Scientia

https://drive.google.com/file/d/1u2Osjd5Fv0nhuGEY0_o70S8yXKPcwbJr/view

https://docs.google.com/document/d/1TCLwTKWSqIeh1Uvjp_1bbfXgoGmUj18D0eFC8bx6Xtk/edit?usp=sharing

To illustrate how our project works, we decided to choose a real case and demonstrate how it could have been avoided through the procedures we devised.

In 2011, Rio de Janeiro suffered the biggest natural disaster in the country, between January 11th and 12th, intense rains affected the mountainous region of the state, caused by the entry of air masses from the Convergence Zone of the South Atlantic in the Region Mountain range.

The strong and intense rains that hit the local reached precipitation of approximately 900 mm in the municipality of Nova Friburgo (with monthly averages around 300 mm), which caused a large number of landslides. Mountainous compartments and mountainous alignments made it particularly vulnerable to the entry of air masses and associated storms. In addition, the mountainous region contains a high density of accidents involving landslides, which would already place this territory as a risk area that would need more attention.

Our solution will be active 24 hours every day, monitoring climate change and crossing geomorphological information from across the country. We will also have data sent by users that will be used to improve the AI. In addition, our project will return feedback to the population about the photos sent, indicating if this is a sign of risk and which agencies to contact.

Crossing normal rainfall data with those predicted by meteorology, the developed model would identify anomalies in real time, thus analyzing the possibility of the occurrence of the disaster. With such a risk identified, he would issue a warning signal to the government, which would send messages to the resident population through the city's signal towers, assisting in the evacuation of the affected area.

These data and alerts allowed the government to safely plan effective strategies to evacuate the population, as well as what standards should be applied to civil construction, which would preserve thousands of lives, the local economy and also avoid the need for high costs to rebuild the affected areas.

We use Adobe Xd to illustrate how the application interface would be made available to the population.

Much information is needed for an AI algorithm to be able to establish relations between regional data, understand patterns and improve the algorithm for predicting geographic catastrophes such as landslides. To accomplish it, several data sources, with historical and current records are crucial for the development of our prototype and the improvement of machine learning. 

All data listed are available in NASA DataBase.

NASA's Land Atmosphere Near real-time Capability for EOS (LANCE) provides a huge variety of natural and man-made phenomena using near real-time data. All the data of the session of Hazards and Disasters are valuable for structure our analysis, with emphasis on the Floods, Drought, Severe Storms and Vegetations data.

Giovanni allow us to filter data by some geophysicists parameters selected and also by a time period, all information could be selected based on the need of the user and it includes information about the whole globe. It has vegetation and precipitation information, which would be useful for the app, especially when we think about to replicate the idea. 

The database of Integrated Multi-satellite Retrievals for GPM (IMERG) – Early Run could predict precipitation based on information provided by various satellites in the globe almost in real-time.

The base has information from June 2000 until the present, which supports better predictions, and the resolution in the view is equal to 10 km for each 0.1 degree. 

Important DataSources: 

PPS Near Real-Time – (Precipitation Rate – mm/hr)/ PrecipitationCal; 

PPS Near Real-Time (Precipitation Accumulation – mm)

The Landslide Hazard Assessment for Situational Awareness (LHASA) is precipitation measurement mission on (PMM) which provides valuable registers of past events related to landslides, grouped by the type, the impact, the number of victims and could indicate places with more probability of new incidents. The Landslide Viewer was accessed. 

Data from the Land Information System of NASA also could be used. For example the view of Soil Moisture Percentile (between 0 and 200 centimeters) which provides an analysis of the soil moisture in different regions.

Space apps was the first hackathon of all our team members, which certainly made the experience even more special. Participating in an event with so many companies involved in developing and spreading science was really amazing.

During the development of the project, we learned a lot about several subjects that we had superficial knowledge, or even no knowledge at all. We discovered several databases, in addition to tools that were new to us until then, we gained experience in setting up and developing a project and how to optimize our work by dividing tasks according to each one's strengths.

We would like to thank everyone involved in making this event possible, for encouraging scientific outreach and supporting new ideas coming from young scientists around the world.

Data bases and Space Agency Data:

https://earthdata.nasa.gov/earth-observation-data/near-real-time/hazards-and-disasters

https://giovanni.gsfc.nasa.gov/giovanni/#service=TmAvMp&starttime=&endtime=&dataKeyword=IMERG%20Final

PPS Near Real-Time – (Precipitation Rate – mm/hr)/ PrecipitationCal; PPS Near Real-Time (Precipitation Accumulation – mm); both in https://gpm.nasa.gov/data/directory

https://maps.nccs.nasa.gov/arcgis/apps/webappviewer/index.html?id=824ea5864ec8423fb985b33ee6bc05b7

https://maps.disasters.nasa.gov/arcgis/home/webmap/viewer.html?useExisting=1&layers=0d2840617fb04a2993962af6491999d3

https://www.intel.com.br/content/www/br/pt/analytics/machine-learning/machine-learning-data-and-predictive-analytics.html

To learn about landslides, their causes and consequences:

http://g1.globo.com/rio-de-janeiro/chuvas-no-rj/noticia/2011/01/chuva-na-regiao-serrana-e-maior-tragedia-climatica-da-historia-do-pais.html

https://agenciadenoticias.ibge.gov.br/agencia-sala-de-imprensa/2013-agencia-de-noticias/releases/26118-estudo-inedito-do-ibge-mostra-sul-e-sudeste-como-regioes-que-concentram-as-maiores-areas-de-suscetibilidade-a-deslizamentos#:~:text=Rio%20de%20Janeiro%2C%20Esp%C3%ADrito%20Santo,Catarina%20(33%2C7%25).

https://exame.com/ciencia/temporal-mudou-geografia-da-regiao-serrana-do-rio/

https://www.ibge.gov.br/busca.html?searchword=deslizamentos

https://biblioteca.ibge.gov.br/index.php/biblioteca-catalogo?view=detalhes&id=2101684

https://ceos.org/ourwork/workinggroups/disasters/landslide-pilot/

https://www.weather.gov/wrn/wea

https://gpm.nasa.gov/applications/landslides

https://gpm.nasa.gov/applications/disasters/modeling-landslide-threats-near-realtime

https://tethys.byu.edu/apps/geoglows-hydroviewer/

https://maps.disasters.nasa.gov/arcgis/apps/MapSeries/index.html?appid=ab7723584fe847449faaa2e62d3bef74

http://www.fiocruz.br/omsambiental/cgi/cgilua.exe/sys/start.htm?from%5Finfo%5Findex=406&infoid=256&sid=13

https://g1.globo.com/rj/regiao-serrana/noticia/2021/01/11/confira-imagens-marcantes-da-tragedia-de-2011-na-regiao-serrana-do-rj.ghtml

https://portal.inmet.gov.br/

https://www.scielo.br/j/rbmet/a/LjvGBPxGBBkhtH6WGHvFjCc/?lang=pt

http://ppegeo.igc.usp.br/index.php/anigeo/article/view/5950

https://tethys.byu.edu/apps/geoglows-hydroviewer/

https://repositorio.unb.br/bitstream/10482/31119/1/2017_Ver%C3%B4nicaMoreiraRamos.pdf

http://www.ufrgs.br/grid/noticias/ibge-desastres-naturais-atingiram-40-9-dos-municipios-do-pais-nos-ultimos-anos

Book: Suscetibilidade a deslizamentos do Brasil: Primeira aproximação

Book: O manual de deslizamento – um guia para a compreensão de deslizamentos

Tesis: Mapeamento áreas susceptíveis à ocorrência de escorregamentos no Brasil e suas relações com aspectos socioeconômicos

MAPEAMENTO DE ÁREAS SUSCEPTÍVEIS À OCORRÊNCIA DE ESCORREGAMENTOS NO BRASIL E SUAS RELAÇÕES COM ASPECTOS SOCIOECONÔMICOS. XXIV 156p (UnB/IH/GEA, Doutorado, Gestão Ambiental e Territorial, 2017).  

Suscetibilidade a deslizamentos do Brasil : primeira aproximação / IBGE, Coordenação de Recursos Naturais e Estudos Ambientais. - Rio de Janeiro : IBGE, 2019.

#landslide #database #ai #population #risks #government

This project has been submitted for consideration during the Judging process.