Our project pulls real-time data for numerous environmental factors which is then presented in a practical application that is constantly updated so that all users stay informed on how the virus is responding to the environment. The application contains several tabs with many functionalities. Live case counts, recoveries, and weather data for every county in California are displayed on the Home tab. This home tab provides a daily overview of live numbers of weather and COVID-19 statistics. The environment tab displays graphs which update constantly to show relationships between COVID-19 and temperature, wind speeds, pressure, and other climate elements. Users can view the environment tab to learn trends of how the environment is affecting COVID-19 spread in a given California county. The world tab is similar to the environment tab, however, it shows the correlation of COVID-19 and environment on a world scale. When viewing the world tab, users are able to compare the local county trends with those of the entire world in order to better understand the progression of COVID-19 in their local community. The final tab of our application is one that predicts the probability of infection in different California counties. The predictions are based on trained models created by a machine learning algorithm which becomes more accurate as more data is analyzed. This unique feature enables users to examine the severity of COVID-19 in their area or when traveling outside of their community.
The Wahoo Hackers’ extensive background in computer and data science made the “Where There’s a Link, There’s a Way” challenge particularly appealing to our team considering our experience and interests. We also realized with this challenge that we would be able to spread factual information rather than opinionated information. In addition, with this challenge, we were excited by the opportunity given to create relationships between the environment and COVID-19 which has not previously been thoroughly investigated.
Our approach to the project was a “Scrum” approach where we strove to deliver a great product and maintain accountability by having constant check-in meetings with the entire team. Each team member played several roles, from Scrum master to Product owner. Throughout our daily scrum meetings, we made sure the application was answering the desires of the customer, while also maintaining a clear To-do list for that block of the day. The most useful aspect of the “Scrum” approach were the sprints where each team member had a certain task to complete before moving onto the next. These sprints kept up the quality and accountability of each team member. In the end, the values of commitment, openness, and respect were ingrained into us and we were proud of our final product.
The Space Agency Data we used was a map from NASA that contained Land Surface Temperature Anomalies (https://neo.sci.gsfc.nasa.gov/view.php?datasetId=MOD_LSTAN_M&date=2020-05-08) for each week of the current year. The map contained color gradients to represent temperature deviations from average temperatures in the period from 2001-2010. The color gradient showed temperature differences of up to 12°C warmer or cooler than the given average, with blue signifying cooler than average and red signifying warmer than average. In a majority of the dates, the maps showed a large percentage of the world with regions of blue, meaning the temperatures were cooler than in years past. With this knowledge, we arrived at the conjecture that these cooler than normal temperatures allowed for a wider spread of COVID-19. Many scientists had previously predicted that the warm summer months would lead to a decline in the expansion of COVID-19 with its typically high temperatures. As evidenced by the ever-growing cases throughout the world, the summer months have done little to lessen the growth of COVID-19 and the Land Surface Temperature Anomalies Map shows that the cooler than expected temperatures are a possible cause for such an occurrence.
Using a variety of tools, from Python to Figma, we were able to paint an informative picture of how the environment affects the spread of COVID-19 across time. Python was used for data extraction and creation, while R and Microsoft Excel were used for statistical analysis and graphing. R was also used in the creation of our Machine Learning predictive model, which became more accurate after receiving larger quantities of data. Once all background information was obtained, we utilized Figma to design a blueprint of our application and showcase our findings.
We ran into numerous problems during the development of our project. A minor conflict we faced after gathering our data was how to present our findings. After collecting a vast amount of data, we ran into issues on how to present our material in an informative yet viewable size in the mobile application. Another major conflict we faced during the initial stages was determining the scope and audience of the data we wished to present. We began by planning to examine environmental data across the entire world in order to inform the largest population possible of the environmental effects, however we soon realized that extracting and interpreting such a large amount of data within the given time frame would prove an insurmountable task. This led us to reduce our analysis to just California and its counties in the time being for the prototype, with interests to expand our application to a world-level in the future.
In addition to our team managing to create an organized application which is able to gather relevant, real-time environmental data and display it in a clean and informative view for Californians, we were also given the chance to overcome setbacks, push our limits, and apply ourselves to the fullest on a challenge every team member was fully invested in. A byproduct of completing the challenge was the feeling of accomplishment, diligence, and companionship between each team member in a time of quarantine and isolation.
NASA Land Surface Temperatures Anomalies: https://neo.sci.gsfc.nasa.gov/view.php?datasetId=MOD_LSTAN_M&date=2020-05-08
NASA Moon Phases Video: https://svs.gsfc.nasa.gov/4768
Los Angeles Climate Data: https://www.wunderground.com/history/monthly/us/ca/los-angeles/date/2020-5
County COVID-19 Data: https://raw.githubusercontent.com/nytimes/covid-19-data/master/us-counties.csv
World COVID-19 Graphs Data: https://ourworldindata.org/coronavirus-source-data
US Census Bureau: https://www.census.gov/data/datasets/time-series/demo/popest/2010s-counties-total.html
Fips Codes: https://www.nrcs.usda.gov/wps/portal/nrcs/detail/ca/home/?cid=nrcs143_013697