Scholarlink

<p>Since June 2016, the National Centre for Medium-Range Weather Forecasting (NCMRWF) has utilized NEMOVAR, which is a variational data assimilation system of the Nucleus for European Modelling of the Ocean (NEMO), to produce ocean analyses for initializing its coupled atmosphere-ocean model. In this study, we evaluate the accuracy of these analyses using quality-controlled ship-based ocean observations, focusing on both diurnal and monthly scales over the global and regional ocean from 2017 to May 2022. An average of over 72,000 ship-based observations per month was used for validation. The Root Mean Square Errors (RMSE) of 0.65&#xa0;°C to 1.15&#xa0;°C were found across different oceanic regions. Further, the model accurately describes the mean and variability of Sea Surface Temperature (SST) on a monthly scale. Both the model and observations consistently show lower SST variability in the Indian Ocean and tropical Atlantic and Pacific regions compared to higher latitudes. At diurnal scales, the model tends to overestimate SST variability globally. Regionally, the diurnal SST cycle is overestimated in the Indian Ocean and underestimated in the North Atlantic during the January to March (JFM), April to June (AMJ), and October to December (OND). The diurnal SST RMSE is also relatively higher (lower) in the North (South) Pacific Ocean. Overall, the ocean analysis effectively represents the diurnal SST cycle, achieving RMSE values between 0.98&#xa0;°C and 1.06&#xa0;°C and exhibiting a strong spatiotemporal correlation exceeding 0.99 with ship-based observations globally.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Assessment of the Diurnal Cycle of Sea Surface Temperature from Model Analysis Against Ship-Based Ocean Observations

  • 期刊论文

  • 发表日期：

    2026年1月2日

查看全文
  • Imranali M. Momin,
  • S. K. Sahoo,
  • Radharani Sen,
  • Anitha Gera

摘要

Since June 2016, the National Centre for Medium-Range Weather Forecasting (NCMRWF) has utilized NEMOVAR, which is a variational data assimilation system of the Nucleus for European Modelling of the Ocean (NEMO), to produce ocean analyses for initializing its coupled atmosphere-ocean model. In this study, we evaluate the accuracy of these analyses using quality-controlled ship-based ocean observations, focusing on both diurnal and monthly scales over the global and regional ocean from 2017 to May 2022. An average of over 72,000 ship-based observations per month was used for validation. The Root Mean Square Errors (RMSE) of 0.65 °C to 1.15 °C were found across different oceanic regions. Further, the model accurately describes the mean and variability of Sea Surface Temperature (SST) on a monthly scale. Both the model and observations consistently show lower SST variability in the Indian Ocean and tropical Atlantic and Pacific regions compared to higher latitudes. At diurnal scales, the model tends to overestimate SST variability globally. Regionally, the diurnal SST cycle is overestimated in the Indian Ocean and underestimated in the North Atlantic during the January to March (JFM), April to June (AMJ), and October to December (OND). The diurnal SST RMSE is also relatively higher (lower) in the North (South) Pacific Ocean. Overall, the ocean analysis effectively represents the diurnal SST cycle, achieving RMSE values between 0.98 °C and 1.06 °C and exhibiting a strong spatiotemporal correlation exceeding 0.99 with ship-based observations globally.