Positive indirect effects outweigh negative direct effects of ungulate grazers on population growth of a grassland herb

Herbivores can affect plant population growth both directly through the damage they inflict on a focal species and indirectly by moderating conditions for plant recruitment, nutrient cycling, competition, and other biotic interactions, for example, through trampling, defecation, and grazing on surrounding vegetation. Still, the relative importance of direct vs. indirect effects of herbivores on plant vital rates (establishment, survival, fecundity), size, and population growth rate is poorly known. We quantified direct and indirect effects of ungulate grazers on the population growth rate of the short-lived perennial herb Primula farinosa, using integral projection models based on demographic data collected over 7 years in exclosures and open control plots in nine grassland populations in southern Sweden. Grazers had negative direct effects on P. farinosa population growth rate, but these were, on average, more than balanced by positive indirect effects. As predicted, the strength of the direct negative effect tended to increase with grazing intensity. The positive indirect effect was mainly linked to improved conditions for plant recruitment and was strongest in populations where vegetation height differed most between exclosure and control. Simulations indicated that indirect effects of ungulate grazers on population growth rate via interactions with pollinators, seed predators, and small herbivores were weak. Our study illustrates how both the overall, direct, and indirect effects of grazing on plant population growth rate can be identified and quantified, and thereby provides a more complete understanding of how grazers influence plant fitness, abundance, and distribution. Such insight will be crucial for predictions of the effects of environmental change on population viability and the management of declining species.

Methods
This data set includes data from a demographic study conducted from 2010 to 2016 in control plots and exclosures in nine populations of the short-lived perennial herb Primula farinosa in alvar grasslands on the island Öland, southern Sweden. During annual censuses, we recorded the size, survival, reproductive status (vegetative or flower-producing), flower production, and grazing damage to individually marked and mapped plants in 1 m2 squares. For flowering plants escaping grazing, we in addition recorded the total number of fruits produced, and the number of fruits consumed by the seed predator. In addition, we marked and mapped any new recruits. The data have been analyzed with code produced in R. Files with code are deposited with the data set.