||A study to develop a new wind estimation method to elucidate the general circulation of the Venus atmosphere
Venus is covered with thick clouds of sulfuric acid that are present at 45-70 km above its surface. The winds on Venus are faster than the planet's rotation at all altitudes above its surface, in the same direction as its rotation.This phenomenon is known as super-rotation. In particular, at the cloud top level (65-70 km) of Venus, the winds reach up to 100 m s-1, which is 60 times the speed of the planet's rotation at the equatorial region. One of the proposed mechanisms for the generation of super-rotation is the Gierasch mechanism, which has been supported by numerical simulations. In the Gierasch mechanism, super-rotation is generated by angular momentum transport in the poleward direction via meridional circulation and in the equatorward direction via eddies. To verify this mechanism, it is necessary to analyze the eddy transport of angular momentum on a horizontal scales of less than several thousand kilometers. However, previous studies have proven this to be a difficult task. The purpose of this study is to develop a method of cloud tracking and accuracy evaluation that is able to provide the datasets required to analyze this phenomenon on the horizontal scale. Here, the Venus Monitoring Camera onboard the European Space Agency's Venus Express, which captures UV images at 365 nm, is used for the cloud tracking. In the conventional cloud tracking methods, the cross-correlation coefficients between the sub-images of two images are computed, and cloud motion is estimated by selecting the maximum of the cross-correlation coefficients under the condition that the cloud morphology is advected by the wind. However, the estimation accuracy is reduced by noise and the time-evolving cloud morphology in the brightness image, etc. The cloud tracking method developed here resolves this problem by utilizing multiple images acquired simultaneously in a short time interval. The superposition improves the accuracy of wind velocity measurements and reduces false pattern matches that cause large errors by increasing the number of images. The improved results obtained in this study helped to clarify the Gierasch mechanism. In previous studies, a method of evaluating the accuracy of wind velocity measurements had not been established. Moreover, in several of these studies, the evaluation was performed based on the natural variability of the wind velocities. As a result, such evaluation methods are not suitable when the error is much smaller than the natural variability, because natural variability is regarded as part of the error. In dynamical studies, reliable error estimation in wind velocity is necessary at each grid point. In the present study, we developed evaluating (1) the precision in the wind measurements based on the lower con dence bound of cross-correlation and (2) the error in each wind measurement based on the two estimates of wind measurement. In (2), multiple images obtained during an orbit can be used to estimate the error by subdividing them into two groups, performing cloud tracking for each group, and comparing the results. The statistical error was obtained from screening using the two methods described above to evaluate the accuracy and error of each wind (< 45◦ S) about 210-298 days and 436-530 days after the spacecraft entered the Venusian orbit on 20 April, 2006. At low latitudes, the median of accuracy obtained using method (1) was about 8 m s-1, and the error obtained from comparing winds using method (2) was about 2 m s-1. The error in the wind measurement, which was less than 10 m s-1, was estimated by visually tracking clouds over successive images. Based on the results, we show that, using the winds derived by the proposed method, it will be possible to analyze physical phenomena taking place in less than several thousands of kilometers. The proposed method is now being applied to cloud tracking using the observation data of the Venus meteorological satellite, known as the Akatsuki orbiter, which was inserted into orbit on 7 December, 2015.
Hokkaido University（北海道大学）. 博士(環境科学)