Gradual replacement of wild bees by honeybees in flowers of the Mediterranean Basin over the last 50 years

Abstract

Current evidence for generalized pollinator decline largely originates from mid-latitude regions in North America and Europe. Unacknowledged geographical heterogeneity in pollinator trends, in combination with geographical biases in pollinator studies, can produce distorted extrapolations and limit understanding of pollinator responses to environmental changes. In contrast to the severe declines experienced in some well-investigated European and North American regions, honeybees seem to have increased in recent years over large expanses of the Mediterranean Basin. Since honeybees can have negative impacts on wild bees, the hypothesis was formulated that if honeybees are actually increasing in the Mediterranean Basin, then an extensive, biome-wide alteration in the composition of bee pollinator assemblages may be currently underway there, involving a progressive reduction in the importance of wild bees as pollinators. This hypothesis was tested using a large data sample gathered from published investigations on the composition of bee pollinators of wild and cultivated plants conducted between 1963-2017 in the Mediterranean Basin. Over this period, honeybee colonies increased exponentially and wild bees were gradually replaced by honeybees in flowers of wild and cultivated plants. Mean estimated proportion of wild bees at flowers roughly quadruplicated that of honeybees at the beginning of the period considered, the proportions of both groups becoming roughly similar fifty years later. The Mediterranean Basin is a world biodiversity hotspot for wild bees and wild bee-pollinated plants, and the ubiquitous rise of honeybees to dominance as pollinators could in the long run undermine the diversity of plants and wild bees, as well as their mutualistic relationships in the region.

“El sur también existe”

Joan Manuel Serrat, singer and songwriter

Introduction

The structure and dynamics of ecological communities can vary tremendously across biomes and continents. Critical elements of ecological knowledge will thus be closely tied to the particular location where it is attained, and attempts at generalizations which are based on limited, spatially biased ecological data may produce distorted or erroneous inferences (Martin et al. 2012, Culumber et al. 2019). Unawareness of geographical sampling biases has been pointed out as one possible cause of erroneous extrapolations related to the notions of “pollinator decline” and “pollination crisis” (Ghazoul 2005, Archer 2014, Herrera 2019, Jamieson et al. 2019), two topics that have recently elicited considerable academic and popular interest because of the importance of animal pollination for the reproduction of many wild and crop plants (Ollerton et al. 2014, Senapathi et al. 2015, Breeze et al. 2016, Ollerton 2017). Evidence for the widely held view of a generalized pollinator decline is strongly biased geographically, as it mostly originates from a few mid-latitude regions in Europe and North America (Rodger et al. 2004, Ghazoul 2005, Winfree et al. 2009, Archer 2014, Hung et al. 2018, Nicholson and Egan 2019). Mounting evidence indicates, however, that pollinator declines are not universal; that the sign and magnitude of temporal trends in pollinator abundance may differ among pollinator groups, continents or regions; and that taxonomic and geographical biases in pollinator studies in combination with unrecognized patterns of geographical or taxonomic differences in pollinator trends are bound to limit a realistic understanding of pollinator responses to environmental changes and the causal mechanisms involved (Aizen and Harder 2009, Potts et al. 2010, Hofmann et al. 2018, Herrera 2019, Jamieson et al. 2019, Thomson 2019).

Even for well-studied social bees, data supporting a general decline are geographically biased (Archer et al. 2014, Ollerton 2017, Hung et al. 2018). For example, in thoroughly studied North America and mid-western Europe the number of honey bee colonies has experienced severe declines, but the trend is apparently reversed in the less investigated areas of southernmost Europe, where honeybees seems to be increasing over large expanses in the last few decades (Aizen and Harder 2009: Fig. S1, Potts et al. 2010, vanEngelsdorp and Meixner 2010). Honeybees can have strong negative impacts on wild bee populations in both natural and anthropogenic scenarios (Shavit et al. 2009, Lindström et al. 2016, Torné-Noguera et al. 2016, Herrera 2019, Ropars et al. 2019, Valido et al. 2019). I thus formulated the hypothesis that, if the abundance of managed honeybees has been actually increasing in the Mediterranean Basin over the last decades, then a profound biome-wide alteration in the composition of bee pollinator assemblages could be currently underway there, involving a progressive replacement of wild bees by honeybees in flowers. This paper verifies this hypothesis using data from a large sample of published investigations on the composition of bee pollinators of wild and cultivated plants, conducted during the last 50 years throughout the Mediterranean Basin. Results of this study stress the importance of broadening the geographical scope of current investigations on pollinator trends, while at the same time issue a warning on the perils of uncritically importing to Mediterranean ecosystems honeybee conservation actions specifically designed for the contrasting situations that prevails in temperate-climate European or North American countries.

Material and methods

The data

The literature on floral biology, pollination ecology, plant-pollinator interactions and crop pollination was searched for field studies conducted during 1960-2019 in the Mediterranean Basin and providing quantitative data on the relative abundance of honeybees and wild bees at flowers of insect-pollinated plants, either wild-growing or cultivated. Preliminary searches showed that studies conducted before 1960 quite rarely reported quantitative data on bee abundance at flowers. The literature screening used searches in Web of Science, Google Scholar and my personal database of plant-pollinator studies. To improve the chances of obtaining a representative, geographically comprehensive coverage of all regions surrounding the Mediterranean Sea (i.e., African, Asian and European shores), literature searches were conducted using terms in English, French, Italian, Portuguese and Spanish. For inclusion in this study I considered exclusively field investigations where (1) quantitative data were provided on relative numbers of wild bees and honeybees based on direct visual counts or standardized collections at flowering individuals or flowering patches of single plant species. Investigations at the plant community level or providing semiquantitative or subjective abundance scores of bee abundance were thus excluded; and (2) the year(s) on which bee abundance data had been originally collected in the field was unambiguously stated. In a few publications where information from two or more study years had been pooled into a single estimate of wild bee and honeybee abundances, but the data were otherwise suitable, the average year was used. A total of 336 estimates of wild bee and honeybee abundance at the flowers of 200 plant species were gathered from 136 different literature sources. Each data record corresponded to a unique combination of plant species x sampling year x sampling location. The data had been collected in the field between 1963– 2017 in 13 different countries surrounding the Mediterranean Sea (Fig. 1). Information on plant type (wild-growing vs. cultivated) and taxonomic affiliation (plant family) was also incorporated into the data set.

Fig. 1.

Distribution among 13 circum-Mediterranean countries of the N = 336 published estimates of wild bee and honeybee abundance in flowers of cultivated and wild-growing plants for the period 1963-2017 considered in this study (Table S1, electronic supplementary material).

The complete data set including literature sources is presented in Table S1, electronic supplementary material. Most records originate from Spain, Italy, Algeria and Egypt (159, 59, 33 and 21, respectively; Fig. 1). The median of the distribution of study years was 1996 (interquartile range = 1986-2008). There were 106 and 230 records for cultivated and wild-growing plants, respectively. A total of 54 plant families were represented in the sample, with most species belonging to Fabaceae, Lamiaceae, Asteraceae, Rosaceae and Cistaceae; 51, 34, 32, 30 and 25 records, respectively).

Trends in honeybee abundance in the Mediterranean Basin over the period considered in this study were assessed using information gathered from the Food and Agriculture Organization (FAO) of the United Nations databases (FAOSTAT; http://www.fao.org/faostat). This data source has been used previously in some historical reviews of honeybee abundance (e.g., Aizen and Harder 2009, vanEngelsdorp and Meixner 2010). Number of honeybee colonies per country and year for the period 1963-2017 was obtained from FAOSTAT (accessed 25 September 2019) for each of the 13 Mediterranean countries with estimates of wild bee and honeybee relative abundances in my data set (Fig. 1). Comparable abundance figures were obtained by dividing the number of honeybee colonies by the land surface of the country (obtained also from FAOSTAT), which provided estimates of honeybee colonies/km² per country and year.

Statistical analyses

The original figures of bee abundance at flowers found in the literature were transformed to proportions of wild bees (p_wb) and honeybees (p_hb = 1 - p_wb) relative to all bees combined. For the purpose of statistical analyses, the log-odds that one randomly chosen bee found at flowers was a wild bee rather than a honeybee was estimated for each data record using the logit transformation, logit(p_wb) = log(p_wb/p_hb). Since the logit function is undefined for p = 0 or 1, proportions were remaped to the interval (0.05, 0.95) prior to the transformation.

The null hypothesis that the relative proportions of wild and honeybees at flowers were unrelated to the year on which the data had been collected was tested by fitting a linear mixed effect model. Logit(p_wb) was the response variable, and data collection year (treated as a continuous numerical variable), plant type (two-level factor, wild-growing vs. cultivated) and their interaction were included as fixed effects. Country of origin, plant family and plant species were included as random effects to statistically control for, on one side, the effects of likely taxonomic and geographical correlations in the data and, on the other, the unbalanced distribution of data across countries and plant taxonomic groups. The existence of a long-term trend in honeybee abundance in the Mediterranean Basin as a whole was tested by fitting a linear mixed model to the FAOSTAT colony density data (log-transformed). Year (as a numerical variable) was the single fixed effect, and country was included in the model as a random effect to account for the correlated data of the same country. Linear mixed models allow drawing conclusions on fixed effects with reference to a broad inference space whose scope transcends the specific samples studied (McLean et al. 1991, Bolker 2015). In the present instance, the universe of all countries and plant species in the Mediterranean Basin that could have been sampled for this study represents the broad inference space (Schabenberger and Pierce 2001). Conclusions on long-term trends in honeybee abundance and logit(p_wb), including predicted marginal effects, will thus refer to such inference space.

All statistical analyses were carried out using the R environment (R Core Team 2018). Linear mixed models were fitted with the lmer function in the lme4 package (Bates et al. 2015). Statistical significance of fixed effects was assessed using analysis of deviance-based, Type II Wald chi-square tests using the Anova function in the car package (Fox and Weisberg 2011). The function ggpredict from the ggeffects package (Lüdecke 2018) was used to compute marginal effects of year on logit(p_wb) separately for wild-growing and cultivated plants.

Results

Estimated density of managed honeybee colonies tended to increase steadily over the 1963-2017 period in most Mediterranean countries considered in this paper (Fig. 2). The linear mixed model fitted to colony density data (log-transformed), with year as fixed effect and country as random effect, revealed a highly significant linear effect of year on colony density (Chi-squared = 412.9, P < 10⁻¹⁶). The estimated linear trend for the whole Mediterranean Basin obtained from this model is depicted by the blue line in Fig. 2. Linearity of the estimated relationship on the logarithmic scale reveals an exponential increase in the density of honeybee colonies in the region over the period considered.

Fig. 2.

Variation over 1963-2017 in density of honeybee colonies in the 13 circum-Mediterranean countries considered in this study (gray lines), and overall relationship for the Mediterranean Basin as a whole (blue line; estimated from parameters obtained by fitting a linear mixed model to the data with country as a random effect). Note the logarithmic scale on vertical axis.

For all years, countries and plant species combined, the logarithm of the ratio between proportions of wild bees and honeybees at flowers [logit(p_wb)] encompassed the whole range of possible values, and there was extensive overlap between cultivated and wild-growing plants (Fig. 3). Wild bees tended to be proportionally more abundant in flowers of wild-growing plants, with mean logit(p_wb) differing significantly between cultivated and wild-growing plants (Chi-squared = 18.96, P = 0.000013, Kruskal-Wallis rank sum test). For all the data combined (“naïve” least-squares regression fitted to the data; Fig. 4A), there existed a statistically significant, negative relationship between logit(p_wb) and year of study (r_s = -0.139, N = 336, P = 0.011, Spearman rank correlation), thus suggesting a declining temporal trend in the importance of wild bees at flowers relative to honeybees over the period considered (Fig. 4A). The reality of this trend was corroborated and strenghtened after accounting statistically for correlations underlying the data and unbalanced distribution across plant types, countries, plant families and plant species.

Fig. 3.

Frequency distribution of logit(p_wb), the logarithm of the ratio between proportions of wild bees and honeybees at flowers, in the N = 336 unique combinations of plant species x sampling year x sampling location considered in this study. Bars to the left and right of the vertical dashed line [logit(p_wb) = 0] correspond to situations of numerical dominance at flowers of honeybees and wild bees, respectively.

Fig. 4.

A. Relationship between logit(p_wb), the logarithm of the ratio between proportions of wild bees and honeybees at flowers, and year of study. Each dot corresponds to a unique combinations of plant species x sampling year x sampling location (N = 336). The black line is the “naïve” least-squares regression fitted to the data, all countries, plant species and plant types (cultivated and wild-growing) combined. B. Mean marginal effects of year on logit(p_wb) for cultivated and wild-growing plants, as predicted from the linear mixed model with country, plant family, and plant species as random effects (Table 1).

Results of the linear mixed model testing for the effect of year of study on logit(p_wb) are summarized in Table 1. After statistically accounting for plant type (wild-growing vs. cultivated), country, plant family and plant species, there was a highly significant negative effect of study year on logit(p_wb). The effect was similar for wild-growing and cultivated species, as denoted by the statistical nonsignificance of the year x plant type interaction. After statistically accounting for the rest of effects in the model, the effect of plant type on logit(p_wb) was only marginally significant (Table 1). Mean predicted marginal effects of year on logit(p_wb), computed separately for wild-growing and cultivated plants, illustrate a linear decline in logit(p_wb) over the study period (Fig. 4B). In 1963, the data-predicted proportion of wild bees at flowers roughly quadruplicated that of honeybees, while the proportions of both groups had become roughly similar in 2017. This long-term replacement of wild bees by honeybees at flowers occurred at similar rates in wild and cultivated plants, as shown by the parallel predicted marginal effects (Fig. 4B).

Table 1.

Summary of results of the linear mixed model testing for the significance of supra-annual variation in logit(p_wb), the log of the quotient between proportions of wild bees and honeybees, in flowers of wild-growing and cultivated plants of the Mediterranean Basin.

Discussion

Previous studies that have examined long-term trends in honeybee colony numbers from a wide geographical perspective have shown that (1) there is not any hint of honeybees declining at a planetary scale, but instead considerable evidence that the total number of colonies is increasing globally and in almost every continent; (2) well documented cases of honeybee decline are few and fairly restricted geographically, being mostly circumscribed to parts of Europe and North America; and (3) in the thoroughly-investigated European continent, severe honeybee declines affect just a few countries (e.g., Austria, Germany, Sweden), while stability or increases predominate in the rest (e.g., Finland, Greece, Spain) (Aizen and Harder 2009, Potts et al. 2010, vanEngelsdorp and Meixner 2010). As an illustration, Fig. 5 depicts the inverse trajectories of honeybee colony density over the last half century in Spain and Germany, two representative countries of Mediterranean and mid-western Europe. The analyses presented in this study show that honeybee colonies have increased exponentially over the last 50 years in the Mediterranean Basin, comprising areas of southern Europe, the Middle East and Northern Africa. The latter two regions are prominent examples of ecologically understudied areas (Martin et al. 2012) and, as far as I know, have been never considered in quantitative analyses of bee population trends. The empirical evidence available, therefore, supports the view that, to the extent that broad extrapolations on “pollinator decline” or “pollination crisis” were inspired or supported by honeybee declines (see, e.g., Ghazoul 2005, Potts et al. 2010, Ollerton 2017, for reviews), such generalizations exemplify distorted ecological knowledge arising from geographically biased data (Ghazoul 2005, Martin et al. 2012, Archer et al. 2014, Culumber et al. 2019).

Fig. 5.

Variation over 1963-2017 in density of honeybee colonies in Germany an Spain (gray lines), based on FAOSTAT data (see text). These two countries were chosen as representatives, respectively, of thoroughly-studied, mid-western, temperate-climate Europe, and insufficiently-studied, southern, Mediterranean-climate Europe. The blue line depict least-squares fitted linear regressions.

Correlative and experimental evidence alike has recently shown that beyond certain density threshold honeybees can have strong negative impacts on wild bee populations in both natural and anthropogenic scenarios (Shavit et al. 2009, Lindström et al. 2016, Torné-Noguera et al. 2016, Ropars et al. 2019, Valido et al. 2019), and that in well-preserved natural areas honeybee absence is associated with substantial long-term increases in wild bee populations (Herrera 2019). Much of the direct or circumstantial evidence on the harmful effects of honeybees on wild bees originated in the Mediterranean Basin, which motivated the hypothesis formulated in this paper of a possible replacement of wild bees by honeybees in the Mediterranean running parallel to the increasing abundance of honeybees. This hypothesis has been tested using literature data from highly heterogeneous sources, and originally collected using an enormous variety of field procedures. The data were also imbalanced with regard to observation year, country of origin or plant species identity, which unavoidably combined to produce a “messy” dataset. Despite these limitations, the data have verified the prediction of a gradual long-term replacement of wild bees by honeybees in flowers of the Mediterranean Basin. This conclusion persisted regardless of whether the hypothesis was tested “naïvely” (i.e., simple correlation on all data pooled) or by fitting a linear mixed model where major sources of data “messiness” were appropriately handled by treating them as random effects. Estimated marginal effects predicted from the mixed model revealed that, on average, the proportion of wild bees at Mediterranean flowers roughly quadruplicated that of honeybees at the beginning of the period considered (logit(p_wb) ∼ 1.5) while fifty years later the proportions of both groups had become roughly similar (logit(p_wb) ∼ 0).

On average, model-predicted importance of wild bees relative to honeybees was slightly lower in flowers of cultivated plants throughout the period considered, a finding that seems logically related to the traditional practice of placing honeybee colonies in the vicinity of orchards or cultivated land to ensure crop pollination. More difficult to interpret is the close similarity between wild and cultivated plants in average replacement rate of wild bees by honeybees in flowers, as denoted by parallel slopes of mean predicted marginal effects of year on logit(p_wb) and the statistical nonsignificance of the year x plant type interaction effect. A tentative interpretation of this finding is that the causal mechanism behind temporal trends in bee composition at flowers was one and the same for cultivated and wild plants, or in other words, that increasing honeybee colony density affected in similarly negative ways to wild bees in flowers from anthropogenous and natural habitats. Irrespective of the causal mechanism accounting for it, however, parallel trends in the decline of wild bees relative to honeybees in wild and cultivated plants corroborate in a broader geographical context previous findings at a regional scale showing that natural Mediterranean habitats are not exempt from the negative impact of increasing honeybee densities in anthropogenous habitats nearby (Magrach et al. 2017).

Results of this study are important because the Mediterranean Basin is a world biodiversity hotspot for both wild bees and wild bee-pollinated plants (Petanidou and Vokou 1993, Dafni and O’Toole 1994, Michener 2000, Petanidou and Lamborn 2005, Harrison and Noss 2017). Predicting the global consequences for the Mediterranean flora of the proportional decline of wild bees as floral visitors documented in this paper will require extensive data, e.g., on the pollinating effectiveness of different groups of bees on different plants. Nevertheless, studies conducted so far on the effectiveness of honeybees and wild bees as pollinators of cultivated and wild species in the Mediterranean Basin have found that wild bees generally are better pollinators than honeybees (Herrera 1987, Obeso 1992, Bosch and Blas 1994, Vicens and Bosch 2000, Potts et al. 2001, Monzón et al. 2004). If these limited findings are corroborated in the future by more extensive investigations, then the gradual replacement of wild bees by honeybees currently underway in Mediterranean flowers could translate into impaired fruit and seed production and, in the case of pollen-limited wild plants, reduced population recruitment.

It does not seem implausible to suggest that, because of its colossal magnitude and spatial extent, the exponential flood of honeybee colonies that is silently taking over the Mediterranean Basin can pose serious threats to two hallmarks of the Mediterranean biome, namely the extraordinary diversities of wild bees and wild bee-pollinated plants (Blondel et al. 2010). The Mediterranean Basin is home to ∼3300 wild bee species, or about 87% of those occurring in the whole Western Palaearctic region (data taken from Discover Life, https://www.discoverlife.org/, accessed 1 November 2019; and Kuhlmann 2019). Large as that percent may seem, it is likely an underestimate given the imperfect knowledge of Mediterranean Africa wild bee faunas. From a conservation perspective, actions advocated for promoting apiculture or enhancing honeybee populations in European regions where the species is actually declining (de la Rúa et al. 2009, Cayuela et al. 2011) should not be transferred uncritically to the Mediterranean Basin, as Fig. 5 should make clear to anyone. In the Mediterranean, such actions would be aiming at the wrong conservation target and, even worse, could be indirectly threatening the unique regional diversity of wild bees, wild bee-pollinated plants and their mutualistic relationships.

Acknowledgements

This study was prompted by the troubling discrepancy between allusions to the honeybees’ impending demise so often found in popular media, and my subjective impression in the field that managed honeybees are quickly displacing wild bees from the flowers of many species in the Iberian Peninsula. Assistance from the Red de Bibliotecas y Archivos del CSIC was essential for procuring old publications from rather oscure sources. I am grateful to Oscar Aguado, Angel Guardiola, Fernando Jubete and Alejandro Martínez Abraín for stimulating discussion, and Mónica Medrano for suggestions on the manuscript. The research reported in this paper received no specific grant from any funding agency.