Temperature and salinity (T&S) are fundamental in physical oceanography. Objective analysis can generate highly reliable gridded T&S datasets. However, most existing representative products have relatively coarse resolution. This study develops a global 1/4° high-resolution product by integrating spatiotemporal correlations and physical constraints. Results show that the dataset achieves excellent overall accuracy and unbiased performance, while effectively capturing abundant mesoscale features.