Ometeotl

https://github.com/Oscar6647/SpaceApps-2022-Ometeotl

https://drive.google.com/file/d/1QWwP7SG6py8EyWiH0IRyxO52G0_y9S-i/view?usp=sharing

Ejecutive Summary

To solve the visibility and accessibility of Earth Observation data (EO), we have created a gateway available on the web (Chrome, Safari, Opera, etc) and mobile devices (Android, iOS, Tablets, etc). This, with the purpose of having the greatest reach of users of all kinds of backgrounds. Our gateway filters the most relevant information for the user, we want the information available to be understandable without prior experience in data science. 

In Figure 1 we can observe a brief summary of the research steps we took in order to reach our desired project proposal

Figure 1

Functionality of our gateaway

Our team is composed of 6 members: Juan, Cris, Oscar, Jocelyn, Noemi and Jp. Cris and Noemi worked on the project proposal and deliveries, Jocelyn and Jp worked in the front-end of the application, Juan filtered data from NASA and the Mexican Space Agency in order to create a prediction algorithm, and Oscar worked with the integration of the prediction algorithm with the front-end. 

Project proposal and deliveries 

For Phase 1 we divided our research in two main streams and created two PPTs to address each one of them. 

• How is EO data delivered to citizens and the current strategies? Link to ppt: https://drive.google.com/file/d/1Mg0e9BN2QOQmLXzajKteKuU1XCLkAQvu/view?usp=sharing
• How do policy makers take decisions based on EO observations? Link to ppt: https://docs.google.com/presentation/d/1JUC6neSCw55OijlICBVz2izye0sPIPTvUsp-vHQycUk/edit usp=sharing

For Phase 2 we created a Docs that contains all the methodology and design of our gateaway. Link to the docs: https://docs.google.com/document/d/1mM5A0OwcLn7aHOcfrJ6XoHACWE8reCnyykki21ERhkY/edit?usp=sharing

For Phase 3 we used several business tools such as SWOT, PEESTLE, Ansoff Matrix, etc in order to predict the viability of our project. The different marketing and financial tools can be found in the following PPT. Link to PPT: https://docs.google.com/presentation/d/1jmo_Vu3YxfWrcnU4XgF3rhs5w5N5Xsi4mjJ3Qi-BTv8/edit?usp=sharing

For Phase 4 we presented our pitch to the judges. They really liked it! Even if the transmission kept loosing signal. Link to the pitch video: https://drive.google.com/file/d/1bXx5klOVjoslFSfAy-l-mE3VVg7atWxg/view?usp=sharing

Front end of the application

In Figure 2 we can observe briefly the different steps and tools we used in order to develop the front-end for our gateaway. We decided to use Flutter framework because it provides a native code that runs in all devices and all the preset designs are developed to target a minimalist UI.

Figure 2

Creation of the Prediction Algorithm

The prediction algorithm targets: Air Pollution and Rain. On Figure 3 we can observe how we conducted our data base research.

Figure 3

On Figure 4.1 we can observe the methods we use in order to filter all the data obtained.

Figure 4.1

On Figure 4.2 we can observe the prediction model process and its accuracy results.

Figure 4.2

Integration of the prediction algorithm with the front-end

In Figure 5 we can observe the undertaken process to deliver the arrays from the databases into the front-end solution

Figure 5

Final Application Available in Web and Mobile Devices

Figure 6 shows our gateaway in iOS

Figure 6.1 shows our gateaway in Web

Gateaway Goal

We believe that our gateaway will provide accurate data to users who wish to maintain informed in a daily basis. Let us place this as an example: Imagine a farmer who is not familiarized with data science or any other kind of database. This farmer wants to know what are the best days to plant crops. But, hold on. Our farmer also has asthma so he also needs to know the air pollution levels in his location. With our gateaway Ometeotl Ambiental Solutions the farmer can filter data such as Air pollution and Rain Forecasting to only se what its more relevant. After this filter is applied he then introduces the date in MM/DD/YY to know what are the best days to plant crops and go outside. Our prediction algorithm will give him precise and easy to understand data for him to make the best decision using EO in an easy and interactive way! (Always considering his location in order to make decisions).

In Figure 7 we can observe an example of the displayed reccomendation message for the farmer.

Figure 7

We hope to achieve MORE!

We were really fascinated with the simple gateway that we created. And let me tell you this. This is only the beginning… We had 48 hours to develop a solution, but imagine what we can achieve in a month! We are planning to add earthquake notifications, tsunami alerts, tornado predictions, covid 19 cases depending on locations, etc. We have many big ideas for this project, and we are sure that when we reach the end of the Nasa Space Apps challenge we will create a complete EO data display prediction model for all users. 

Note

If you wish to visit our Drive Folder with all of the evidences for each of the Phases you can visit the following link. Link to Drive: https://drive.google.com/drive/folders/1qpLF2R9fVb5T67rI1O0d8EI42lluU-Qr?usp=sharing

For the Rain Forecasting prediction algorithm we used a CONACYT (Consejo Nacional de Ciencia y Tecnología) database that gathers data from the Mexican Space Agency. Link to the database: http://clicom-mex.cicese.mx

For the Air Pollution we used a NASA database. Link to the database: https://search.earthdata.nasa.gov/search/granules?p=C2184005745-GES_DISC!C2184005745-GES_DISC&pg[1][a]=2184563973!GES_DISC&pg[1][v]=t&pg[1][gsk]=-start_date&tl=1664679526!3!!&fs10=Particulates&fsm0=Air%20Quality&fst0=Atmosphere&zoom=0

We verified this data with WHO (World Health Organization) in order to provide profitability in our prediction model on what is considered to be "good air quality". Link to database: https://www.who.int/data/gho/data/themes/air-pollution/who-air-quality-database

We believed that all of our data bases where a crucial source of information. Even though there was a huge amount of complex information delivered in these databases, we managed to filter the information that we specifically required using data science as shown in the Detailed Project Description section. 

In the link below you can find our open source data base with the filtering, classification, model prediction, accuracy rating and metrics used to implement all the data given by the mentioned databases. Link to OUR databases: https://colab.research.google.com/drive/1Z6RN6l7pTDfRHKKinJkLIWdUBBGu3W6b?usp=sharing

Space Apps was a great challenge for everyone, having members from different disciplines, we wanted to share and learn from everyone's progress and share different ideas and solutions. For our team, being able to contribute significantly to the planet was through information and the correct use of it. Despite the short time, the passion we had for generating real solutions capable of generating a great impact was what prompted us to give everything of ourselves. We believe that gratitude falls short not only to the team in general and our families, but also to SPACEAPPS for providing us with a space where ideas will be heard, and that through our advisors and judges we can improve more and more our Project.

Here is a list of resources that we did not cite during some of the phases. Still for the majority of PPTs and Docs citations can be found at the end or as footnotes.

Secretaría de Medio Ambiente y SIMA. (2015). Mapa de promedio móvil en el área metropolitana de Monterrey. Gob.mx. https://sih.conagua.gob.mx/

Comisión Nacional del Agua. (2019). SIH. Gob.mx. https://sih.conagua.gob.mx/

Indicadores Básicos del Desempeño Ambiental - Atmósfera - Calidad del aire - Consumo final de petrolíferos - datos.gob.mx/busca. (2019). Gob.mx. https://datos.gob.mx/busca/dataset/indicadores-basicos-del-desempeno-ambiental--atmosfera--calidad-del-aire/resource/457c8ce1-454e-48b8-8a9a-1a61bb89bae9?inner_span=True

CLICOM. (2012). Cicese.mx. http://clicom-mex.cicese.mx/

How do you measure air pollution? (s/f). Plumelabs.com. Recuperado el 3 de octubre de 2022, de https://air.plumelabs.com/learn/en/how-do-you-measure-air-pollution

Citations for Python libraries:

umPy

Harris, C. R., Millman, K. J., van der Walt, S. J., Gommers, R., Virtanen, P., Cournapeau, D., … Oliphant, T. E. (2020). Array programming with NumPy. Nature, 585, 357–362. https://doi.org/10.1038/s41586-020-2649-2

Pandas

McKinney, W., & others. (2010). Data structures for statistical computing in python. In Proceedings of the 9th Python in Science Conference (Vol. 445, pp. 51–56).

Sklearn

Pedregosa, F., Varoquaux, Ga"el, Gramfort, A., Michel, V., Thirion, B., Grisel, O., … others. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12(Oct), 2825–2830.

Pickle and Random 

Van Rossum, G. (2020). The Python Library Reference, release 3.8.2. Python Software Foundation.

Random Forest Classifier 

Ho, T. K. (1995). Random decision forests. In Proceedings of 3rd international conference on document analysis and recognition (Vol. 1, pp. 278–282).

#app #webapp #flutter #python #prediction #algorithm #nasa #aem #mexico