This study explores the updraft and downdraft properties of mature stage mesoscale convective systems (MCSs) in terms of draft core width, shape, intensity, and mass flux characteristics. The observations use extended radar wind profiler (RWP) and surveillance radar data sets from the U.S. Department of Energy Atmospheric Radiation Measurement program for midlatitude (Oklahoma, USA) and tropical (Amazon, Brazil) sites. MCS drafts behave qualitatively similar to previous aircraft and RWP cloud summaries. The Oklahoma MCSs indicate larger and more intense convective updraft and downdraft cores, and greater mass flux than Amazon MCS counterparts. However, similar size‐intensity relationships and draft vertical profile behaviors are observed for both regions. Additional similarities include weak positive correlations between core intensity and core width (correlation coefficient r ∼ 0.5) and increases in draft intensity with altitude. A model‐observational intercomparison for draft properties (core width, intensity, and mass flux) is also performed to illustrate the potential usefulness of statistical observed draft characterizations. Idealized simulations with the Weather Research and Forecasting model aligned with midlatitude MCS conditions are performed at model grid spacings (△x ) that range from 4 km to 250 m. It is shown that the simulations performed at △x = 250 m at similar mature MCS lifecycle stages are those that exhibit draft intensity, width, mass flux, and shape parameter performances best matching with observed properties.

中文翻译：

---

雷达风廓线仪揭示的上，下气流核心尺寸和强度：MCS观测和理想模型比较

这项研究探讨了成熟阶段的中尺度对流系统（MCS）的上升和下降特性，包括吃水宽度，形状，强度和质量通量特性。观测使用来自美国能源部大气辐射测量计划的扩展雷达风廓线仪（RWP）和监视雷达数据集，用于中纬度（美国俄克拉荷马州）和热带地区（巴西亚马逊）。MCS草图的质量在本质上类似于以前的飞机和RWP云摘要。俄克拉荷马州的MCS显示出比Amazon MCS对应的更大，更强的对流上升气流和下降气流核心，以及更大的质量通量。但是，在两个区域都观察到了相似的尺寸强度关系和吃水垂直剖面行为。r〜0.5），吃水深度随海拔高度而增加。还对吃水特性（核心宽度，强度和质量通量）进行了模型观测比对，以说明统计观察到的吃水特征的潜在有用性。使用天气研究和预报模型与中纬度MCS条件对齐的理想化模拟是在模型网格间距（△ x）为4 km至250 m的范围内进行的。结果表明，在类似的成熟MCS生命周期阶段，在△ x = 250 m处进行的仿真显示出吃水强度，宽度，质量通量和形状参数性能与观察到的特性最匹配。