A global dataset of spatiotemporally seamless daily mean land surface temperatures: generation, validation, and analysis

[摘要] Daily mean land surface temperatures (LSTs) acquired frompolar orbiters are crucial for various applications such as global andregional climate change analysis. However, thermal sensors frompolar orbiters can only sample the surface effectively with very limitedtimes per day under cloud-free conditions. These limitations have produced asystematic sampling bias ( Δ T sb ) on the daily mean LST( T dm ) estimated with the traditional method, which uses the averages ofclear-sky LST observations directly as the T dm . Several methods havebeen proposed for the estimation of the T dm , yet they are becoming lesscapable of generating spatiotemporally seamless T dm across the globe.Based on MODIS and reanalysis data, here we propose an improved annual anddiurnal temperature cycle-based framework (termed the IADTC framework) togenerate global spatiotemporally seamless T dm products ranging from 2003to 2019 (named the GADTC products). The validations show that the IADTCframework reduces the systematic Δ T sb significantly. Whenvalidated only with in situ data, the assessments show that the mean absoluteerrors (MAEs) of the IADTC framework are 1.4 and 1.1 K for SURFRAD andFLUXNET data, respectively, and the mean biases are both close to zero.Direct comparisons between the GADTC products and in situ measurements indicatethat the MAEs are 2.2 and 3.1 K for the SURFRAD and FLUXNET datasets,respectively, and the mean biases are −1.6 and −1.5  K for these twodatasets, respectively. By taking the GADTC products as references, furtheranalysis reveals that the T dm estimated with the traditional averagingmethod yields a positive systematic Δ T sb of greater than 2.0 Kin low-latitude and midlatitude regions while of a relatively small value inhigh-latitude regions. Although the global-mean LST trend (2003 to 2019)calculated with the traditional method and the IADTC framework is relativelyclose (both between 0.025 to 0.029 K yr −1 ), regional discrepancies in LSTtrend do occur – the pixel-based MAE in LST trend between these twomethods reaches 0.012 K yr −1 . We consider the IADTC framework can guide thefurther optimization of T dm estimation across the globe, and thegenerated GADTC products should be valuable in various applications such asglobal and regional warming analysis. The GADTC products are freelyavailable at https://doi.org/10.5281/zenodo.6287052 (Hong etal., 2022).

[学科分类] 眼科学