基于扩散滤波的多尺度三维变分研究

将扩散方程引入三维变分分析,揭示了传统3D-VAR不能有效提取多尺度观测信息的根本原因,即观测导致的目标函数梯度在空间的不连续分布.将扩散滤波融入基于梯度的最优化算法,发展了基于扩散滤波的多尺度3D-VAR.海表面温度数据同化试验结果表明,新方法可从长波至短波有效地提取多尺度的观测信息.     

An energy-controlling boundary condition for partial wave reflections in the mild slope equation

An energy-controlling technique to actively manage the reflective property of waves from solid boundary is presented. As linear waves propagate through an energy-controlling area, a reduction in wave heights occurs due to energy dissipation, which can be placed under direct control through the imaginary part of the wavenumber and phase velocity. Based on this relationship, the present study investigates a new method to control reflected waves with desired heights in the mild slope equation model. The method is validated through numerical tests for various reflection coefficients and the results confirm the promising use of energy-controlling boundary condition for partial wave reflections.     

秦皇岛地区雾天气气候特征及预报

随着交通线路的快速发展,雾对交通运输的影响越来越大,浓雾天气常常引发交通事故,给人民的生命财产造成严重损失,危害性极大.应用秦皇岛1957-2006年"气表-1"资料,统计分析秦皇岛内陆、沿海地区雾天气时空分布特征,返查1981-2000年历史天气图,得出本地区出现雾的主要天气类型,筛选出影响雾的主要天气要素,建立MOS预报方程,为雾预报及临近预警信号的发布、解除提供有利依据,达到监测、预报之目的.     

拉萨紫外辐射特征分析及估算公式的建立

本文利用2005~2010年的辐射观测资料对拉萨地区紫外辐射的时间变化特征及紫外辐射与总辐射比值的变化特征进行了分析,结果表明,紫外辐射与太阳总辐射的变化规律基本一致,日变化表现为正午大、早晚小;季变化特征是夏季6月最大,冬季1月最小.紫外辐射日累积值6年平均为0.87 MJ·m-2·d-1;紫外辐射有逐年递减的趋势.紫外辐射与总辐射比值也存在着明显日变化,表现为正午大、早晚小的规律;其季节变化也是夏季最大,冬季最小.紫外辐射与总辐射比值6年平均为0.0418;紫外辐射与总辐射比值也呈现逐年递减的趋势.利用2010年大气质量数和晴空指数,建立了适合于拉萨紫外辐射估量的公式,估算值的瞬时值与观测值的平均相对误差最大为8.66%,紫外辐射日累积重构值与观测值平均相对误差仅为5.5%.     

双平方根单程波动方程叠前τ偏移方法

本文将常规双平方根（DSR）单程波动方程从深度域变换到双程垂直走时（τ）域,由此推导出可从数学上实现“沉降观测”的单程波DSR传播算子. 其递归波场延拓算法包含波数域针对常速背景的相移处理和空间域针对横向速度扰动的相位校正,可以应对上覆地层速度横向变化对构造成像的影响. 结合零炮检距、零时间成像条件,提出了在τ域进行波场延拓与成像的DSR方程叠前偏移新方法. 为了克服其全三维偏移算法在实际应用中可能面临的困难,本文采用稳相近似,在crossline常炮检距偏移理论基础上推导了实用的共方位角叠前τ偏移方法. 数值试验表明,DSR方程叠前τ偏移在强横向非均匀介质中的成像精度与分辨率优于传统的时间域成像技术.     

On the scavenging of SO2 by cloud and raindrops: II. An experimental study of SO2 absorption and desorption for water drops in air

For the purpose of testing our previously described theory of SO₂ scavenging a laboratory investigation was carried out in the UCLA 33 m long rainshaft. Drops with radii between 250 and 2500 m were allowed to come to terminal velocity, after which they passed through a chamber of variable length filled with various SO₂ concentrations in air. After falling through a gas separating chamber consisting of a fluorocarbon gas the drops were collected and analyzed for their total S content in order to determine the rate of SO ₂ absorption.The SO₂ concentration in air studied ranged between 1 and 60% (v). Such relatively large concentrations were necessary due to the short times the drops were exposed to SO₂ in the present setup. The present experimental results were therefore not used to simulate atmospheric conditions but rather to test our previously derived theory which is applicable to any laboratory or atmospheric condition. Comparison of our studies with the results from our theory applied to our laboratory conditions led to predicted values for the S concentration in the drops which agreed well with those observed if the drops had radii smaller than 500 m. In order to obtain agreement between predicted and observed S concentrations in larger drops, an empirically derived eddy diffusivity for SO₂ in water had to be included in the theory to take into account the effect of turbulent mixing inside such large drops.In a subsequent set of experiments, drops initially saturated with S (IV) were allowed to fall through S-free air to determine the rate of SO ₂ desorption. The results of these studies also agreed well with the results of our theoretical model, thus justifying the reversibility assumption made in our theoretical models.In a final set of experiments, the effects of oxidation on SO₂ absorption was studied by means of drops containing various amounts of H₂O₂. For comparable exposure times to SO₂, the S concentration in drops with H₂O₂ was found to be up to 10 times higher than the concentration in drops in which no oxidation occurred.     

Contamination of a well in a uniform background flow

In this paper we describe the transport of pollution in groundwater in the neighbourhood of a well in a uniform background flow. We compute the rate at which contaminated particles reach the well as a function of the place of the source of pollution. The motion of a particle in a dispersive flow is seen as a random walk process. The Fokker-Planck equation for the random motion of a particle is transformed using the complex potential for the advective flow field. The resulting equation is solved asymptotically after a stretching transformation. Finally, the analytical solution is compared with results from Monte Carlo simulations with the random walk model. The method can be extended to arbitrary flow fields. Then by a numerical coordinate transformation the analytical results can still be employed.     

On the general character of coastal flow

It is well-known that some coastal currents such as the Florida Current have a recirculating region on their coastal side, while other currents such as the East Australian Current have no such region. Under the hypothesis that the form drag on a coastal current by an irregular coastal topography reduces the momentum transport of the current to a minimum for the given flow within the current, the hydraulic theory of coastal currents is used to explain this behaviour.     

波动方程模拟P ML吸收影响因素分析

地震数值模拟中,完全匹配层( PML)边界能有效地吸收衰减地震波,得到无边界反射干扰的波场快照和地震记录。在前人研究的基础上,进一步分析了分裂完全匹配层( SPML)的衰减机制,通过波动方程模拟,分析了震源主频、空间网格间距、介质速度等参数对SPML边界吸收衰减特征的影响。得到了不同条件下,PML边界对地震波的吸收效果。通过分析对比,得出了震源主频对PML吸收效果无直接影响,空间网格间距与PML吸收效果成反比,高速层PML吸收效果缓慢等结论。并通过Marmousi2模型测算,对结论进行了验证。最后给出了复杂模型、不同尺度数值模拟中SPML参数的选取方法,为地震数值模拟中SPML边界参数的定量选取奠定了基础。