pure and applied geophysics最新文献

Traditional rainfall-runoff modeling techniques require large datasets and often an exhaustive calibration process, which is challenging, especially in poorly-gauged basins and resource-limited settings. Therefore, it is necessary to examine new ways of constructing predictive models for runoff that can achieve satisfactory results, while also minimizing the data requirement and model construction time. In this study, the effectiveness of integrating the Random Forest (RF) as an important feature identifier with novel gradient boosted trees to achieve satisfactory results was examined for two adjacent catchments in Vietnam. Antecedent daily runoff in combination with daily and one-day antecedent rainfall was found to significantly influence the runoff at the outlet of the catchments. Categorical Boosting (CatBoost) and Extreme Gradient Boosting (XGBoost) were effective in predicting day-ahead runoff. For instance, CatBoost with NSE, d, r, and R² values of 0.92, 0.98, 0.96, and 0.92, respectively, and XGBoost with NSE, d, r, and R² values of 0.91, 0.98, 0.96, and 0.92, respectively, are well suited for predicting runoff. A comparative analysis of their results with previous studies revealed that the models were very effective since they were able to better reduce generalization errors at different calibration and validation phases. This study presents the integration of RF and gradient boosted trees as a simplified alternative to computationally expensive and data-intensive physically-based rainfall-runoff models. The practitioners can build upon the experimentation presented in this study to minimize the computational time requirement, construction process complexity, and data requirement, which are often serious constraints in physically-based rainfall-runoff modeling.

A new method of GNSS gravity leveling is introduced to determine precisely normal height differences, Both the principle and application of the method are elaborated. Leveling surveying, gravity measurements, and GNSS measurements are carried out in a special region (including slopes, valleys and mountain ridges) to verify its accuracy by combining with gravity potential model. The results show that the precision by this method is mainly influenced by ellipsoidal height differences, gravity potential models, and gravity observations. However, the error by this method exhibits a clear linear relationship with the height difference, while it is independent of the length of the survey line. Within a specific range of height differences (within 360 m), the precision of the GNSS gravity leveling can reach the level of ± 10 mm. This method can, to some extent, provides a modern solution for height measurement which can replace the high-precision leveling surveying. The advantages of GNSS gravity leveling include high precision and high efficiency. It has a promising application prospect in geodesy, hydraulic engineering, earthquake and volcano monitoring.

In this paper, a two-dimensional numerical model for simulating the generation and propagation of tsunami waves caused by upthrust bed movement is developed. To consider the nonlinearity as well as save the computational cost, a Navier–Stokes equation solver is used for the generation zone, and a Serre equation solver is adopted for the downstream evolution of the tsunami waves. The solution of the Navier–Stokes equation solver is extracted and transferred as the initial solution of the Serre solver, which means a one-way coupling is achieved. In this way, a one-way coupled Navier–Stokes-Serre model is obtained. After a detailed validation of the individual solvers, the coupled model is utilized for simulating the generation and propagation of tsunami waves caused by the upthrust bed movement in shallow water of uniform depth. It is found that the coupled model is comparable to the traditional Boussinesq equation model. Finally, the capacity of the coupled model for simulating wave-breaking cases is demonstrated.

Tsunamis generated by the Hunga Tonga–Hunga Ha’apai volcanic eruption on January 15, 2022 were recorded on ocean bottom pressure and tide gauges around the Pacific Ocean, earlier than the expected arrival times calculated by tsunami propagation speed. Atmospheric waves from the eruption were also recorded globally with propagation speeds of ~ 310 m/s (Lamb wave) and 200–250 m/s (Pekeris wave). Previous studies have suggested that these propagating atmospheric waves caused at least the initial part of the observed tsunami. We simulated the tsunamis generated by the propagation of the Lamb and Pekeris waves by adding concentric atmospheric pressure changes. The concentric sources are parameterized by their propagation speeds, initial atmospheric wave amplitudes that decay with the distance, and a rise time. For the Lamb wave, inversions of the observed tsunami waveforms at 14 U.S. and nine New Zealand DART stations indicate the start of the positive rise at 4:16 UTC, the peak amplitude of 383 hPa, and the propagation speed of 310 m/s, assuming a rise time of 10 min. The later phases of the observed tsunami waveforms can be better reproduced by adding another propagating concentric wave (Pekeris wave) with a negative amplitude (− 50 hPa) and propagation speeds of 200–250 m/s. The DART records around the Pacific indicate that the Pekeris wave speed is faster toward the northwest and slightly slower toward the northeast. The synthetic waveforms roughly reproduced the far-field tsunami waveforms recorded at tide gauge stations, including the later phases, suggesting that the large amplitude in the later phase may be due to the coupling of the Pekeris wave and the tsunami, as well as resonance around tide gauge stations.

The movement of the Earth's surface mass, including the atmosphere and oceans, as well as hydrology and glacier melting, causes the redistribution of surface loads, deformation of the solid Earth, and fluctuations in the gravity field. Global Navigation Satellite Systems (GNSS) provide useful information about the movement of the Earth's surface mass. The impact of surface loading deformation over 145 GNSS sites in Africa was investigated using vertical height time series analysis. The study investigates and quantifies the impact of surface loading on the GNSS coordinates utilizing GNSS Precise Point Positioning (PPP) approach. The German Research Center for Geosciences (GFZ) EPOS.P8 software was used to process and analyze eleven years of GPS data from all the stations, as well as dedicated hydrological and atmospheric loading correction models given by the Earth System Modeling group at Deutsches GeoForschungsZentrum (ESMGFZ). The results of the hydrological loading corrections arising from the surface-deformation were analysed to determine the extent of station improvements. The results revealed about 40% of the stations showed improvement with an average Root Mean Square Error (RMSE) residual of 7.3 mm before the application of the hydrological loading corrections and 7.1 mm Root Mean Square Error (RMSE) after the application of the hydrological loading corrections. Similarly, the atmospheric loading corrections gave an improvement of about 57%. Furthermore, the amplitude values decreased from 4.1–8.1 mm to 3.5–6.2 mm after atmospheric loading corrections. This finding presupposes that applying loading corrections to the derived time series reduces amplitude in some African regions.