Sensor networks are playing an increasingly important role in ecology.Continual advances in affordable sensors and wireless communication are making the development of automated sensing systems with remote communication (i.e.,sensor networks) affordable for many ecological research programs (Porter et al.2005)[1].These in situ instruments provide high-frequency data of key variables that previously were measured intermittently and by hand.A number of federal research organizations have realized the potential of environmental sensor networks, and large-scale initiatives are under development.Independent of these initiatives,small sensor networks have emerged to meet the needs of the individual or small teams of ecologists.Ecologists are entering (or already have entered,in some cases) an era in which high temporal and spatial resolution in situ measurements are generating data at unprecedented rates.The use of sensor networks will dramatically increase the volume of ecological data generated in the next decade.
Sensor networks are playing an increasingly important role in ecology.Continual advances in affordable sensors and wireless communication are making the development of automated sensing systems with remote communication (i.e.,sensor networks) affordable for many ecological research programs (Porter et al.2005)[1].These in situ instruments provide high-frequency data of key variables that previously were measured intermittently and by hand.A number of federal research organizations have realized the potential of environmental sensor networks, and large-scale initiatives are under development.Independent of these initiatives,small sensor networks have emerged to meet the needs of the individual or small teams of ecologists.Ecologists are entering (or already have entered,in some cases) an era in which high temporal and spatial resolution in situ measurements are generating data at unprecedented rates.The use of sensor networks will dramatically increase the volume of ecological data generated in the next decade.
全 文 ：第27卷第5期
2008年lO月
生态科学
EcologicalS ience
27(5)：300-302
Oct．2008
NewecologicalinsightsthroughtheGlobalLake
EcologicalObservatoryNetwork(GLEON)
PaulC．Hanson
UniversityofWisconsin,Centerfo，_Limnology
ABSTRACT
Sensornetworksareplayingallincreasingly
importantroleinecology．Continualadvancesin
afrordablesensorsandwirelesscommunicationare
makingthedevelopmentofau omatedsensing
systemswithremotecommunication(i，ese sor
networks)affordableormanyecologicalresearch
programs(Portereta1．2005)【l】．Thesein situ
instrumentsprovidehigh—frequencydataofkey
variablesthatpreviouslyweremeasuredintermittently
andbyhand．Anumberoffederalresearch
organizationshaverealizedthepotentialof
environmentalsensornetworks，andlarge-scale
initiativesar underdevelopment．Independentof
theseinitiatives，smallsensornetworkshaveemerged
tomeettheneedsoftheindividualorsmallteamsof
ecologists．Ecologistsarentering(oralreadyhave
entered，insomecases)anerainwhichhightemporal
andspatialresolutionin situmeasurementsare
generatingdatatunprecedentedrates．Theusof
sensornetworkswilldramaticallyncreasethevolume
ofecologicaldatageneratedin henextdecade．
Thefocustodateonthetechnologyofsensor
networkshascauseddatagatheringcapacitytoleap
aheadofthemodelsandquestionsrequiredto xploit
thesedata．Ecologicalrese rchsaparadigmcanbe
visualizedas theinextricablelinksbetween
observations，models，andquestions．Whenanyon
nodeintheparadigmispushedtoanewtimeorspace
domain，theo rtwomustfollow．Sensornetworks
havepushedobservationstoanewdomaininwhich
high—frequencydatarecollectedOVerextended
spatialextents，requiringUStoexplorenewaysof
modelingecosystemsandchallengingUStoidentify
themost．compellingscientificquestionsgiventhese
newdata．Tofacilitatethisdevelopment，weneedto
improveecologicaldiscussionandtransferofideas
amongecologistsandbetweenecologistsand
informationtechnologyexperts．Integrationac oss
theseareasisextremelydifficultfortraditionalsmall
researchteamsbecausefewgroupshaveallofthe
requiredexpertise．Furthermore．researchatthe
regionalorglobalscaleswillrequireconnecting
individualnetworks．Thislevelofresearchrequiresa
collectionofscientificexpertise，models，diverse
approachesforcapacitybuilding，andinformation
tech ologythatistypicallyscatteredamongdisparate
researchprogramsindifferentfields．Thesec allenges
arecommont allsub-disciplinesofecology．
111escientificcommunityoflakeecologistsis
well—poisedtoa dressthesechallengesbecause
wirelesssensortechnologiesthatm asurekeylake
variablesarebeingdevelopedand／ordeployed
independentlybymanysmallresearchgroups
worldwide．Coordinatingtheselakecologistswi h
informationtech ologistswillacceleratetherateat
whichnewsensornetworktechnologyisintegrated
withkeyquestionsa dmodelsto increase
understandingoflakedynamicsatlocaltoglobal
scales．
Understandinghowchangesin land·use，
humanpopulation，andclimateinteractwi hlake
dynamicsatlocal，regional，continental，andglobal
scalesi oneofthegreatestchallengesforlake
ecologistsoverthenextdecade．Developingthis
万方数据
．5期PaulC．Hanson．NewecologicalinsightsthroughtheGlobalLakeEcologic——al——Obse—r—va——to—r—yNetwork(GLEON)———————————————3——．0．—1．
understandingtsuchscalesisdaunting，butismade
easierbyanumberof ecentdevelopments．First，
sensorscapableofmeasuringkeyfeaturesoflakes，
suchaswatertemperature，watermovem nt，dissolved
gases，pH，conductivity，andchlorophyllfluorescence
havebeendevelopedv rrecentdecadesandare
beingdeployedfora varietyofscientificand
managementobjectives．Second，advancesin
eyberinfrastructure，suchaswirele ssensornetworks，
haveledtotheincreasingprevalenceof／nsitu
continuousmea urementsiIllakeworldwide．Third,
duringthepastdecadeanincreasedimportancehas
beenplacedonunderstandingthecouplingofphysical
andbiologicallakeprocesses，forexample，how
circulationpatterns，internalwavesandstream
intrusionsinfluenceutrientcycling，lake·wide
metabolism,andthewax waneofalgalb oomsin
lakes．Asaresultofheseadvancesandinparticular
theimprovementsindatainputtosimulationmodels，
thereis greaterpotentialtopredicthowlake
ecosystemsr spondtonaturalandanthropogenic
mediatedev nts．
Todevelopecologicalstudiesr levanttothe
world’sdiversityoflakesandtoaddresstheroleof
lakesinglobalphenomena，anintemationalnetwork
oflakeswithsensorshaemergedastheGlobalLake
EcologicalObservatoryNetwork(GLEON；Kratzeta1．
2006)[21Scientificssuescriticaltosociety，sucha
thechangeinthequalityandquantityoffreshwater
resourcesandtheimportanceoflakesandreservoirs
inregionalandglobalcarbonbalancestranscend
nationalboundaries．Understandingdynamicsof
importantl keprocesses，suchaslakemetabolism,
canbenefitimmenselyfromcomparatives udiesof
lakesthathavedifferentclimatic，geologic，
morphometric，hydrologic，andculturalinfluences．
Strongpartnershipsamonglakescientists，engineers，
computerscientists，educators，andinf mation
technologyandmanagementexpertsfrommultiple
institutionshroughouttheworldarcmakingthe
visionofaglobalnetworkflakeobservingsystemsa
reality．TheoverarchingmissionofGLEONisto
buildaninternationalmultidisciplinarycommunityof
researcherstoadvancethegloballakesciencemade
possibleya globallakecologicalobservatory
n twork．
Examplesofnewscience
lakesensornetworks：
Example1：Lakemetabolism．Lake
metabolismisthebalancebetweenprocessesofgross
primaryp oduction(GPP)andecosystemrespiration
(R)，andhastraditionallybeendescribedbymass
balanceofdissolvedoxygen，bothaproduct(ofGPP)
andasubstrate(ofR)ofmetabolism(Odum1956)‘引．
Quantifyingmetabolismatthecosystemcaleisvery
difficult，becauseitcanotbeobserveddirectlybut
mustbeinferredfromeasurementsofrelated
variables．Traditionalapproachesthatrelyonchanges
inbiomassorbottlePPandRareproblematicbecause
hydrologic，thermal，chemical，andphysical
complexityinlakesconfoundextrapolationfrom
experimentstothewholeecosystem．Sensornetwork
datacanhelp．Changesinfr e—waterdissolvedoxygen
(DO)，asmeasuredbysensors，integratelake
complexityg vingusa beRerunderstandingof
me abolicbalanceattheecosystemcale．Results
indicatethatlakestendtohaveRinexcessofGPP，
leadingtonetheterotrophyandlakesactingasnet
sourcesofC02totheatmosphere[41(Hansoneta1．
2003)．
Example2：Multi-scaleanalysis．Ecological
datarcnotoriouslycomplex，showingimportant
patternatmultiplespaceandtimescales．The
inferenceswedrawfromecosystemsdependsonthe
scalesatwhichwemakeourmeasurementsand
developourmodels．Dissolvedoxygeninlakesi an
excellentexampleofhowmultiplerocessescontrol
thevariable，dependingonthescale．Atshorttime
scales，intemalwaves。horizontalandveaicalw ter
万方数据
302 生态科学EcologicalScience 27卷
movementscontrolmeasurements．Atdailytime
scales，GPPandRcontrolDO，andatweeklyto
monthlyscales．weatheriSimportant．AtdecadaI
scales，fluctuationsintemperatureinducedbyclimate
variabilitycontrolDO．Newanalyticaltechniques，
suchaswavelettransforms(TorrenceandCampo
1998)【5i】，allowUStoextractdatabyscaleproviding
fornewinsights．
Example3：Surprises：Sometimes．itisnot
thetrendsorrepeatingpatternsindatathataremost
important，butthesurprisingchanges．Forexample，
theonsetofalgalb ooms，crashesinfishstocks，or
abruptchangeslong-termicecoverCallOCCur
relativelyrapidlyandmaybedifficultorimpossibleto
predict，yetremedialactiondemandsthatwerespond
rapidly．SurpriseTheoryprovidesanewconstructin
whichtoviewdatandetectunexpectedchange，
evenwhenlittlehistoricaldataexist0 tiandBaldi
20051f6】．Withinthecontextofsensornetworks，
SurpriseTh oryisbeingusedtOdetectsensor
malfunctionandshowspromisefordetectingrapid
changesinwaterqualityvariables，suchaschlorophyll
concentration．
Summary：Aquaticecologyis evolving
rapidlyasnewtechnologiesprovideforobservations
andmodelsunavailableevenadecadeago．Yet，we
areonlybeginningtounderstandheinformation
contentofsensornetworkdataandexploititsvalue
forgaininecologicalnference．AttimeswhenOur
freshwaterresourcesarunderunrelentingpressures
leadingtorapiddegradationinwaterquality，weneed
morethaneverapidassessmentandremediation．
Thiswillonlyhappenwithaninfusionofyoung,
talentedscientistswhoarewillingtocrosstraditional
disciplinary，national，andculturalboundariestos ek
novelapproachestostudyinga dunderstandingOur
ecosystems．Thecombinationofnewtechnologiesand
newideas，drivenbyn wscientists，holdsthepromise
forabetterfutureforOurfreshwaterresources．
REFERENCES
1111 Porter，J．，P．Arzberger，P．Hanson，T．Kratz，S．
Gage，T．Williams，S．Shapiro，P．Bryant，F．Lin，
H．King，T．Hansen，H．Braun，W．Michener．
2005．Wirelesssensornetworksforecology．
Bioscience．55：561 572．
12JKratz，T．K．，P．Arzberger，B．J．Benson，C．Y．
Chiu，K．Chiu，L．Ding，T．Fountain，D．Hamilton，
P．C．Hanson，Y．H．Hu，F．P．Lin，D．F．McMullen，
S．Tilak,C．Wu．2006．TowardsaGlobalLake
EcologicalObservatoryNetwork．Publications
of the KarelianI stitute145：51．63．
(http：／／www．gleon．org／GleonKratzetal2006．p
固
’
13l Odum,H．T．1956．Primaryproductionin
flowingwaters．Limn01．Oceanogr．1：103—17．
14l Hanson，P．C．，Bade，D．L．，Carpenter,S．R．，and
T． K． Kratz．2003．Lakemetabolism：
Relationshipswithdi solvedorganiccarbonand
phosphorus．Limn01．Oceanogr．48：1112-11 9．
f515 Torrence，C．，andG．P．Compo．1998．Apractical
guideto waveletanalysis．Bulletinofthe
AmericanMeteorologicalSoc ety．79：61-78．
161 IttiLandBaldiP．(2005)Aprincipledapproach
todetectingsurprisingeventsinvideo．In：Proc．
IEEEConferenceo ComputerVisiona d
PattemRecognition，Volume1．CVPR．
171Hanson，P．C．Agrassrootsapproachtosensor
andscienceetworks．2007．FrontiersinEcology
andtheEnvironment．5(7)：343．
万方数据
New ecological insights through the Global Lake
Ecological Observatory Network (GLEON)
作者： Paul C.Hanson
作者单位： University of Wisconsin,Center for Limnology
刊名： 生态科学
英文刊名： ECOLOGICAL SCIENCE
年，卷(期)： 2008,27(5)
参考文献(7条)
1.Odum,H.T Primary production in flowing waters 1956
2.
Kratz,T.K;P.Arzberger;B.J.Benson;C.Y.Chiu,K.Chiu,L.Ding,T.Fountain,D.Hamilton,P.C.Hanson,Y.H.Hu,F.P.
Lin,D.F.McMullen,S.Tilak,C.Wu Towards a Global Lake Ecological Observatory Network 2006
3.Hanson,P.C A grassroots approach to sensor and science networks[外文期刊] 2007(07)
4.Itti L;Baldi P A principled approach to detecting surprising events in video 2005
5.Torrence,C;G.P.Compo A practical guide to wavelet analysis[外文期刊] 1998
6.Hanson,P.C;Bade,D.L;Carpenter,S.R;T.K.Kratz Lake metabolism:Relationships with dissolved organic
carbon and phosphorus[外文期刊] 2003
7.
Porter,J;P.Arzberger;P.Hanson;T.Kratz,S.Gage,T.Williams,S.Shapiro,P.Bryant,F.Lin,H.King,T.Hansen,H.B
raun,W.Michener Wireless sensor networks for ecology[外文期刊] 2005(7)
本文链接：http://d.g.wanfangdata.com.cn/Periodical_stkx200805006.aspx