Biogeochemical ocean models are useful tools but subject to uncertainties arising from simplifications, inaccurate parameterization of processes, and poorly known model parameters. Parameter optimization is a standard method for addressing the latter but typically cannot constrain all biogeochemical parameters because of insufficient observations. Here we assess the trade-offs between satellite observations of ocean color and biogeochemical (BGC) Argo profiles and the benefits of combining both observation types for optimizing biogeochemical parameters in a model of the Gulf of Mexico. A suite of optimization experiments is carried out using different combinations of satellite chlorophyll and profile measurements of chlorophyll, phytoplankton biomass, and particulate organic carbon (POC) from autonomous floats. As parameter optimization in 3D models is computationally expensive, we optimize the parameters in a 1D model version and then perform 3D simulations using these parameters. We show first that the use of optimal 1D parameters, with a few modifications, improves the skill of the 3D model. Parameters that are only optimized with respect to surface chlorophyll cannot reproduce subsurface distributions of biological fields. Adding profiles of chlorophyll in the parameter optimization yields significant improvements for surface and subsurface chlorophyll but does not accurately capture subsurface phytoplankton and POC distributions because the parameter for the maximum ratio of chlorophyll to phytoplankton carbon is not well constrained in that case. Using all available observations leads to significant improvements of both observed (chlorophyll, phytoplankton, and POC) and unobserved (e.g., primary production) variables. Our results highlight the significant benefits of BGC-Argo measurements for biogeochemical parameter optimization and model calibration.

中文翻译：

---

评估生物地球化学Argo配置文件与海洋颜色观测值对墨西哥湾生物地球化学模型优化的价值

生物地球化学海洋模型是有用的工具，但存在因简化，过程参数设置不正确以及模型参数未知而引起的不确定性。参数优化是解决后者的一种标准方法，但由于观测不足，通常无法约束所有生物地球化学参数。在这里，我们评估了在海洋颜色和生物地球化学（BGC）Argo概况的卫星观测之间的权衡，以及在墨西哥湾模型中结合两种观测类型以优化生物地球化学参数的好处。使用卫星叶绿素的不同组合以及来自自动浮标的叶绿素，浮游植物生物量和颗粒有机碳（POC）的剖面测量，进行了一系列优化实验。由于3D模型中的参数优化在计算上很昂贵，因此我们在1D模型版本中优化参数，然后使用这些参数执行3D仿真。我们首先表明，使用最佳一维参数并进行一些修改可以提高3D模型的技能。仅针对表面叶绿素进行了优化的参数无法重现生物场的地下分布。在参数优化中添加叶绿素的分布可显着改善表面和亚表面的叶绿素，但不能准确捕获地下浮游植物和POC的分布，因为在这种情况下，叶绿素与浮游植物碳最大比例的参数没有得到很好的限制。使用所有可用的观察结果，可以显着改善所观察到的（叶绿素，浮游植物和POC）和未观察到的变量（例如初级生产力）。我们的结果突出了BGC-Argo测量对于生物地球化学参数优化和模型校准的显着优势。