The fresh insight came from ancient air samples trapped inside ice cores taken from Greenland and Antarctica – some of which were analysed at Niwa's Wellington lab.
Further analysis was carried out in a machine called a mass spectrometer which pin-pointed methane sources according to the measured content of the rare, radioactive isotope radiocarbon.
For every single measurement, the team had drilled out one tonne of ice while working in polar environments.
From that, they extracted the equivalent of five household buckets of air, and then isolated the equivalent volume of one water drop of methane.
In that methane, only one in a trillion molecules is radiocarbon and they measured the changes in that minuscule content.
Study co-author and Niwa principal atmospheric technician, Tony Bromley, said the degree of accuracy provided by the analysis meant it was possible to determine exactly how much methane was produced by humans.
Until now, that had been difficult for researchers to determine exactly where methane emissions originated.
Biological methane, for instance, could be released from wetlands, rice fields and livestock, while fossil methane could stem from natural geological seeps, or industries extracting and using fossil fuels like oil, gas and coal.
The two types of methane could be distinguished, because biological methane contained measurable amounts of radiocarbon.
In fossil methane, all the radiocarbon had decayed away, while the gas was stored in underground reservoirs.
Yet radiocarbon measurements in modern air could not separate fossil methane that was emitted naturally from industrial sources.
"As a scientific community we've been struggling to understand exactly how much methane we as humans are emitting into the atmosphere," said study author Dr Vasilii Petrenko, of the University of Rochester in the US.
"We know that the fossil fuel component is one of our biggest component emissions, but it has been challenging to pin that down because in today's atmosphere, the natural and anthropogenic components of the fossil emissions look the same, isotopically."
By measuring the methane radiocarbon from more than 200 years ago when there were no industrial sources, the researchers knew that all fossil methane had to be emitted naturally.
They found that almost all of the methane emitted to the atmosphere was biological until about 1870.
That was when the fossil component began to rise rapidly – and the timing coincided with a sharp increase in the use of fossil fuels.
They also discovered the levels of naturally released fossil methane are about 10 times lower than previous research reported.
Methane being emitted by human activity was the second largest contributor to global warming, after carbon dioxide.
But, compared to carbon dioxide, methane has a relatively short shelf-life; it lasted an average of only nine years in the atmosphere, while carbon dioxide can persist in the atmosphere for about a century.
That made methane an especially suitable target for curbing climate change in a short time frame.
"If we stopped emitting all carbon dioxide today, high carbon dioxide levels in the atmosphere would still persist for a long time," said co-author Dr Benjamin Hmiel, also of Rochester.
"Methane is important to study because if we make changes to our current methane emissions, it's going to reflect more quickly."
In New Zealand, where methane makes up about 43 per cent of gross greenhouse gas emissions, the new Zero Carbon Act would treat biogenic methane emissions differently to CO2.
While the act strived to push net emissions of all other all greenhouse gases down to zero by 2050, its goal for biogenic methane was a reduction of 24 to 47 per cent below 2017 levels – 10 per cent of which would need to be cut this decade.
