Multidimensionality of plant defenses and herbivore niches: implications for eco-evolutionary dynamics

Abstract

Plant defenses are very diverse and often involve contrasted costs and benefits.Quantitative defenses, whose protective effect is dependent on the dose, are effective against a wide range of herbivores, but often divert energy from growth and reproduction.Qualitative defenses often have little allocation costs. However, while deterrent to some herbivores, they often incur costs through other interactions within the community (eg, decrease in pollination or attraction of other enemies). In the present work, we model the evolutionary dynamics of these two types of defenses, as well and the evolutionary dynamics of the herbivore niche. We assess the effects of such evolutionary dynamics for the maintenance of diversity within the plant-herbivore system, and for the functioning of such systems under various levels of resource availability. We show that the two types of defenses have different implications. Evolution of quantitative defenses often helps to maintain or even increase diversity, while evolution of qualitative defenses most often has a detrimental effect on species coexistence. From a functional point of view, increased resource availability selects for higher levels of quantitative defenses, which reduces topdown controls exerted by herbivores. Resource availability does not affect qualitative defenses, nor the evolution of the herbivore niche. The growing evidence that plant defenses are diverse in types, benefits and costs has large implications not only for the evolution of these traits, but also for their impacts on community diversity and ecosystem functioning.

Attracting pollinators vs escaping herbivores: eco-evolutionary dynamics of plants confronted with an ecological trade-off

“…To focus on how trophic vs mutualistic interactions shape selection and its ecological consequences, we assumed constant competition rates. Considering similarity-dependent competition would likely favor the maintenance of dimorphism (Kisdi & Geritz 1999; Doebeli & Dieckmann 2000; Loeuille & Hauzy 2018). We show that only 10% of niche-overlap competition may suffice to maintain dimorphism ( Fig.…”

Extinction in complex communities as driven by adaptive dynamics

“…Conversely, when a pathogen (perhaps the same pathogen) kills all but resistant individuals, a previously finite interaction parameter drops to zero. Amending population dynamics equations to include such adaptive changes to interspecies interactions is referred as 'adaptive dynamics' (AD) (Abrams, 1999;Abrams, 2000;Cortez et al, 2020;Dieckmann et al, 1995;Gandon et al, 2008;Gavrilets, 1997;Hochberg & Holt, 1995;Jian et al, 2016;Lehtinen & Geritz, 2019;Lion, 2018;Loeuille & Hauzy, 2018;Marrow et al, 1996;Rosenzweig et al, 1987;. We should caution that AD is applicable only to asexual, well-mixed communities, in the low mutation rate limit.…”

“…While standard LV equations keep A ij unchanging, AD considers parameters A ij ⃗ s that depend on some traits ⃗ s = s 1 , s 2 , … . The specific functional form of A ij ⃗ s depends on the ecological details of the system of interest (Abrams, 1999;Abrams, 2000;Cortez et al, 2020;Dieckmann et al, 1995;Gandon et al, 2008;Gavrilets, 1997;Hochberg & Holt, 1995;Jian et al, 2016;Lehtinen & Geritz, 2019;Lion, 2018;Loeuille & Hauzy, 2018;Marrow et al, 1996;Rosenzweig et al, 1987;. However, the defining aspect of all AD models is that novel strains with differing traits (and therefore different interaction matrix elements) will be introduced, and subsequently, LV equations will govern the fate of these strains, thereby playing the role of natural selection.…”

Extinction in complex communities as driven by adaptive dynamics

“…Amending population dynamics equations to include such adaptive changes to interspecies interactions is referred as "adaptive dynamics" (AD) [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. We should caution that AD is applicable only to asexual, well-mixed communities, in the low mutation rate limit.…”

“…The specific functional form of A ij ( s) depends on the ecological details of the system of interest. [8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. However the defining aspect of all AD models is that novel strains with differing traits (and therefore different interaction matrix elements) will be introduced, and subsequently, LV equations will govern the fate of these strains, thereby playing the role of natural selection.…”