During the past decades, the aquaculture industry has developed rapidly, due to drop in wild fish catch. Water quality variables play major role in aquaculture operations, specifically seawater temperature has major impact on the metabolism of the fish species and therefore on the growth rate too. Since the fish farming business relies on the growth rate of the species to plan and operate the farm, seawater temperature becomes crucial information. With the availability of hydrodynamic modeling tools and global ocean information source such as Copernicus Marine Environment Monitoring Service (CMEMS), seawater temperature can be simulated for practically any coast with dynamic downscaling approach. However, the simulated data needs to be assessed for uncertainties for enabling informed decision making using such model predictions. In this paper, a coastal 3D hydrodynamic model aiming at simulating seawater temperature is developed for the southern Aegean Sea, Greece using the Delft3D Flexible Mesh modeling tool. Seawater temperature is impacted by atmospheric forces; therefore, uncertainties are assessed for seawater temperature using ensemble atmospheric forcing functions of the European Centre for Medium-Range Weather Forecasts (ECMWF) ERA5. Spatial analysis of the uncertainty indicates regions of different seawater temperature behavior within the model domain. Seasonal behavior of the vertical temperature gradient suggests that farms need to adapt different operational strategies in different seasons to make best use of the seawater temperature. The application of CMEMS data along with ECMWF ERA5 ensemble atmospheric forcing members proves to be beneficial in analyzing the uncertainties both in spatial and vertical gradient of seawater temperature.

中文翻译：

---

海水温度的 3D 整体模拟——在水产养殖作业中的应用

在过去的几十年中，由于野生鱼类捕捞量下降，水产养殖业发展迅速。水质变量在水产养殖操作中起着重要作用，特别是海水温度对鱼类的新陈代谢有重大影响，因此对生长速度也有重大影响。由于养鱼业务依赖于物种的增长率来规划和经营养殖场，因此海水温度成为至关重要的信息。随着水动力建模工具和全球海洋信息源（如哥白尼海洋环境监测服务 (CMEMS)）的可用性，几乎可以使用动态降尺度方法模拟任何海岸的海水温度。但是，需要评估模拟数据的不确定性，以便使用此类模型预测做出明智的决策。在本文中，使用 Delft3D 柔性网格建模工具为希腊爱琴海南部开发了一个旨在模拟海水温度的沿海 3D 流体动力学模型。海水温度受大气力的影响；因此，使用欧洲中期天气预报中心 (ECMWF) ERA5 的集合大气强迫函数评估海水温度的不确定性。不确定性的空间分析表明模型域内不同海水温度行为的区域。垂直温度梯度的季节性行为表明，养殖场需要在不同季节调整不同的运营策略，以充分利用海水温度。