Open Access to the GTS (Open-GTS)

Owned by Xiaoxia Chen
Last updated: Jun 22, 2021
2 min read

Introduction

Currently, oceanographic and marine meteorological data is distributed globally to forecast centres and other operational data users. This is done using the WMO Global Telecommunication System (GTS). The availability of real-time data is crucial for forecasters, emergency managers, and other scientific purposes. The format of choice for distribution on the GTS is BUFR, which is a table-driven binary format that has very rarely been utilized in research oceanography. This format requirement becomes an imposing barrier for distributing and accessing data via the GTS.

The goal of the Open Access to the GTS (Open-GTS) project is to develop and implement improved methodologies for the distribution and access of near-real-time ocean and marine meteorological data through the GTS. As the Open-GTS project workflow embraces several of the WIS 2.0 principles, it seems highly compatible as a WIS 2.0 demonstration project.

The project has been developed and supported by the GOOS Observations Coordination Group (OCG).

Project description

As mentioned, there are two facets to the Open-GTS Project: data distribution and data access. For data distribution, the Open-GTS project leverages an open-source tool called ERDDAP to connect data producers with National Data Centres (NDC). The benefit of this connection is that data producers can work in the data formats they are most used to, with the main requirement being that the data is served by ERDDAP and contains sufficient metadata. The NDC will then harvest the data in whatever format they prefer, encode that data into BUFR using standard templates, and distribute those messages via the GTS.

For accessing data from the GTS, the workflow is reversed. NDC’s will harvest the data from the GTS using their connection to the WIS, decode the BUFR messages, save the data in a consumable format and load the data into ERDDAP. It is also possible to push the data into ERDDAP directly as part of the decoding process, thereby skipping the need to create intermediate files. Once in ERDDAP, data consumers are free to access and use the data in a variety of formats, ensuring that they will be able to use the data of the clients they are most familiar with, without being burdened by the cumbersome process of reformatting BUFR data. It is worth noting that the ERDDAP data platform also supports federation, thereby creating easy access to distributed datasets as well. This capability ensures that data consumers who may not have a direct connection to the WIS would still be able to access and use the near-real-time data.

Project Lead

Kevin O’Brien (US)