Search

Traditionally, geoid models have been validated using GNSS-levelling benchmarks on land only. As such benchmarks cannot be established offshore, marine areas of geoid models must be evaluated in a different way. In this research, we present a marine GNSS/gravity campaign where existing geoid models were validated at sea areas by GNSS measurements in combination with sea surface models. Additionally, a new geoid model, calculated using the newly collected marine gravity data, was validated. The campaign was carried out with the marine geology research catamaran Geomari (operated by the Geological Survey of Finland), which sailed back and forth the eastern part of the Finnish territorial waters of the Gulf of Finland during the early summer of 2018. From the GNSS and sea surface data we were able to obtain geoid heights at sea areas with an accuracy of a few centimetres. When the GNSS derived geoid heights are compared with geoid heights from the geoid models differences between the respective models are seen in the most eastern and southern parts of the campaign area. The new gravity data changed the geoid model heights by up to 15 cm in areas of sparse/non-existing gravity data. [ABSTRACT FROM AUTHOR]

Accurate sea-level forecasting is crucial for navigation, engineering and coastal conservation. One of the major obstacles in obtaining accurate sea-level data, both at coastal and offshore areas, has been in defining a realistic vertical datum. The Baltic Sea countries, however, have collaborated in calculating a high-resolution regional geoid model (NKG2015). This now paves the way for determining accurate sea-levels over the entire sea. Accordingly, this study explores the application of a deep learning two-dimensional Convolutional Neural Network (Conv2D) technique along with using essential inputs (e.g. accurate dynamic topography (DT), wind speed and direction, surface pressure and temperature). The method was tested for a three-year 2017–2019 period in the Baltic Sea. The evaluation was based on two statistical criteria: spatial root mean squared error (RMSE) and R-squared (R 2). Results revealed that the proposed Conv2D model allows predicting the DT of the Baltic Sea, with an R 2 of 0.91. The spatial RMSE plot also confirms accurate DT predictions for most of the Baltic Sea points, with the discrepancies distributed within ±4 cm. Spatially, some larger RMSE values (∼10 cm) were obtained at particular locations in south-eastern Baltic Sea, which may be due to the input sources utilized. Examination of sea level maxima also showed that the Conv2D model reproduced the maxima events in most scenarios (9 of 10 and 8 of 11 in Gulf of Finland and Riga, respectively) with residuals that varied up to 7 cm–18 cm. For the higher residuals, the Conv2D tended to underestimate the sea level. This suggests the importance of considering the necessary inputs (e.g waves) for forecasting storm surges and that the Conv2D model can still be improved. External validation was also performed using along-track sea level satellite altimetry data, with differences between forecasted model and satellite being within 5 cm. This confirms the validity of the forecasted model and the occurrences of model biases to be minimum. The method utilized shows the potential to contribute toward operational sea level forecasting in the Baltic Sea. • To forecast absolute sea level (DT) in the Baltic Sea, a multivariate conv2D algorithm is proposed using hydro-meteorological characteristics as inputs. • A spatial RMSE within ±4 cm was obtained for majority of points, thus confirming accuracy of DT forecast model. • Spatially, some larger RMSE values were obtained at particular locations in the south-eastern Baltic Sea (RMSE of over 10 cm), which may be due to the input sources utilized. • External validation using along-track satellite altimetry data showed a difference between forecasted and SA within 5 cm. • The model effectively captured sea level maxima dates in the Gulf of Finland and Riga. [ABSTRACT FROM AUTHOR]

Accurate and compatible sea level data are now more important than ever before, especially in semi-enclosed sea areas that are highly exploited and surrounded by many countries, such as the Baltic Sea. Obtaining accurate sea level data is however, not only hindered by resolution deficiencies and systematic and random errors from the various available sources (e.g. satellite altimetry (SA), tide gauges (TG), hydrodynamic models (HDM), etc.), but most importantly by variations in and insufficient knowledge of the vertical reference datums. This study demonstrates that by incorporating the geoid (equipotential surface of the earth, that represents a stable vertical datum) along with a network of tide gauges and hydrodynamic models it is possible to obtain accurate and realistic sea level data. A simplified method is developed that calculates the bias between TG and HDM and identifies an optimum time period to be utilized, given the associated accuracy required. A bias period of 0–6 h results in a standard deviation of less than 5 cm at all participating TG stations. The method is tested in the estuarine water body of the Gulf of Finland (in the eastern section of the Baltic Sea). Results show that without the bias correction, the mean dynamic topography from a westerly to easterly direction along the gulf, varied from −12.7 to −8.2 cm (a difference of 4.5 cm) whilst after bias correction the model varied from 18 cm to 25.4 cm (a difference of 7.4 cm). Both these scenarios demonstrate an increasing eastward trend. Nevertheless, a major difference in quantification exists and to a first approximation, this may vary by as much as a factor of almost 2. The analysis also intrinsically identified critical areas where drastic changes in dynamic topography occur and the associated seasons. Thus, the utilization of more stable vertical reference, such as the geoid, displays promising results, that essentially allows better quantification of more realistic parameters (e.g. sea level trends, extreme value analysis etc.). In addition, it allows identification of hydrodynamic modelling imperfections and that a coherent compatibility with other sources of sea level data (e.g. SA, Global Navigational Satellite Systems etc.) is now possible. This accuracy and conformity in sea level data are urgently required for a comprehensive understanding of climate change, marine engineering and navigation applications. • Utilization of a stable vertical reference datum such as the geoid (equipotential surface of earth) allows realistic determination of sea level heights. • Synergy of tide gauge, geoid and hydrodynamic models allow estimates of a spatial and temporal bias correction. • For hydrodynamic models a bias correction of 0–6 h results in an accuracy of better than 5 cm. • The developed method improves mean dynamic topography by a factor of almost two. • Method allows consistency and accuracy amongst various sea level sources. [ABSTRACT FROM AUTHOR]