Multiscale Characteristics of the Surface Sensible Heat Enhancement Process in the Central and Eastern Tibetan Plateau in Spring: A Comparative Analysis Based on Multi-source Datasets

Thermal anomalies of the Tibetan Plateau (TP) plays a critical role in driving the Asian monsoon system and global atmospheric circulation. However, the selection of heat source data for the TP has been controversial in its climate impact research. Therefore, the evaluation of the suitability of multi-source data on the TP is of particular importance. Based on station observations and reanalysis datasets (JRA-55, ERA5, and MERRA2), this study systematically analyzed the multi-scale characteristics of the springtime surface sensible heat (SH) enhancement process in the central-eastern TP (CETP) from 1982 to 2020 and evaluated the applicability of multi-source daily surface SH flux data in the TP region. The results have shown that: (1) On average, the time (T1) when the surface sensible heat (SH) over the CETP changes from weak to strong within a year occurs during the 14th - 17th pentads (early and mid - March). Subsequently, it keeps increasing. The peak time (T2) always occurs before the 30th pentad. The average duration of its increasing process in spring is approximately 14 pentads (70 days), and the increasing intensity (Q) ranges from 45 to 60 W/m2. In terms of intraseasonal evolution, the SH enhancement process exhibits a pronounced quasi-biweekly oscillation with a dominant periodicity of 10–20 days, accounting for 30% to 50% of the total seasonal variance. Regarding phase evolution, the pattern is primarily characterized by a single-pole mode. Both T1 and T2 undergo a trend shift around 2000, characterized by an initial decrease followed by an increase. Meanwhile, Q displays a notable weakening in its climatological state around the same period. (2) Evaluation of three surface sensible heat (SH) flux reanalysis datasets reveals that, compared to JRA-55 and ERA5, MERRA2 exhibits significantly higher correlations with station observations in both the intensity and phase evolution of quasi-biweekly oscillations during the spring SH enhancement process and the interannual variations of T1, T2, and Q. This indicates the best applicability of MERRA2 over the CETP, followed by JRA-55, whereas ERA5 performs relatively poorly. (3)Evaluation of spring surface wind speed and surface-air temperature difference increments shows that all three reanalysis datasets capture the interannual variability and spatial distribution of surface wind speed well, with over 50% of stations showing significant correlations with station observations. ERA5 has a mean value roughly half the observed, while JRA-55 and MERRA2 show anomalously large values over the southwestern Tibetan Plateau. For the temperature difference increment, MERRA2 has the most similar spatial distribution to station observations; JRA-55 and ERA5 show climatologically small and large values, respectively, due to inconsistent deviations in surface and air temperature increments. MERRA2 outperforms the other two in both variables, which may explain its better performance in intraseasonal and interannual variations of the surface sensible heat enhancement process. The findings provide a valuable reference for selecting daily SH flux data over the TP and lay the foundation for a more comprehensive understanding of the climatic impacts of anomalies in the spring surface sensible heat enhancement process in this region.