Fire effects on pollination and plant reproduction: a quantitative review

Abstract

Background and aims: Fire may favour plant flowering by opening up the vegetation and increasing abiotic resource availability. Increased floral display size can attract more pollinators and increase absolute fruit and seed production immediately after a fire. However, anthropogenic increases in fire frequency may alter these responses. We aim to assess the effects of fire on the pollination and reproductive success of plants at the global scale.

Methods: We performed a systematic literature review and meta-analyses to examine overall fire effects as well as different fire parameters on pollination and on plant reproduction. We also explored to what extent the responses vary among pollinators, pollination vectors, plant regeneration strategies, compatibility systems, vegetation types and biomes.

Key results: Most studies were conducted in fire-prone ecosystems. Overall, single fires increased pollination and plant reproduction but this effect was overridden by recurrent fires. Floral visitation rates of pollinators were enhanced immediately following a wildfire, and especially in bee-pollinated plants. Fire increased the absolute production of fruits or seeds but not fruit or seed set. The reproductive benefits were mostly observed in wind-pollinated (graminoids), herbaceous and resprouter species. Finally, fire effects on pollination were positively correlated with fire effects on plant reproductive success.

Conclusions: Fire has a central role in pollination and plant sexual reproduction in fire-prone ecosystems. The increase in the absolute production of fruits and seeds suggests that fire benefits to plant reproduction are probably driven by increased abiotic resources and the consequent floral display size. However, reproduction efficiency, as measured by fruit or seed set, does not increase with fire. In contrast, when assessed on the same plant simultaneously, fire effects on pollination are translated into reproduction. Increased fire frequency due to anthropogenic changes can alter the nature of the response to fire.

Keywords: Angiosperms; anthropic changes in fire regime; fire ecology; flowering plants; meta-analysis; plant fitness; plant sexual reproduction; plant–pollinator interaction; reproductive success.

Fig. 1.

Distribution of studies evaluating fire effects on pollination and plant reproductive success. Studies reporting data on pollination, plant reproduction and both simultaneously are indicated with different colours. The number of effects per study is indicated with different circle sizes.

Fig. 2.

Effect of fire on the pollination process. The weighted‐mean effect sizes and 95% bias‐corrected confidence intervals of overall (A), pollination variables (B), fire frequency (C), post-fire time (D), fire type (E), pollinator group (F) and vegetation type (G). Mean estimates with confidence intervals that do not overlap the vertical dotted line (Hedge’s d = 0) have a significant positive effect (black circles). White circles indicate non-significant effects. Sample sizes (n) for each category are shown on the right side of the panels. The size of each individual effect is proportional to its weight or contribution to the overall mean calculation (1/s.e.). Full statistical information is provided in Tables S2 and S3.

Fig. 3.

Effect of fire on plant reproductive success. The weighted‐mean effect sizes and 95% bias‐corrected confidence intervals of overall (A), reproduction variables (B), fire frequency (C), post-fire time (D), fire type (E), pollination vector (F), life form (G) and post-fire regeneration strategy (H). Mean estimates with confidence intervals that do not overlap the vertical dotted line (Hedge’s d = 0) have a significant positive (black circles) or negative (green circles) effect. White circles indicate non-significant effects. Sample sizes (n) for each category are shown on the right side of the panels. The size of each individual effect is proportional to its weight or contribution to the overall mean calculation (1/s.e.). Full statistical information is provided in Tables S2 and S3.

Fig. 4.

Effect of fire on fruit set or seed set (A) and on fruit or seed number (B) for self-incompatible (SI) and self-compatible (SC) plant species; and effect of fire on reproductive success of plant species in different vegetation types (C). Mean estimates with confidence intervals that do not overlap the vertical dotted line (Hedge’s d = 0) have a significant positive (black circles) or negative (green circles) effect. White circles indicate non-significant effects. Sample sizes (n) for each category are shown on the right side of the panels. The size of each individual effect is proportional to its weight or contribution to the overall mean calculation (1/s.e.).

Fig. 5.

Relationship between fire effects on pollination and on plant reproductive success (fruit set or seed set). Each point represents a unique plant species where fire effects were assessed simultaneously on both the pollination process and fruit set or seed set as a measure of female reproductive success. The relationship is positive and significant (Pearson’s r = 0.489, P = 0.046, N = 17 effect size pairs, representing 15 plant species from 13 studies).