ENSO as a climate conductor for global crop yields
This is a guest post by Weston Anderson (@HydroClim) who is an assistant research scientist with the Earth System Science Interdisciplinary Center at the University of Maryland and the NASA Goddard Space Flight Center, where he works with the Famine Early Warning System team.
A couple of weeks ago, NOAA issued a La Niña Watch based on the possible development of a La Niña this fall that could persist through Northern Hemisphere winter 2021-22. At times like this, we climate scientists spend a great deal of time discussing how ENSO will affect precipitation and temperature, but we often neglect discussion of what those changes will mean for vegetation and crops. That’s a mistake.
As a scientific community, we’re doing a lackluster job of using what we already know about ENSO (El Niño/Southern Oscillation, the entire El Niño-La Niña system) to monitor and predict global patterns of crop yield surpluses and deficits. If I had to give us a grade, I’d say we get a C+. That may sound harsh, but remember that in the 1980s we already understood many of the ways that ENSO affects the global atmospheric circulation. By the 1990s, we had connected ENSO to crop yields in many parts of the world (see references at the bottom of this post).
And yet, for over 20 years afterwards, despite that meticulous study of how ENSO affects global precipitation, many scientists continued to focus on how ENSO impacts crop yields at regional levels, while ignoring the question of how ENSO impacts crop yield anomalies on global scales. We did this even as strong ENSO events determined the spatial pattern of crop yields on nearly every continent in some years. But it wasn’t until after 2010 that we started to seriously study the question (e.g. Iizumi et al., 2014). We failed to see the forest for the trees.
Maize yield anomalies 1982-83
Observed maize yield anomalies (top panel) expressed as percent deviations from the long-term trend for 1983. A very strong El Niño began developing during Mar-May of 1982, peaked during Oct-Feb of 1982-83 and had transitioned into a weak La Niña by the following Aug-Oct of 1983. Estimates of the extent to which ENSO affected maize yield anomalies in 1983 are shown in the bottom panel. Figure adapted from Anderson et al., 2019, which contains the methods used to attribute yield anomalies to ENSO. Figure modified by climate.gov.
Our Food System is Global
Why does it matter if there’s any kind of pattern to global crop yield anomalies in ENSO years? To understand that, we’ll need to take a quick detour to discuss our global food system.
When I say the “global food system,” I’m referring to the fact that roughly a quarter of all calories consumed by humans are traded internationally today. This is a marked increase from 30-40 years ago, prior to widespread agricultural trade liberalization. While more people have access to an adequate number of calories today than ever before, people are increasingly reliant on global trade for that access (1).
Relying on internationally traded food is not necessarily a bad thing, it’s simply a different food system than one in which you rely on food produced entirely in your immediate region. One benefit of a global food system is that by geographically distributing crop production, your access to a crop isn’t necessarily wiped out by any single drought. For example, one country may depend on another country, like Australia, to provide them enough wheat. If Australia doesn’t produce enough wheat due to a poor harvest, then that country can get their wheat from somewhere else, like Argentina. However, this stability assumes that major wheat producing countries like Australia and Argentina won’t simultaneously have poor harvests.
Perhaps it’s a good time to focus on that forest again.
ENSO is a control freak
The assumed independence of crop yields in distant breadbasket regions can break down during ENSO events. As I mentioned before, we know that ENSO organizes precipitation, temperature, and cloud cover on global scales in many seasons, so it’s only logical that it likewise affects crop yields (2). And by influencing the growing season climate to be more or less favorable, ENSO is picking agricultural winners and losers.
So how widespread is the influence of ENSO? After all, many crops are grown in Northern Hemisphere summer, when ENSO tends to have weaker amplitude and a weaker influence on the global climate. Well, for better or worse, ENSO is estimated to affect yields on over a quarter of global croplands. And while it doesn’t affect crop yields every year, it can have an outsized influence in certain years when ENSO is particularly prominent relative to other weather and climate patterns (3). The massive El Niño of 1982–83 that subsequently swung into a La Niña, for example, had a tremendous influence on global maize yields. It was responsible for affecting the maize yields on nearly every continent that year.
But when I say ENSO is a control freak, I don’t just mean that it organizes the spatial pattern of global crop yields. It can also tie good or bad years together across time.
La Niñas bring a friend
One feature of ENSO that’s particularly relevant to agriculture is that La Niñas often come in pairs and occur back to back, like those terrifying twins from The Shining. The avid reader of the ENSO Blog may already know this, given that last year was a La Niña year, and here we are again in a La Niña Watch. And, unfortunately, just like those nightmarish ghoul-girls, more isn’t always better in regions where La Niña tends to cause drought. In areas like the southern US, Horn of Africa and southeast South America, multiple La Niñas may mean consecutive years of poor crop yields (4). In these regions, multi-year droughts can deplete water reservoirs used for agriculture and could draw down grain reserves two years in a row. In southeast South America, for example, there is good evidence that multiple La Niñas often mean consecutive years of below expected yields (Anderson et al., 2017). So, these regions may understandably want to prepare for the possibility of a second consecutive year of poor yields given the current La Niña Watch.
sst and circulation changes linked to maize anomalies
Average observed sea surface temperature and geopotential height anomalies for Nov-Jan during La Niñas, which tend to be preceded by El Niños and followed by weaker La Niña-like conditions, for the years 1950-2010 (top panel). Corresponding maize yield anomalies, expressed as percent deviations from the long-term trend, in Argentina, Uruguay, and Southern Brazil (Paraná, Santa Catarina, and Rio Grande do Sul). Figure provided by Weston Anderson and modified by climate.gov.
Aside from worry, what should we do?
Going forward, the climate science community would do well to strengthen working relationships with the food systems and food security community. And I’m not just talking about agronomists. We need to be working with those that study pastoralists, aquaculture, and global trade as well (among others). By doing so, we can make good use of the truly remarkable knowledge amassed by the climate science community with respect to the way that ENSO organizes global climate, and, therefore, presents a structured risk to the global food system.
Lead Editor: Michelle L’Heureux (NOAA CPC)
« Back to index