Our project addresses this challenge by finding an effective way to utilize NASA satellite data (Reference 1) to predict risk assessment for Covid-19 and (we believe) zoonotic Corona Viruses in general. For example, this could help local governance initiate quarantine beforethey become “hotspots”, and aid travelers to help them avoid potential “hotspots” (and subsequently help them avoid bringing the virus back to their home cities).
It is well-documented (ref. 2,3) that the original host animal of Covid-19 (and many other Corona viruses) is most likely bats; in particular horseshoe bats (Genus Rhinolophus) carry viruses 80+% similar in genetic code and demonstrate high prevalence of antibodies to these viruses. As with most bat species, this genus is largely nocturnal. It is logical to conclude that Corona Viruses have evolved to maximize their survival in the host, without killing the host nor being detrimental to the ability of the virus to thrive. It is therefore logical to hypothesize that sunlight may have a detrimental effect on a virus that has evolved to thrive in a nocturnal host.
• To evaluate our hypothesis, we started by utilizing NASA Earthsearch data and Giovanni to calculate OMI UV index for 2° x 2 ° grids around the globe, with a particular focus on areas of high-risk based on modeled air-travel from cities in China near the outbreak epicenter (ref. 4,5). We utilized the coordinates for a particular city of interest, then create a box that was +/- 1° in latitude and longitude as the input grid to Giovanni. Utilizing the Time-series Area-Averaged parameter, we calculated UV index for 25 global cities worth of 2x2 grids over the period of January 22-March 22 and exported the data into Excel. Utilizing Excel, we then calculated 7-day moving average OMI UV index for all 25 cities over this 2- month timeframe. We then calculated time-lapse Covid-19 cases (ref. 6) occurring within these 2x2 grids, sorting data by latitude and longitude and matching to the Giovanni output data.
• We created our plots, data analysis, and correlations based on the above. We focused on this date range because this was pre-mass quarantine and prior to mass global contagion. We utilized the outstanding features of Excel for the data analysis, and PowerPoint for creating the Slides.
• Our youngest member, 16 year-old Henry Bushong, wrote the Python code to form the basis of the initial predictive algorithm. The algorithm would take user input (the city/region of interest), call latitude and longitude for that city from a suitable database, use this as input to Giovanni OMI (auto-creating the +/- 1 grid for the input), utilize the current date as the “end date” with the “start date” auto-filling as 7 days prior to the current date, and output the average 7-day UV index for that 2x2 grid. A risk assessment (on a Scale of 1-10 for example) can be assigned (based on our correlation between UV Index and Covid-19 cases shown in our plot “Total Covid-19 Cases within UV Ranges”) and returned to the user as the practical and useful output of the model. Of course the Code will only be fully functional when integrated with Giovanni; it's meant as a concept-demo for now. It would be great for Henry to be able to work with NASA experts on how to enhance and integrate into Giovanni!
• Our main problems were what data to collect (region, timeframe, grid size), what wavelength (UV was available though we would like to consider the full spectrum of light for the future use of this model / analysis), and sorting / organizing / how to analyze an enormous amount of localized data to draw global conclusions that were presentable and easy to understand!
https://docs.google.com/presentation/d/1lc3qGADwVQlrFMw6ZI9WxMHoSuI7WIemz-7Py0xAHlQ/edit?usp=sharing
Reference 1: J. G. Acker and G. Leptoukh, “Online Analysis Enhances Use of NASA Earth Science Data”, Eos, Trans. AGU, Vol. 88, No. 2 (9 January 2007), pages 14 and 17. We also acknowledge the MODIS mission scientists and associated NASA personnel for the production of the data used in this research effort.
Reference 2: Bats are Natural Reservoirs of SARS-like Corona Viruses (Wendong Li et al., Science 310, 676 (2005) DOI: 10.1126/Science.1118391
Reference 3: Discovery of a rich gene pool of bat SARS-related coronaviruses provides new insights into the origin of SARS coronavirus. Hu B, Zeng L-P, Yang X-L, Ge X-Y, Zhang W, Li B, et al. (2017) PLoS Pathog 13(11): e1006698.https://doi.org/10.1371/ journal.ppat.1006698
Reference 4: Preliminary risk analysis of 2019 novel coronavirus spread within and beyond China. Shengjie Lai1*, Isaac I. Bogoch2, Alexander Watts3,4, Kamran Khan2,3,4, Andrew Tatem1* 1 WorldPop, School of Geography and Environmental Science, University of Southampton, UK 2 Department of Medicine, University of Toronto, Toronto, Canada 3 Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto, Canada 4 Bluedot, Toronto, Canada
Reference 5: https://systems.jhu.edu/research/public-health/ncov-model-2/
Reference 6: https://www.covidtracker.com/