Data-driven optimization of coastal sea level monitoring: Leveraging known patterns for enhanced reconstruction

Elif Kartal*, Abdüsselam Altunkaynak, Anıl Çelik

*Bu çalışma için yazışmadan sorumlu yazar

Araştırma sonucu: Dergiye katkıMakalebilirkişi

1 Atıf (Scopus)

Özet

Efficiently configuring sea level monitoring stations is crucial for obtaining accurate spatiotemporal data while managing operational and maintenance costs and addressing the challenges posed by missing data. This study focuses on optimizing the selection of stations within Turkey's coastal sea level monitoring network by leveraging the inherent lower dimensionality in data. The network consists of 18 stations distributed along Turkey's coastline. To identify dominant patterns in historical sea level data, Empirical Orthogonal Function (EOF) analysis was employed, followed by the application of the QR decomposition with column pivoting algorithm. Model performance is assessed using the Nash-Sutcliffe coefficient of efficiency (CE) and root mean square error (RMSE). Remarkably, the results demonstrated that reconstructing the entire dataset, encompassing all 18 stations was possible with a CE of 0.94 and an RMSE of 0.06. Even utilizing data from two or three stations alone achieves acceptable reconstruction accuracy. The effectiveness of EOF analysis combined with QR with column pivoting algorithm suggests promising applications in various scientific fields.

Orijinal dilİngilizce
Makale numarası103878
DergiRegional Studies in Marine Science
Hacim80
DOI'lar
Yayın durumuYayınlandı - 30 Ara 2024

Bibliyografik not

Publisher Copyright:
© 2024

Parmak izi

Data-driven optimization of coastal sea level monitoring: Leveraging known patterns for enhanced reconstruction' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Alıntı Yap