Horizontal Pressure Gradient Errors of the Monterey Bay Sigma Coordinate Ocean Model with Various Grids
Horizontal Pressure Gradient Errors of the Monterey Bay Sigma Coordinate Ocean Model with Various Grids
复制标题
不同网格蒙特利湾西格玛坐标海洋模型的水平压力梯度误差
DOI：
--
复制
发表时间：
1999
期刊：
影响因子：
--
通讯作者：
Lin Jiang
中科院分区：
文献类型：
--
作者：
L. Ly;Lin Jiang
利用普林斯顿海洋模式（POM）和网格生成技术（GGT）建立了一个具有真实海底地形和海岸线的蒙特利湾（MOB）近岸海洋σ坐标模式，研究了MOB陡峭地形引起的水平压力梯度误差。MOB的海底峡谷是世界海洋中最陡峭的地形之一。MOB网格是使用EAGEAL View和GENIE++网格生成系统设计的。在本研究中使用了Ly和Luong（1993）开发的网格包来将网格耦合到模型。用正交和曲线近正交（CNO）网格对MOB模型进行了测试。CNO网格的水平分辨率为300 ~ 2km，而正交网格的分辨率为1.25km和1.38km。这些网格覆盖了180 × 160 km的区域，网格点数量相同，为131 × 131。垂直分辨率为25，35和45垂直西格玛水平进行了测试。根据不同网格的平均动能和速度，垂直、水平分辨率和σ分布，以及海底地形平滑，对MOB中的误差进行了评估。不同网格下的模拟结果表明，GGT可以作为海岸海洋模拟中除提高分辨率和平滑海底地形外的另一种减小σ坐标误差的工具。地形平滑不仅降低了地形坡度，而且改变了真实的地形。与网格点数量相同的矩形网格相比，沿着陡坡和Monterey Submarine峡谷填充的网格密度较高的CNO网格可将误差降低40%。CNO网格比矩形网格更有效，因为它的大部分网格都在水面上。模拟结果表明，所提出的MOB σ坐标模型可以使用的水平压力梯度误差的信心。
A coastal ocean σ-coordinate model of Monterey Bay (MOB) with realistic bottom topography and coastlines is developed using the Princeton Ocean Model (POM) and grid generation technique (GGT) to study the horizontal pressure gradient errors associated with the MOB steep topography. The submarine canyon in MOB features some of the steepest topography encountered anywhere in the world oceans. The MOB grids are designed using the EAGEAL View and GENIE++ grid generation systems. A grid package developed by Ly and Luong (1993) is used in this study to couple grids to the model. The MOB model is tested with both orthogonal and curvilinear nearly-orthogonal (CNO) grids. The CNO grid has horizontal resolution which varies from 300 m to 2 km, while the resolution of the orthogonal grid is uniform with δx = 1.25 km and δy = 1.38 km. These grids cover a domain of 180 × 160 km with the same number of grid points of 131 × 131. Vertical resolutions of 25, 35 and 45 vertical sigma levels are tested. The error in the MOB are evaluated in terms of mean kinetic energy and velocity against various grids, vertical, horizontal resolution and σdistributions, and bottom topography smoothing. Simulations with various grids show that GGT can be used as another tool in reducing σ-coordinate errors in coastal ocean modeling besides increasing resolution and smoothing bottom topography. Topographical smoothing not only reduces topographic slope, but changes realistic topography. A CNO grid with a high grid density packed along steep slopes and Monterey Submarine Canyon reduces the errors by 40% compared to a rectangular grid with the same number of grid points. The CNO grid is more efficient than the rectangular grid, since it has most of its grids over water. The simulations show that the presented MOB σ-coordinate model can be used with a confidence regarding horizontal pressure gradient error.