白令海冰间湖的数值模拟及影响模拟准确度的关键因素

A Modeling of Bering Sea Polynyas and Analysis of Key Factors Impacting Simulation Accuracy

冬季在北白令海陆架区域频繁地出现潜热冰间湖,对当地的生态系统和北极盐跃层贡献很大。将CICE海冰模式应用到该区域,采用高分辨率(6.37 km)网格,模拟2002年11月至2003年4月的海冰变化过程,模拟的海冰总面积和海冰密集度与AMSR-E/Aqua卫星遥感结果吻合很好,其中两者日平均海冰总面积在模拟期间的相关系数达到0.97。模拟结果表明,东北风将海冰向南输运在东西走向的海岸南部形成冰间湖,反映了潜热冰间湖形成和演化的动力过程。对卫星观测数据,将海冰密集度<75%作为冰间湖的判据;而对数值模拟结果,确定海冰密集度<70%为冰间湖的判据。据此讨论白令海4个区域的冰间湖形成过程,与卫星数据进行比较,大部分冰间湖得到很好的模拟。深入讨论了影响冰间湖模拟准确度的主要因素,认为选用恰当的阈值、提高气象强迫场的空间和时间分辨率有助于提高模拟效果。对部分海域的冰间湖模拟效果不佳,需要发展冰海耦合模式才能最终解决。

In winter, a lot of latent polynyas appear regularly in the shelf of Northern Bering Sea. They have made great contributions to local ecological system and the Arctic halocline. A Los Alamos Sea Ice Model （CICE） with a horizontal resolution of 6.37 km has been implemented to simulate a full year of sea-ice growth and decay starting on November 1,2002 in the Bering Sea. The total sea ice area from the model results and one from AMSRE/Aqua satellite observations has a good consistency. Their correlation coefficient of daily mean total sea ice area equals to 0. 97 during the modeling period. Model results show that polynyas in southern domains of east-west coasts are formed by means of southward movements of sea ice, which are mainly forced by offshore northeast wind. So the CICE allows us to reproduce some key dynamic processes of latent polynyas opening and closing events during January-April 2003. For the satellite observation data, polynyas are defined as regions covered by 〈 75 % sea ice concentration and for model results we use 〈 70% sea ice concentration as polynyas criterion. Accordingly formation processes of polynyas are discussed in four districts of the Bearing Sea. Comparing with satellite data, most polynyas are very well simulated. The article profoundly discusses key factors, which impact simulation accuracy of polynyas. It is concluded that selecting suitable threshold and increasing the spatial and temporal resolution of atmospheric forcing are very favorable to improving the simulation precision. We need to use a general ocean model coupled to the sea ice model to solve the problem of deviations of some polynyas.