Search

Reanalysis data are extensively employed in investigation of the tropical large-scale climate variability. In this study, we attempt to use the Advanced Microwave Sounding unit-A (AMSU-A) window channel brightness temperature (TB) observations to examine their relationship with the sea surface temperature (SST). A close relationship is found between the AMSU-A window channel TB and SST in the tropical oceans. A TB Niño Index (TNI) of AMSU-A window (sounding) channels is found to sometimes reach its maximum for about 2 months ahead of Oceanic Niño Index (ONI) for half of El Niño cases from 1998 to 2020. The AMSU-A sounding channels TNI had minimum values about 1–3 months behind ONI peak month. Spatial correlation maps of AMSU-A window channel TB anomalies with TNI are similar to those of SST anomalies with ONI, featuring a horseshoe-shaped pattern in the tropical Pacific and a dipole-shaped pattern in the tropical Indian Ocean. The AMSU-A retrieved liquid water path (LWP) is also highly correlated with the ONI, with larger values in El Niño years. An active development of tropical convective disturbances and their movement associated with westerly wind bursts may be the reason for the peak TNI to occur 1–3 months earlier than the peak ONI. The potential for AMSU-A TB observations to reveal El Niño signals adds additional satellite data source for El Niño research. [ABSTRACT FROM AUTHOR]

Satellite microwave temperature sounding data have been widely used in the study of climate trends over the past decades. When merging advanced microwave sounding unit-A (AMSU-A) data from 1998 to 2017 from different satellites, brightness temperature observations can be affected by differences in the center frequency, incidence angle, and local equator crossing time (LECT) among the instruments. Atmospheric drag and gravity variations with latitude can also cause orbital drift that leads to changes in LECT of the same instrument. Inter-sensor calibration and diurnal correction are thus necessary before applying AMSU-A data to climate studies. In this study, AMSU-A data from the National Oceanic and Atmospheric Administration's (NOAA's) -15, -18, and -19 satellites and the European Organization for the Exploitation of Meteorological Satellites MetOp-A/-B collected during 1998–2017 were first inter-calibrated by a double difference method to remove inter-sensor biases. AMSU-A data from NOAA-18 was used as the reference for the double difference inter-sensor calibration. A diurnal correction was then applied to data over the Amazon rainforest to eliminate the effects of different LECTs. Finally, linear and nonlinear climate trends were calculated to show that warming (cooling) trends over the Amazon rainforest for window and tropospheric AMSU-A channels 1–8 and 15 (stratospheric channels 9–14) significantly decreased (increased) if inter-sensor calibration and a diurnal correction was applied. The nonlinear climate trends reveal more rapid warming trends for tropospheric sounding channels 3–7 and 15 and less rapid cooling trends for stratospheric channels 10–12 during 1998–2008 than during 2008–2017. Channels 1–2 (channel 13) have cooling (warming) and warming (cooling) trends before and after 2008, respectively. [ABSTRACT FROM AUTHOR]