Late rainy season reliably predicts drought in regions prone to food insecurity

New indicator could help mitigate food insecurity in Sub-Saharan Africa


Research News


The onset date of the yearly rainy season reliably predicts if seasonal drought will occur in parts of Sub-Saharan Africa that are particularly vulnerable to food insecurity, and could help to mitigate its effects. Shraddhanand Shukla and colleagues at the University of California, Santa Barbara’s Climate Hazards Center, present these findings in the open-access journal PLOS ONE on January 20, 2021.

Climate-driven seasonal drought can impact crop yields and is among major contributors to food insecurity, which can threaten people’s lives and livelihoods. In the last five years, parts of Sub-Saharan Africa have experienced a significant rise in food insecurity, sometimes requiring emergency food assistance. Early warning systems that reliably predict conditions likely to lead to food insecurity could help drive timely actions to mitigate these effects.

Shukla and colleagues hypothesized that the onset date of the rainy season, as calculated from precipitation data, could serve as such a warning. To explore this possibility, they analyzed the relationship between the onset date, drought conditions observed via satellite images of vegetation cover, and the risks of food insecurity based on quarterly reports on food insecurity in across Sub-Saharan Africa from April 2011 through February 2020.

The analysis showed that a delay of about 10 days from the median date of onset of the rainy season was associated with a significantly higher likelihood of seasonal drought in regions with the highest risk of acutesevere food insecurity. A 20-day delay indicated a 50 percent chance of drought in those regions. Further analysis confirmed the predictive relationship between rainy season onset date and drought risk across Sub-Saharan Africa as a whole, and particularly in East Africa.

These findings suggest that the onset date of the rainy season could be an important component of an early warning system for droughts likely to lead to food insecurity in Sub-Saharan Africa. Further research could examine the relationship between onset date and other food insecurity indicators, such as high-resolution data on crop yields and prices or mid-season livestock prices.

The authors add: “Timing of rainfall onset can be tracked using remotely sensed observations and forecasted using climate models, and the results of this study show that it can be a reliable indicator of agricultural droughts, particularly in the most food insecure regions of Sub-Saharan Africa, which makes it a simple yet powerful tool to support effective early warning of food insecurity, thus saving lives and livelihood.”


Citation: Shukla S, Husak G, Turner W, Davenport F, Funk C, Harrison L, et al. (2021) A slow rainy season onset is a reliable harbinger of drought in most food insecure regions in Sub-Saharan Africa. PLoS ONE 16(1): e0242883.

Funding: SS supported by #80NSSC18M0039 from the NASA Harvest Consortium, GH, WT, LH supported by #72DFFP19CA00001 from the United States Agency for International Development (USAID) cooperative agreement, CF received support from the USGS Drivers of Drought program and the NASA Precipitation Monitoring Mission grant 80NSSC19K0686, SS and FD received support from W911NF-18-1-0018 by the Defense Advanced Research Projects Agency (DARPA) World Modelers Program under Army Research Office (ARO). FD supported by #80NSSC20K0159 from NASA SERVIR.

Competing Interests: The authors have declared that no competing interests exist.

In your coverage please use this URL to provide access to the freely available article in PLOS ONE:

From EurekAlert!

5 3 votes
Article Rating
Newest Most Voted
Inline Feedbacks
View all comments
Ed Zuiderwijk
January 21, 2021 2:09 am

‘Climate driven droughts’. Remarkable droughts that can be driven.

I always thought droughts are an aspect of climate, but you learn something every day.

Last edited 1 month ago by Ed Zuiderwijk
Reply to  Ed Zuiderwijk
January 21, 2021 9:12 am

Yes, despite some of the sequential relationships being traceable the original post hype of climate being the driver seems a superficial insertion. For example, the Anasazi people in what is now juncture region of Utah/Arizona/Colorado/NewMexico experienced 50 years of drought from the years ~1130 to 1180; & then the 23 year drought from the years ~1276 to 1299, where some suggest no rain actually fell.

January 21, 2021 3:04 am

The main take away from the article is:- If the rainy season starts three weeks late there will be less rain in the rainy season.
This sounds reasonable and it’s also fair to say less rain on average produces less crops.
However, on a local level less rain does not always equal drought or worse crop yields.
Every valley is different with different soil types, slopes, percentage, evaporation rates, cultivation coverage, timing of rain and type of crop.

January 21, 2021 4:41 am

“and forecasted using climate models”

ho hum.

Janice Moore
Reply to  lee
January 21, 2021 11:55 am


You identified the only reason this “analysis” was written up:

(pseudo) – evidence “validating” the proven-unskilled climate simulation models.


1) state the obvious
2) “prove” it using climate models
3) assert (falsely) that, therefore, climate models are skilled vis a vis human CO2 emissions.


They keep the solar, electric vehicle, wind, “carbon storage”, etc., scams going (because the models are valid!)

Disgusting to see scientists behave like this.

Feynman (and a mighty chorus of others) must be yelling in his grave….. “THAT’S NOT SCIENCE!”

January 21, 2021 5:35 am

So if seasonal rain is delayed, the chance of drought increases. If it is delayed more the chance of drought increases more. Who would have thought?

Michael in Dublin
January 21, 2021 6:58 am

This is far too simplistic. Africa is continent of droughts and floods. There is plenty of water if it can be dammed during the wet years in time for the dry years. Parts of Africa, like Namibia, have had very good rainfall with dams overflowing and more rain on the way. Droughts and climate are not the problem. People who fail to plan ahead and properly maintain and manage their infrastructure are the real problem.

Reply to  Michael in Dublin
January 21, 2021 7:07 am

Sounds like California.

Reply to  RelPerm
January 21, 2021 9:28 pm

It’s been acknowledged as a misprint. The authors are blaming autospell. The more letters you get wrong at the start of the text…

Reply to  Michael in Dublin
January 21, 2021 9:36 am

In the arid USA south west the cultivation of crops by NativeAmericans was historically an engineering feat. Arable plots were laid out in a diamond shape & with a graded slope. One feature was of the apex having the lowest profile so that gravity naturally directed any rain to within what constituted the soil inside the diamond shaped plot.

Then, depending on the rains any one arable diamond plot received it was planted to the extent that tribal experience showed was appropriate.For example in low rainfall years only the lowest portion of a plot would be planted in order to take advantage of the water gravity had delivered & conserve seeds.

Arable plots were not necessarily contiguous with each other. Any cultivation sites in proximity to fresh water channels usually required less engineering than the diamond shaped plots.

Reply to  gringojay
January 21, 2021 9:57 am

edit: highest, not “lowest”

Michael in Dublin
Reply to  gringojay
January 21, 2021 12:41 pm


Years ago a friend was involved with promoting what was called “appropriate” or “intermediate” technology and first introduced me to the subject. What you describe is a good illustration. Can you please give a link for further reading?


Reply to  Michael in Dublin
January 21, 2021 1:44 pm

This is from my memory of 20 years ago & consequently I do not recall a source.

Reply to  Michael in Dublin
January 21, 2021 2:35 pm

Below is diamond field picture from W.E. Doolittle’s (2004) “The Safford Valley Grids: prehistoric cultivation in the southern Arizona Desert”; published by University of Arizona Press. I’ve also seen the arid technique where a plot was laid out described as a “wedge” – simply wide at higher level tapering to the lower point. Rock borders of various sizes conserved moisture underneath them.

January 21, 2021 7:38 am

so Less Rain = Drough ??? these folks are BRILLIANT … as Sam Kinison used to say … don’t send them food, send them luggage so they CAN MOVE … they live in a desert …

Peta of Newark
January 21, 2021 7:48 am

Just look at the map, any map, of where Sub-Saharan Africa is.
Slap bang between the 2 tropics with the Equator going right through the middle

It is The Land Of Our Birth.

It should be Dense Tropical Rainforest
Yet the only defining thing about it now is ‘tropical’

We trashed it.
We cut, burned, grazed, overgrazed, tilled and tilled and tilled until we made a desert out of it.

Unless a few people:

  • Wake up,
  • Stop passing the buck
  • Stop counting Dancing Angels
  • Quit admiring The Emperor and admit his undress
  • Grow A Pair
  • Grow Up in fact
  • Pack in all the arguments & childish tr0llery about the Modern Phlogiston
  • Actually read & understand what scientists of the last 200 years told us
  • Venture out into The Real World

and Admit As Much.
We will turn The Entire Planet into ‘Sub-Saharan Africa’ and, discover that deserts are Cold Places, no matter what thermometers say abut them.

That deserts are Hell and; we are all going there unless we acknowledge and act upon what’s in that list…

Last edited 1 month ago by Peta of Newark
Peta of Newark
Reply to  Peta of Newark
January 21, 2021 8:17 am

I put up a link the agnet just before, here is the source of it, again.

It is mentioned in there, that Soil Erosion is “A Natural Thing”

Yes it is.
As long as:
…. the metals that make up ‘rock’ have alkaline chemistry
….CO2 remains being acidic when dissolved in water
….some/any amount of (water) rain continues to fall,
Erosion will happen and continue to happen.

Regarding Ice Ages – who is alert and awake enough to see where I might go on that subject………….
In any case, I’d be repeating myself.

The Burning Question is thus:
Is tillage and agriculture that uses (highly corrosive/erosive) Nitrogen Fertilser = “natural”

Last edited 1 month ago by Peta of Newark
January 21, 2021 9:52 am

In order to fully understand the cycle of droughts in any one region, any study needs to look beyond the immediate area and include the conditions in the oceans that provide the sources of moisture. In the case of East Africa this is the Indian Ocean, and as in the Pacific Ocean where the El Nino/La Nina cycle is well studied, it is the Indian Ocean Dipole that should be central to any study of droughts, and that study should include how the varying ocean conditions can bring below average rains to the land bordering one side of the ocean whilst bringing above average rains to the land bordering other side. Whilst East African may be subject to a cycle of below and above average rains, so too will Australia but whilst the cycle may result in a lack of rain on one side of the ocean, the other side will enjoy above average rains, and in such circumstances in the case of Australia, this may result in above average rains carrying right across the continent to Victoria in the south east corner of Australia, in other words far reaching from the source of the moisture.
Just focusing on one particular region may document the cycle of varying rainfall in that region but does not explain what is the real driver of such cycles in the bigger picture.

January 21, 2021 12:59 pm

Speaking of drought Northern California is well below average for rainfall so far this fall/winter. This may last for another 1 to 2 years. …

%d bloggers like this: