A new visibility parameterization for warm-fog applications in numerical weather prediction models
A new visibility parameterization for warm-fog applications in numerical weather prediction models
复制标题
DOI：
10.1175/jam2423.1
复制
发表时间：
2006-11-01
影响因子：
3
通讯作者：
Boybeyi, Z.
中科院分区：
文献类型：
--
作者：
Gultepe, I.;Mueller, M. D.;Boybeyi, Z.
本文的目的是为数值天气预报（NWP）模式提出一种新的暖雾能见度参数化方案。利用辐射和气溶胶云实验期间收集的代表边界层低层云的现场观测资料，建立了能见度与液滴数、浓度N-d和液态水含量（LWC）的组合参数之间的参数化方案。目前的NWP模式通常使用消光系数与LWC之间的关系。将一种新的能见度参数化方案Vis = f（LWC, Nd）应用于NOAA非流体静力中尺度模式。在该模型中，雾的微物理特性采用ID参数化雾（PAFOG）模型，并应用于1.5 km以下的大气。为了测试新的参数化方案，在以瑞士苏黎世独特机场为中心的50公里最内层嵌套模拟域中使用1公里的水平网格间距进行了模拟。10 h的模拟结果表明，新旧参数化方案的能见度差异可达50%以上。得出的结论是，准确的能见度估计需要熟练的LWC和从预报中估计Nd。因此，根据环境条件，目前的模型可能明显高估/低估Vis（不确定性超过50%）。在参数化中包含N-d作为预测（或参数化）变量将显著改进业务预测模型。
The objective of this work is to Suggest a new warm-fog visibility parameterization scheme for numerical weather prediction (NWP) models. In situ observations collected during the Radiation and Aerosol Cloud Experiment, representing boundary layer low-level clouds, were used to develop a parameterization scheme between visibility and a combined parameter as a function of both droplet number concentration N-d and liquid water content (LWC). The current NWP models usually use relationships between extinction coefficient and LWC. A newly developed parameterization scheme for visibility, Vis = f(LWC, Nd), is applied to the NOAA Nonhydrostatic Mesoscale Model. In this model, the microphysics of fog was adapted from the ID Parameterized Fog (PAFOG) model and then was used in the lower 1.5 km of the atmosphere. Simulations for testing the new parameterization scheme are performed in a 50-km innermost-nested simulation domain using a horizontal grid spacing of 1 km centered on Zurich Unique Airport in Switzerland. The simulations over a 10-h time period showed that visibility differences between old and new parameterization schemes can be more than 50%. It is concluded that accurate visibility estimates require skillful LWC as well as Nd estimates from forecasts. Therefore, the current models can significantly over-/ underestimate Vis (with more than 50% uncertainty) depending on environmental conditions. Inclusion of N-d as a prognostic (or parameterized) variable in parameterizations would significantly improve the operational forecast models.