An enormous amount of data is collected, analyzed, and visualized on maps for vessel operators worldwide. How do developers work with such real-time weather data?

As our development teams contribute to digital tools that provide such services, we decided to share our knowledge.

Importance of weather data for better routing and voyage planning

Data is essential for proper vessel route optimization, and according to IMO, weather routing saves at least 3% in fuel consumption. Hence, industries’ focus on this data is huge while achieving decarbonization goals. How do we understand data optimization in this case? How weather data is used to help vessel operators. All of this will be revealed in this article.

To quote Søren Christian Meyer, CEO, ZeroNorth: “Anyone involved with ship operation knows how vital weather routing is.” source: ZeroNorth. For those unfamiliar with this term: weather routing is a commercial service offered to vessels to maximize their journey efficiency.

As Søren continues his speech, he admits data has matured to the point where advanced, integrated technologies can genuinely provide a clear picture of the reality that a vessel faces on the oceans – and we can optimize vessels in real-time.

Data is matured due to many tools and technologies used to provide these quality forecasts that support vessel operators in better weather routing and voyage planning. To achieve this goal, there are dedicated digital tools like the one provided by the mentioned earlier ZeroNorth.

Many worldwide development teams contribute here, including Setapp. One of our team’s expertise is in weather data optimization. Before we dig into this part in more detail, let me explain the whole process, from collecting to optimizing and visualizing this weather data on maps.

The purpose of this process is to support vessel operators in better voyage planning, resulting in less fuel consumption, lower carbon footprint emissions, and higher earnings.

Collecting, acquiring, gathering, and optimizing weather data

1. To visualize the process of collecting weather data, it is worth learning that the World Meteorological Organization collect them from 15 satellites, 100 stationary buoys, 600 drifting buoys, 3,000 aircraft, 7,300 ships, and some 10,000 land-based stations.
2. As OER services explain: a 5-day weather forecast today is as reliable as a 2-day forecast was 20 years ago! The quality is higher than 20 years ago. – courses.lumenlearning.com
   First, you must acquire data from providers like WNI. There are several options, and the quality of these services is relatively even.
3. When you have your weather data provider, the next step is to gather periodic forecast data. Each forecast covers measurements for 16 days, and this is updated every 6 hours. The amount of this data can be measured in thousands.
   One task is to gather it all in a document database (db) using integration with an external weather forecast provider.
4. Further, teams like us use API to process this data in many ways depending on the purpose we have in mind: either data visualization on maps or weather routing, etc.
5. Our role, among others, is to optimize weather data, so they do not weigh thousands of bytes, which would slow down the process of visualization on maps.
   Our development teams seek the balance between the data accuracy (meaning the number of digits after a comma) and its size.
6. Having said that, we provide the best system performance for its users as they need access to this data in real-time to work and make proper decisions.
7. In other words, weather data optimization aims to achieve top vessel route planning for ship operators.
   Above all, visualization on maps allows vessel operators to make decisions based on proper data. This is the particular business goal of our development.

What do we mean by weather data optimization?

As explained earlier, it refers to rounding out values of, for instance, wind strength which is presented as a number with many digits after a comma. In order to provide proper visualization on maps, we do not need that much accuracy, so we simply cut it out.

567,77649527325325425.

Consequently, as you can see, data Optimization means suffering the accuracy of a particular data to achieve lighter files. The whole process has a minor influence on the quality of the way data is visualized on maps or for other purposes planned for this data.

Weather Data Optimization – what if no one optimized it?

Processing weather data is one of the most problematic and challenging, as it is constantly changing over time. Each new forecast is available every 6 hours. Secondly, the amount of this data would be massive stress for a digital tool if no one optimized it.

Weather data are collected in enormous amounts in thousands of bytes. Visualizing the weather data on the map and adjusting views and layers would not be efficient without proper optimization. That is why our team takes care of this. – Paweł Paszkiewicz, Managing Director of Setapp.

Is weather data saved and optimized as an image?

That is correct. It is one of the ways we can do it. We used it while working on ZeroNorth’s project. Our teams are saving all this big weather data as an image, simple and well-known png files. Thanks to this, we get two additional values – x and y position of a point on an image. This point translates to the longitude and latitude of a sphere. Further, developers can display this data on our map and animate it. In real world scenario, it should be an infinite amount of images with minor differences between them – exactly like a flipbook.

In general, the process can be summarized in these 3 steps:

1. Query data by forecast date and coordinates.
2. Encode and save data to png files.
3. Query png files from the browser to allow presentation on deckgl layers*

*Side note: Deckgl is a framework for visual exploratory data analysis of large datasets.

 
If you are interested, below, we share what tools we use while helping ZeroNorth and their clients achieve their essential # decarbonization goals by optimizing weather data and visualizing them correctly on maps.
 
Deck.gl - rendering weather layers
ASP.NET - data encoding in a 6 hours interval
AWS S3 - data storage
MapTiler - masking data over lands

In case you are a developer, We know that projects and tasks like this are a better pick-up line for you than building another e-commerce store. If you wish to join our team, check our Career page.

What can vessel operators do with weather-optimized data on maps?

• Operators can verify the weather forecasts while monitoring vessels’ position in time on the simulated route.
• Users can compare weather conditions on different, alternative routes.
• Users can change the date of the displayed weather.
• The wind is displayed according to the standard way used in industry – wind barbs.
• Data from all layers need to be visible on a single tooltip.
• Data is not overlapping with the land.
• The solution needed to be usable on maps in various applications.

This is a list of functionalities available in the ZeroNorth tool that Setapp development teams contributed to and still are a part of worldwide teams enhancing this product.

Being a part of this and the ability to constantly learn and improve for the betterment of our planet and the maritime industry is just cool from our developer’s perspective.

Read more about: our ZeroNorth case study or Decarbonizing maritime transport.

Finally, if you have a question about weather data optimisation in your development project, just use the contact form.