资源描述:
《the r book references and further reading》由会员上传分享,免费在线阅读,更多相关内容在工程资料-天天文库。
1、28SimulationModelsSimulationmodellingisanenormoustopic,andallIintendhereistodemonstrateafewverysimpletemporalandspatialsimulationtechniquesthatgivetheflavourofwhatispossibleinR.Simulationmodelsareusedforinvestigatingdynamicsintime,inspace,orinbothspaceandtimetogether.Fortemporaldynamicswem
2、ightbeinterestedin:thetransientdynamics(thebehaviourafterthestartbutbeforeequilibriumisattained–ifindeedequilibriumiseverattained);equilibriumbehaviour(afterthetransientshavedampedaway);chaos(random-looking,butactuallydeterministictemporaldynamicsthatareextremelysensitivetoinitialcondi
3、tions).Forspatialdynamics,wemightusesimulationmodelstostudy:metapopulationdynamics(wherelocalextinctionandrecolonizationofpatchescharacterizethelong-termbehaviour,withconstantturnoverofoccupiedpatches);neighbourrelations(inspatiallyexplicitsystemswheretheperformanceofindividualsisdeterm
4、inedbytheidentityandattributesoftheirimmediateneighbours);patterngeneration(dynamicalprocessesthatleadtothegenerationofemergent,butmoreorlesscoherentpatterns).28.1Temporaldynamics:ChaoticdynamicsinpopulationsizeBiologicalpopulationstypicallyincreaseexponentiallywhentheyaresmall,butindivi
5、dualsperformlesswellaspopulationdensityrises,becauseofcompetition,predationordisease.Inaggregate,theseeffectsonbirthanddeathratesarecalleddensity-dependentprocesses,anditisthenatureofthedensity-dependentprocessesthatdeterminesthetemporalpatternofpopulationdynamics.Thesimplestdensity-depen
6、dentTheRBook,SecondEdition.MichaelJ.Crawley.©2013JohnWiley&Sons,Ltd.Published2013byJohnWiley&Sons,Ltd.894THERBOOKmodelofpopulationdynamicsisknownasthequadraticmap.Itisafirst-ordernon-lineardifferenceequation,N(t+1)=λN(t)[1−N(t)],whereN(t)isthepopulationsizeattimet,N(t+1)isthepopulationsize
7、attimet+1,andthesingleparameter,λ,isknownastheper-capitamultiplicationrate.Thepopulationcanonlyincreasewhenthepopulationissmallifλ>1,theso-calledinvasioncriterion.Buthowdoesthesystembehaveasλincreasesabove1?Webeginbysimulatingtimeseriesofpopulationsfordifferentvalue