Bridging the Gap between Field Experiments and Machine Learning: The EC H2020 B-GOOD Project as a Case Study towards Automated Predictive Health Monitoring of Honey Bee Colonies.

Honey bee colonies have great societal and economic importance. The main challenge that beekeepers face is keeping bee colonies healthy under ever-changing environmental conditions. In the past two decades, beekeepers that manage colonies of Western honey bees (Apis mellifera) have become increasingly concerned by the presence of parasites and pathogens affecting the bees, the reduction in pollen and nectar availability, and the colonies' exposure to pesticides, among others. Hence, beekeepers need to know the health condition of their colonies and how to keep them alive and thriving, which creates a need for a new holistic data collection method to harmonize the flow of information from various sources that can be linked at the colony level for different health determinants, such as bee colony, environmental, socioeconomic, and genetic statuses. For this purpose, we have developed and implemented the B-GOOD (Giving Beekeeping Guidance by computational-assisted Decision Making) project as a case study to categorize the colony's health condition and find a Health Status Index (HSI). Using a 3-tier setup guided by work plans and standardized protocols, we have collected data from inside the colonies (amount of brood, disease load, honey harvest, etc.) and from their environment (floral resource availability). Most of the project's data was automatically collected by the BEEP Base Sensor System. This continuous stream of data served as the basis to determine and validate an algorithm to calculate the HSI using machine learning. In this article, we share our insights on this holistic methodology and also highlight the importance of using a standardized data language to increase the compatibility between different current and future studies. We argue that the combined management of big data will be an essential building block in the development of targeted guidance for beekeepers and for the future of sustainable beekeeping.

Is the effectiveness of policy-driven mitigation measures on carabid populations driven by landscape and farmland heterogeneity? Applying a modelling approach in the Dutch agroecosystems.

The growing challenges of protecting biodiversity in agro-ecosystems and maintaining high agricultural productivity has become an important issue within the European Union, shaping both European and national agro-policies. The presented study is part of a broader evaluation of the interim targets of the 2013 Dutch policy plan on sustainable use of pesticides, carried out in 2019 by the PBL (Planbureau voor de Leefomgeving) Netherlands Environmental Assessment Agency. We aimed to assess the effectiveness of selected mitigation measures suggested in the policy plan on non-target terrestrial arthropods using a common carabid beetle Bembidion lampros as a model species. We combined dynamic landscape models with detailed agent-based population modelling to simulate impacts of reduction of toxicity of insecticides, reduction of spray drift to the off-crop area, and increase in area of field margins on the beetle population dynamics in ten agricultural landscapes representing different farming systems. Our simulations showed that a shift towards low-risk products should be the priority if the goal is to increase beetle range. To promote local beetle abundance this needs be coupled with increasing amount of field margins in a landscape. Overall, the observed treatment and landscape effects were highly context-specific and therefore we suggest that care is used when defining and interpretting metrics based on population effects of policy measures. This caveat notwithstanding, the use of simulation to assess complex interactions between landscape, ecology and behaviour of species, and policy measures can be a powerful tool supporting innovative policy management. This should include the development of landscape-context specific targets and/or mitigation measures.

The need for weeds: Man-made, non-cropped habitats complement crops and natural habitats in providing honey bees and bumble bees with pollen resources.

In Europe, honey bees and bumble bees are among the most important pollinators, and there is a growing interest in understanding the effects of floral resource availability on their survival. Yet, to date, data on nectar and pollen supplies available to bees in agricultural landscapes are still scarce. In this paper, we quantify species-, habitat- and landscape-scale pollen production in the Lublin Upland, SE Poland. The production per unit area was highest (mean = 2.2-2.6 g/m2) in non-forest woody vegetation, field margins and fallows, whilst significantly lower pollen amounts were shown to be available in road verges and railway embankments (mean = 1.3-1.6 g/m2). At landscape scale, natural and semi-natural areas (forests and meadows/pastures) offered ca. 44% of the total pollen resources during the year. Relatively high amounts of pollen (ca. 35% of the year-round total pollen resources) were from winter rape, but this resource was short-term. Man-made, non-cropped habitats added only ca. 18% of the total pollen mass offered for pollinators during flowering season. However, they provided 66-99% of pollen resources available from July to October. There exists an imbalance in the availability of pollen resources throughout the year. Hence, a diversity of natural, semi-natural and man-made, non-cropped areas is required to support the seasonal continuity of pollen resources for pollinators in an agricultural landscape. Efforts should be made to secure habitat heterogeneity to enhance the flower diversity and continual pollen availability for pollinators.

Animal-vehicle collisions during the COVID-19 lockdown in early 2020 in the Krakow metropolitan region, Poland.

The interrelations between human activity and animal populations are of increasing interest due to the emergence of the novel COVID-19 and the consequent pandemic across the world. Anthropogenic impacts of the pandemic on animals in urban-suburban environments are largely unknown. In this study, the temporal and spatial patterns of urban animal response to the COVID-19 lockdown were assessed using animal-vehicle collisions (AVC) data. We collected AVC data over two 6-month periods in 2019 and 2020 (January to June) from the largest metropolis in southern Poland, which included lockdown months. Furthermore, we used traffic data to understand the impact of lockdown on AVC in the urban area. Our analysis of 1063 AVC incidents revealed that COVID-19 related lockdown decreased AVC rates in suburban areas. However, in the urban area, even though traffic volume had significantly reduced, AVC did not decrease significantly, suggesting that lockdown did not influence the collision rates in the urban area. Our results suggest that there is a need to focus on understanding the effects of changes in traffic volume on both human behaviour and wildlife space use on the resulting impacts on AVC in the urban area.

Phenology and production of pollen, nectar, and sugar in 1612 plant species from various environments.

To predict the quantity and quality of food available to pollinators in various landscapes over time, it is necessary to collect detailed data on the pollen, nectar, and sugar production per unit area and the flowering phenology of plants. Similar data are needed to estimate the contribution of plants to the functioning of food webs via the flow of energy and nutrients through the soil-plant-nectar/pollen-consumer pathway. Current knowledge on this topic is fragmented. This database represents the first compilation of data on the various food resources produced by 1612 plant species belonging to 755 genera and 133 families, including crop plants and wild plants, annuals and perennials, animal- and wind-pollinated plants, and weeds and trees growing in different ecosystems under various environmental conditions. The data set consists of 103 parameters related to the traits of plant species and geographical and environmental factors, allowing for precise calculations of the amounts of nectar, pollen, and energy provided by plants and available to consumers in the considered flora or ecosystem on a daily basis throughout the year. These parameters, gathered by us and extracted from the available literature, describe pollen, nectar, and sugar production (where applicable, in mass, volume, and concentration units), honey yield, the timing and duration of flowering, flower longevity, number of plants and flowers per unit area, weather conditions (temperature and precipitation), geographical location, landscape, and syntaxonomy. The data were obtained from various, mostly European, pedoclimatic zones, and the majority of the data were available for plant species and communities present in Central Europe, especially in Poland, where research on floral resources has a long tradition. These data are representative of the whole continent and may be used as a reference for plant communities occurring on continents other than Europe since the database allows for the consideration of differences in the production of resources by a single plant species growing in different communities. This data set provides a unique opportunity to test hypotheses related to the functioning of food webs, nutrient cycling, plant ecology, and pollinator ecology and conservation. The data are released under a CC-BY-NC-SA license, and this paper must be properly cited when using the database.

Homogeneity of agriculture landscape promotes insecticide resistance in the ground beetle Poecilus cupreus.

The intensification of agriculture leads to increased pesticide use and significant transformation from small fields towards large-scale monocultures. This may significantly affect populations of non-target arthropods (NTA). We aimed to assess whether the multigenerational exposure to plant protection products has resulted in the evolution of resistance to insecticides in the ground beetle Poecilus cupreus originating from different agricultural landscapes. Two contrasting landscapes were selected for the study, one dominated by small and another by large fields. Within each landscape the beetles were collected at nine sites representing range of canola coverage and a variety of habitat types. Part of the collected beetles, after acclimation to laboratory conditions, were tested for sensitivity to Proteus 110 OD-the most commonly used insecticide in the studied landscapes. The rest were bred in the laboratory for two consecutive generations, and part of the beetles from each generation were also tested for sensitivity to selected insecticide. We showed that the beetles inhabiting areas with medium and large share of canola located in the landscape dominated by large fields were less sensitive to the studied insecticide. The persistence of reduced sensitivity to Proteus 110 OD for two consecutive generations indicates that either the beetles have developed resistance to the insecticide or the chronic exposure to pesticides has led to the selection of more resistant individuals naturally present in the studied populations. No increased resistance was found in the beetles from more heterogeneous landscape dominated by small fields, in which spatio-temporal diversity of crops and abundance of small, linear off-crop landscape elements may provide shelter that allows NTAs to survive without developing any, presumably costly, resistance mechanisms.

Effects of agricultural landscape structure, insecticide residues, and pollen diversity on the life-history traits of the red mason bee Osmia bicornis.

Agricultural landscapes have changed substantially in recent decades, shifting from the dominance of small fields (S) with diverse cropping systems toward large-scale monoculture (L), where landscape heterogeneity disappears. In this study, artificial nests of the red mason bee, Osmia bicornis, were placed in S and L landscape types on the perimeter of oilseed rape fields representing different oilseed rape coverages (ORC, % land cover). The local landscape structure around each nest was characterised within a 100, 200, 500, and 1000 m radius using ORC and 14 landscape characteristics, which were then reduced by non-metric multidimensional scaling (nMDS) to two axes: nMDS1 characterised the dataset primarily according to land fragmentation and the main crop, whereas nMDS2 captured the prevalence of more natural areas in the landscape. Pollen diversity and insecticide risk levels in the pollen provisions collected by the bees were analysed, and their dependence on the landscape structure was tested. Thereafter, the effects of pollen diversity, insecticide risk, and landscape structure on the life-history traits of bees and their sensitivity to topically applied Dursban 480 EC were determined. Pollen taxa richness in a single nest ranged from 3 to 12, and 34 pesticides were detected in the pollen at concentrations of up to 320 ng/g for desmedipham. The O. bicornis foraging range was relatively large, indicating that the landscape structure within a radius of ~1000 m around the nest is important for this species. Pollen diversity in the studied areas was of minor importance for bee performance, but the ORC or landscape structure significantly affected the life-history traits of the bees. Contamination of pollen with insecticides affected the bees by decreasing the mass of newly emerged adults but their sensitivity to Dursban 480 EC was not related to environmental variables.

Supporting non-target arthropods in agroecosystems: Modelling effects of insecticides and landscape structure on carabids in agricultural landscapes.

Intensification of agricultural practices is one of the most important drivers of the dramatic decline of arthropod species. We do not know, however, the relative contribution to decline of different anthropogenic stressors that are part of this process. We used high-resolution dynamic landscape models and advanced spatially-explicit population modelling to estimate the relative importance of insecticide use and landscape structure for population dynamics of a widespread carabid beetle Bembidion lampros. The effects of in-crop mitigation measures through the application of insecticides with reduced lethality, and off-crop mitigation measures by increasing abundance of grassy field margins, were evaluated for the beetle along the gradient of landscape heterogeneity. Reducing the insecticide-driven lethality (from 90 to 10%) had larger positive impacts on beetle density and occupancy than increasing the abundance of field margins in a landscape. The effects of increasing field margins depended on their width and overall abundance in the landscape, but only field margins 4 m wide, applied to at least 40% of fields, resulted in an increase in beetle population density comparable to the scenario with the smallest reduction of insecticide-driven lethality we considered. Our findings suggest the importance of field margins rather as a supporting not stand-alone mitigation measure, as they generally improved effects of reduction of insecticide-driven lethality. Therefore, adding sufficiently broad off-field habitats should help to maintain viable beetle populations in agricultural landscapes even with moderate use of insecticides. In general, the less persistent the insecticides are in the environment, the larger positive impacts of applied mitigation measures on beetle populations were found. We also showed that the effectiveness of applied mitigation measures strongly depends on landscape and farmland heterogeneity. Thus, to achieve the same management or mitigation target in different landscapes might require different strategies.