Федеральное государственное бюджетное учреждение науки Федеральный исследовательский центр
«Морской гидрофизический институт РАН» Научно-техническая библиотека
Результаты научной деятельности сотрудников ФГБУН ФИЦ МГИ
Influence of large-scale atmospheric circulation modes on the danube runoff anomalies during the flood period id статьи: 856
Тип публикации
глава в монографии
Язык
En
Журнал
Springer Geology
ISSN:21979545
ISSN:21979545
Год
2021
Выходные данные
том
0
выпуск
страницы
335-341
Авторы
Абстракт
The statistical model for the anomalies of mean monthly runoff of the Danube River forecast in the spring season is proposed. An artificial neural network is used as a model. The predictors are the mean monthly values of the atmospheric circulation indices in wintertime. The time interval between 1950 and 2910 is considered. The model is tested using two criteria: correlation coefficient and Willmott’s “agreement index”. Their values in the most successful forecasts are 0.59 and 0.75 respectively. It is shown that the greatest impact on the Danube River runoff is caused by the Eurasian with a lead time of up to three months. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2021.
Ключевые слова
Artificial neural networks, Atmospheric circulation indices, Danube runoff
Дата занесения
2021-03-17 15:53:00