A climate data record of atmospheric moisture and sea surface temperature from satellite observations

Abstract. We developed a climate data record (CDR) of atmospheric column water vapor (CWV) and sea surface temperature (SST) under oceanic rain-free conditions using over two decades of observational records from three satellite instruments: the Advanced Microwave Scanning Radiometer for Earth Observing System (AMSR-E), the MicroWave Radiation Imager (MWRI), and the Advanced Microwave Scanning Radiometer-2 (AMSR2). The AMSR-E and AMSR2 satellites operated in near-stable orbits, while the MWRI experienced orbital drifts of nearly 1 h during its operational period. The CWV and SST products were retrieved from well-recalibrated level-1 brightness temperatures observed by common channels with the same frequencies on these instruments, designed for measuring these properties. Adjustments for diurnal drifting errors caused by orbital drift were applied to MWRI data using a semi-physical model developed in previous studies. The combination of prior recalibration and diurnal drift adjustment ensured inter-satellite consistency in the CDRs. Compared to in-situ radiosonde and buoy observations, the biases and root mean square errors of the CDRs are within 0.1 and 4.4 mm for CWV, and 0.2 and 1.6 K for SST, respectively. Long-term trends of the retrieved CWV and SST align with observations from the Global Navigation Satellite System (GNSS) and the Global Tropical Moored Buoy Array (GTMBA) products. The global oceanic trends of CWV and SST were 0.39 mm per decade and 0.16 K per decade, respectively, over the period 2002–2022. Inter-consistency between CWV and SST, as well as layer-mean temperatures derived from satellite microwave sounder observations, was examined and compared with climate model simulations from phase 6 of the Coupled Model Intercomparison Project (CMIP6). It was found that the trend ratio of the retrieved CWV to SST was 9.9 % K−1 in the tropics, which closely aligns with CMIP6 models. These validation results indicate that the presented CDR has high accuracy and is suitable for long-term climate change research. The CDR dataset is publicly available at https://doi.org/10.5281/zenodo.14539414 (Fu et al., 2024).