首页> 外文会议>Offshore Technology Conference;ExxonMobil;FMCTechnologies;Schlumberger >Loop Current Operational Forecasting in 2015: Skill Assessment and Lessons Learned
【24h】

Loop Current Operational Forecasting in 2015: Skill Assessment and Lessons Learned

机译:2015年当前运营预测循环:技能评估和经验教训

获取原文
获取原文并翻译 | 示例

摘要

2015 was a year of unprecedented Loop Current (LC) activity. From February to June 2015, the LC andrnits associated eddies were monitored at a significantly higher rate in southern Walker Ridge in support ofrncurrent-sensitive offshore operations. A privately operated ocean forecast system for the Northern Gulf ofrnMexico, a.k.a. the Eddy Forecast System (EFS), was actively employed as an integral part of the currentsrnmonitoring program. A total of 38 forecasts were performed with the EFS over this four-month period,rnmore than half of these were dedicated to support of operations in the southern Walker Ridge. In thisrnstudy, we evaluate the site specific EFS skill by comparing current speed forecasted at the site withrnobserved currents. Despite very complex LC-eddy configurations that presented a major challenge forrnforecasting and most demanding model skill definition, the EFS demonstrated skill up to 6 days onrnaverage for the entire four months period. The average site dependent skill was determined to vary fromrnabout 5 days to over 9 days, depending on the location relative to the LC and major eddies. Simplernensembling techniques, i.e. combining multiple forecasts, were shown to improve the EFS skill by 2-3rndays. Individual forecast failures over the same time period were also analyzed. An additional operationalrndata collection guided by "crucial uncertainties" identified by the EFS is proposed as an effective methodrnof reducing the number of failed forecasts and improving the overall model skill.
机译:2015年是史无前例的回路电流(LC)活动的一年。从2015年2月到2015年6月,在Walker Ridge南部对LC及其相关涡旋进行了显着监测,以支持对电流敏感的海上作业。墨西哥北部海湾的一个私人经营的海洋预报系统,又称涡流预报系统(EFS),被积极地用作当前监测计划的组成部分。在这四个月的时间内,EFS总共进行了38次预报,其中一半以上专门用于支持Walker Ridge南部的行动。在本研究中,我们通过比较在现场观察到的电流速度和观测到的电流来评估现场特定的EFS技能。尽管非常复杂的LC涡流配置对预测和最苛刻的模型技能定义提出了重大挑战,但EFS在整个四个月的时间里平均每6天显示出技能。根据相对于LC和主要涡流的位置,确定的平均位置依赖性技能从5天到9天以上不等。简单组合技术(即组合多个预测)已显示可以将EFS技能提高2-3天。还分析了同期的单个预测故障。提出了一种以EFS识别的“关键不确定性”为指导的额外运营数据收集方法,作为减少失败预测数量并提高整体模型技能的有效方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号