2 research outputs found
Using the Kalman Filter with Satellite Altimetry to Estimate the Water Level of Inland Water
No full textThe Euphrates River extends for approximately 2,700 km, making it the longest river in Southwest Asia. Reliable water level measurements are obtained through the integration of an advanced outlier rejection system with Kalman filter technology. This study employs water level data from the Database for Hydrological Time Series over Inland Waters (DAHITI) and validates them using in situ measurements collected from gauging stations along the Euphrates River. To improve the accuracy of water level time series across the study area (Lat: 31.9676, Lon: 44.9306 to Lat: 31.0955, Lon: 46.0942), the research incorporates multibeam altimetry data from Envisat, Jason-2, and Sentinel-3A/B/B. Validation of the altimetry techniques is carried out by comparing DAHITI water level records with in situ measurements and other satellite-based datasets. Both the Kalman filter and Hydroweb methods yield Unbiased Root Mean Square Difference (ubRMSD) values ranging between 0.2961–0.3922 cm and 0.536–0.577 cm, respectively. The Nash-Sutcliffe Efficiency coefficient for DAHITI-derived water levels varies between 0.5971 and 0.9831, while Hydroweb produces values from –0.871 to 0.567. Overall, DAHITI-based altimetry height estimates demonstrate superior accuracy compared to other altimeter datasets in most parts of the Euphrates River, with precision strongly influenced by river topography. The application of Kalman filtering further enhances water level monitoring, particularly in regions characterized by complex inland water structures
Integration of local mean sea level and land vertical datum over peninsular Malaysia via transformation model
No full textIntegration of land and marine vertical datums is an important aspect of geospatial reference systems. Therefore, this
study has been conducted to identify an optimum approach to integrate the marine and land vertical datums. Two hybrid geoid models have been developed and fitted to the the land levelling datum at benchmark and to the tide gauge-benchmark station (TGBM). The differences between the two hybrid geoid models were computed to establish a vertical datum transformation model (VDT). Among the 305 GNSS-levelling points, 295 have been used in the hybridization process and 10 have been used for validation. Based on the comparison, the geoidal differences at the 10 points range from -7.2 to 7.0 cm while the mean and RMSE of differences are 1.3 cm and ± 4 cm, respectively. The second hybrid
geoid, which was fitted to local MSL, was developed by directly adding to the offset between the gravimetric geoid and local MSL at nine TGBM stations. The result indicates that the offset derived at Tanjung Gelang is the optimum one with an RMSE of ±0.045 m. The VDT model developed shows a transformation accuracy of approximately ± 4 cm
