The application of a neural network-based ragweed pollen forecast by the Ragweed Pollen Alarm System in the Pannonian biogeographical region

Abstract

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 was 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 long validated data set of ragweed pollen were selected for the study from the PBR. Variables including forecasted 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 datadriven method to forecast the behaviour of complex systems. In our case, it has three layers, one of which is hidden. MLP utilizes a supervised learning technique called backpropagation for training to get better

Estimating soil organic carbon stock change at multiple scales using machine learning and multivariate geostatistics

“…In this study we chose supports which are frequently used in Hungary and that are of interest to stakeholders. We used 1 km × 1 km and 10 km × 10 km blocks, which could be useful for countrywide crop simulation (Fodor et al, 2014) and terrestrial ecosystem process modelling (Hidy et al, 2016), in national ragweed (Ambrosia artemisiifolia L.) pollen forecasting (Csépe et al, 2020), in natural vegetation mapping and in national contributions to global/continental assessment of soil organic carbon stock (Yigini et al, 2018). In addition, we also aggregated to the Hungarian counties and entire Hungary (Fig.…”

Biologic Pollution Due to Ambrosia (Ragweed) Pollen in Urban Environment of Bucharest

“…Since 2017, our laboratory is a partner in the regional project Ragweed Pollen Alarm System (R-PAS), coordinated by the Hungarian Aerobiology Society and the Institute of Public Health from Budapest, running since 2014. This project aims to provide pollen information from countries included or close to the Pannonian biogeographical region, by using a neural network-based ragweed pollen forecast [ 36 ].…”

Recent developments in monitoring and modelling airborne pollen, a review

“…The application of ANNs have assessed a variety of aspects relating to the modelling and forecasting of pollen. For example, Multilayer Perceptron ANNs (MLPs) have been used to predict the daily Ambrosia concentration over different cities up to 7 days ahead (Csépe et al 2014) and in the construction of a Ambrosia pollen alarm system in the Pannonian biogeographical region (Csépe et al 2020) Support Vector Machines also express complex non-linear relationships by learning but do so by using Vapnik-Chervonenkis dimensional theories (Du et al 2017). Unlike, ANNs and RFs, very few studies have exclusively employed SVMs for pollen forecasting and are generally compared alongside other modelling methods.…”