Use of SSU/MSU Satellite Observations to Validate Upper Atmospheric Temperature Trends in CMIP5 Simulations

• Published in: Remote Sensing
• Main Authors: Lilong Zhao, Jianjun Xu, Alfred M. Powell, Zhihong Jiang, Donghai Wang
• Format: Article in Journal/Newspaper
• Language: English
• Published: MDPI AG 2015
• Subjects: climate change; SSU/MSU satellite observation; upper atmospheric temperature; CMIP5 simulation; Science; Q; Antarctic; Arctic; The Antarctic; Antarc*; Climate change
• Online Access: https://doi.org/10.3390/rs8010013; https://doaj.org/article/ae28948e4e3e46a0a7dda3a8131e2d18

The tropospheric and stratospheric temperature trends and uncertainties in the fifth Coupled Model Intercomparison Project (CMIP5) model simulations in the period of 1979–2005 have been compared with satellite observations. The satellite data include those from the Stratospheric Sounding Units (SSU), Microwave Sounding Units (MSU), and the Advanced Microwave Sounding Unit-A (AMSU). The results show that the CMIP5 model simulations reproduced the common stratospheric cooling (−0.46–−0.95 K/decade) and tropospheric warming (0.05–0.19 K/decade) features although a significant discrepancy was found among the individual models being selected. The changes of global mean temperature in CMIP5 simulations are highly consistent with the SSU measurements in the stratosphere, and the temporal correlation coefficients between observation and model simulations vary from 0.6–0.99 at the 99% confidence level. At the same time, the spread of temperature mean in CMIP5 simulations increased from stratosphere to troposphere. Multiple linear regression analysis indicates that the temperature variability in the stratosphere is dominated by radioactive gases, volcanic events and solar forcing. Generally, the high-top models show better agreement with observations than the low-top model, especially in the lower stratosphere. The CMIP5 simulations underestimated the stratospheric cooling in the tropics and overestimated the cooling over the Antarctic compared to the satellite observations. The largest spread of temperature trends in CMIP5 simulations is seen in both the Arctic and Antarctic areas, especially in the stratospheric Antarctic.