An evaluation of uncertainties in monitoring middle atmosphere temperatures with the ground-based lidar network in support of space observations

The capability of the longest lidar data sets to monitor long-term temperature changes have been evaluated through comparisons with the successive Stratospheric Sounder Units (SSU) onboard NOAA satellites. Cross-consistency investigations between SSU and the lidar network can be considered as a first attempt to demonstrate how the synergistic use of space and ground-based instruments could provide reliable monitoring of the temperature of the middle atmosphere. The breakdown of the temperature cooling trend, and the following flattening observed in the satellite temperature series, is qualitatively confirmed by the lidars. However, there are still large differences that can either be due to SSU continuity (orbit drifts or weighting function modifications) or lidar operation changes (time of measurements, accuracy, sampling, etc.). SSU vertical weighting functions have been taken into account for comparisons. Some discontinuity events cannot be explained by the SSU weighting function drifts due to CO2. For the upper channels of SSU (peaking around 50 km), the results are probably sensitive to the mesospheric part of the lidar profiles that can explain some discontinuities. Tropical lidar stations show clear inter-annual differences with the SSU channels covering the lowest altitude range that needs further investigations to understand if the origin is instrumental or geophysical. An attempt to derive non-linear trends with combinations of linear, hockey stick, and quadratic functions has been made. While the quadratic term is not highly significant, this approach allows the derivation of a better quantification of the linear trend terms.

Research highlights
► The synergy between space observations using successive similar payloads and ground references for insuring a global temperature monitoring of the middle atmosphere was demonstrated with the lidar network and the SSU series.
► Three proxies have been proposed to monitor lidar time evolution: time data distribution, uncertainty at 50 km and estimated tidal anomalies.
► The temperature trend exhibits a reduction after 1997. The reason for such slowdown is not yet clear, while the impact of ozone recovery is expected later.