The Man Who Bred Bread

It is almost 2 AM and we are at the rooftop bar of the Trademark Hotel, off Limuru Road.

After a hectic week at the 2018 CGIAR Big Data in Agriculture Platform Convention, the after party was a welcome relief. I was seated on a comfy blue couch, overlooking a calm infinity pool.

Beyond it, the Nairobi skyline shone in all its glory.

Courtesy: Mutua Matheka

A little tipsy, I was just seated there smiling sheepishly at the dark sky and brilliant lights. Just as I’m about to slip into a stupor from all the free-flowing drinks, she stands in front of me. There’s just something about a charming young woman that sobers one up.

“Do you know that we have a wheat shortage to thank for all this?”

She asks, while pointing at the beautiful cityscape. She’s wearing a yellow kitenge dress with silhouetted motifs and a fishtail curvature that sensually grabs her figure. I’m still not listening as she repeats the statement.

I am focusing more on her cute, tiny eyes. They are what drew me to her at the gala dinner a couple of hours ago. I want to say something, but the stupor returns, I fall back on the couch.

My father always warned me and my siblings against alcohol. Now I know why.

She walks away. As she does so, she shouted, in her mellow voice:

“Go and find out who bred bread!”

I woke up the next day tired, recovering from a hangover I should never have had in the first place. Its around midday and I am at my place, with no memory of how I even got there. My near-empty stomach is crying, wailing, for help.

Quickly, I jump out of bed, stumble a little, but find my balance and head to the kitchen.

The moment I saw that a loaf of bread on the kitchen table, all the memories came trooping back. The hunger somehow disappears and I rush towards my computer. Her last statement made no sense given that bread cannot be bred. I wanted to find out what she really meant.

A couple of clicks later I came across a name: Norman Borlaug.

It was surprised that I had never heard the name before. This is despite the significant role the American agronomist has played in plant breeding. It is impossible to speak about the science of it without mentioning his name.

I was even more excited given that the blog series I am currently on is entirely about breeding. Opening several other web pages revealed a ton of information about the Nobel Laureate.

Through what is often termed as the Green Revolution, he was able to save an estimated 1 billion lives. That is not an easy feat, and it takes a special kind of person to pull it off. That got me even more psyched up and I went to YouTube to view a couple of documentaries on him.

So, how did he do it?

Borlaug on a field visit.

Wheat, alongside maize and rice, are the three most important food crops. They serve as the staple foods all over the world and a shortage in any one of them is of great concern. The East African famine of 2011 was due to maize shortage. The Great Chinese Famine of 1959-61 was due to a severe shortage of rice.

In the 1940s, Mexico was heading towards the same direction. The wheat production was extremely low and farmers could not manage to produce enough to even feed their own families. A combination of factors including fungal diseases and land degradation led to low produce that left the country in a precarious situation.

He moved to Mexico in 1944 and took charge of a wheat improvement program jointly funded by the Mexican Government and the Rockefeller Foundation. Dubbed the Cooperative Wheat Research and Production Program, it served as a precursor to the International Maize and Wheat Improvement Center (CIMMYT).

Borlaug quickly set to work and created a team that was focused on achieving the three primary objectives of breeding: Disease resistance, robustness, and high productivity. In under five years, they were able to breed a variety of wheat that met all these characteristics.

Furthermore, they further refined it by breeding for semi-dwarfism which made the stalks stronger and hence capable of holding more produce. This lower height also made it suitable for mechanization of wheat production. Within some 20 years, Mexico became self-sufficient in wheat production, and soon began exporting.

He replicated this success in Pakistan and India which were also facing similar situations. Most of the wheat that is farmed commercially these days bears characteristics that are the result of his work. Production of wheat products, therefore, still heavily relies on his pioneering work.

This means that, in a way, he is the man who bred bread as we know it.

Alongside the International Rice Research Institute (IRRI), CIMMYT became among the first agricultural research centers brought under the umbrella of CGIAR (Consultative Group for International Agricultural Research). The event that she and I met in was one organized by CGIAR.

Turns out she was right about this, too. We did have a wheat shortage to thank for that beautiful night.

Crop Breeding: From Svalbard to Zuhura

“The year 2128 was a terrible one”.

The robot seemed to cry as it said those words.

We were pacing all over one of Zuhura’s numerous decks as we inspected the rows and rows and rows of rice plants. The white basins on which they grew were filled with a dense bluish mist that nourished them. Piled up to 300 meters into the air, production was in full force.

As she spoke, Ms. Robot and I were being lifted on a platform that would allow us to see the crops on the upper decks. She glided along the rows, took mental notes and reported that the crops were doing fine. I agreed, and we moved on.

The platform lowered us and we were soon moving on solid ground again.

The snake-like, 5 foot robot led the way as I followed a short distance behind. For such a long machine, it was slender given that it did not measure more than 2 inches at its widest point. Its cylindrical body was filled with all sorts of sensors, making her the perfect assistant during crop inspection runs.

We were soon at the orchard.

This was one of my favorite spots, it reminded me of the world we lost. Of all the planting decks, it had the what appeared to be the closest resemblance to the vineyards of centuries past. The mangoes, apples, cherries, papayas, loquats, berries, avocadoes, and many many more.

I stood at one end and beamed the inspection gun. Its blue-yellow light filled the deck as it gathered data on each fruit. In the meantime, Ms. Robot moved from one tree to another at lightning speed. When she came back, her results matched mine.

They were all doing fine, too.

Our inspections done for the day, I bid Ms. Robot goodbye and headed to the observatory. It was raining, and the pink, brown and colorless drops of sulfuric acid covered that thick windows that protected the deck from the chaos that is Venus’ atmosphere.

Zuhura is one of the two dozen farm-cities that float some 50 kilometers above the surface of Venus. They are part of hundreds of similar but much larger cities akin to the space crafts I dreamt about all those years ago. Humans moved here after the catastrophe.

Zuhura, as it stands today.

 

Sitting at the observatory reminded me of that eventful day. Even after five decades since the doom, I remember it clearly, as if it just happened yesterday. It was in February, on Friday the 13th, the year 2128.

An apprehensive day, it was.

Unlike previous in years when most people would have been preparing for Valentines Day, evacuations to the Antarctic and outer space were in had already ended. Not all people could be carried on the space arks and therefore priority was given to those with expertise in needed areas.

The frozen continent was thought to be the next best alternative. That was a tough and controversial choice, I often wonder whether the situation could have been handled differently.

There was a huge risk of a meteor colliding with the planet.

“Asteroids and meteors are just jealous of Earth because they didn’t become planets!”, were the last words Ivar, one of my colleagues in the 1000-strong team said before we were asked to do the necessary. Some in the multidisciplinary team laughed, nervously.

There were fears that should the meteor hit the Earth, it could cause mass extinction. We had been preparing to evacuate the Global Seed Bank at Svalbard for years since the first reports of the risk were announced. The bank was made for Dooms day, and here it was.

The billions of seeds in its valuts had to be evacuated to the moon.

The evacuation was done secretly for several weeks eight months before the collision happened. We separated the vaults into appropriate sets, and placed them in massive shipping containers. Each one was labelled with the logo and of the Crop Trust.

The Norwegian cold stood in sharp contrast with the heat that would soon sweep through the world. That heat killed more than 8.5 billion people. Just slightly over a billion people safely made it to the Venus-bound floating cities like Zuhura and the Mars colony.

The last memories I have of Earth are of what happened on that Friday the 13th. The telescopes orbiting the planet beamed us the images on our moon station. The impact was, ironically, near the Cape of Good Hope. A huge blinding light was seen followed by a loud bang that could be heard all over the galaxy!

Even at thousands of kilometers away, we sweated, profusely the extreme heat that scorched the Earth. We trembled upon seeing the violent winds swept up sky-crappers like leaves in a storm. Tsunamis that were a hundred times the height of Mount Everest wrecked havoc. We wept. All of us lost family and friends in the doom.

Earth died a violent death.

The collision, as it happened.

Before that, the billions of seeds were shipped to the Kennedy Space Center on the Florida coast. Our team then split into twenty groups of 50 each and were airlifted to several rocket launch sites around the world.

A few days later we were blasted into the lunar surface to set up facilities for preserving and breeding the seeds. Since then, moon has acted as the new Svalbard and all space farms rely on it for their seeds.

The extensive diversity in that gene pool has been vital to humanity’s survival in this new environment.

Even in the face of such a catastrophe, crop breeding still exerts its importance. That is why in the next couple of weeks I’ll share with you stories on:

1. The History of crop breeding
2. The art & craft of crop breeding
3. The science of crop breeding
4. The economics of crop breeding

Keep breeding, and yeah, pun intended!

The Svalbard Global Seed Bank, as it once stood.

Artificial Intelligence & Machine Learning in Agriculture

Over the past couple of weeks I have largely been talking about data and its usage in agriculture. Data on its own, however, is useless. What makes it useful and in this case for agricultural decision-making, is how it is managed. That is why Artificial Intelligence (AI) and Machine Learning (ML) are such essential technologies.

What is AI and ML?

The term intelligence simply refers to the ability to gather and apply knowledge and skills. On the other hand, learning is described as the acquisition of knowledge or skills either through study, experience, or by being taught.

Intelligence and learning are, naturally, animalistic attributes, but have in the past few decades become synonymous with computers too. Therefore, when referring to computerized systems, they are known as artificial intelligence and machine learning, respectively.

How do They Work?

Riding a bicycle is one of the easiest skills anyone can possess. It simply involves handling the bars in the right manner, pedaling in a certain pattern, and knowing when to brake or not. For this to happen, your brain needs to gather data from all these points, process it, and simultaneously give feedback to your muscles.

This is done in a split second and ensures that one remains stable while heading in the right direction. Over time, one can be able to cycle in tougher terrains and those who are intent on it get to learn various stunts.

When done continuously and consistently enough, it becomes almost instinctive as muscle memory allows one to cycle or do stunts without even thinking about it.

AI and ML systems work in the same way. Most of the systems currently in place are designed to perform a narrow range of operations that traditionally would require human intellect.

An AI system gathers and processes data about a certain function, allowing it to make decision-making about something. That’s how the face-recognition feature on your phone works and is able to know that it is you and not an intruder.

If, as you probably do, use Google, it is highly likely that the pop-ups you see are a reflection of your interests. The search engine has a powerful AI system that gathers data about your internet searches and the websites you visit.

This system, over time, and by analyzing this data learns certain patterns about your interests and makes appropriate inferences. That is ML in action with the objective of driving decision-making. These conclusions are what determine the adverts you see.

How Are They Useful in Agriculture?

Last week I wrote about the internet of things. In the post, I described how the system we named UDAS collected data on soil conditions and sent it to a database on the internet. This is where AI sets in motion to analyze the data and determine what decision to make. It is the same way your brain picks signals from the eyes and tells your arms to shift the handle bar so that you don’t fall into a ditch.

I’ll use the example of UDAS to further explain this. Once the system gets this data, it runs a series of algorithms on it to identify the prevailing soil humidity. This varies from one type of soil to another. The system was also fed with information on the water requirements for the crops we had on the miniature farm and the soil types therein.

With these sets of data at hand, the AI system can determine whether the moisture within the soil is able to sustain the plant or not. If not, it takes the necessary action by triggering irrigation.

Furthermore, the infiltration rates vary from one soil to another and hence the systems decides on the volume of water that is required and the frequency of irrigation.

As I also mentioned in the previous posts, a system such as UDAS heavily relies on a series of sensors that collect large volumes of data.

The system is therefore provided with numerous Big Data points that it analyzes continuously.Over time, the machine learns certain patterns such as the average water consumption per plant, and the water requirements at a certain time of the day.

Such predictive analytics is an illustration of how ML works. This makes it able to make better inferences in future and hence improve decision-making. It’s just like learning how to do stunts on a bicycle.

Some Examples of the Current Applications in Agriculture

AI and ML are presently used for the following applications in agriculture:

1. Diagnosing deficiencies and defects in the soil
2. Predicting weather patterns
3. Automating irrigation and fumigation systems
4. Recommending precise times for sowing
5. Optimizing weed control.