Green Sahara: how? pt.1 — Model good

If you haven't yet, read this post to find out what "green Sahara" means.

---

The Sahara, a place that we know to be a near-lifeless expanse of hot desert, was once a humid and highly vegetated environment? Only six thousand years ago? How could this be? I know, reader: the information in my last post shattered your reality. Please pull it back together, and make yourself a nice cup of tea. In this post I'm going to outline the natural processes that drive North Africa to flip between green and desert states, and discuss how humans have used modelling to understand these mechanisms.

At the heart of this issue is Milankovitch theory, named after the Serbian astrophysicist who developed it in the early 20th century. It holds that the Earth's orbital parameters have cycles of variation which determine not only the amount of solar radiation Earth receives, but also the spatial and temporal distribution of this energy (if you want a refresher, click here and here for an excellent, in-depth primer). Their combined rhythm is what drives many of our planet's climatic trends on a long term basis.

The pacing of the Earth's glaciation cycles by the Milankovitch cycles.

In the image above you can see that the approximately 100 ky cycles of the Quaternary glaciations appear to be paced by eccentricity. However, of the three orbital forcing components, it is 20 ky precessional signal which has the most dominant influence on the state of Sahara. Land surfaces heat up more quickly under summer insolation than the ocean, and the resultant atmospheric pressure difference drives moist marine air over the African continent, where it rises and cools to produce the rainfall that we call a monsoon. Land warming is increased when the perihelion is closest to the boreal summer, and the combined effect of the 9 ka orbital configuration is a ~6% higher monsoon season insolation over northern Africa compared to today. The resultant increase in the land-sea thermal contrast intensifies the West African monsoon, bringing this seasonal precipitation deeper and more northwards into Africa and painting the Sahara green with life — or so goes the theory.

When evidence for the green Sahara was first emerging, atmosphere general circulation models (AGCMs) were used to explore the effect of mid-Holocene orbital forcing on climate in North Africa. Since these models could be run repeatedly with minor changes, they were used to determine the sensitivity of the results to orbital parameters and to changes in boundary conditions (such as sea surface temperatures and the ice sheet configuration), and the experiments were were largely successful in simulating the degree of rainfall intensification suggested by the paleoclimatic evidence. One model's increase in mid-Holocene monsoon wetness in North Africa (approximately 0 – 30 °N) compared to modern can be seen below.

Modelled latitudinally-averaged precipitation – evaporation anomaly (9ka – 0ka). Dashed line shows results of a run which prescribed a North American ice sheet. Reproduced from Kutzbach & Otto-Bliesner (1982)

By showing that the monsoon trend was representative of an overall northward shift of the inter-tropical convergence zone (ITCZ), these experiments improved the mechanistic understanding of the processes in the context of large-scale atmospheric circulation. AGCMs were therefore very useful in confirming the primacy of the Milankovitch cycles in modulating rainfall conditions over northern Africa. They also showed that the influence of on climate is considerably weaker in the tropics than in higher latitudes, as shown in the graph above.

By the end of the 1980s, it seemed like climate models forced by mid-Holocene conditions were able provide a fairly "realistic" simulation of a green Sahara, reaffirming the theories —  developed more than half a century previously — of the control of long term climate by insolation cycles. In the my next post I'll expand on why the initial understanding of the green Sahara described here isn't quite as rosy as it seemed.