Radiation is one of the main ways in which energy is transported through the atmosphere and radiation to space is the only way in which Earth can get rid of excess energy. The CO2 we emit into the atmosphere makes it more difficult for Earth to radiate energy to space. Thus, this additional energy accumulates which causes temperatures to rise. In a nutshell, this is the so-called "greenhouse effect", the driving factor behind climate change.
But how do so-called "greenhouse gases" inhibit the radiation escaping to space?
To understand this, you have to consider that radiation comes in a range of different wavelengths, ranging from the short-wavelength light we receive from the sun to the very-long-wavelength microwave radiation we use to warm up our leftovers. In between lies the long-wavelength infrared radiation that Earth's surface and atmosphere emit. Infrared radiation is not visible to the naked eye, but it is crucial for Earth's energy balance. Greenhouse gases in Earth's atmosphere - such as CO2, but also water vapour - absorb this infrared radiation, thus preventing it from escaping to space.
But greenhouse gases only absorb infrared radiation with certain wavelengths. For water vapour, these wavelengths are called water vapour bands. At other wavelengths however, the greenhouse gases let radiation pass through. Parts of the spectrum where little to no absorption takes place are commonly referred to as windows. Just like real windows that let through the sunlight from outside, windows in the infrared let radiation emitted by Earth's surface pass through the atmosphere. On the other hand, greenhouse gases act like dirt on these windows that does not let any light through.
Now imagine you are a satellite looking down at Earth. Then you would "see" a lot of radiation in the infrared windows, because the radiation there directly comes from Earth's warm surface. In the "dirty" water vapour bands however, radiation from the surface is absorbed on the way through the atmosphere, so the satellite only sees the radiation emitted by the upper atmosphere. Because the upper atmosphere is much colder than the surface, it also emits much less radiation.
But what happens during climate change?
As Earth warms, the radiation it emits increases, but not equally at all wavelengths. This change in the spectrum of infrared radiation Earth emits is also called "spectral longwave feedback parameter", and it is a central quantity in our understanding of climate change. Using climate models, some previous studies predicted that when Earth warms, the infrared radiation it emits strongly increases in the window region (which corresponds to a negative spectral longwave feedback parameter). At the same time, they predicted the infrared radiation Earth emits would barely change in the water vapour bands (which corresponds to a spectral longwave feedback parameter of zero). In contrast, other studies have predicted an increase in emitted radiation in both the window region and the water vapour bands. However, all of those studies were based on climate models, which have their limitations and biases.
So we wanted to know: How does the spectrum of emitted infrared radiation actually change with warming?
Luckily, the infrared radiation Earth emits can be directly observed by satellites orbiting Earth. In our study, we use satellite observations by the so-called IASI instrument which measures the outoing radiation at more than 8,000 different wavelengths and analyse how the radiation Earth emits has changed with the temperature of Earth's surface between 2007 and 2020. We find that the emitted infrared radiation seems to increase in both the window and the water vapour bands, i.e., the spectral longwave feedback parameter is negative throughout the spectrum (Figure 1).
But why do some climate models disagree with our findings?
We argue that there are two main reasons: The first reason has to do with relative humidity. Usually, it is assumed that as Earth warms its average relative humidity does not change. However, in the time period we look at, relative humidity tends to be slightly lower when temperature was higher and vice versa. Using idealised simulations with a climate model, we show that with warming the increase in radiation in the water vapour bands measured by the satellite can be partly attributed to this decrease in relative humidity (Figure 2).
The second reason has to do with the polar regions. You might remember that I referred to the water vapour bands as "dirt on a window", which does not let through any light. However, the cold polar regions contain almost no water vapour in their atmosphere - in our analogy, the windows there are almost clean. This means that even in the water vapour bands, some radiation from the polar surface can directly pass through the atmosphere. As the surface warms, this radiation increases, which can be observed by the satellite (Figure 3). In the very simple climate models that were often used to investigate the spectral longwave feedback parameter before, this effect is not accounted for.
In conclusion, we directly observe the spectral longwave feedback parameter. In contrast to some model predictions, our observations reveal that as Earth warms, it can actually radiate more energy to space in the water vapour bands. We now also understand better how changes in relative humidity contribute to this, highlighting the importance of further study on this topic. Going forward, our findings can be used to evaluate the accuracy of climate models to better predict future warming.