In a country where voting is not compulsory and when the margin of victory is small, the weather has the potential to have an influence on the election results.
Research has shown fair weather in 1960 aided Democrat John F Kennedy, while rain in 2000 helped secure Florida and an eventual win to Republican George W Bush.
The race to the White House this year between Donald Trump and Kamala Harris is a classic case where the atmosphere could be the decisive factor, with a 3,000-kilometre band of heavy rain and thunderstorms forming from Texas to Canada, which is predicted to bring a swathe of severe weather during the next 48 hours over multiple states.
Non-resident senior fellow at the University of Sydney's United States Studies Centre Professor Benjamin Reilly said the contest between the candidates is one of the tightest in US election history.
"It's too close to call. Opinion polls put it very close," he said.
Rosalind Dixon, Professor of Law and Director of the Gilbert + Tobin Centre of Public Law at the University of NSW, said the race was "extremely tight".
"I think that all the polling suggests it is down to a few states, most of which are within the margin of error," she said.
Professor Reilly said while the severe wet weather would "negatively affect" voter turnout, it was too difficult to determine how significant the impact would be for the current election.
So could the weather make a difference? During periods of inclement weather, like rain or snow, turnout numbers are impacted for marginal voters who are less politically active.
And since people on the periphery of politics are more likely to be working class, the drop in turnout is higher for one party over another.
Two swing states could be impacted by weather
Even without the backdrop of a US election, the current weather system is newsworthy.
A very moist airstream of the Gulf Of Mexico is feeding into a low pressure system over the central US.
This system is already causing severe weather with excessive rainfall and severe thunderstorms across most of south-central US on Monday, leading to areas of flooding.
This low is slowly moving east and will extend a band of rain and storms from Canada to Texas on Tuesday (US time).
The heaviest rain is likely to fall over relatively safe states, including Republican stalwarts in the south and the Democrat-leaning Illinois.
However up to 50 millimetres is also possible in critical swing states Wisconsin and Michigan.
Professor Dixon said these two states were the ones to "watch" because of this reason.
While Detroit should remain east of the main rainband, northern Michigan and southern Wisconsin, including Milwaukee, are likely to be blanketed by the rainband for the majority of voting hours on Tuesday.
How a wet day impacts voter turnout
Several studies looking into the weather's influence on elections across the world and the results are clear — rain and snow have a significant impact on voter turnout.
A 2023 study on Denmark elections showed rainfall dropped participation by 1 per cent per 10mm.
A 2007 study analysed US elections and found that when compared to normal conditions, rain reduces voter participation by a rate of just less than 1 per cent per inch, or 1 per cent per 25mm.
That could mean a drop of 2 per cent in participation in Michigan, Wisconsin and other counties affected by the heaviest falls (and even higher in states further south where 100mm is possible).
A separate weather system dropping snow on Montana is also likely to lower turnout, but its unlikely to prevent a safe win for Trump.
Republicans pray for rain no more
So does rain or snow favour one party over another? The old adage "Republicans should pray for rain", has been found to be accurate during past US elections.
As the theory went, bad weather may be the last straw for peripheral voters, especially young people who have not developed a voting habit, and they were disproportionately inclined to support the Democratic candidate.
This link was tested in the Gomez study and the data indicated for every 25mm increase in rain, the Republican presidential candidate received just over 2 per cent of the vote — easily enough to swing a state, and therefore an entire election.
Professor Reilly said due to the fact generally staunch Republican states in the south — Texas, Oklahoma and Arkansas — will be the worst affected by the rainband, it would impact the overall vote share for the party.
However, going by academic studies including the study by Gomez, Professor Reilly said the Republican Party has been "speculated" to benefit more from poor weather historically.
Professor Dixon said she suspected the weather would have "a very modest impact".
"If you're asking me to bet on it, I would say this won't turn out to be significant. Could it be? Absolutely," she said.
"Weather adds just one more element of uncertainty in an already enormously close, uncertain election."
Professor Dixon said "more determined" and "politically engaged" voters would turn up, regardless of the weather.
"There are definitely motivated voters on both sides, and I think it is very relevant to say weather is a factor and we should be paying attention to it.
"But I think it's very ill-advised to be making firm predictions about which way it will cut."
Indeed, the weather may have contributed to two past electoral college outcomes.
In a very tightly contested 1960 election, it is estimated that very rainy or snowy weather would have gained Richard Nixon an additional 106 electoral college votes over John F Kennedy.
The reverse results – if it was dry as opposed to wet, revealed two instances which would have changed an electoral college outcome — Bill Clinton to win North Carolina in 1992 and Al Gore to win Florida in 2000.
This latter change would have swung the incredibly close 2000 election in Mr Gore's favour over George W Bush.
However recent evidence shows this pattern may have reversed, as working class voters now lean more towards the Republican Party and college-educated voters to the Democrats.
A paper by Spencer Goidel and co-authors in 2023 showed the shift in turnout by studying US federal elections from 2010 to 2020.
"Our findings show that increases in turnout greatly benefit the Democratic Party in the pre-Trump era. However, this pattern has drastically changed. In 2016, 2018, and 2020, the Democratic Party sees a much smaller gain in vote share as simulated turnout increases, but also a large vote share advantage when voter turnout is extremely low."
This implies Mr Trump should perhaps pray for sunshine rather than rain over Michigan and Wisconsin on Tuesday.
