Hive weight time series prediction using meteorological factors during R. pseudoacacia nectar flow

Honey bees (Apis mellifera L.) are one of the most important pollinators worldwide, primarily managed by beekeepers. Beyond their role in pollination, they produce a range of valuable products, including honey, bee bread, pollen, propolis, royal jelly, venom, apilarnil and wax. It is a well-known fact that the weight is a good indicator of the health, strength, food and foraging capabilities of the colony despite its complex structure. Various external and internal factors could influence these variations, yet until today, this is the most popular way of observing bees. Nowadays, the use of digital hive scales is becoming more widespread, offering precise and high temporal resolution monitoring of bee activity.

Our study aims to forecast honey bee foraging behaviour during a heavy, monocultural nectar flow (Robinia pseudoacacia L.) in the period between April 15 and June 15 in a Hungarian active apiary, incorporating meteorological factors as exogenous variables. To achieve this, we deployed hive scales under two hives from 2021 to 2024, recording weight data at 1-hour and 30-minute intervals. After pre-processing the data, we applied time-series models using a 24-hour rolling forecast method and defined four cases: 1) without the exogenous variables, 2) with meteorological variables, 3) with diurnal patterns, and 4) combining all factors. Linear and non-linear models were tested with and without the seasonal component, utilizing ARIMAX (AutoRegressive Integrated Moving Average with eXogenous variables), SARIMAX (Seasonal ARIMAX) and additional LSTM (Long Short-Term Memory) approaches. Beyond model development, we seek to quantify the impact of weather conditions on hive weight fluctuations and explore the biological responses of honey bees to environmental changes.

These insights are crucial for advancing modern precision apiculture, optimizing hive management strategies and preparing for climate-driven challenges in beekeeping.

This work has been implemented by the National Multidisciplinary Laboratory for Climate Change (RRF-2.3.1-21-2022-00014) project within the framework of Hungary's National Recovery and Resilience Plan supported by the Recovery and Resilience Facility of the European Union.