【作者】 黄倩；

【导师】 张镭； John H. Marsham； Doug J. Parker；

【作者基本信息】 兰州大学 ， 大气物理学与大气环境， 2008， 博士

【摘要】 边界层对流在热量、水汽、动量以及气溶胶的传输中起着重要作用。不同的下垫面不仅影响空气中的物理特性,而且还能影响边界层对流的形式及其发展高度。边界层对流卷是平坦地形上较常见的边界层对流。当控制深对流触发的中尺度辐合较弱时,边界层对流可以触发深对流的发生。地形的变化能改变边界层的气流,如在起伏不平的山地,对流的分布往往是由迎风坡的坡度和宽决定的。而在极端干旱的沙漠地区,较强的地表感热通量引起的深厚边界层干对流使沙漠地区夏季的边界层厚度可以达到5km以上的高度。深厚的边界层对流对沙尘的传输也有重要影响。本文利用一大涡模式（LEM）模拟研究了平坦地形上边界层对流卷形成和消散的地表热通量和风切变的条件,比较了对流卷和非对流卷在深对流的触发中的不同。还选取了一实际的半干旱复杂山区用复杂地形上的边界层模式（BLASIUS）,在不同的天气条件下模拟研究了半干旱山区上空的垂直速度场分布和对流特征以及地形对热力对流活动的影响。最后在极端干旱的沙漠地区,用LEM模式模拟了撒哈拉沙漠地区特有的边界层结构,并探讨了非均匀的地表热通量对撒哈拉对流边界层和撒哈拉残留层、以及对流边界层到残留层之间传输的影响。边界层中有不同形式的对流活动。而水平对流卷是边界层中普遍存在的对流形式之一。它在深对流的触发中起着重要作用。对流卷不仅可以与中尺度辐合带相互作用为深对流的发展提供有利地点,而且它本身也可以触发深对流。本文利用英国气象局的三维大涡模式（LEM）模拟研究了Small Cumulus Microlohysical Study（SCMS）外场试验期间8月15日观测的水平对流卷。模拟的边界层变量与实测的探空廓线和飞机观测结果相比,尽管模拟的虚位温和水汽混合分别比观测的大0.5K和减小0.2gkg^-1,但是模拟和观测的垂直速度场具有相似的变化范围(-2或-3 ms^-1到2.5ms^-1),而且模式能较合理地模拟出实测的边界层对流卷。利用此模式我们还考察了对流卷的形成对地表热通量和地转风切变值的大小依赖关系。模拟的个例结果显示对流卷维持所需要的地表热通量的上限是110Wm^-2（由于模式运行所需的自旋（spin-up）时间的限制,我们无法判断对流卷维持所需的地表热通量值的下限）。当边界层风切变大于5×10^-3 s^-1时,有对流卷形成。模拟结果还表明,当5≤－Zi/L≤45时,对流卷开始消散。这个临界值的大小与包含风切变、对流边界层高度和对流速度尺度的无量纲量（如（?））有关。我们还对垂直速度场和余弦函数进行了二维相关分析,从而定量地评估了边界层对流卷的线性（用0.25作为对流卷形成的临界值）。为了探讨边界层对流卷在深对流触发中的重要性,我们比较了边界层对流卷和非对流卷的个例（对流卷和非对流卷的实验中,除了风和风切变不同,其它模拟条件都相同）。结果表明,在对流卷和非对流卷的个例中,是初始气块的变化对对流抑制能量（CIN）的大小有主要贡献,而初始气块抬升逆温层顶的过程对CIN的变化影响较小。另外,非对流卷具有较强的湿上升气流,而在对流卷中上升和下沉运动的分布较对称,而且它的下沉气流比非对流卷的强。因此非对流卷具有更小的CIN值,这使非对流卷的云顶高度较对流卷的早15分钟开始增加。而对流卷的个例中云层的加深较非对流卷的迅速,从而使两者的云顶高度约在同一时间达到最大（2.7km）。本文还利用高分辨率的数值模式（BLASIUS）,从西北地区选取了一实际的复杂山区进行了一系列的理想数值模拟。分析了在不同天气条件下（不同的大尺度地转风和大气静力稳定度下）山区上空的垂直速度场分布和对流特征以及地形对热力对流活动的影响。模式结果表明,当地转风速较小(如风速为5ms^-1和10ms^-1的情况)（Froude数小于0.5的情况）的条件下,气流往往被山峰阻塞而在迎风坡造成地形强迫和辐合性抬升,从而易在迎风坡触发深对流活动;在背风坡则由于迎风坡的绕流重新辐合而造成垂直运动。绕流的辐合是触发深对流活动的另一重要因子。当地转风速较大时(20ms^-1),气流很容易越过山脊,地形重力波容易在山地下游被激发。另外还分析了复杂地形上空热力对流的特征。结果显示,模式积分3h后,有对流卷在山脊的周边产生;模式积分6h后,一些强度较弱的对流卷开始消失或者与其它的对流卷合并,而且对流卷的波长也有所增加。另外,在大气静力稳定度较小时,对流卷的强度也较强。因此对流活动在稳定度较小时可以延伸到更高的高度。如果对流上升运动正好与重力波相拌的上升运动区相合,在此区域内的垂直上升运动将得到加强。相反地,如果对流上升运动正好与重力波相伴的下沉运动区相重合,在此区域内的对流上升运动将受到抑制。重力波除了与对流活动相耦合,使气团上升到较高的高度外,如果重力波的走向与对流卷的走向夹角不大时,重力波对对流卷走向的影响较显著。因此我们认为重力波的走向很可能会影响到深对流系统的传播路经。实测结果表明撒哈拉对流边界层（CBL）增温2K与撒哈拉地表反照率分布不均匀（如尺度约为20km的岩石区地表反照率是0.2,而周围沙漠的地表反照率为0.45）密切相关。进而猜测地表反照率分布不均匀也可能影响撒哈拉残留层（SRL）的垂直混合。由于缺少SRL的观测资料,这一猜测只能由数值模拟结果来证实。因此我们GERB Intercomparison of Longwave and Shortwave Radiation（GERBILS）外场试验期间两个观测个例为基础,利用大涡模式进行了大量的敏感性数值模拟实验,研究了非均匀的地表热通量对撒哈拉对流边界层、撒哈拉气层及CBL到SRL的传输的影响。模拟结果表明,地表较热的区域里（在模式的底层取一小块,使它表面的热通量高于周围区域表面的热通量）CBL中产生的上升和辐合运动,加强了CBL到SRL的传输;而热区周围的下沉运动又抑制了CBL的增长。在环境风速较小,非均匀的地表热通量加强的条件下,如果保持模拟区域的地表热通量保持不变,即增加热区地表热通量但减少热区周围的地表热通量时,导致对流边界层变冷、厚度减小;如果是热区周围区域的地表热通量保持不变,则对流边界层变暖,且厚度减小。为了进一步研究地表热通量的非均匀性对CBL和SRL之间物质传输的影响,我们在模式的底层（200m高度以下）加入了被动示踪物（值为100）。通过分析不同的（?）或（?）（这里M和D分别是地表热通量和热区水平尺度的增大倍数,U是1000m高度上的初始风速）。对应的示踪物浓度的垂直廓线表明,（?）或MD增大时（风速较小,非均匀的地表热通量强度较大时）不仅使对流边界层（CBL）的厚度减小,而且可以加强示踪物从对流边界层（CBL）到撒哈拉残留层（SRL）的传输。模拟结果还显示,当环境风速增加时,由于热区引起的辐合运动加强了它西侧的风速,从而导致示踪物在热区西侧的抬升有所加强。而且当环境风速较低时(小于15ms^-1),由于非均匀的地表热通量可以引起局地风速的增大,从而增加了总体示踪物抬升的趋势。更多还原

【Abstract】 Boundary layer convection plays a significant role in transport of heat, moisture, momentum and aerosols. Different underling surfaces affect not only the physical properties of the air but also the forms and developments of convection. Convective rolls have been shown to be common in the boundary layer. When the mesoscale forcings that often controls the deep convection are weak, convective rolls can provide a significant control on convective initiation. Variety terrains can change the flow within the boundary layer. The distribution of convection over an orographic area is determined by the depth and the width of a mountain. In extremely dry Saharan area, the depth of the boundary layer can reach upto 5km due to the very deep dry convection which is caused by the strong surface heat fluxes. Deep boundary layer convection also has important effects on the transport of dust. In this paper, a large eddy model （LEM） was used to simulate the necessary conditions of surface heat flux and wind for the formation and dissipation of boundary rolls, with comparing the differences of rolls and nonroll convection on initiation of deep convection. A high resolution boundary layer model （BLASIUS） was also utilized to investigate the characteristics of vertical velocity fields and thermal convection over complex terrain and effects of orography on them. The LEM model was applied over the Saharan desert to simulate the boundary structure in this area. The effects of surface flux anomalies on Saharan CBL, Saharan SRL and transport from the CBL into the SRL were investigated.Boundary layer convection can take variety forms. Convective rolls have been shown to be common in the boundary layer. Rolls play an important role in the deep convection initiation. Not only the interactions of rolls with the mesoscale convergence zones create preferred locations for convective development, but rolls themselves can initiate storms. In this paper we use the Met Office large eddy model （LEM） to explicitly resolve boundary layer rolls observed on 14 August 1995 during the Small Cumulus Microphysical Study （SCMS） field campaign. Comparing with the sounding and the aircraft data, average the simulations are 0.5K warmer and 0.2gkg^-1 drier than the observations, but the modeled and observed vertical velocity fields have similar range (-2 or -3 to 2.5ms^-1). And the observed boundary layer rolls were reproduced by the model. We also investigated how the rolls relate to the surface heat fluxes and the magnitude of the geostrophic wind shear. For the case modeled, rolls persisted for surface buoyancy fluxes less than 110Wm^-2 （because the limitation of the spin-up time, the lower value of surface heat fluxes for rolls persistence was not determined）. Rolls formed for boundarylayer wind shears greater than 5×10^-3s^-1. Rolls started to decay when - Z_i, /L exceededa threshold, with a value between 5 and 45. The value was found to depend on a non-dimensional combination of the low-level wind shear, the height of the CBL, and theeddy velocity scale （e.g.（?））. Two dimensional correlation between w and a cosinefunction was performed to measure the linearity of the rolls （a threshold of 0.25 used as the criteria for linear convection）.In order to investigate the importance of the rolls on the initiation of deep convection, simulated rolls convection was compared with nonroll simulation, which was identical except for the wind and wind shear used. In both the roll and nonroll cases the variability in convective inhibition （CIN） was dominated by the source air, rather than the lifting of the top of the boundary layer by the convection. Stronger moist updrafts existed in the nonroll convection, whereas roll convection gave a more symmetrical distribution of up and downdrafts, with stronger downdrafts than the nonroll case. The nonroll convection simulations have lower minimum values of CIN and clouds develop 15 min earlier this case. However, in the roll case clouds deepen slightly more rapid than in the nonroll case, so the cloud tops reach 2.7km at approximately the same time.In this paper, a series of idealised model simulations have been performed over a real hilly terrain located in north west region using a high resolution boundary layer model. The model used is Met Office boundary layer model called Boundary Layer Above Stationary, Inhomogeneous Uneven Surface （BLASIUS）. Based on those simulations, the characteristics of vertical velocity fields and thermal convection over complex terrain are investigated and effects of orography on them are diagnosed. The simulated results show when the geostrophic wind is low (which is 5ms^-1 or 10ms^-1) （Froude number is less than 0.5）, air flow tends to be blocked and upward motion can be forced by orography and convergence over windward slope. Under such circumstances deep convection is likely to be triggered. Due to the division of airflow over windward slope the vertical motion can also be induced in the lee because of convergence. The convergence in the lee of mountains is another important factor which may cause deep convective systems. When the gostrophic wind is high, air can climb over the mountain easily, and gravity waves can be induced at the downwind of mountains.We also analysed the characters of thermal convection over the complex terrain. The simulated results show that the rolls appeared around the mountains after 3 hours of model running. After 6 hours of the model run, weak rolls started to decay or combined with others, with the increase of rolls wavelength. And rolls tend to be strong with the lower static stability value. So the convection can develop to the higher altitude with small static stability value. The vertical updrafts were enhanced at where the convective updrafts combined with the upward motion of the gravity wave. Conversely, the vertical updrafts were inhibited when convective updrafts corresponded with the downward motion of the gravity wave. Not only the gravity wave moved the air to the upper altitude, but it altered the direction of the rolls when the direction of gravity wave was close to that of rolls. So we suggest that the gravity wave might affect the transmitting of the deep convection.The observation showed that a rocky area approximately 20km across, with an albedo of approximately 0.2 compared with 0.45 for the surrounding desert, was linked to a CBL temperature increase of approximately 2K. Such variations may significantly affect the vertical mixing of the SRL. This assume has to be confirmed by the numerical simulation due to the lack of the observed data in SRL. Using two cases based on observations from the GERBILS （GERB, Intercomparison of Longwave and Shortwave radiation） field campaign, large eddy model （LEM） simulations have been used to investigate the effects of surface flux anomalies on the growth of the summertime Saharan convective boundary layer （CBL） into the Saharan Residual layer （SRL） above, and transport from the CBL into the SRL.Modeled results show that hot surface anomalies （increase the surface flux over the warm patch） generated updraughts and convergence in the CBL that increased transport from the CBL into the SRL. The induced subsidence in regions away from the anomalies inhibited growth of the CBL there. Under stronger surface flux anomalies and lower ambient wind conditions, if the domain averaged surface flux kept constant this led to a shallower, cooler CBL, while if fluxes outside the anomaly were kept fixed this gave a warmer, shallower CBL.In order to further understand the effects of surface anomalies on the transport between CBL and SRL, a passive tracer with a fixed value of 100 was added below the 200m model level in all simulations. Vertical distributions of the horizontally-averaged passivetracer concentrations at different values of （?） or MD show that not only do strongersurface flux anomalies with lighter winds tend to decrease the CBL depth （using balanced or unbalanced surface fluxes）, but they also enhance the vertical transport from the CBL into the SRL. The simulations also show that the uplift of tracer was enhanced at the west of the warm patch because the convergence increase the wind at the west of the patch, with the ambient wind increased. When the ambient wind speed is low (< 15ms^-1), anomaly tend to increase tracer uplift due to anomalies lead to locally increase wind speeds.更多还原