Our team was inspired to tackle this challenge because of our personal desires and experience to raise  awareness of the effects of COVID-19 on the environment and local businesses. Thus, we wanted to see the how the economic impact of local businesses could be Influenced by the decrease of major economic sectors such as transportation but also explore any environmental recovery with less human interaction. 

UX/UI

When considering the users, the website was designed to encourage and excite users to contribute financial help to the local economy through the purchase of pizzas.

To gather insight into user research, we considered indirect competitors in the food delivery (i.e. Yelp, Uber Eats, Papa Johns) to consider overall trends in food delivery website interfaces along with possible target audiences. This competitive research then informed our “Environmentally Aware Business Analyst” archetype target audience. This persona cares about the economy and environment with alike enthusiasm in both pre-COVID and COVID scenarios.

The website information architecture required a features priority list, sitemap and user flow (built in Figma) when considering the landing page structure and hierarchy. We wanted to prioritize a landing page where users could easily find the desired location, but also view important data charts per location quickly in a left side panel.

A style guide was created to inform the high fidelity wireframes built in Figma that include a landing page with a map and left panel that prompts users to click the call to action button. From there, users are taken through an onboarding panel that informs users of necessary steps required to contribute a pizza purchase commitment. Users are also provided an information panel about different economic indicators in their selected region before being prompted to select a numerical pizza commitment. Once users choose their number, the charts respond reactively to their contribution and that of other users, revealing new projected economic indicators.

Ideally, product prototypes would undergo usability testing for further wireframe reiterations but the limited time restraints did not allow for users to test the prototype and provide feedback.

Back End

EO:

• Time series data from different satellites were preprocessed and plotted to see if any evident sign of anomaly that could be associated with the massive slowing-down of the economy exists.
• If yes, then we would follow up by using numerical analysis, such as statistical analysis. However, our preliminary analysis, even the one in which we used medium resolution data (30-m), showed no convincing evidence that the ongoing crisis caused any significant perturbation that is detectable from space.

APIs

• App: WebAPI backend using C#.  The frontend connects to the backend's API endpoints to get the calculated data.  The data used for the calculations are stored in the backend app and came from various sources we used during the project.

Future possibilities

UX/UI

• With the time restraints, next steps for this prototype would be to scale the project nationally and globally, while including other local businesses.
• After users add their pizza commitment numbers, users would be prompted to then take action immediately and insert their credit card information in the following page. The financial commitment would then be transferred to partner pizza shops who would then re

EO

• We have found that the European Space Agency (ESA) once deployed a trial on the Road Traffic Monitoring by Satellite (RTMS). Such information would add an important insight into an economic assessment, especially in an unprecedented circumstances like this, because transportation is closely associated with economic activities and is an economic sector on its own. Unfortunately, no data from the trial covered the present period.

Challenges

• Figuring out what data sets to use required a lot of time at the beginning to determine the scope of the project based on what possible data could be produced.
• Communication between different experts required more effort to get everyone up to speed about the general background, purpose and solution. We had video calls often during the weekend with a busy Slack workplace where team members were constantly updating each other.
• The UX/UI Designer struggled with assessing what information could be used since the user (tangible user-focused solutions) and scientific goals (more broad high-level abstract concepts) was very different and at times hard to communicate across to the other unfamiliar expert
• The idea of using a pizza as an indicator of economic demand can be weird to connect, but the goal to make the information relatable to users was important to making the complex data usable for everyday people. So we chose an object that users could purchase during COVID and support locally.
• Finding recent Input-Output data was a daunting task. Thankfully, Prof. Adam Rose from the University of Southern California generously provided the IO data from 2016 for us to use in the analysis. That was not the end of the issue because we had to translate it to better fit the most recent situation. An extensive online data search was conducted and finally, we were able to deliver our results with all the data sourced from diverse platforms stitched together.
• No evident difference between the pre- and post-COVID is found on the satellite images used in the analysis. First, we thought that it is due to the coarse resolution of the data (1 km for VIIRS and MODIS), so we tried using the land surface temperature data from Landsat-8. Unfortunately, even after working at a finer resolution, no strong pattern was unveiled. This kinda reminded us of Carl Sagan’s research that demonstrated from space, humans are vaguely detectable (Sagan et al. 1993: https://www.nature.com/articles/365715a0). An endearing reason to embrace humility, it is.
• In this project, we were seeking the possibility to integrate earth observation data to provide proxies for the slowing down of the economic activity that has been happening due to the global pandemic. Some potential earth observation instruments considered and analyzed in the study were Landsat-8, MODIS, and VIIRS. These instruments were expected to be able to display some differences between the pre- and post-COVID situation. However, our preliminary analysis showed little difference depicted by time-series land surface temperature and night light intensity data extracted from these sources. This may be due to the spatial resolution of the data that are not fine enough to capture the reduction of economic activity on the ground.
• The back end developer ran into some challenges where he had trouble figuring out how the calculations were going to occur.  He considered doing it in python or R but it might take more time for us to figure out how to connect to a R or python script from the backend API.  Therefore, we decided to do the calculation in the backend API, but needed help from other teammates to visually figure out the calculations.
• Another back end challenge was massaging the JSON files from those data sources to make sure they were useful for the backend API. Looking back, we probably could have used a database to store the data.

https://youtu.be/4pM_as28iOA

https://space-apps-hibird-theory.now.sh/

Resources

• Charts: https://www.chartjs.org/
• Maps: GoogleMaps
• Map Style: https://snazzymaps.com/
• Technologies used: JavaScript, HTML/CSS, React

References

U.S. Energy Information Administration. "Energy consumption estimates in 2017" (2018), "Carbon Dioxide Emissions Coefficients" (2016).

U.S Department of Commerce-Bureau of Economic Analysis. "Input-Output Table for the State of California 2016".

Walmsley, T., Rose, A., and Wei, D., 2020, "Impacts on the U.S. Macroeconomy of Mandatory Business Closures in Response to the COVID-19 Pandemic". in-press

Gorelick, Noel, Matt Hancher, Mike Dixon, Simon Ilyushchenko, David Thau, and Rebecca Moore. 2017. “Google Earth Engine: Planetary-scale geospatial analysis for everyone.” Remote Sensing of Environment202: 18–27.https://doi.org/10.1016/j.rse.2017.06.031.

NOAA, "VIIRS Stray Light Corrected Nighttime Day/Night Band Composites Version 1". Courtesy of Earth Observation Group, NOAA National Centers for Environmental Information (NCEI)

The U.S. Geological Survey. "Landsat-8 Collection 1 Tier 1 TOA Reflectance". Courtesy of the U.S. Geological Survey

Wan, Z., S. Hook, G. Hulley. MOD11A1 MODIS/Terra Land Surface Temperature/Emissivity Daily L3 Global 1km SIN Grid V006. 2015, distributed by NASA EOSDIS Land Processes DAAC, https://doi.org/10.5067/MODIS/MOD11A1.006. Accessed 2020-06-01.