Imagine you were an extraterrestrial creature trying to understand how to get away from a car, from a ship, or from a plane chasing you. Imagine you didn’t know how they worked, and you couldn’t see them, but you could track where they showed up. Look at enough data points on where the ship turned up, and you’d figure out to avoid water. Look at enough data points on where the car turned up, and you’d learn to avoid roads. Gather enough data on where the airplane, and you’d learn to stay underground, or at least avoid airports.

Similarly, a pandemic affords scientists millions of data points on where the virus goes ,  and where it doesn’t; where it is most deadly , and where it isn’t. This kind of data can yield clues on how to avoid it, and how to fight it. I call this approach global comparative pandemics. It is one that can offer great insights on this pandemic that has already claimed more than 800,000 lives worldwide.

In the past several months, I have engaged in such an exercise. (The original research papers expanding on the findings referenced here, some of which have been presented at the WMO Climatological, Meteorological and Environmental factors in the Covid-19 pandemic Symposium, can be found here.) I have learned that Covid-19 spreads more slowly in places with more sunlight; learned that Covid-19 case counts grow less rapidly following one incubation cycle after sufficient sunlight exposure. I have also learned that this pandemic originated in Wuhan during the darkest month of the year there, on the darkest January in over a decade.

African Americans have been dying from Covid-19 in disproportionately high numbers. My research shows that this happens with more frequency in places with less sunlight. In Milwaukee, for example, they are dying at 670% of the rate of the rest of the population. Yet in Florida, they are dying only at a rate 8% higher than the rest of the population. This is consistent with an important role of sunlight in fighting Covid-19: In places with a lot of sun, even dark-skinned people get enough sunlight, but in places with less sunlight this time of year, dark-skinned people are at a more marked disadvantage.

Indeed, my research shows that one ethnicity shows a much lower confirmed case fatality rate (cCFR) for Covid-19 than the rest of the population: Latinos. And although Latino blacks are more likely to be diagnosed with Covid than non-Latino blacks, Latino blacks show a cCFR 75% lower than that of non-Hispanic African Americans. We found the same trend was true for every other race examined: Latino Native Americans die less than non-Latinos, as do Latino Asians. What explains Latino’s innate defense against dying from Covid-19? The answer may have to do with the sun once again: our research shows that Latinos spend more time outdoors than any other ethnicity in the U.S.

There is some historic precedence here: During the 1918 Spanish Flu, open-air hospitals in Boston that exposed patients to the sun saw a reduction in fatality rates: to 13% from 40%.

We’re in the middle of a solar cycle minimum, when the sun’s magnetic field flips and the total solar irradiance reaching the Earth is at a minimum, as are the number of sunspots. The last pandemic, H1N1, happened during the last solar cycle minimum. The 1918 Spanish flu pandemic happened during a solar cycle maximum. When it comes to viral pandemics, the flipping of the Sun’s magnetic field seems to matter, on either extreme, when electromagnetic radiation reaching the Earth is changing most slowly and virus evolution can keep up with the rate of change. While higher solar activity and irradiance typically correlate with more sunspots, the passage of a sunspot temporarily reduces irradiance reaching the Earth. Indeed, my research found that Covid-19 cases grow faster one viral incubation cycle after days with a higher number of sunspots. It turns out the coronavirus’ name may have proved uncannily prescient. Perhaps isn’t just the virus’ crown we should be looking at, but the sun’s, as well.

Smokers in France have been shown to be less affected by Covid-19. Whilst the French are testing giving medical personnel nicotine patches, the answer may be simpler: with all the smoking restrictions indoors, smokers spend more time outdoors, in the sunlight. Differences in the time spent outdoors could underlie some of the increased prevalence of Covid-19 among the elderly, who spend less time outdoors, and perhaps even some of its gender bias.

The virus that causes Covid, Coronavirus-COV-2, is an RNA virus, made of the single stranded nucleic acid that is also used as intermediaries between DNA and proteins. That means they don’t have a backup copy of the genetic information that can be used to repair damage caused by ultraviolet radiation. That makes Covid-19 particularly sensitive to the sun’s rays. Indeed, the virus evolved in a bat, a nocturnal animal, so it did not get selected for resistance to sunlight.

Sunlight also stimulates production of vitamin D, which is known to boost immune response and also to reduce the production of pro-inflammatory cytokines, which could also explain some of the benefit of sunlight, since Covid-19 infection, like H1N1, gives rise to a cytokine storm, an autoimmune response where the immune system attacks the body’s own cells. Indeed, countries with lower prevalence of vitamin D deficiency among elderly females have seen half the Covid-19 fatality rate than those with higher prevalence of vitamin D deficiency.

My research also showed that women in countries with larger fractions of Islamic population show a higher Covid-19 morbidity relative to men’s than those with smaller relative Islamic populations, and a lower Covid-19 confirmed case fatality rate relative to men’s than those with smaller relative Islamic populations. The effect cannot be explained by the gender-specific fatality rates for several other diseases. Indeed, we observed a significant negative correlation between the female/male morbidity ratio and the female/male case fatality rate ratio, consistent with the possibility that the higher the proportion of the population that is Islamic, the more women are socially distant and the lower their sunlight exposure, leading to reduced morbidity (infection) and increased fatality, respectively.

Indeed, the homeless population throughout California has done much better than people expected throughout the Covid-19 pandemic—most likely because they spend so much time outdoors, in the sunlight.

Architects, government officials, and urban planners have an important role to play in applying these findings to plan for pandemic and post-pandemic life. This will not be the last pandemic. Outdoors spaces and spaces with plenty of sunlight, especially schools, will help society return to daily activities faster, and we should provide them wherever possible.

The way the Sun influences the epidemic’s growth could be even more interesting and novel. This doesn’t mean you can’t get sick in the light ;  it’s a game of chances. The spread of the disease slows significantly when the sun shines stronger. This theory correctly predicted the pandemic would get worse in the southern hemisphere during the Southern winter months of June, July, and August, and predicts it may take on new strength in the north starting in October, peaking in January.

The lesson here is clear: This is a disease of darkness. While social distancing indeed saves lives , we may need to learn how to keep apart at safe distances—not locked away in the shadows, but in the sunlight.