Evaluating predictive performance of statistical models explaining wild bee abundance in a mass‐flowering crop

Abstract

Wild bee populations are threatened by current agricultural practices in many parts of the world, which may put pollination services and crop yields at risk. Loss of pollination services can potentially be predicted by models that link bee abundances with landscape-scale land-use, but there is little knowledge on the degree to which these statistical models are transferable across time and space. This study assesses the transferability of models for wild bee abundance in a mass-flowering crop across space (from one region to another) and across time (from one year to another). The models used existing data on bumblebee and solitary bee abundance in winter oilseed rape fields, together with high-resolution land-use crop-cover and semi-natural habitats data, from studies conducted in five different regions located in four countries (Sweden, Germany, Netherlands and the UK), in three different years (2011, 2012, 2013). We developed a hierarchical model combining all studies and evaluated the transferability using cross-validation. We found that both the landscape-scale cover of mass-flowering crops and permanent semi-natural habitats, including grasslands and forests, are important drivers of wild bee abundance in all regions. However, while the negative effect of increasing mass-flowering crops on the density of the pollinators is consistent between studies, the direction of the effect of semi-natural habitat is variable between studies. The transferability of these statistical models is limited, especially across regions, but also across time. Our study demonstrates the limits of using statistical models in conjunction with widely available land-use crop-cover classes for extrapolating pollinator density across years and regions, likely in part because input variables such as cover of semi-natural habitats poorly capture variability in pollinator resources between regions and years.

Mass-flowering monoculture attracts bees, amplifying parasite prevalence

“…In agricultural systems, animals dependent on floral resources will spatially and temporally track bloom events [8]. Repeated mass-bloom events have been shown to increase animal population sizes and species richness [4,[8][9][10]. This may result in host aggregation at a resource-increasing exposure between infected individuals and increasing parasitism (amplification) [11].…”

Characterization Factors to Assess Land Use Impacts on Pollinator Abundance in Life Cycle Assessment

“…For example, annual crops, which are usually intensively managed, were linked in several studies to the lowest expected abundance and richness of pollinator communities, , while natural grasslands were commonly found to harbor the highest abundance rates, helping increase species richness in comparison with annual crops . However, it is important to notice that the method proposed in this study is based on averaged relative values and may not always be comparable to results from local measurements or predictions performed in a site-specific area . Moreover, the high divergence observed for multiple low-abundance estimates may highlight the need for further field and on-site research to verify the state of pollinator communities and allow for a better comparison of relative differences.…”

“…When dealing with data derived from expert elicitation methods, there are generally three main sources of uncertainty. These are generally described as within-expert uncertainty, between-experts uncertainty, and the uncertainty that can be attributed to the data itself (e.g., due to real heterogeneity, misclassifications, etc. ). , Within-expert uncertainty occurs when an expert is unsure about the state or assessed quality of a particular land category (described as well as imperfect knowledge).…”

“… 8 However, it is important to notice that the method proposed in this study is based on averaged relative values and may not always be comparable to results from local measurements or predictions performed in a site-specific area. 42 Moreover, the high divergence observed for multiple low-abundance estimates may highlight the need for further field and on-site research to verify the state of pollinator communities and allow for a better comparison of relative differences. While no confidence scores were provided for low and high estimates, the consistently high divergence of scores for low-abundance estimates could indicate intrinsic regional and management variations or a general lack of certainty and knowledge regarding the extent of pollinator abundance decrease in poor-quality areas and intensively managed landscapes.…”

A novel approach for surveying flowers as a proxy for bee pollinators using drone images

“…Topography, meteorological condition and the right timewindow can also influence negatively the image acquisition and the successively correlation analysis. When applying the methodology explained in our study, it must be clear whether relationships determined in one area or year can be generalized to other areas and/or years (Blasi et al, 2021). If UAV based images indicate high flower cover, does that mean that a site is suitable for bees throughout the growing season or just during a few days or weeks before or after the day the image was obtained.…”