Abstract We studied geoid validation using ship-borne global navigation satellite systems (GNSS) on the Baltic Sea. We obtained geoid heights by combining GNSS–inertial measurement unit observations, tide gauge data, and a physical sea model. We used two different geoid models available for the area. The ship route was divided into lines and the lines were processed separately. The GNSS results were reduced to the sea surface using attitude and draft parameters available from the vessel during the campaign. For these lines, the residual errors between ellipsoidal height versus geoid height and absolute dynamic topography varied between 0 and 15 cm, grand mean being 2 cm. The mean standard deviations of the original time series were approximately 11 cm and reduced to below 5 cm for the time series filtered with 10 min moving average. We showed that it is possible to recover geoid heights from the GNSS observations at sea and validate existing geoid models in a well-controlled area.

中文翻译：

---

使用船载 GNSS 数据对波罗的海大地水准面进行验证

摘要 我们在波罗的海使用船载全球导航卫星系统 (GNSS) 研究了大地水准面验证。我们通过结合 GNSS 惯性测量单元观测、潮汐测量数据和物理海模型获得了大地水准面高度。我们使用了该地区可用的两种不同的大地水准面模型。将航线划分为线段，对线段进行单独处理。使用活动期间船舶提供的姿态和吃水参数将 GNSS 结果减少到海面。对于这些线，椭球高度与大地水准面高度和绝对动态地形之间的残差在 0 到 15 厘米之间变化，总平均值为 2 厘米。原始时间序列的平均标准偏差约为 11 厘米，对于用 10 分钟移动平均值过滤的时间序列，平均标准偏差降低到 5 厘米以下。