Assessing the biodiversity and the impact of pollinators on carob production

that was of lower quality. In addition to the most abundant pollinator, the honeybee Apis melli- fera, we report for the first time five different species of wild bees visiting carob flowers, and potentially providing pollination services. We also note the importance of carob flowers as a feeding source for wild bees, especially during periods of low floral availability. Bees represented the majority of insects visiting carob flowers during blooming, with A. mellifera being the most common visitor. Several studies conducted in neighbouring Israel [26], Jordan [45], Spain [30] and Portugal [28] reported A. mellifera as the most frequent visi- tor to carob flowers. Even though we did not observe any beehives in the immediate vicinity of the carob groves, hives are distributed in large numbers throughout the island, which explains the high visitation rates by honeybees. Retana et al. [29] found that flies were the most frequent visitors to carob flowers, whereas we recorded negligible visits by Diptera (Syrphidae) and Lep- idoptera (butterflies). Although carob flowers are visited by nocturnal insect species belonging to Lepidoptera and Neuroptera [26], the current work assessed only diurnal insect visitors. Wild bees were the second most frequent visitor to carob flowers, especially on male trees. Only two prior studies reported wild bees visiting carob flowers. Halictus sp. and Amegilla sp. were recorded from carob flowers in Israel [26] while a companion study by our group reported seven species of wild bees from carob trees in Cyprus: Lasioglossum anellum, Ceylalic-

tus variegatus, Colletes brevigena, C. creticus, C. cyprius, Hylaeus cypricola and H. imparilis. C. Table 2. Statistical analyses for the generalized mixed-effects model evaluating the effect of region, year, week, tree sex, tree side, A. mellifera or wasp presence on wild bee presence in two regions in 2016 and 2017. Fixed Effects df F-value P-value Region (NW Coast, S. Plateau) 1 37 0.07 0.79 Year 1 1550 0.74 0.39 Week 4 1550 1.95 0.10 Tree sex (male or female) 1 37 22.08 < 0.001 Tree side (N, S, E, W) 3 1550 0.62 0.60 Apis mellifera presence 1 1550 59.64 < 0.001 Wasp presence 1 1550 21.83 < 0.001 *SD for Random Effect of Carob Tree (n = 40) 2.82 Residual degrees of freedom 1586 * SD: standard deviation. See Materials and Methods and Results for more information on statistical analyses. https://doi.org/10.1371/journal.pone.0291431.t002 Table 3. Results of statistical analyses for the generalized mixed-effects model evaluating the effect of treatment, region and their interaction, and the number of individual inflorescences per inflorescence group on the number of pods produced in the open and wind pollination treatment. Fixed Effects df χ2-value p-value Treatment (open or wind pollination) 1 52.94 <0.001 Region (NW Coast, S. Plateau, S. Coast) 2 10.13 0.006 Treatment: Region 2 1.13 0.56 No inflorescences per group 1 1.78 0.18 SD* for Random Effect of Tree (n = 60) 0.003 SD for Random Effect of Branch nested within Tree (n = 120) 0.67 Residual degrees of freedom 230 Dispersion parameter for negative binomial family 3.08

* SD: standard deviation. See Materials and Methods and Results for more information on statistical analyses. https://doi.org/10.1371/journal.pone.0291431.t003 PLOS ONE Biodiversity and the impact of pollinators on carobs PLOS ONE | https://doi.org/10.1371/journal.pone.0291431 October 20, 2023 9 / 16