Skip to main content

Relationships among functional traits in experimental microcosms of the protist Tetrahymena thermophila by Nils Svendsen

eli
    • 11 Mar
  • Not accessible
teams
More information

Functional traits are phenotypic traits that affect an organism's performance and shape ecosystem-level processes. However, estimatingfunctionaldiversity(i.e. thesumoffunctionsinducedbythe diversity of functional traits) as ameasure of biodiversity remains difficult inpractice. Oneof themainchallenges is tochoosewhich phenotypictraitsshouldbeconsideredfunctionalandmeasured, since effortandmoneyarelimited. Asonewayofdealingwiththis,Hodgsonetal. (1999) introducedthe ideaof twotypesof traits,withsoft traits thatareeasyandquickto quantify, and hard traits that are directly linked to ecosystem functioningbutdifficulttomeasure. Ifalinkexistsbetweenthetraits, thenonecouldusesofttraitsasaproxyforhardtraitsforaquickbut meaningfulassessmentofbiodiversity.However, this isusuallylimited by two factors: (1) traitsmust be tightlyconnected toallowreliable predictionof oneusingtheother; (2) therelationshipbetweentraits mustbemonotonicandlinear tobedetectedbythemostcommonly usedstatisticaltechniques(e.g. linearmodel,PCA). Followingthatlogic,myaimduringthisthesiswastotestthepresence of suchrelationshipsby focusingonsix functional traitsof theprotist speciesTetrahymenathermophila. Inthefirstexperiment, I testedthe presenceoftheserelationshipsusinglinearandnon-linearrelationship detectionmethods in a stable environment. Then, in the second experiment, I testedhowtheserelationshipswerevaryingalongtwo environmentalgradientsusingsimilardetectionmethods. Bothtimes, thetraitswereprovedtoberatherindependent, indicating that eachrepresents adistinct aspect of functional diversity for this organism,andahighnumberofnon-linear relationshipsandpatterns betweenthetraitsweredetected,highlightingtheneedtobecareful aboutwhat statistical techniquesoneuses toestimate relationships betweentraits.

  • Monday, 11 March 2024, 08h00
    Monday, 11 March 2024, 17h00