Changes in lake water levels are closely related to climate change and can also reflect information about local human activities. Therefore, obtaining high temporal resolution time series of lake water levels is necessary for accurately analyzing hydrological changes. However, the existing methods mainly focus on the long-term changes in lake water levels, with less attention paid to short-term changes in lake water levels. In this article, we proposed a new method to construct high temporal resolution lake water level time series by fusing multisource altimetry satellite data based on Kalman filtering and using the MissForest algorithm to combine meteorological data Kalman Fusion-MissForest water level (KF-MFWL). The accuracy of KF-MFWL was validated using gauge data, as well as compared with HYDROWEB and DAHITI. Finally, a dataset of daily lake water level time series for the Qinghai-Tibet Plateau from 2019 to 2021 has been compiled, and the driving factors influencing water level changes were analyzed. Our result shows that the KF-MFWL time series is comparable to that of HYDROWEB and DAHITI, but with a much higher temporal resolution. The annual rate of water level change for 264 lakes in the Qinghai-Tibet Plateau is 0.021 m/y. Among them, the water level of 82 lakes has significantly increased with an average annual change rate of 0.171 m/y, while that of 55 lakes exhibits a remarkable decrease with an average annual change rate of −0.145 m/y. This study can provide an important data basis for water resource management in the Qinghai-Tibet Plateau region.

The global ionospheric model is the most effective way to study the structure and variation of the global ionosphere. However, the current global ionosphere maps (GIMs) have the defect of low spatiotemporal resolution and cannot reflect the short-term nonlinear changes and small-scale structures of the vertical total electron content (VTEC). This article proposes a new method for establishing a global ionospheric model with high spatiotemporal resolution. The spherical harmonic (SH) expansions are used to model the VTEC observations, and the Kalman filter is used to estimate the model’s coefficients with high accuracy. The method calculates SH coefficients every 5 min, increasing the spatial resolution to 7.2°. Precise determination methods for the state noise covariance matrix and the observation noise covariance matrix in the Kalman filter are also proposed. The model with a high spatiotemporal resolution was established using the observations of 300 global navigation satellite system (GNSS) tracking stations worldwide. The results show that the model with high spatiotemporal resolution can more finely reflect the nonlinear changes of VTEC in a short period and the small-scale structure of the ionosphere. The accuracy of the high-resolution model is validated using high-precision differential slant total electron content (dSTEC) observations from three sets of tracking stations and compared with the final product of three international GNSS service ionosphere associate analysis centers (IGS IAACs). The results show that the accuracy of the high spatiotemporal resolution model in this article is better than the products of IGS IAACs. The research in this article provides a new idea for establishing GIMs with higher spatiotemporal resolution and accuracy.