Humans did affect the atmosphere – even before industrialisation



The past is the key to the future. When snow falls on polar ice sheets, in Greenland and Antarctica for example, air is trapped between the snowflakes. Year after year, the snow compacts under its own weight to become ice, and air is enclosed in small bubbles. These are the best atmospheric archives on Earth.

By analysing the air enclosed in polar ice, we can reconstruct past changes in the atmosphere’s composition. We have reconstructed past methane (CH4) variations back to the Roman Empire period, and discovered that humans emitted significant amounts of methane 2,000 years ago.

Nobel Prize winner in chemistry Paul Crutzen described the current geological period as the “Anthropocene”, or the human era. It started with the onset of the industrial revolution in the second half of the 19th century, when human-related emissions of atmospheric trace gases strongly increased.

But did it really start then? For how long have humans influenced the composition of the atmosphere?

A decade ago, famous climatologist William Ruddiman suggested humans influenced the climate much earlier than previously thought. Ruddiman’s hypothesis was criticised, but a few years later isotope measurements on methane trapped in polar ice cores indicated strong biomass burning – likely related to human activity – had increased atmospheric methane levels before the 16th century.

The strength of isotope measurements is that they allow distinguishing variations in methane emitted from various sources. Indeed, every type of sources produces methane with a characteristic isotopic signature.

Because methane is emitted by multiple sources, understanding its budget is not straightforward. These sources can be divided into three categories: biogenic, pyrogenic and fossil sources, with each category comprising both natural and anthropogenic methane emissions.

Biogenic methane is formed by tropical and boreal wetlands, ruminants, rice paddies, landfills and waste water treatment. Pyrogenic emissions include biomass burning and the combustion of bio fuel. And fossil sources can be divided into two groups: fossil fuel and geological sources (methane formed deep in the Earth’s crust, which then travels upward through the sediment to the atmosphere).

In order to better understand the contribution of the different methane sources to the atmospheric burden, we have analysed the methane isotopic signature on ice core samples from Greenland. These samples date back to the Roman empire period.

Our results were interpreted with the help of atmospheric models. They show that the centennial-scale variations in isotope ratios can be attributed to changes in pyrogenic and biogenic sources. They also reveal several distinct periods of higher methane emissions from biomass burning within the last two thousand years.

Variations in biomass burning coincided partly with climate variability (changes in temperature and precipitation), but also with changes in human population and land use. For example, the burning-related emissions decreased during the decline of the Roman empire and the Han dynasty, and increased during the population expansion of the Medieval period. This is attributed to increased deforestation during those periods, but also to the burning of wood for heating purposes and metallurgy.

Heavy metals in dust were recorded in the sediment of several lakes in Asia and in Greenland ice. This shows metallurgy, especially to produce weapons and tools, was important during these periods.

Further analysis of changes in land use was carried out by EPFL’s Atmosphere Regolith Vegetation group in Lausanne, Switzerland. Their results, together with the methane data, suggest that the long-term increase in methane concentration over the last two thousand years is caused at least partly by agricultural activities. For example, the development of rice paddies and irrigation of agricultural fields, thereby providing experimental evidence for the Ruddiman hypothesis.

It seems humans had a global impact on the atmosphere long before industrialisation. The results of our study show a need to reconsider the benchmark of natural versus “anthropogenic” eras, while aiming to predict future climates.