Tuesday, June 25, 2019

Africa always wins

Belladonna's friend from Ghana made a statement something to the effect that "Western science does not work in Africa." The context was using science to improve crop yields.

Like all broad generalizations, there are elements of truth to the statement as well as some important caveats. In general, Bella's friend was mostly right.

The literature pretty clearly supports the contention that ag experts from temperate regions cannot go to tropical farms and use the same boiler-plate soil chemistry targets.

For example, in temperate regions most people target a pH of 6.5. pH is a measure of soil acidity/alkalinity. 6.5 is just slightly acid of neutral.
The southwest corner of Ghana gets a lot of rain. 1300mm is over four feet of rain.

The challenge in high rainfall, tropical areas is that most of the soil is a gritty clay with lots of kaolin. In temperate regions the clay is greasy with large percentages of bentonite and montmorillonites.

The tropical kaolins are highly leached and acid. The nutrients that are not leached away are either bound up in plants or "locked" in complex, insoluble compounds with aluminum and phosphate.

Dumping enough limestone on the soil to jack it all the way to a pH of 6.5 results in releasing toxic levels of those elements, like manganese, that are locked up.

The same southwest corner that gets dumped on with lots of rain is deficient in calcium. No surprise there.

The "smart money" does not recommend wholesale application of limestone on acid, tropical soils. It recommends an incremental, feel-your-way along approach where limestone is treated like a nutrient and relatively tiny amounts are added and the results monitored.

Variable versus attribute data
Another major impediment is that most African farmers don't believe in higher math like division.

In the United States farmers actively seek incremental yield/incremental input data. They want to know, "How much nitrogen must I add if I am planning for 200 bushel/acre harvest?" They understand that each bushel of corn removes a set amount of nitrogen, about 0.7 pounds. Some more is lost to leaching and vaporizes. Some more is tied up in the leaves and stalks...so most farmers expect to apply one pound of nitrogen for every sixty pounds of corn they expect to harvest.

A pound of nitrogen costs $0.30 (spot price anhydrous ammonia, by the ton) while the spot price for corn at the time of this writing is $4.60 a bushel (almost 60 pounds). That is like a machine that you feed one dollar bills into and it spits out a fifteen perfectly legal dollar bills back out. Why would you not buy fertilizer?

Lack of trust. How does one know that the pale powder that looks like dirt is what the seller says it is?

This is a two-factor experiment. For example, the red shaded area might be ten pounds of limestone per tree and white area a quarter pound of potassium. Or, red might be five pounds of calcium and the white ten pounds of calcium. There is also a freebie hiding in the setup. If care is taken and the trees beside the road are harvested separately the scientist can also evaluate the effect of lower density planting.
Even if an enterprising agent were to set up field trials most farmers would struggle to assimilate the information. Why should they pay cash money out-of-pocket...which is very difficult to come by...when they get some harvest for free. Any finite number divided by zero is infinite. It is hard to be more cost effective than that. The fundamental watershed in thinking is to divide incremental yield by incremental input.

This "divide by zero input" thinking was common in the United States in the 1920s. "Why should I pay money to buy Hereford calves and pay money to feed them five pounds of grain a day? I can raise the native cattle on grass alone to market weight."

Well, Sodbuster, it took you seven years to raise that sway-backed, cat-hammed steer to market weight on unimproved pasture. Improved cattle plus a little bit of grain means you can take that calf to market in 20 months, not 84 months. That is four times as many gross receipts in your pocket

One intermediate step is to put the nuts harvested into each tree into its own, individual basket. It helps if the baskets are color coded. Then, weigh the baskets and rank order them from heaviest yield to lightest yield. It should be clear if most of the heavy end is one color and most of the light end is a different one. This is a variation of the Tukey End Count test.

It is beyond dispute that fertilizer application levels are suboptimal in Africa. Some studies indicate that yields could be tripled if recommended levels were applied.

Africa and other tropical areas are rapidly being deforested to plant oil palms, even as they import edible oils. Applying fertilizer seems far more benign than leveling the remnants of tropical rain forests.

The tribal thing
The other issue is that little trust exists between tribes. Since fertilizer looks like dirt, who is to say that the rascal from the other tribe didn't divert most of the fertilizer to his cousin, mix in dirt and made you pay full price?

It happens.

Consequently, you buy only from your uncle or cousin or a person from your own tribe. There are few economies of scale to be had, especially when you layer on every level where mordita occurs. It shouldn't be hard to find limestone in Ghana, for instance. Deposits occur west of Accura, Oterkpolu and Jama. For all practical purposes there is no part of southwestern Ghana that is more than 160 miles from a commercially viable limestone/dolomite deposit.

Money funny business
Ghana, like many other third world countries, got squeezed by the world banking organization. They were put on an austerity diet and stopped subsidizing fertilizer purchases. Crop yields dropped. Exports dropped at the same time prices tanked. The currency traded at 2000 to the USD before austerity and 7000 to the USD afterward.

The issue of bringing rigorous science to Africa is not a problem that can be attacked in isolation. In the end, Africa wins.

Post Script on tribalism: Tribalism is like the system of water-tight compartments on a warship. While it incurs costs it also firewalls epidemics. It may be that Africa's obdurate tribalism is the most viable strategy for dealing with certain diseases. It will never minimize the number who die but it guarantees that a sustainable population will survive somewhere.

Under the conditions of Africa, virulent tribalism may be the only rational system of organization.

1 comment:

  1. Yeah, but Tribalism exists elsewhere, and isn't the biggest issue preventing success elsewhere....there is another actor....maybe cultural, maybe genetic, maybe something else. I doubt if it is simply melanin either..., but the issues are more than (although probably include) Tribalim.


Readers who are willing to comment make this a better blog. Civil dialog is a valuable thing.