An empirical noise model for the benefit of model-based hydrodynamic leveling

Abstract The main objective of this study is to develop and analyze an empirical noise model for model-derived coastal summer mean water levels (SMWLs) use that obtain a more realistic quality impact combining hydrodynamic leveling Unified European Leveling Network (UELN) data in realizing the Vertical Reference System (EVRS). We considered three state-of-the-art models Northeast Atlantic Ocean, including North Sea Wadden Sea; AMM7, DCSMv6-ZUNOv4, 3D DCSM-FM. Moreover, we assess spatiotemporal performance these representing SMWLs. are determined from differences between observation- SMWLs at tide gauges. All show indeed correlated over sea distances up hundreds kilometers. At same time, they all relatively large discontinuity origin (i.e., nugget effect); 12.1 cm $$^2$$ 2 (3D DCSM-FM) 16.3 (DCSMv6-ZUNOv4). variance covariance zero distance) two 15.3 21.7 , respectively. Averaging summers, lowered effect DCSM-FM 12.7 10.0 Our analysis also showed 30 50% has be attributed errors vertical referencing lacked information what proportion observed covariances should errors. assessments revealed significant variations both space time as well among models. results suggest there still room improvement. In final experiments, used best overall performing reassess UELN EVRS. not leads overestimation total by $$7\%$$ 7 % $$8\%$$ 8 when average one periods,