NASA-SPACE-APPS-2020
NEED (Natural Events Exposure & Detection.)
Countless phenomena such as floods, fires, and algae blooms routinely impact ecosystems, economies, and human safety. Your challenge is to use satellite data to create a machine learning model that detects a specific phenomenon and build an interface that not only displays the detected phenomenon, but also layers it alongside ancillary data to help researchers and decision-makers better understand its impacts and scope.
Project Demo
Summary
Through a website, the user can access NASA and ESA platforms, which contain satellite data. There they will select the place in the world to be analyzed and the corresponding satellite. We work on Tsunamis in this case, but the webpage will have a menu for the user to select among various natural disasters. It also has a button where they can access news about the phenomenon, an option to communicate with organizations devoted to mitigating the impact, a forum where researchers can exchange data, information resources and a simulator where you can see the evolution of a phenomenon under certain conditions. How We Addressed This Challenge
We develop an interface method that, with the use of available satellite information, through mapping and data recognition, can analyze it and provide an automated diagnosis about of natural phenomenon, in order to provide relevant information to researchers and organizations.
With the data obtained by different tools and satellites it is possible for us to develop a machine learning model that allows us to evaluate and predict phenomena. This is a very useful tool for researchers and political actors to have the necessary information and in advance of the event in order to take preventive action and to have the opportunity to save as many lives as possible. We will base ourselves on the detection and use of data to warn about the possible origin of tsunamis and other phenomena.
This model includes the use of a double-entry cross-matrix which will be responsible for assessing the input of data on the one hand and on the other the impact analysis generated by previous events and of similar magnitude in order to give a frame of reference to our artificial intelligence and thus it can be possible to estimate the magnitude of the event and possible consequences if this were the natural course of phenomenon and thereby enable the relevant actor to take preventive action and alert the population to considerable time in advance.
The idea is to integrate satellite data mainly from NASA into our application by means of an automatic learning method which will process the data and yield statistical results and modelling using dynamic graphics and simulator.
How We Developed This Project
Our planet is constantly exposed to severe climatic and non-climatic events that threaten the health, security, economy and productivity of nations and their inhabitants. For this reason, we decided to tackle this problem, developing a method that gives automatic information for the detection of threats and imparts solutions to natural disasters, providing the people in charge with a solution and immediate resources to take action and save lives.
We can access the Copernicus Open Access Hub software through a web page that we designed. There will be indicated the place to be analyzed and the satellite that contains the data needed. Files with .NC extension are downloaded and we use Matlab to open them, they contain the statistical information of the variables on the natural phenomenon to be analyzed. In turn, to analyze the images in a timelapse, we use EO Browser, which is a powerful tool for viewing satellite images, and the place is analyzed by modifying layers and bands.
For the development of the system, tools such as laptops, web systems such as marvelapp were used, Canva (canva.com) and Google Drive.
The distance was a problem factor but it did not prevent us from working and developing, thanks to the technology and virtual system.
How We Used Space Agency Data in This Project
We’ve used the Copernicus Open Access Hub software. There, we draw a polygon on a place in the world to analyze the natural phenomena and the software shows those satellites that contain the data we need and the entire surface to study. Files with extension .NC are downloaded and through Matlab we access the statistical information and images of the variables on the natural phenomenon to be analyzed. In turn, to analyze the images in a timelapse, we use EO Browser, which is a powerful tool to view satellite images. In turn, modifying layers and bands, we can understand the parameters and variables that are in that site
Bibliography and resources consulted
MISR. You can see the sunlight reflecting on the waves
MISR Website (Jet Propulsion Laboratory)
MODIS website (Goddard Space Flight Center)