MODELLING THE IMPACT OF CLIMATE CHANGE ON THE HUNGARIAN WINE REGIONS USING RANDOM FOREST

This paper aims to simulate and analyse the impact of climate change on the Hungarian wine regions using spatial layers of temperature-based bioclimatic indices. Random forest classification was used to analyse the similarities between the present and future climate of the wine regions. The model was firstly calibrated for the present period then applied for the expected future climatic conditions simulated by the RegCM3 model with A1B scenario. Results show that in the near future (2021-2050) the grapevine regions typical of the southern may expand in greater part of the country, while at the end of the century (2071-2100) only the northern part of the country shows some similarities with the present climate. Despite these results, Hungary is expected to remain amongst the regions with good quality grapevine growing conditions, but the structure of the cultivation and/or varieties should be changed.