Launch of Ocean Weather Forecasts 1km resolution

1. Introduction

We have launched a new website “Ocean Weather Forecasts”, which provides current ocean conditions around Japan and forecasts for the next few days at 1km resolution and hourly intervals.

To combine satellite ocean observations and ocean models, JAXA and the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) Application Laboratory (APL) have been releasing Ocean Weather Forecasts with a spatial resolution of approximately 3 km around Japan (JCOPE-T DA) since 2018. The forecast data was created by combining JAXA’s satellite sea surface temperature data with JAMSTEC’s ocean circulation model, JCOPE-T.

A new Ocean Weather Forecasts website (Figures 1 and 2) with improved spatial resolution of approximately 1 km (JCOPE-T 1ks) has been available since March 27, 2025. JAMSTEC developed JCOPE-T 1ks (see Section 2), and JAXA validated and evaluated the data using satellite and in-situ observations (see Section 3) and built the visualization website (see Section 1).

The new website also includes the following features:

• When zoomed in, ocean current vectors can be displayed in detail without thinning out.
• Minimum and maximum values of the color bar for the current vector can be changed
• Time series graph of any point can be displayed

The 3km resolution Ocean Weather Forecasts website will also continue to operate in parallel.

Figure 1. Water temperature (top) and ocean current vectors (bottom) at a depth of 0 m in Tokyo Bay and off the coast of Choshi using the current JCOPE-T DA (3 km resolution). Ocean current vectors are thinned out.	Figure 2. Water temperature (top) and ocean current vectors (bottom) at 0m depth in Tokyo Bay and off the coast of Choshi using the new JCOPE-T 1ks (1km resolution)

2. JCOPE-T 1ks System

JAMSTEC APL is conducting research and development of Ocean Weather Forecasts [1] as part of the Japan Coastal Ocean Predictability Experiment (JCOPE). JCOPE-T 1ks (1km scale: approximately 1km resolution) [2] is an Ocean Weather Forecasts system that is a higher-resolution version of the currently available JCOPE-T DA (Data Assimilation: approximately 3km resolution) [3]. The underlying ocean circulation models for both are JCOPE-T [4]. A comparison of JCOPE-T DA and JCOPE-T 1ks is shown in Table 1. For more information about the JCOPE-T 1ks system, please see here.

3. Spatial distribution validation using GCOM-C/SGLI ocean products

For the heavy rainfall in the Kyushu-Kanto region caused by Typhoon No. 10 in 2024, we used the light absorption coefficient of dissolved organic matter (a_CDOM)[5], an SGLI ocean standard product, as a tracer in freshwater to compare the spatial distribution of sea surface salinity (SSS) from JCOPE-T 1ks (Figure 3).

JCOPE-T 1ks correctly reproduced the trend of decreasing salinity caused by river flooding and was able to reproduce detailed horizontal gradients, especially in semi-enclosed bays such as Tokyo Bay and Ise Bay, which were difficult to reproduce with conventional model resolution. Such high-resolution analysis results from JCOPE-T 1ks are expected to be utilized as a valuable tool for delivering ocean condition information to coastal environmental conservation and to the fisheries sector. On the other hand, problems such as underestimation of sea surface temperatures due to errors in the river boundary conditions have been identified (See here for more details). Currently, it is recommended that the data be used in conjunction with actual observation data from satellites, etc., but JAXA and JAMSTEC APL plan to continue collaborating on research to improve it.

Reference：
[1] H. Hayashida, Y. Miyazawa, T. Miyama, Y. Baba and S. Kido, 2024: Ocean weather forecasts, Oceanography in Japan, 33, 89-101.
https://kaiyo-gakkai.jp/jos/uminokenkyu/vol33/33-5_6/33-56-3_hayashida.pdf
[2] Wang, S., S. Kido, X. Guo, Y. Miyazawa, and S, M. Varlamov, 2025: Topographically generated submesoscale processes in the Tokara Strait revealed by a high-resolution model, in revision.
[3] Miyazawa, Y., S. M. Varlamov, T. Miyama, Y. Kurihara, H. Murakami, and M. Kachi, 2021: A nowcast/forecast system for Japan’s coasts using daily assimilation of remote sensing and in situ data Remote Sensing, 13, 2431.
https://www.mdpi.com/2072-4292/13/13/2431
[4] Varlamov, S. M. , and Miyazawa (2021) High-performance computing of ocean models for Japan Coastal Ocean Predictability Experiment: a parallelized sigma-coordinate ocean circulation model JCOPET. Annual Report of the Earth Simulator, April 2020-Feb 2021, 1–3.
https://www.jamstec.go.jp/es/jp/output/publication/annual/annual2020/pdf/2project/chapter2/II-3_miyazawa.pdf
[5] Takafumi, Hirata and Yohei, Yamashita “Derivation of the absorption coefficient of Colourd Dissolved Organic Matter (CDOM)” SGLI Algorithm Technical Background Document (2020).