基于4D-VAR的动态非对称台风初值化方法试验研究

EXPERIMENTAL STUDY ON A DYNAMIC ASYMMETRICAL TYPHOON INITIALIZATION SCHEME BASED ON 4D-VAR

经典BDA台风初值化方案中台风模型为一个不能充分反映个别台风具体特征的轴对称海平面涡旋场,且在同化窗口内强迫发展着的台风向这个静止涡旋场适应。本文针对该特点,提出一种充分融合分析场信息和实况信息并考虑副高影响的非对称台风Bogus方法,利用PSU/NCAR的MM5模式及其伴随模式,通过四维变分同化(4D-VAR)技术,建立动态非对称BDA方案,对多个不同时次的Bogus资料进行同化。通过对夏季受副热带高压影响较大的6个台风的数值模拟试验对比发现,新方案的路径模拟普遍优于经典方案,且对于经典方案没能改善路径模拟的时段,新方法的模拟效果却能得以改进;新方法模拟的强度普遍弱于经典方案,但这种趋势使得新方法的中后期模拟强度较经典方案更接近于实况。另外,当实况中台风强度已大为减弱时,同化试验模拟的台风仍然保持很强的强度,造成气旋的过度发展,这种情况对BDA方案模拟效果的限制作用还需进一步研究。

An axisymmetric bogus vortex at sea level was used in the traditional bogus data assimilation （BDA） scheme. This vortex could not accurately describe the specific characteristics of the typhoon and the evolving real typhoon was forced to unreasonably adapt to this changeless vortex in that scheme. To address these problems, an asymmetrical typhoon bogus blended with information from the analysis and observation is brought forward while the impact of the subtropical anticyclone is considered as well. With the fifth-generation Penn State/NCAR Mesoscale Model（MM5） and its adjoint model, a four-dimensional variatiaonal data assimilation（4D-VAR） technique is employed to build a dynamic asymmetrical BDA scheme to assimilate different asymmetrical bogus vortexes at different time. The track and intensity of six summer typhoons that are affected by the subtropical anticyclone are simulated. The results show that, the improvement of track simulation in the new scheme is more significant than that in the traditional scheme. Moreover, the track simulation can be improved in the new scheme while it cannot in the traditional scheme. The results also reveal that the typhoon intensity of the simulation in the new scheme is weaker than that in the traditional scheme, which indicates the advantage of the new scheme： its intensity simulation during the later period is closer to the observation. In addition, the simulated typhoon intensity of both two BDA schemes at the later period is still very large while the intensity from observation has been largely weakened. It makes the typhoon overly develop during the simulation. The BDA scheme＇s limitation for long-term simulation course needs further study.