**5. Conclusions**

The winter 2018/2019 represented an average year in terms of overwinter honey bee colony losses in Austria. Significantly different loss rates were observed among the states, which could be led back to a divergent composition of landscapes or regional differences in weather and varroa pressure [19]. Low colony losses were associated with indicators for advanced beekeeping: migratory beekeeping or keeping a larger number of colonies. The majority of operational decisions that beekeepers can make in Austria, like certified organic beekeeping or having insulated hives in winter, had no effect on winter mortality. The only operational decision leading to lower losses was the self-sustaining operation of wax cycles, again a sign of a certain degree of professionalism. Queen problems during the summer resulted in a higher queen-related loss rate. Young queens showed a beneficial effect on both colony survival and queen-related losses. The observation of a notable number of bees with deformed wings in summer can be interpreted as an alarm signal for beekeepers, as this resulted in high colony losses. Access to various crops or honey flows was confirmed to impact colony survival—the exact drivers behind these need further investigation [17]. Varroa treatment with biotechnical methods in summer had a favourable effect on winter survival. For the first time, we investigated different commonly applied combinations of varroa control methods on winter colony losses. The combination of long-term

evaporation of formic acid in summer and oxalic acid usage in winter, the dominant and officially recommended varroa control strategy in Austria, resulted in an average loss rates. The lowest loss rates were observed for biotechnical methods and short-term evaporation of formic acid in summer followed by oxalic acid trickling in winter. The best treatment combination without formic acid was oxalic acid sublimation in summer and winter. It still needs to be considered that the *V. destructor* population cycle is very complex and that treatment methods do not return the anticipated results each year, which could be led back to weather conditions or other environmental factors. Therefore, a multiple year analysis on treatment combinations is required to provide more reliable insights.

Our study supports how diverse the factors influencing honey bee colony mortality can be and how difficult it is to determine one factor alone to improve colony survival. It is substantial to repeat these studies and to make multi-year and multi-country analyses in order to maintain relevancy in light of current trends in beekeeping practices and climate change. Our goal is to assist beekeepers with statistical analysis and to improve beekeeping and the honey bee population in Austria with empirically gained knowledge. One measure, based on this study, is to increase professionalism in beekeeping to reduce winter colony losses.

**Author Contributions:** Conceptualization, R.B.; methodology, R.B.; software, H.O.; validation, H.O.; formal analysis, H.O.; investigation, R.B. and H.O.; resources, R.B.; data curation, H.O.; writing—original draft preparation, H.O.; writing—review and editing, H.O. and R.B.; visualization, H.O.; supervision, R.B.; project administration, R.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by ZUKUNFT BIENE 2 gran<sup>t</sup> number 101295/2.

**Acknowledgments:** We want to thank all beekeepers that participated in our study or helped with discussions or suggestions to improve the study aims. We also want to thank the monitoring core project of the nonprofit honey bee research association COLOSS (www.coloss.org).

**Conflicts of Interest:** The authors declare no conflict of interest.
