The pollen calendar is the simplest forecasting method for pollen concentrations. As pollen concentrations are liable to seasonal variations due to alterations in climate and land-use, it is necessary to update the pollen calendar using recent data. To attenuate the impact of considerable temporal and spatial variability in pollen concentrations on the pollen calendar, it is essential to employ a new methodology for its creation. Methods: A pollen calendar was produced in Korea using data from recent observations, and a new method for creating the calendar was proposed, considering both risk levels and temporal resolution of pollen concentrations. A probability distribution was used for smoothing concentrations and determining risk levels. Airborne pollen grains were collected between 2007 and 2017 at 8 stations; 13 allergenic pollens, including those of alder, Japanese cedar, birch, hazelnut, oak, elm, pine, ginkgo, chestnut, grasses, ragweed, mugwort and Japanese hop, were identified from the collected grains. Results: The concentrations of each pollen depend on locations and seasons due to large variability in species distribution and their environmental condition. In the descending order of concentration, pine, oak and Japanese hop pollens were found to be the most common in Korea. The pollen concentrations were high in spring and autumn, and those of oak and Japanese hop were probably the most common cause of allergy symptoms in spring and autumn, respectively. High Japanese cedar pollen counts were observed in Jeju, while moderate concentrations were in Jeonju, Gwangju and Busan. Conclusions: A new methodology for the creation of a pollen calendar was developed to attenuate the impact of large temporal and spatial variability in pollen concentrations. This revised calendar should be available to the public and allergic patients to prevent aggravation of pollen allergy.

PurposeOak is the dominant tree species in Korea. Oak pollen has the highest sensitivity rate among all allergenic tree species in Korea. A deep neural network (DNN)-based estimation model was developed to determine the concentration of oak pollen and overcome the shortcomings of conventional regression models.MethodsThe DNN model proposed in this study utilized weather factors as the input and provided pollen concentrations as the output. Weather and pollen concentration data were used from 2007 to 2016 obtained from the Korea Meteorological Administration pollen observation network. Because it is difficult to prevent over-fitting and underestimation by using a DNN model alone, we developed a bootstrap aggregating-type ensemble model. Each of the 30 ensemble members was trained with random sampling at a fixed rate according to the pollen risk grade. To verify the effectiveness of the proposed model, we compared its performance with those of models of regression and support vector regression (SVR) under the same conditions, with respect to the prediction of pollen concentrations, risk levels, and season length.ResultsThe mean absolute percentage error in the estimated pollen concentrations was 11.18%, 10.37%, and 5.04% for the regression, SVR and DNN models, respectively. The start of the pollen season was estimated to be 20, 22, and 6 days earlier than that predicted by the regression, SVR and DNN models, respectively. Similarly, the end of the pollen season was estimated to be 33, 20, and 9 days later that predicted by the regression, SVR and DNN models, respectively.ConclusionsOverall, the DNN model performed better than the other models. However, the prediction of peak pollen concentrations needs improvement. Improved observation quality with optimization of the DNN model will resolve this issue.

No abstract

No abstract