Warm Temperatures Reduce Flower Attractiveness and Bumblebee Foraging

Abstract

(1) Background: Plants attract pollinators using several visual signals, mainly involving the display, size, shape, and color of flowers. Each signal is relevant for pollinators foraging for floral rewards, pollen, and nectar. Changes in floral signals and rewards can be induced by an increase in temperature, drought, or other abiotic stresses and are expected to increase as global temperatures rise. In this study, we explored how pollinators respond to modified floral signals and rewards following an increase in temperature; (2) Methods: We tested the effects of warmer temperatures on bee-pollinated starflower (Borago officinalis, Boraginaceae) and determined the behavior of one of its main pollinators, the buff-tailed bumblebee (Bombus terrestris). We measured visual floral traits (display and size) and rewards (nectar and pollen) for plants cultivated at 21 °C or 26 °C. We investigated bumblebee behavior by tracking insect visits in a binary choice experiment in an indoor flight arena; (3) Results: Plants cultivated at 26 °C exhibited a smaller floral area (i.e., corolla sizes summed for all flowers per plant, 34.4 ± 2.3 cm² versus 71.2 ± 2.7 cm²) and a greater flower height (i.e., height of the last inflorescence on the stem, 87 ± 1 cm versus 75 ± 1 cm) compared to plants grown at 21 °C. Nectar production per flower was lower in plants grown at 26 °C than in plants grown at 21 °C (2.67 ± 0.37 µL versus 4.15 ± 0.22 µL), and bumblebees visited flowers from plants grown at 26 °C four times less frequently than they visited those from plants grown at 21 °C; (4) Conclusions: These results show that warmer temperatures affect floral signals and reduce overall floral resources accessible to pollinators. Thus, the global increases in temperature caused by climate change could reduce plant pollination rates and reproductive success by reducing flower visitation.

Keywords: bumblebees; floral signals; flower size; nectar; plant–pollinator interactions; pollen; temperature rise.

Figure 1

Flower development during anthesis. (a,b) Stage 1, male phase just after flower opening, with pink or purple petals. (c,d) Stage 2, transition to female phase with blue petals and stigma inserted. (e,f) Stage 3, flower with blue petals and protruded receptive stigma. (g,h) Stage 4, wilting flower before petal abscission. Scale bar, 10 mm.

Figure 2

Effects of higher temperature on the visual floral traits of Borago officinalis. (a) Flower height (N = 88); (b) number of open flowers (N = 88); (c) floral area, as defined by the sum of all corolla surfaces per plant (N = 77); and (d) corolla surface (N = 909) in plants grown at 21 °C (left-side plots) or 26 °C (right-side plots). Data are presented as means ± standard errors (SEs) as boxplots and with raw data points included in light gray.

Figure 3

Effects of higher temperature on nectar production in Borago officinalis. (a,b) Nectar volume per flower (N = 178); (c,d) sugar concentration of nectar; and (e,f) total amount of nectar sugar (N = 173 for c–f) at the various flower developmental stages of plants grown at 21 °C (a,c,e) or 26 °C (b,d,f). Data are presented as means ± standard errors (SEs) as boxplots and with raw data points included in light gray.

Figure 4

Effects of temperature on bumblebee visits to Borago officinalis flowers. (a) Number of flowers visited per plant (N = 60); (b) time spent per flower visit (N = 653); (c) number of flowers of plants grown at 21 °C or 26 °C that were visited according to the order of visit in each flight sequence (sum of all bumblebees flight sequences, N = 34 sequences); and (d) proportion of constant flights (from a flower from a plant grown at 21 °C (or 26 °C) to another flower from a plant grown at 21 °C (or 26 °C), divided by the total number of flights) in relation with the proportion of visits to flowers from plants grown at 21 °C. The C_O/E coefficient equals 1 along the black line (a sequence above this line, with a C_O/E larger than 1, will show more constancy than expected by chance).