科研之友
微信
新浪微博
Facebook
分享链接
摘要
Warm target effect and diurnal drift errors are the main sources of uncertainties in the trend determination from the NOAA Microwave Sounding Unit (MSU) observations. Currently, there are two methods to correct the warm target effect: 1) finding a best root-level (level-1c) calibration nonlinearity using simultaneous nadir overpass (SNO) matchups to minimize this effect for each scene radiance, and 2) finding a best-fit empirical relationship between the correction term of the end-level gridded brightness temperature and warm target temperature and then removing the best fit from the unadjusted time series. The former corrects the warm target effect before the diurnal drift adjustment and provides more accurate, warm target effect-minimized, level-1c scene radiances for reanalysis applications. The latter corrects the warm target effect at the end-level merging step, which depend on the diurnal drift correction that occurred at a previous step. Although minimized, the first method still leaves small residual warm target-related errors due to imperfect calibrations. This study demonstrates that when the diurnal drift effect is negligible, a combination of the two methods completely removes warm target effect and produces an invariant trend that is independent of the level-1c calibration in the SNO framework. The conclusion is directly applicable to the MSU channel-2 oceanic midtropospheric temperature (T-2) and global channel-3 upper-tropospheric temperature (T-3) and channel-4 lower-stratospheric temperature (T-4), which satisfy the condition of negligible diurnal drift effect. On the basis of these results, version 1.2 of the National Environmental Satellite, Data, and Information Service (NESDIS)-Center for Satellite Applications and Research (STAR) multisatellite MSU time series was constructed, including all T-2, T-3, and T-4 products. In addition, a diurnal drift correction based on the Remote Sensing Systems diurnal anomalies was applied to the T-2 product, which produces consistent climate trends between land and ocean. The global long-term climate trends for T-2 and T-4 derived from the STAR V1.2 dataset are, respectively, 0.18 +/- 0.05 and -0.39 +/- 0.36 K decade(-1) during 1979-2006; the T-3 trend is 0.11 +/- 0.08 K decade(-1) for 1981-2006.
- 出版日期2010-11
