Weather-related solutions

Weather has always had a tremendous impact on us and everything we do. While there is no way for us to control the weather, we can now be better equipped for whatever Mother Nature throws at us thanks to recent advances in science and technology.

Below are five weather- and environment-related Finnish solutions designed to make life easier and safer for both professionals and individuals.

A METHOD TO PREDICT BLACKOUT-CAUSING STORMS

Stormy weather often brings about power outages, especially in highly forested countries such as Finland. This makes predicting the impact of storms crucially important for energy companies and power grid operators. To tackle the issue, researchers at Aalto University and the Finnish Meteorological Institute (FMI) have teamed up to develop a method that makes it possible to predict blackout-causing storms up to 10 days in advance.

Based on artificial intelligences (AI) and machine learning, the method divides storms into three categories – no damage, low damage and high damage – and helps to predict the location of the storm to within 15 kilometres, and the time of the storm to within three hours. Although accurate, the model has its limitations, too.

“The prediction will, however, never be perfect,” warned Roope Tervo, PhD candidate at Aalto University and software architect at FMI. “It is also good to remember that, when employing weather prediction data, errors are coming from both weather prediction and the outage prediction models.”

FORECASTS OF EXTREME TROPICAL CYCLONES

With climate change making intense storms more common and lack of adequate warnings resulting in extensive damage, Silo AI and Hurricane Unwinder have developed an award-winning computer vision-based AI solution that is able to forecast tropical hurricane intensity at world-class accuracy.

Leveraging high-resolution satellite images and parameters such as sea water temperature and weather forecasts, the solution can predict tropical cyclones and their intensity in the coastal areas of the Atlantic, the Pacific and the Indian Ocean up to 24 hours in advance. In the future, the warning time could be increased to 96 hours.

Since knowing the intensity of tropical cyclones is one of the major factors for preparation and safety procedures, Peter Sarlin, CEO of Silo AI, is convinced that the solution can make a real difference and “has the potential to help governments and cities of some of the world’s most populous areas”.

ESTIMATING THE PROBABILITY OF EXTREME EVENTS

Natural disasters, such as storms, floods and earthquakes, have been around since the beginning of time. Despite constantly learning more about the many types of weather calamities, scientists still grapple with the issue of how to predict them as precisely as possible.

To estimate the probability of extreme events, statistics of previously observed extremes are normally subjected to a so-called extreme value analysis. Two researchers at VTT Technical Research Centre of Finland, Lasse Makkonen and Maria Tikanmäki, have improved the method, making it not only more accurate, but also suitable for a wide range of applications, including economics and medical data analysis.

“Extreme value analysis of natural phenomena aims at appropriate preparation against hazardous extreme events,” Makkonen summarised. “Therefore, it forms the foundation of all regulations that aim at securing the safety of buildings and infrastructure. Stronger constructions and protection arrangements increase costs, so that economic [optimisation] is another issue here.”

A SIMPLE WAY TO CREATE RELIABLE AIR QUALITY DATA

As air pollution continues to be a pressing problem in many parts of the world, researchers at the University of Helsinki have found an innovative way to produce reliable data on air quality in urban regions using inexpensive measuring equipment. Having been tested in prototype form, the technique can potentially be used to improve urban planning and reduce health hazards associated with air pollution.

Complementing fixed air quality monitoring stations, low-cost sensors can be installed in great numbers, for example, in offices or public transport to accumulate masses of data. The sensors are combined with technological solutions that automatically adjust measuring accuracy by means of AI and specific mathematical models.

“With the technique, we are proposing, sufficiently accurate data can be gained from low-cost sensors,” said Martha Arbayani Zaidan, a postdoctoral researcher at the University of Helsinki’s Institute for Atmospheric and Earth System Research. “Our technique could be used to significantly extend the network for measuring air quality, with the data accumulated used for the benefit of humanity.”

AN AI-DRIVEN WEATHER APP FOR FAMILIES

With its handy AI-driven app for families, children’s activewear company Reima is showing time and again that there is no such thing as bad weather if the gear is right. Available in English, Finnish, German and Swedish, the Reima Weather app provides a daily forecast and recommends the best clothes and footwear to match. Furthermore, the app is full of useful tips and Nordic knowhow on active family life.

“The key reason we created the Reima Weather app is to help parents toward hassle-free parenthood,” explained chief experience officer Johanna Sarviharju. “We have a long expertise of the harsh Finnish climate, and […] we’re exactly in the spot for offering young parents an easy way to learn how to dress their kids for adventure.”

Thanks to its functionality and ability to address children and parents’ needs in an appealing way, the app has found global recognition, receiving a German Design Awards 2021 along with several fellow Finns.