GSO,1995, 2010)(Fig.1 Tourism isnow themostimportant economic activity in the area, and it generated 58% of Sa Pa

decreasedgraduallyfrom36%in2000to21%in2009(GSO,2000,

2010) Local inhabitants that potentially benefit from tourism

activitiesarehotelandrestaurantownersandshopkeepersinSaPa

town;tourguides,traditionalcraftsellersfromtheruralvillages

and farming households that offer rooms for homestays The

tourism infrastructure is dominantly controlled by the Kinh

majority,whiletheotherminoritiesmainlydeliverlabourforce

torunthetourismindustry

Materialsandmethods

Mappinglandcoverandlandcoverchanges

Inordertoevaluatethepotentialimpactoftourismactivitieson

forestcoverinSaPa,threelandcovermapswerecompiledbased

on LANDSAT images available from the U.S Geological Survey

archives (http://glovis.usgs.gov) One LANDSAT-patch(path/row

128/45)coversthewholeSaPadistrictwitharesolutionof30mby

30m.The LandsatimagesdatefromFeb1,1993(just afterthe

opening for international tourism), Nov 4, 2006 (midst of the

evaluation period)and Jan02, 2014 (currentstate) All images

weretakeninthepost-harvestperiodwhenthearablefieldsare

bare.AllLandsatimagesinthefreelyavailableUSGSarchiveare

orthorectified with precision terrain correction level L1T

(Vanonckelen et al., 2013) All images were then corrected for

atmosphericandtopographiceffectsusingtheMODTRAN-4code

and the semi-empirical topographic correction implemented

in ATCOR2/3 (Richter, 2011; Balthazar et al., 2012) Then, a supervisedmaximumlikelihoodclassificationwascarriedoutto mapthefollowing5landcovercategories(Fig.2):forest,shrub, arableland,waterbodyandurbanarea.Spectralsignaturesforthe differentlandcovertypeswereidentifiedbydelineatingtraining areasonthebasisoffieldworkcarriedoutin2010(Fig.5

Theaccuracyofthelandcovermapswasassessedbycomparing theclassifiedlandcoverwithvisualinterpretationsofveryhigh resolution remote sensing data For 1993,the comparison was done withaerial photographs(MONRE, 1993);for 2006witha VHR-SPOT4image(MONRE,2006)andfor2014withaVHR-SPOT5 image(MONRE,2012).Randomsamplingofvalidationpointswas donewithn=219forthe1993map,n=315forthe2006map,and

n=306forthe2014map.Thenumberofsamplepointsperland coverclassvariedfrom3to111,dependingonthearealcoverof theclasses.Forallrandomlyselectedpoints,thelandcoverwas comparedwiththeclassifiedlandcover.Thiscomparisonallowed

to assess the overall accuracy, quantity disagreement and allocationdisagreement(in%)followingtheproceduresdescribed

byPontiusandMillones(2011)

In order to analyze land cover change trajectories over 3 timeperiods,thechangetrajectoriesweregroupedin6classes:(1) deforestation(changefromanyclassofforesttonon-forest),(2) reforestation(change fromnon-forest toforest),(3) land aban-donment(change fromagriculturalland toshrub orforest),(4) expansionofarableland(conversionfromshrubtoarableland),(5)

Fig 1 Location of Sa Pa district, with graph on the evolution and distribution of tourist visits.

otherchanges,and (6)nochange(Table1 Theoriginalclasses

‘waterbody’and‘urbanarea’thatonlyoccupyaminorfractionof

thelandwerenottakenintoconsideration.Amajorchallengein

mappinglandcover changeisthedetectionof permanentland

abandonmentinshiftingcultivationsystemsinwhich fieldsare

regularly taken out of production for a short time span only

Therefore in this study we defined land abandonment as a

transition fromagricultural land (observedin 1993)to natural

regrowthofshrub(observedin2006)onconditionthattheparcel

wasnottakenagaininproductionin2014.Pixelswithobserved

transitionssuchasA-A-SandA-A-F(Table1)ofwhichitisnotsure

that they are permanently abandoned were classified into the

group‘Otherchange’

Analysisofthecontrolsonlandcoverchangepatterns

Inordertounderstandtheobservedlandcoverchangepatterns,

socio-economicandbiophysicaldatawerecollectedatthelevelof

villages.InSaPadistrict,themajorityoftheethnicgroupslivesin

ethnicallyhomogeneousvillages(ba:northoˆninVietnamese).Only

4ofthe85villagesareinhabitedbymultipleethnicgroups,and

theyaretypicallylocatedinthecommune(xa˜)centres.Therefore,

thevillagelevelisconsidered asthemostdetailedandrelevant

scale levelfor theanalysis of human–environmentinteractions

Fig 2 Land cover maps derived from Landsat images for(A) 1993, (B) 2006, (C) 2014 and (D) proportions of the main land cover classes in 1993, 2006, 2014.

Table 1 Land cover change trajectories and their descriptions for period 1993–2014 Order Land cover change Category Change trajectory

1993 2006 2014

1 Deforestation F-F-A Forest Forest Arable land

F-F-S Forest Forest Shrubs F-A-A Forest Arable land Arable land F-A-S Forest Arable land Shrubs F-S-A Forest Shrubs Arable land F-S-S Forest Shrubs Shrubs

2 Reforestation A-F-F Arable land Forest Forest

S-F-F Shrubs Forest Forest S-S-F Shrubs Shrubs Forest

3 Land abandonment A-S-S Arable land Shrubs Shrubs

A-S-F Arable land Shrubs Forest

4 Arable land expansion S-S-A Shrubs Shrubs Arable land

S-A-A Shrubs Arable land Arable land S-F-A Shrubs Forest Arable land

5 Other change A-A-S, A-F-S, A-F-A, S-A-S, A-S-A, S-F-S, F-A-F,

F-S-F, S-A-F, A-A-F

6 No change F-F-F, A-A-A, S-S-S

not officially delineated because the commune is the lowest

administrativeunit(Castellaetal., 2005) Therefore, thevillage

boundaries(n=85)inSaPadistrictweredelineatedbymeansof

participatory mapping following the procedure described by

Castella et al (2005) and Meyfroidt (2009) Cadastral officers

wereoffereda1/10.000scalecolourprintofthe2006VHR-SPOT4

image(printedintruecolours,5mresolution)andwereaskedto

drawthevillagebordersonatransparentsheetontop

Tables 2 and3 showall thevariables that werecollected at

the village level Socio-economic variables were derived from

theyearbook of 1989 and 2006, and from the Vietnam Rural,

Agricultural, and FisheryCensus conducted in 2006 under the

leadershipoftheDepartmentofAgriculture,ForestryandFishery

StatisticsandtheGeneralStatisticsOfficewithsupportfromthe

WorldBank.Theoriginalcensusdataavailableathouseholdlevel

wereaggregatedtovillagelevel,andthefollowingvariableswere

calculated:thepercentageofhouseholdsinvolvedintourism(%),

theethnicgroup(categorical),thepopulationgrowthrate(%/year),

thepovertyrateexpressedaspercentageofhouseholdsunderthe

nationalpovertythresholdof2400,000VND/person/yearandthe

involvementincardamomcultivation(ha/household)(Table3 In

ordertoevaluatethepotentialeffectofthelandusepolicyinside

and outside the National park, one more categorical variable

(inside/outsidethepark)wastakenintoaccounttoexaminethe

effect of public policy Six biophysical variables were firstly

collectedatpixel level(3030m): theelevation(m abovesea

level),theslopegradient(degree),thedistancetomainroad(m),

thedistancetorivers(m),thedistance toSaPa town(m), the

distance to nearest market (m) and were then aggregated to

villageleveltomatchwiththescaleofsocio-economicvariables

(Table3

Multiple regression analysis using ANCOVA (analysis of covariance) was performed to detect possible associations betweenlandcoverchange,andsocio-economicandbiophysical variablesatthelevelofindividualvillageswhichcanconsideredas homogeneousunitsintermsofethnicity,livelihoodand biophysi-cal setting ANCOVA is a widelyappliedtechnique asit allows evaluatingthecombinedeffectofarangeofbothcategoricaland numericalpredictors(ManeeshaandBajpai,2013).ANCOVAwas performed for each one of the four land cover change types (deforestation,reforestation,landabandonment,andexpansion

of arable land) as the dependent variable A multicollinearity testwascarriedout todetect correlationbetweenexplanatory variables Multicollinearity diagnostics were performed by calculatingtheVariationInflationFactors(VIF)andtheTolerance (TOL).Inthisstudy,variableswithVIFgreaterthan2 andTOL less than 0.6 are excluded from the analyses as proposed by

Allison(1999).Thefinalmodelsincludedethnicityandeffectof preservation as categoricalvariables; engagement in tourism, cardamom cultivation, poverty rate,population growth, slope, distancetorivers, distancetomainroadanddistancetoSaPa townasnumericalvariables(Table3).ANCOVAmodelparameters were estimated using XLSTAT software, and the explanatory poweroftheANCOVAmodelswasassessedbytheGoodnessoffit statistics,R2

Results Landcoverchangepatterns

Fig.2showsthelandcovermapsfortheyears1993,2006and

2014 Theoverall accuracy of theland cover classification was assessedat80.0%,86.4%and84.6%(quantitydisagreementof5.0%,

Table 3

Independent variables used for the ANCOVA analysis All variables were collected or averaged at the village level.

Variable Acronym Description Type Unit Source

Engagement in tourism TOUR The percentage of households engaged in tourism in 2006 Numerical % Vietnam Rural, Agricultural,

and Fishery Census conducted in 2006 with the support of the World Bank ( GSO, 2006b )

Y Yao

T Ta`y

G Gia´y

XP Xa Pho MIX Mixed ethnic groups Poverty Rate PR The percentage of households under the poverty level

of Vietnam in 2006

Numerical % Cardamom cultivation CC Surface area of cardamom cultivation per household Numerical ha/hh

Population growth POPGR Population growth rate in the period 1989–2006 Numerical %/year Yearbook in 1989 and 2006

( GSO, 1989, 2006a ) Effect of preservation policy InsideNP The villages are located inside the National park Categorical Topographic maps of 2009

published by Ministry of Natural Resources and Environment at the scale 1:50,000 ( MONRE, 2009 )

Outside NP The villages are located outside the National park Elevation EL Average elevation calculated at pixel-level Numerical m

Slope SL Average slope calculated at pixel-level Numerical Degree

Distance to roads DROAD Average of the distance to main roads calculated at pixel-level Numerical m

Distance to rivers DRIVER Average of the distance to rivers calculated at pixel-level Numerical m

Distance to Sa Pa town DTOWN Average of the distance to Sa Pa town calculated at pixel level Numerical m

Distance to market DMARKET Average of the distance to markets calculated at pixel level Numerical m

Table 2

Dependent variables used for the ANCOVA analysis All variables were averaged at the village level.

Variable Acronym Description Type Unit Source

Deforestation DEFOREST Total area of deforestation of each village divided

by the surface area of the village and multiply by 100

Numerical % Landsat images 1993,

2006, 2014 Reforestation REFOREST Total area of reforestation of each village divided

by the surface area of the village and multiply by 100 Land abandonment ABANLAND Total area of land abandonment of each village divided

by the surface area of the village and multiply by 100 Expansion of arable land EXPLAND Total area of arable land expansion of each village divided

by the surface area of the village and multiply by 100

thelandcovermapsof1993,2006and2014,respectively

ThelandcoverpatterninSaPadistrictisstronglydetermined

bythetopography.Valleysaregenerallycultivated.Steepslopes

and mountain peaks are predominantly covered by forests or

shrubs Patches of forest are concentrated on the Hoang Lien

mountainrangeinthesouthernpartofSaPadistrict,andarealso

foundonremotesteepslopes.Shrubsarewidelydistributed,and

canbefoundinvalleys,mountainpeaksoronsteepslopes

Between1993and2014,theoverallareacoveredbyforestand

arable land increasedslightly (with respectively +3% and +2%)

whileshrubsdecreasedwith 5%(Fig.2D).However,landcover

changesarenotlinearinSaPadistrict,andthereexistsubstantial

temporal differences During the first period (1993–2006), the

study area experienced a general trend of deforestation for

expansionofarableland.Between1993and2006theareacovered

byforestdecreasedby 1%whilearable landincreasedby+4%,

respectively The deforestation tendency seems to be reversed

after2006inSaPadistrict.Theareacoveredbyforestsincreasedby

+4%whilearablelanddecreasedby 2%between2006and2014

Theareacoveredbyshrubsdecreasedcontinuouslybetween1993

and2014.Aforesttransitioncouldbeobservedinthestudyareaas

a shift from a net deforestation to a net reforestation, and it

occurredatthemidofthe2000s

Fig.3showsthespatialpatternoflandcoverchangebetween

1993 and 2014 Most of the deforestation took place in the

northern and southeastern part of the district which can be

explainedby thefact that forestsin thesouthwestern partare

mainlysituatedwithintheHoangLienNationalPark.Accordingto

thenationallaw,farmlandexpansionisforbiddenwithinnational

parks Nevertheless,someforestloss canbe observedwhich is

probablyduetoforestfiresandillegallogging

Descriptionofthepredictorsoflandcoverchange

Fig.4showsthespatialpatternoftheindependentvariables

thatwereevaluatedinthisstudy.ItisclearthatKinhpeopleare

livinginSaPatown,whileHmongandTa`yethnicgroupsoccupy

the rural area Hmong ethnic groups are settled on higher

elevations,andTa`yaregenerallysettlednearbytheriversinthe valleys.ThevillagesoftheYaoaresituatedintheperipheralareas

in thenorthand southofSaPa district.Fig.4Ashows thatthe householdinvolvementintourismishighestinSaPatown(>50%) Involvementintourismintheperipheralareasisrestrictedtoafew isolatedvillages.ThepovertyratemapshowsthatthetownofSa

Paanditssurroundingvillagesarericherthanthemoreperipheral areas.Thesouthernpartofthedistrictisalsoricherbecausemany local households receive anadditional income from cardamom cultivationunderforest.Cardamomismainlygrownundertreesof theHoangLienNationalParkinthesouthernpartofthedistrict Thepopulationgrowthispositiveinthewholedistrictandhighest

inSaPatownanditsimmediatesurroundings

Table4showstheresultsoftheANCOVAanalysisforfourland covertrajectories:deforestation,reforestation,landabandonment and expansion of arable land The explanatory power of the ANCOVAmodelsisassessedbytheR2values(Table4 Between55 and72%ofthevarianceinlandcoverchangeisexplainedbythe selected predictors Land cover change is controlled by a combinationofbiophysicalandsocio-economicalfactors.Forests aretypically betterpreservedin villageswithpooraccessibility (steepslopes,farfrommainroads,andpoormarketaccess),anda lowornegativepopulationgrowth.Theinfluenceofenvironmental and demographicdriverson forestcoverchangehaspreviously beendescribed forotherareas offrontiercolonization(Castella

etal.,2005;Hieteletal.,2005;Getahunetal.,2013;Vuetal.,2013)

negativelyassociatedwithdeforestationandpositivelywithland abandonment.When theinvolvement ofhouseholdsin tourism activitiesincreasedwith10%,deforestationis predictedtohave decreased with resp 0.61% and land abandonment to have increased with0.45% Deforestationis higher in villages in the northandsoutheastofSaPa district,thatarelocatedatgreater distancefromthetourism centre.Landabandonmentis mostly observedinSaPatownandinthecommunesofTaPhin,SanSaHo, LaoChai,TaVanandBanHo(Figs.1and3 Insomevillages(SaPa town;TaChaivillage,belongingtoTaPhincommune;LyLaoChai village, belongingtoLaoChaicommuneandHoangLienvillage, belongingtoBanHocommune),morethan8%ofthesurfacearea wasabandonedbetween1993and2014.Overtheperiod1995–

2009,thenumberoftouristsinSaPadistricthasincreasedby25 times(Fig.1 Giventhecurrenteconomicpolicy,itisexpectedthat

Fig 3 Land cover change map for the period 1993–2014.

Table 4 The results of ANCOVA at village level (85 villages) Only the controlling factors that are retained at a 10% significance level are listed.

LCC categories Controlling

factors

Value Standard deviation

Pr > |t| Goodness

of fit of model (R 2 ) Deforestation SL 0.494 0.148 0.001 0.59

TOUR 0.061 0.065 0.034 POPGR 1.543 0.723 0.036

CC 0.975 1.141 0.100 Inside NP a

4.985 1.802 0.007 Reforestation SL 0.356 0.191 0.067 0.55

Inside NP a 6.183 2.304 0.009 Land abandonment DRIVER 0.003 0.001 0.039 0.63

TOUR 0.045 0.029 0.093 Inside NP a

1.567 0.688 0.026 Ethnic-MIX b

2.443 0.998 0.017 Expansion of

arable land

DROAD 0.001 0.000 0.087 0.72

SL 0.599 0.174 0.001

CC 3.444 1.330 0.012 Inside NP a 3.464 2.007 0.089 Ethnic-H b

3.693 1.737 0.037

a

Reference category is Outside NP.

b

future(MichaudandTurner,2006)

Thestatisticalresultsindicatethatthecultivationofcardamom

isnegativelyassociatedwithdeforestationandexpansionofarable

land.Thismeansthattheinvolvementincardamomcultivation

(under forest) slows down deforestation and expansion of

cultivatedland,ascardamomplantationsarenotclassifiedhere

as agricultural land Cardamom production provides higher

incomes than traditional crop farming (Sowerwine, 2004a)

Recently, cardamom is emerging as an important cash crop in

northernVietnamthatrequireslittleinvestmentandlabourbut

mayofferhigherincomelevels(Tugault-LafleurandTurner,2009)

Becauseoftherequirementofadenseforestcanopyforoptimal production,thevillagersnotonlyprotecttheremainingoldforest butalsoallowregenerationofsomeoftheswiddenlandsinorder

tocreatethenecessaryecologicalconditionstoplantandharvest cardamom(Sowerwine,2004b).Itsimpactonforestconservation

issimilartothesystemofshadecoffeecultivationinforestthat alsocontributedtoapreservationoftheafromontaneforestsin, e.g.,thesouthofEthiopia(Getahunetal.,2013)

Theroleofethnicityiscomplex.Aftercontrollingfor biophysi-calandsocio-economicsettings,Hmongvillagesarecharacterized

byhigherexpansionratesofarablelandcomparedtoYaovillages ThiscanbeexplainedbythefactthatHmongvillagesaremore

Fig 4 Spatial pattern of the independent variables: (A) percentage of households engaged in tourism, (B) surface area of cardamom cultivation per household, (C) ethnic distribution in 2006, (D) poverty rate in 2006, (E) population growth rate in the period 1989–2006, (F) distance to Sa Pa town at pixel-level, (G) slope gradient, (H) distance to the nearest road at pixel-level and (I) distance to the rivers at pixel-level.

denselypopulatedthanYao villages(Jadinetal., 2013)sothey

needtoexpandtheirarablelandmoretosupplythefooddemand

Invillageswithmixedethnicities,thelandabandonmentrateis

higherthaninYaovillages,whichcanbeexplainedbythefactthat

mixedethnicitiesonlyoccurintheaccessiblecommunecentres

thataremoreinvolvedinoff-farmactivities

Theeffectof preservationpolicyis certainly reflectedin the

differenceinlandcoverchangesinsideandoutsidetheNational

park The estimated coefficients for the explanatory variable

‘Inside NP’ are negative for all land cover change categories

wherebythe‘OutsideNP’istakenasareferencevalue.Thismeans

that land units with the same physical and socio-economic

propertieshavelowerdynamicswhentheyarelocatedinsidethe

HoangLienNationalpark

Discussion

In Vietnam,therapidincrease in forestarea since theearly

1990s resultedina reversalof thenationaldeforestationtrend

(Meyfroidt and Lambin, 2008b) The national-scale assessment

masksawiderangeofotherlandusedynamicsthatexistatthe

localscale,andthatarenotnecessarilyconformtothetrendsin

forest cover change at national scale In the Sa Pa district,

reforestationwasobservedatthemidofthe2000s,someyears

laterthanwasobservedatnationalscale.Thistimepointroughly

correspondstothestrongincreaseinnumberoftouriststoSaPa

(Fig.1 Thereisawidevarietyofhuman-inducedchangeinforest

cover Forest cover changes are different in villages that are

stronglyinvolvedintourismactivities.Theyarecharacterizedby

significantlyhigherratesoflandabandonmentandlowerratesof

deforestation.Thiscanbeexplainedbyrecentchangesinlabour

divisionandincomeinruralhouseholds.Inthetraditionalethnic

society, labour was mainly divided by gender (Duong, 2008b)

Traditionally,womenwereprimarilyresponsibleforhousework,

agriculturallabour and firewood collectionwhile men werein

charge of the heavy works such as logging, plowing, building

housesand processingtools (Cooper,1984; Sowerwine, 2004a;

Symonds,2004).Thistraditionallabourdivisionwaschallengedby therapidgrowthofthetourismindustryinSaPatown(Duong,

strongly and tradeopportunitiesappeared,women fromethnic minoritiesengagedintheseactivities(MichaudandTurner,2000) Today, many young female from rural villages act as trekking guides,andyoungandoldwomenfromethnicminoritiesalikesell textilecommoditiestotourists(Turner,2011).Someofthemhave becomeprofessionaltourguidesandarehiredbyhotelsandtravel agenciesin town,and cangainhigherincomes(Duong,2008a) Withthisextraincome,theycanliveindependently,maketheir own money and are able to provide financial support totheir families(Duong,2008a)

Thedevelopmentoftourismactivitiesmainlyofferednew off-farmopportunitiesforwomenfromethnicminorities,havingasa direct consequence that women are now less involved in agriculturalactivitieswhilemenaremoreinvolvedintohousehold management As there is less labour available for agricultural activities,cuttingorclearingoftrees,marginalagriculturalfields withlowproductivityarepreferentiallyabandoned(Fig.5D)and deforestationisreduced.Our resultssuggestthattheadditional incomefromtourismissufficientlyhightoexceedtheaddedvalue that can be gained fromsteep land agriculture or from forest extraction The fallowedfields will regenerate into shrubs and secondary forests that can develop the optimal ecological conditions for cardamom cultivation Despite the fact that it may takeuptoa decadeor longerbefore thehouseholds will achieve economic returns ontheir investment, manyfamilies expectanincomefromcardamomfortheirchildren(Sowerwine,

cultivationbecauseofincreasinglandconstraints,loweryields, loss ofsoilfertilityandlackof labouravailability(Sowerwine,

‘‘watershed protection land’’, and swidden rice varieties are rapidlyabandonedasmoretimeisdevotedtowetriceproduction (Sowerwine, 2004a) Because of diversification in alternative economicactivities,ruralhouseholdsarebecomingless

depen-Fig 5 Major land cover types that were identified in Sa Pa district: (A) Forest, (B) Shrubs, (C) Arable land and (D) Abandoned steep farmland.

reduced

Thisdecreaseinlandpressureaftertourismdevelopmentisnot

confirmed by previous studies in Southeast Asia, where the

presence of alternative income sources has increased the

frequency of cultivation through hired rural labour and/or the

expansionofthecultivatedareathroughlandpurchase(e.g.,Forsyth

(1995)fornorthernThailand).Thissuggeststhatlocalandnational

landusepolicylikelyplaysanimportantroleindirectingtourism

developmenttowardssustainablenaturalresourcemanagement.In

Sa Pa, conservation policy has had a positive effect on forest

protectionasmostoftheforestswithintheNationalparkremained

intactduringlastthe21years.Thismakestheareaattractivefor

tourists,andtouristsarefurthersupportingbiodiversity

conserva-tionbyprovidingextrarevenueforconservation.Directrevenueis

presentlybeingraisedbytheHamRongproject,andbythecharging

offeesforclimbingFansipanmountainorvisitingexclusivesites

withinSaPadistrict(FrontierVietnam,1999)

Conclusions

This paper aimed at better understanding of the human–

environmentinteractionintheSaPadistrictaftertheadventand

growth of the tourism industry A land cover change analysis

between1993and2014showedthattheSaPadistrictasawhole

experienceda foresttransition,withanobserved turningpoint

around mid 2000s However, trends at district level mask

substantialheterogeneity at villagelevel The resultsfrom this

papershowthatforestcoverchangesaredifferentinruralvillages

that have access to alternative income sources, either from

cardamom cultivation under forest canopy or from tourism

activities.Theseruralvillagesaretypicallycharacterizedbyhigher

ratesoflandabandonmentandlowerratesofdeforestation.Because

ofdiversificationinalternativeeconomicactivities,ruralhouseholds

arebecominglessdependentonnaturalresourcesandagricultural

productsfortheirsurvival.Ourresultssuggestthatthecreationof

off-farmjobsinthetourismsector,constructionormanufacturing

canbeadriverofshiftsincoupledhuman–environmentalchanges

Thetourismactivitiesin SaPacantherefore beconsideredasa

specialpathwaytowardsaforesttransitionwherebythegeneration

ofoff-farmjobshasapositiveimpactonforestconservation.Inour

view,themainchallengeistofinda balancebetweentherapid

development of tourism activities and the preservation of the

authenticsocio-culturalelementsoftheethnicminoritiesthatmake

theareaattractivefortouristsinthefirstplace

Acknowledgements

This researchwas part of the bilateralscientific project on

‘Land-usechangeunderimpactofsocio-economic development

anditsimplicationsonenvironmentalservicesinVietnam’funded

bytheBelgianSciencePolicy(BELSPO)(GrantSPPPSBL/10/V26)

and the Vietnamese Ministry of Science & Technology (MOST)

(Grant42/2009/HÐ-NÐT).PatrickMeyfroidt,IsalineJadin,Francois

Clapuyt have provided valuable suggestions for this research

project We are thankful to all ministries and institutions in

Vietnam which provided the necessary data to undertake this

research.WealsothankvillageleadersandlocalpeopleinSaPa

districtforfacilitatingthefielddatacollection,andtheanonymous

reviewersfortheirvaluableinput

References

Adams, W.M., Aveling, R., Brockington, D., Dickson, B., Elliott, J., Hutton, J., Roe, D.,

Vira, B., Wolmer, W., 2004 Biodiversity conservation and the eradication of

Allison, P.D., 1999 Logistic Regression Using the SAS System: Theory and Applica-tion SAS Institute Inc., Cary, NC.

Atik, M., Altan, T., Artar, M., 2010 Land use changes in relation to coastal tourism developments in Turkish Mediterranean Pol J Environ Stud 19 (1) 21–33.

Balthazar, V., Vanacker, V., Lambin, E., 2012 Evaluation and parametrization of ATCOR3 topographic correction method for forest cover mapping in mountain areas Int J Appl Earth Observ Geoinf 18, 436–450.

Burgess, J.C., Barbier, E.B., 2001 Sustainable development Int Encycl Soc Behav Sci 15329–15335.

Casse, T., Milhøj, A., Ranaivoson, S., Randriamanarivo, J.R., 2004 Causes of defores-tation in southwestern Madagascar: what do we know? For Policy Econ 6 (1) 33–48.

Castella, J.C., Pham, H.M., Kam, S.P., Villano, L., Tronche, N.R., 2005 Analysis of village accessibility and its impact on land use dynamics in a mountainous province of northern Vietnam Appl Geogr 25, 308–326.

Castella, J.C., Tronche, N.R., Nguyen, V., 2002 Landscape changes in Cho Don District during the doi moi era (1990–2000) and their implications for sustainable natural resource management in Vietnam’s mountainous provinces In: Cas-tella, J.C., Dang, D.Q (Eds.), Doi Moi in the Mountains The Agricultural Pub-lishing House, Hanoi, Vietnam, pp 149–173.

Cooper, R., 1984 Resource Scarcity and the H’mong Response: Patterns of Settle-ment and Economy in Transition National University of Singapore Press, Singapore.

Coulibaly-Lingani, P., Tigabu, M., Savadogo, P., Oden, P-C., Ouadba, J-M., 2009.

Determinants of access to forest products in southern Burkina Faso For Policy Econ 11 (7) 516–524.

Dasgupta, S., Deichmann, U., Wheeler, D., 2005 Where is the poverty–environment nexus? Evidence from Cambodia, Lao PDR, and Vietnam World Dev 33 (4) 617–638.

Dong, J., Xiao, X., Sheldon, S., Biradar, C., Duong, N.D., Hazarika, M., 2012 A comparison of forest cover maps in Mainland Southeast Asia from multiple sources: PALSAR, MERIS, MODIS and FRA Remote Sens Environ 127, 60–73.

Dong, R., Yu, L., Liu, G., 2008 Impact of tourism development on land-cover change

in a matriarchal community in the Lugu Lake area Int J Sustain Dev World Ecol 15, 28–35.

Duong, B.H., 2008a Contesting marginality: consumption, networks, and everyday practice among Hmong girls in Sa Pa, Northwestern Vietnam J Vietnamese Stud 3 (3) 231–260.

Duong, B.H., 2008b S u’ thay p oˆ:i ca´c moˆ´i quan h lao p oˆng o:’ m oˆt ba:n H’moˆng t ai Sa

Pa, Taˆy Ba˘´c Vi t Nam (Changing Labor Relations in a Hmong Village in SaPa, Northwestern Vietnam) In: Sikor, T., Sowerwine, J., Romm, J., Nghiem, P.T (Eds.), Nhu~’ngchuyeˆ:n p oˆ:i kinh teˆ´ – xa˜ h oˆi o:’ vu`ng cao Vi t Nam (Upland Transformations) CRES (Trung taˆm Nghieˆn c u ta`i nguyeˆn va` Moˆi Tru’o`’ng,

Ð ai h oc Quoˆ´c gia Ha` N oˆi), Hanoi, Vietnam, pp 85–97.

Etter, A., McAlpine, C., Wilson, K., Phinn, S., Possingham, H., 2006 Regional pattern

of agricultural land use and deforestation in Colombia Agric Ecosyst Environ.

114, 369–386.

FAO, 2006 Global Forest Resources Assessment 2005: Progress Towards Sustain-able Forest Management Rome, http://www.fao.org

FAO, 2010 Global Forest Resources Assessment 2010 FAO, Rome.

Forsyth, T.J., 1995 Tourism and agricultural development in Thailand Ann Tour Res 22 (4) 877–900.

Fox, J., Truong, D.M., Rambo, A.T., Tuyen, N.P., Cuc, L.T., Leisz, S., 2000 Shifting cultivation: a new old paradigm for managing tropical forests Bioscience 50, 521–528.

Frontier Vietnam, 1997 Biodiversity Value of Hoang Lien Mountains and Strategies for Conservation Society for Environmental Exploration, UK and Institute

of Ecology and Biological Resources, Hanoi, Hanoi, Vietnam, pp 68 Report

nr 13.

Frontier Vietnam, 1999 Hoang Lien Nature Reserve – Conservation Evaluation 1997/98 Society for Environmental Exploration, UK and Institute of Ecology and Biological Resources, Hanoi, Hanoi, Vietnam, pp 107 Report nr 17.

Garcı´a-Martı´nez, A., Bernue´s, A., Olaizola, A.M., 2011 Simulation of mountain cattle farming system changes under diverse agricultural policies and off-farm labour scenarios? Livest Sci 137 (1–3) 73–86.

Gaughan, A.E., Binford, M.W., Southworth, J., 2009 Tourism, forest conversion, and land transformations in the Angkor basin, Cambodia Appl Geogr 29 (2) 212–223.

Geist, H.J., Lambin, E.F., 2001 What Drives Tropical Deforestation? A Meta-Analysis of Proximate and Underlying Causes of Deforestation Based on Sub-national Case Study Evidence CIACO, Louvain-la-Neuve, Belgium LUCC Report Series No 4.

Getahun, K., Van Rompaey, A., Van Turnhout, P., Poesen, J., 2013 Factors controlling patterns of deforestation in moist evergreen Afromontane forests of Southwest Ethiopia For Ecol Manage 304, 171–181.

Grainger, A., 2008 Difficulties in tracking the long-term global trend in tropical forest area Proc Natl Acad Sci U S A 105, 818–823.

GSO [General Statistics Office], 1989 National Vietnamese Census, Year book The people committee of Sa Pa district, Lao Cai, Vietnam.

GSO [General Statistics Office], 1995 National Vietnamese Census, Year book The people committee of Sa Pa district, Lao Cai, Vietnam.

GSO [General Statistics Office], 2000 National Vietnamese Census, Year book in period 1995–2000 The people committee of Sa Pa district, Lao Cai, Vietnam.

GSO [General Statistics Office], 2006a National Vietnamese Census, Year book The people committee of Sa Pa district, Lao Cai, Vietnam.

GSO, 2006b Vietnam Rural, Agricultural, and Fishery Census Department of

Agriculture, Forestry and Fishery.

GSO [General Statistics Office], 2010 National Vietnamese Census, Year book The

people committee of Sa Pa district, Lao Cai, Vietnam.

Hardy, A., 2005 Red Hills Migrants and the State in the Highlands of Vietnam

Institute of Southeast Asian Studies (ISEAS), Singapore.

Hietel, E., Waldhardt, R., Otte, A., 2005 Linking socio-economic factors,

environ-ment and land cover in the German Highlands, 1945–1999 J Environ Manage.

75, 133–143.

Hobbs, R., 2001 Agricultural expansion and tropical deforestation: poverty,

inter-national trade and land use Agric Ecosyst Environ 86 (2) 218–219.

Jadin, I., Vanacker, V., Hoang, T.T.H., 2013 Drivers of forest cover dynamics in

smallholder farming systems: The case of Northwestern Vietnam AMBIO 42 (3)

344–356.

Job, H., Paesler, F., 2013 Links between nature-based tourism, protected areas,

poverty alleviation and crises – the example of Wasini Island (Kenya) J Outdoor

Recreation Tour 1 (2) 18–28.

Jodha, N.S., 1998 Common property resources and rural poor in dry regions of India.

Econ Pol Wkly 21, 1169–1181.

Koning, D., 2000 Accessibility: key factor for development In: Nori, M., et al (Eds.),

Drawing Lessons from Experience Proceeding of Workshop on Sustainable

Rural Development in the Southeast Asian Mountainous Region.

Lambin, E.F., Geist, H.J., 2003 Regional differences in tropical deforestation Issue

Environ 45, 22–36.

Le, V.L., 2004 Hoang Lien National Park National Culture Publishing House, Hanoi,

Vietnam.

Leisz, S.J., Ha, N.T.T., Yen, N.T.B., Lam, N.T., Vien, T.D., 2004 Developing a

method-ology for identifying, mapping and potentially monitoring the distribution of

general farming system types in Vietnam’s northern mountain region Agric.

Syst 85 (3) 40–63.

Maneesha, Bajpai, P., 2013 Multiple regression analysis using ANCOVA in

Univer-sity model Int J Appl Phys Math 3 (5) 336–340.

Mather, A.S., 1992 The forest transition Area 24 (4) 367–379.

Meyfroidt, P., 2009 Forest Transition in Vietnam: Evidence, Theory and

Social-Ecological Feedbacks (PhD dissertation)Universite´ Catholique de Louvain,

Louvain La Neuve, Belgium.

Meyfroidt, P., Lambin, E.F., 2008a The causes of the reforestation in Vietnam Land

Use Policy 25 (2) 182–197.

Meyfroidt, P., Lambin, E.F., 2008b Forest transition in Vietnam and its

environ-mental impacts Glob Change Biol 14 (6) 1319–1336.

Michaud, J., 1997 From Southwest China into upper Indochina: an overview of

Hmong (Miao) migration Asia Pacific Viewp 38 (2) 119–130.

Michaud, J., Turner, S., 2000 The Sa Pa marketplace, Lao Cai province, Vietnam Asia

Pacific Viewp 41 (1) 85–100.

Michaud, J., Turner, S., 2006 Contending visions of a hill-station in Vietnam Ann.

Tour Res 33 (3) 785–808.

MONRE, 1993 Aerial Photos Department of Survey and Mapping, Ministry of

Natural Resources and Environment (MONRE).

MONRE, 2006 VHR-SPOT Image 2006 Remote Sensing Center, Ministry of Natural

Resources and Environment, Vietnam (MONRE).

MONRE, 2009 Topographic Maps at the Scale 1:50,000 Ministry of Natural

Resources and Environment.

MONRE, 2012 VHR-SPOT Image 2012 Remote Sensing Center, Ministry of Natural

Resources and Environment, Vietnam (MONRE).

Mottet, A., Ladet, S., Coque´, N., Gibon, A., 2006 Agricultural land-use change and its

drivers in mountain landscapes: a case study in the Pyrenees Agric Ecosyst.

Environ 114 (2–4) 296–310.

Nyaupane, G.P., Poudel, S., 2011 Linkages among biodiversity, livelihood, and

tourism Ann Tour Res 38 (4) 1344–1366.

Pontius Jr., R.G., Millones, M., 2011 Death to Kappa: birth of quantity disagreement

and allocation disagreement for accuracy assessment Int J Remote Sens 32

(15) 4407–4429.

Ravnborg, H.M., 2003 Poverty and environmental degradation in the Nicaraguan

hillsides World Dev 31 (11) 1933–1946.

Richter, R., 2011 Atmospheric/Topographic Correction for Satellite Imagery— ATCOR-2/3 User Guide, Version 8.0 Switzerland: ReSe Applications Schla¨pfer.

Robinson, J.G., 2006 Conservation biology and real-world conservation Conserv Biol 20 (3) 658–669.

Rudel, T.K., Coomes, O.T., Moran, E., Achard, F., Angelsen, A., Xu, J., Lambin, E., 2005.

Forest transitions: towards a global understanding of land use change Glob Environ Change 15 (1) 22–31.

Scherr, S., 2000 A downward spiral? Research evidence on the relationship between poverty and natural resource degradation Food Policy 25 (4) 479–498.

Sowerwine, J., 2004a The Political Ecology of Dao (Yao) Landscape Transformations: Territory, Gender, and Livelihood Politics in Highland Vietnam (PhD disserta-tion)University of California, Berkeley, America.

Sowerwine, J., 2004b Territorialisation and the politics of highland landscapes in Vietnam: negotiating property relations in policy, meaning and practice Con-serv Soc 2 (1) 98–117.

Symonds, P.V., 2004 Calling in the Soul: Gender and the Cycle of Life in a H’mong Village The University of Washington Press, The United States of America.

Szillassi, P., Jordan, G., Kovacs, F., Van Rompaey, A., Van Dessel, W., 2010 Investi-gating the link between soil quality and agricultural land use change A case study in the Lake Balaton Catchment, Hungary Carpathian J Earth Environ Sci.

5, 61–70.

Teka-Belay, K., Van Rompaey, A., Poesen, J., 2013 Assessing the role of policies on land use change and agricultural development since 1960 in northern Ethiopia Land Use Policy 30, 944–951.

Tugault-Lafleur, C., 2007 Diversifying Livelihoods: Hmong Use and Trade of Forest Products in Northern Vietnam (Master thesis)McGill University, Montreal, Canada.

Tugault-Lafleur, C., Turner, S., 2009 The price of spice: ethnic minority livelihoods and cardamom commodity chains in upland northern Vietnam Singapore J Trop Geogr 30 (3) 388–403.

Turner, S., 2011 Making a living the Hmong way: an actor-oriented livelihoods approach to everyday politics and resistance in upland Vietnam Ann Assoc.

Am Geogr 102 (2) 403–422.

Turner, S., 2012 Forever Hmong: ethnic minority livelihoods and agrarian transi-tion in upland northern vietnam Prof Geogr 64 (4) 540–553.

Van Dessel, W., Van Rompaey, A., Poelmans, L., Szillassi, P., 2008 Predicting land cover changes and their impact on the sediment influx in the Lake Balaton Catchment Landsc Ecol 23, 645–656.

Vanacker, V., Bellin, N., Molina, A., Kubik, P.W., 2014 Erosion regulation as a function of human disturbances to vegetation cover: a conceptual model Landsc Ecol 29, 293–309.

Vanegas, R., Henry, S., 2012 Identifying socio-economic and demographic driving factors at household population level conditioning the local agricultural prac-tices, land use and perception of land degradation in the Paute catchment (Ecuador) In: 5th Conference of the Aso Latinoamericana de Poblacion Proceedings, Uruguay.

Vanonckelen, S., Lhermitte, S., Van Rompaey, A., 2013 The effect of atmospheric and topographic correction methods on land cover classification accuracy Int J Appl Earth Observ Geoinf 24, 9–21.

Vermeiren, K., Van Rompaey, A., Loopmans, M., Serwajja, E., Mukwaya, P., 2012.

Urban growth of Kampala, Uganda: Pattern analysis and scenario development Landsc Urban Plan 106 (2) 199–206.

Vu, K.C., Van Rompaey, A., Govers, G., Vanacker, V., Schmook, B., Nguyen, H., 2013.

Land transitions in Northwest Vietnam: an integrated analysis of biophysical and socio-cultural factors Hum Ecol 41 (1) 37–50.

Wang, J., Liu, Y., 2013 Tourism-led land-use changes and their environmental effects in the southern coastal region of Hainan Island, China J Coast Res 29 (5) 1118–1125.

Zhang, Y., 2000 Deforestation and forest transition: theory and evidence in China In: Palo, M., Vanhanen, H (Eds.), World Forests from Deforestation to Transi-tion? Kluwer Academic Publishers, Dordrecht, p 25.

Zwane, A.P., 2007 Does poverty constrain deforestation? Econometric evidence from Peru J Dev Econ 84 (1) 330–349.