measurable in well defined units and by avoiding shortterm
population dynamics:
Traits
Examples of functional traits (see Glossary) include basal
metabolic rate, beak size, seed or egg size, nutrient
concentrations and stoichiometries, adult body mass,
frost tolerance, potential photosynthetic rate, and leaf
mass per area with associated fast–slow leaf economics
[14]. To be useful to community ecology, traits should vary
more between than within species and preferably be
measured on continuous scales. Although being interested
in the role of traits in ecology is not new [15], community
ecologists have preferred to emphasize a nomenclatural
approach by focusing on species identities, which has
resulted in a loss of ecological generality [16]. For example
[17], the trait-based statement ‘compact plants with
canopy area !30 cm2 and small or absent leaves are
restricted to marshes with !18 mg gK1 soil P’ is more
useful than the nomenclatural statement ‘Campanula
aparinoides is found only in infertile habitats.’ Statements
about traits give generality and predictability, whereas
nomenclatural ecology tends towards highly contingent
rules and special cases [2,3].
Environmental gradients
Much of recent community ecology ignores the fact that
real communities occur on gradients of temperature,
moisture and soil chemistry. This is justified if we believe
that community properties are determined mainly by
interactions among species, but a major goal of community
ecology is to explain why communities change in a
systematic fashion across space. For example, predicting
the ecological impact of global warming requires an
understanding of how communities are affected by the
environment, which is most easily understood by investigating
variation along gradients.
Interaction milieu
Community ecologists must also address biotic interactions
(e.g. competition, predation, etc.), but the key
question is how. The favored approach since the 1960s
focuses on specific pairs of interacting species, their
population dynamics and assembly into a community
matrix. However, for many communities, interactions are
diffuse [1,18,19], and considering each pairwise interaction
as a separate process is difficult [20]. Thus, we
argue that biotic interactions are best modeled as a milieu
or biotic background with which an organism interacts.
Frequency distributions of traits that are important for a
given type of interaction give an operational definition of
this milieu. For example, a histogram of heights of
individuals at a site gives a good first approximation to a
plant light competition milieu. Competition can often best
be conceptualized as a frequency-dependent game-theoretic
model in which an invader plays ‘against the field’ [21]
of strategies or milieu. We can then ask whether a new
strategy can invade depending on the milieu already
present [22], but the dynamical time-course need not be
treated in detail. Predators, herbivores, pathogens and
mutualists might sometimes be as important as
competition in the interaction milieu, but we argue that,
as with competition, the diffuse, game-theoretic approach
will be most productive.
Performance currency
To explore how trait variation affects performance, we
need a common currency that is comparable across species
and along gradients. It has often been assumed (e.g.
[9,23]) that the population increase rate (e.g. instantaneous
rates of increase, r) is the best such currency. But
we argue that these measures become progressively less
useful as the number of species increases, because they
are hard to measure and are inherently phenomenological
and removed from physiology and other connections to the
environment. We favor performance currencies, such as
energy intake and expenditure (optimal foraging), CO2
intake per leaf dry mass invested (plant physiological
ecology) or seed output (reproductive strategies). Such
performance currencies are undeniably related to population-
dynamical measures (a positive rate of population
increase implies that there is an energy budget surplus)
and, moreover, population dynamics have the benefit of
integrating separate performance currencies (e.g. survival
versus growth). Thus, mapping from performance
measures to population dynamics is an important longterm
goal [7,9,24]. However, until this connection is
understood, we favor a greater emphasis on performance
currencies derived from the processes of acquiring,
allocating and spending energy and mineral nutrients,
because these are closely connected to the physical
environment and to interactions in the interaction milieu.
Returning to fundamental and realized niches
The framework that ties these four themes together into a
coherent theory is the idea of the fundamental versus
realized niche [25]. Current efforts to study fundamental
niches focus on measuring growth or growth surrogates in
relation to environmental variables [i.e. physiological
response curves (PRCs); Figure 1a]. Similarly, current
approaches to realized niches involve habitat modeling
[26] and gradient analysis [27] (Figure 1b), whereas
current models of the transformation from fundamental
to realized niche center on community matrix models [1,9]
and species interactions.
We argue that these independent approaches do not
provide a predictive framework for community ecology.
Most PRCs are nomenclatural and are rarely measured
with respect to traits. Habitat modeling and gradient
analysis provide only an observational, correlative view of
the realized niche, with no indications of the fundamental
niche and interaction milieu mechanisms that precede the
realized niche. Community matrix models and studies of
species interactions typically are not positioned on real
geographical gradients, take the list of potential cooccurring
species as given and do not provide information
about the environmental responses of the species that are
potentially present. For example, these separate
approaches cannot explain why species are not necessarily
most abundant at their fundamental-niche optimum
[28–31]. Similarly, these approaches provide limited
predictive ability if the composition of the interaction
วัด ในหน่วยที่กำหนดไว้ และหลีกเลี่ยง shorttermพลศาสตร์ประชากร:ลักษณะตัวอย่างของลักษณะงานที่ (ดูศัพท์) ได้แก่โรคอัตราการเผาผลาญ จะงอยปากขนาด ขนาดเมล็ดหรือไข่ อาหารความเข้มข้นและ stoichiometries มวลร่างกายผู้ใหญ่ยอมรับน้ำแข็ง อัตราเกิด photosynthetic และใบไม้มวลต่อพื้นที่โดยเชื่อมโยงเศรษฐศาสตร์ใบอย่างรวดเร็ว – ช้า[14] เป็นประโยชน์กับชุมชนนิเวศวิทยา ลักษณะจะแตกต่างกันขึ้นระหว่างภาย ในพันธุ์ และควรเป็นวัดบนเครื่องชั่งน้ำหนักอย่างต่อเนื่อง แม้ว่าจะสนใจบทบาทของลักษณะในนิเวศวิทยาไม่ใหม่ [15], ชุมชนecologists ได้ต้องการเน้นเป็น nomenclaturalวิธี โดยเน้นพันธุ์รหัสประจำตัว ซึ่งมีส่งผลให้เกิดการสูญเสียระบบนิเวศ generality [16] ตัวอย่าง[17], งบตามติด ' กระชับพืชด้วยวิตั้ง! 30 cm2 และเล็ก หรือ ขาดใบจำกัด marshes ด้วย! ดิน gK1 18 mg P' มากขึ้นมีประโยชน์กว่างบ nomenclatural ' Campanulaaparinoides พบเฉพาะในช่วงการอยู่อาศัย ' งบเกี่ยวกับลักษณะให้ generality และแอพพลิเคชัน ในขณะที่นิเวศวิทยา nomenclatural มีแนวโน้มไปทางสูงผูกพันกับกฎและกรณีพิเศษ [2,3]ไล่ระดับสีด้านสิ่งแวดล้อมมากล่าสุดชุมชนนิเวศวิทยาละเว้นความจริงที่ชุมชนที่แท้จริงเกิดขึ้นบนไล่ระดับสีของอุณหภูมิเคมีความชื้นและดิน นี้เป็นธรรมถ้าเราเชื่อว่าที่มีกำหนดคุณสมบัติของชุมชนโดยส่วนใหญ่interactions among species, but a major goal of communityecology is to explain why communities change in asystematic fashion across space. For example, predictingthe ecological impact of global warming requires anunderstanding of how communities are affected by theenvironment, which is most easily understood by investigatingvariation along gradients.Interaction milieuCommunity ecologists must also address biotic interactions(e.g. competition, predation, etc.), but the keyquestion is how. The favored approach since the 1960sfocuses on specific pairs of interacting species, theirpopulation dynamics and assembly into a communitymatrix. However, for many communities, interactions arediffuse [1,18,19], and considering each pairwise interactionas a separate process is difficult [20]. Thus, weargue that biotic interactions are best modeled as a milieuor biotic background with which an organism interacts.Frequency distributions of traits that are important for agiven type of interaction give an operational definition ofthis milieu. For example, a histogram of heights ofindividuals at a site gives a good first approximation to aplant light competition milieu. Competition can often bestbe conceptualized as a frequency-dependent game-theoreticmodel in which an invader plays ‘against the field’ [21]of strategies or milieu. We can then ask whether a newstrategy can invade depending on the milieu alreadypresent [22], but the dynamical time-course need not betreated in detail. Predators, herbivores, pathogens andmutualists might sometimes be as important ascompetition in the interaction milieu, but we argue that,as with competition, the diffuse, game-theoretic approachwill be most productive.Performance currencyTo explore how trait variation affects performance, weneed a common currency that is comparable across speciesand along gradients. It has often been assumed (e.g.[9,23]) that the population increase rate (e.g. instantaneousrates of increase, r) is the best such currency. Butwe argue that these measures become progressively lessuseful as the number of species increases, because theyare hard to measure and are inherently phenomenologicaland removed from physiology and other connections to theenvironment. We favor performance currencies, such asenergy intake and expenditure (optimal foraging), CO2intake per leaf dry mass invested (plant physiologicalecology) or seed output (reproductive strategies). Suchperformance currencies are undeniably related to population-dynamical measures (a positive rate of populationincrease implies that there is an energy budget surplus)and, moreover, population dynamics have the benefit ofintegrating separate performance currencies (e.g. survivalversus growth). Thus, mapping from performancemeasures to population dynamics is an important longtermgoal [7,9,24]. However, until this connection isunderstood, we favor a greater emphasis on performancecurrencies derived from the processes of acquiring,allocating and spending energy and mineral nutrients,because these are closely connected to the physicalenvironment and to interactions in the interaction milieu.Returning to fundamental and realized nichesThe framework that ties these four themes together into acoherent theory is the idea of the fundamental versusrealized niche [25]. Current efforts to study fundamentalniches focus on measuring growth or growth surrogates inrelation to environmental variables [i.e. physiologicalresponse curves (PRCs); Figure 1a]. Similarly, currentapproaches to realized niches involve habitat modeling[26] and gradient analysis [27] (Figure 1b), whereascurrent models of the transformation from fundamentalto realized niche center on community matrix models [1,9]and species interactions.We argue that these independent approaches do notprovide a predictive framework for community ecology.Most PRCs are nomenclatural and are rarely measuredwith respect to traits. Habitat modeling and gradientanalysis provide only an observational, correlative view ofthe realized niche, with no indications of the fundamentalniche and interaction milieu mechanisms that precede therealized niche. Community matrix models and studies ofspecies interactions typically are not positioned on realgeographical gradients, take the list of potential cooccurringspecies as given and do not provide informationabout the environmental responses of the species that arepotentially present. For example, these separateapproaches cannot explain why species are not necessarilymost abundant at their fundamental-niche optimum[28–31]. Similarly, these approaches provide limitedpredictive ability if the composition of the interaction
การแปล กรุณารอสักครู่..
