This week in science
November 20th-27th
Thunderstorm Asthma
On Monday, November 21st, a thunderstorm struck Melbourne, Australia which caused a rare phenomenon called “thunderstorm asthma.” According to Robin Ould, chief executive of the Asthma Foundation of Australia, thunderstorm asthma occurs when a storm forms on a day with a lot of pollen and high humidity. Rye grass grains will absorb the water in the air and burst into many tiny pieces. Normally these grains would have been trapped by nose hairs. However, since these grains are 1,000 times smaller, they are able to go into the trachea, and finally, to the lungs. This causes the lungs to be filled with mucus and to inflame. That makes it harder to breathe, thus causing asthma attacks. Thousands of people were rushed to the hospital, and as of November 27th, 6 people died. Of the people affected, 32% never had an asthma attack before. Since this is very rare, hospitals were very unprepared, and people reported having to wait over 40 minutes for an ambulance.
For more Information: http://www.allergy.org.au/patients/asthma-and-allergy/thunderstorm-asthma
http://www.cnn.com/2016/11/23/health/thunderstorm-asthma-australia/
Ice Deposit on Mars
On Tuesday, November 22nd, astronomers working for NASA discovered and verified a massive ice deposit about the size of Lake Superior, just below Mar’s surface. The deposit estimated to be 70% water/ice, with the rest being dust and rocks. At some points, it is 170 meters deep. Cassie Stuurman, of the Institute for Geophysics at the University of Texas, believes this ice cap formed by snow accumulating on a pre-existing ice sheet when Mar’s axis was greater that it is today. Although this is less than 1% of Mar’s known water/ice, it is very important. Because it is located on a very flat area and it is very close to the surface, it can easily be accessed by astronauts on the surface.
For more information: http://www.jpl.nasa.gov/news/news.php?feature=6680
Google’s AI Translates Languages it doesn’t Know
According to Google, Google Translate went from translating a few languages in 2006 to 103 languages as of November 2016. That is due to “Google’s Multilingual Neural Machine Translation System”, also known as GNMT system. GNMT system allows Google Translate to translate two languages it was not taught to translate between explicitly. For example, right now in order for Google Translate to translate between Hindi and Turkish, it needs to be taught how to translate specifically between Hindi and Turkish. That takes a lot of systems to account for these 103 languages. With GNMT system, if Google Translate was taught how to translate between Hindi and French to Turkish, it can now translate between Hindi and Turkish.
For more information: http://www.breitbart.com/tech/2016/11/26/google-ai-creates-its-own-language-to-translate-languages-it-doesnt-know/