THE APPLICATION OF NEURAL NETWORK-BASED RAGWEED POLLEN FORECAST BY THE RAGWEED POLLEN ALARM SYSTEM IN THE PANNONIAN BIOGEOGRAPHICAL REGION

Ragweed Pollen Alarm System (R-PAS) has been running since 2014 to provide pollen information for countries in the Pannonian Biogeographical Region (PBR). The aim of this study is to develop forecast models of the representative aerobiological monitoring stations, identified by analysis based on a Neural Network computation. Monitoring stations with 7-day Hirst type pollen trap·having 10- year Iong validated dataset of ragweed pollen were selected for the study from the PBR. Variables including meteorological data, pollen data of the previous days·and nearby monitoring stations were used as input of the model. We used the·multilayer perceptron model to forecast the pollen concentration. The multilayer perceptron (MLP) is a feedforward artificial neural network. MLP is a data driven method it can use to forecast complex systems. ln our case it has three layers with one hidden layer. MLP utilizes a supervised learning technique called backpropagation for training to get better performance. The Neural Network tests selected different set of variables for predict pollen levels for the next 3 days in each monitoring stations. The predicted pollen Ievels are shown on isarithmic map. We use MAE, RMSE and correlation coefficients to show the forecasting system's performance. Visualization of the results of Neural Network forecast on isarithmic maps is a good tool to communicate pollen information to general public in the PBR.