Answer to Bottleneck

-

Previously

In the article, “AI for Earth observation and numerical weather prediction”, by Boukabara, the community identified the need to boost the growth of data refinement technology. Furthermore, based on the trend and performances observed in other disciplines with similar challenges, Boukabara suggested a strong candidate for the solution: AI. The research by Professor Bue and his colleagues support Boukabara’s argument.

Spectroscopy

The spectroscopy is essentially the study of light crossed with chemistry. When a chemical is treated with certain stress, such as heat, it emits light. This is like a finger print to a person: each chemical emits light with a specific wavelength. The spectroscopy is a branch of science that studies the light wave to the chemical that is associated with it. It is widely used to identify the chemical constituent of an unknown compound or a distant star system, among other things. It is also used to understand the chemical composition of a given segment of air on Earth, but this has proven to be computationally too expensive to perform extensively (Bue et al, 2019).

The Research

Professor Bue and his collogues devised a way to replace the computationally expensive method, called Radiative Transfer Models (RTM) with comparably inexpensive method of emulating it of machine learning. The new method is powered by machine learning, and it has shown comparably accurate results to the traditional counterpart. This means that the use of spectroscopy on satellite images has become far more scalable, increasing the scientists’ capability to distill useful information from the image data. This is certainly a strong step towards answering Boukabara’s urge to the society.

References

  • Bue, B. D., Thompson, D. R., Deshpande, S., Eastwood, M., Green, R. O., Mullen, T., . . . Parente, M. (2019). Neural Network Radiative Transfer for Imaging Spectroscopy. Atmospheric Measurement Techniques Discussions, 1-16. doi:10.5194/amt-2018-436
  • Boukabara, S. (2019, April 17). AI for Earth observation and numerical weather prediction. Retrieved May 26, 2019, from https://spacenews.com/ai-for-earth-observation-and-numerical-weather-prediction/

Comments

Post a Comment

Popular posts from this blog

AI for Earth

-
AI for Earth In the Microsoft blog called “Researchers turn to AI in a bid to improve weather forecasts” by Roach, the author highlights some of the company's contribution to the research that benefits mankind. Through the AI for Earth initiative, the company funds various challenging research projects that studies Earth. The blog also highlights the social activities, such as hackathons, that illustrates some of the current trends of using machine learning to solve difficult problems. References Roach, J. (2019, May 20). Researchers turn to AI in a bid to improve weather forecasts. Retrieved May 26, 2019, from https://blogs.microsoft.com/ai/ai-subseasonal-weather-forecast/
Read more

Faceapp for Houses

-
Image
Personalization The Faceapp is a controversial phone app that manipulates a person’s picture to show how the person would look many years from now (Koetsier, 2019). Controversies aside, it is an interesting app that has sparked an explosive interest from the general population by showing them a personal and likely future of its user. In many ways, Schmidt and his colleagues had a very similar idea in their effort to gain public interest on climate change: show a personal and likely future of its user (Schmidt, et al., 2019). Instead of face, the team used house, and instead of just few years of aging, the team showed the future after 50 years with likely climate change and the related natural disaster in mind (Schmidt, et al., 2019). Figure 1: "Before" and "After" Pictures. From Visualizing the consequences of climate change using cycle-consistent adversarial networks by V. Schmidt, et al., 2019 Conclusion This is a refreshing effort by scientists to nud...
Read more

Early Look at Machine Learning

-
About In this 2018 paper, “Predicting weather forecast uncertainty with machine learning” by professor Scher and professor Messori, shows an early effort to test the feasibility of the technology as a potential replacement to the popular ensemble weather model approach. The ensemble weather model approach, as the name suggest, is similar to having a panel of experts instead of just one expert; the approach produces information that is a composite of different information generated by multiple models given the same input data. Each model caters to different specialization and has their own strength and weaknesses, and by combining information from each model, the approach attempts to generate more accurate information. As such, while it increases the accuracy and precision, it became computationally very expensive to perform in a timely manner. The researchers’ goal was to provide a competitive solution to the computationally expensive ensemble weather model approach by using machine ...
Read more