Szczegóły publikacji
Opis bibliograficzny
Structure of rise in monthly temperature in Europe as estimated by machine learning / Anna FRANCZYK, Robert Twardosz, Adam WALANUS // Pure and Applied Geophysics ; ISSN 0033-4553. — 2025 — vol. 182 iss. 6, s. 2631–2653. — Bibliogr. s. 2652–2653, Abstr. — Publikacja dostępna online od: 2025-06-04
Autorzy (3)
- AGHFranczyk Anna
- Twardosz Robert
- AGHWalanus Adam
Słowa kluczowe
Dane bibliometryczne
| ID BaDAP | 161899 |
|---|---|
| Data dodania do BaDAP | 2025-09-02 |
| Tekst źródłowy | URL |
| DOI | 10.1007/s00024-025-03742-x |
| Rok publikacji | 2025 |
| Typ publikacji | artykuł w czasopiśmie |
| Otwarty dostęp |  |
| Creative Commons | |
| Czasopismo/seria | Pure and Applied Geophysics |
Abstract
The rise in air temperature is a leading research topic. This is not only from the cognitive point of view, but also for practical reasons because it involves many effects that are dangerous to humans and their activities. Although this is not a new issue, it requires continuous monitoring as well as the application of multiple methods, including the latest, apparently most objective methods offered by, inter alia, artificial intelligence. In the present paper, the authors have undertaken to investigate the structure of the rise in mean monthly air temperatures in Europe using unsupervised machine learning methods. The last 70 years can be divided into two periods, one of which is relatively stable and the second of which shows an evident rise in temperature. The correct determination of the year in which that change occurred is crucial. Mean monthly temperatures in Europe and its direct surroundings were used for this purpose. The data originated from 210 meteorological stations and covered the period 1951–2020. The analysis was performed using the hierarchical clustering and k-means clustering methods. The research was conducted in two phases. The first phase involved the analysis of area-average values, followed by the analysis of each station separately. Clear results were obtained, which confirms the usefulness of machine learning as a tool for monitoring temperature change. The quantitative change in the behavior of monthly temperature recorded from 1950 all over Europe is positioned at 1999, when the linear rise started.
