The board game go involves black and white stones placed on a square wooden board that’s checkered with 19 vertical lines and 19 horizontal ones to form 361 intersections. Players try to conquer territory by enclosing vacant points with boundaries made of their own stones. The first player to completely enclose an area wins the game. Historically, go playing was a social and intellectual pursuit among the upper classes in China, Korea, Japan and other countries, but it gained wider popularity after World War II. Today, it’s played in over 200 countries and territories worldwide by more than 2 million people.
Go is a fascinating game for many reasons, but it has also been a crucible for artificial intelligence research. The complexity of the rules, the vast search space, and the need to evaluate boards in a three-dimensional space make it one of the most difficult games for computer programs to master. But new methods, including a deep neural network approach that combines value and policy networks, have pushed AI to the forefront of Go playing.
In the latest development, scientists have used a satellite sensor on NOAA’s Geostationary Operational Environmental Satellite (GOES) system to develop a new view of the Amazon rainforest. Unlike polar-orbiting satellites that pass over the region only a few times per day, GOES’s geostationary orbit allows it to keep its sights trained on the Amazon throughout the day and night. This allows researchers to capture the seasonality of carbon cycle changes, something that’s not always visible from the ground because the forest is often shrouded in persistent clouds.
GOES’s other functions include providing real-time imagery and atmospheric measurements of Earth’s Western Hemisphere, as well as tracking severe storms and monitoring solar activity. It’s also been instrumental in detecting fog and low clouds that impact air and sea travel, as well as spotting volcanic eruptions and the ash and sulfur dioxide they spew into the atmosphere. And its ocean sensors monitor marine heat waves, which affect the food chain and can lead to mass coral bleaching events.
NOAA’s newest series of GOES satellites, known as the GOES-R Series, provides critical atmospheric, hydrologic, oceanic, climate and solar science data. This data improves detection and observation of environmental phenomena that directly impact public safety, property protection and our nation’s economic health and prosperity.
The GOES Image Viewer hosted on the STAR webservers hosts examples of ongoing experimental products. The images and data are not official NOAA operational products, and may not be suitable for use by anyone other than remote sensing researchers or experienced meteorologists/oceanographers.
The American Go Association is a small, close-knit national community that welcomes new members as long-lost family. With over 100 chapters nationwide, there’s likely a club near you. If not, learn how to start your own.