边界条件对北太平洋自然^14C分布的影响

Influences of boundary conditions on natural radiocarbon ^(14)C distributions in North Pacific Ocean

利用一个北太平洋海盆尺度环流模式模拟了北太平洋自然14C的分布和吸收。为了考察开边界与闭边界模式的不同以及不同的开边界条件对模拟结果的影响,相应地做了两个开边界的试验和一个闭边界试验。结果表明,由于自然14C的模拟需要长时间的积分才能达到稳态,因而边界的处理对模拟结果有很大的影响;在闭边界试验中,总体上模拟结果偏大,采用开边界条件打开南边界改变了闭边界中模拟结果偏大的状况。但是,在开边界试验中,边界处观测资料的影响也是显著的。当在边界考虑了出流和入流的不同时,与入流时采用零梯度边界条件的试验相比,入流时边界值给定为观测数据的试验得到了较好的模拟结果,与观测结果更为接近。

A basin-wide ocean general circulation model is used to study the uptake and distribution of natural ^14C in the North Pacific Ocean. Three simulations are conducted to study the model performance with different southern boundary conditions, including a run with the rigid boundary condition and two runs with open boundary conditions （OBCs）, which are Stevens OBC with zero gradients and Stevens OBC with prescribed values along the boundary, respectively. The simulated results show that different treatments of southern boundary conditions have great impacts on the distributions of natural ^14C after a long integration of the model. As a result, the rigid boundary condition does not work well, which makes the simulated value much higher than the observations. The use of an open boundary can improve the simulated natural ^14C results that are overestimated in the rigid boundary experiment. However, it is found that the boundary forcing strongly constrains the results in the whole domain. When the velocity direction （out of or into the model domain） on the boundary is taken into account, compared with the model with zero gradient condition on the boundary when the velocity is into the domain, the model with prescribed values produces results that are in better agreement with observations.