Conservation Farming Woes (aka Nothing's Easy)

It's been awhile; I've been in the field part of this week [more on that], at the plot, at the office, etc., etc. Have been on my feet all day except for yesterday, and it feels like the cartilage in my knees is down to the thickness of felt. Oh well ... they still worked well enough to dodge a runaway taxi this morning.

Anyway, as I mentioned some posts ago, the NGO I work with is promoting a method of farming known as "Conservation Farming" or "Conservation Agriculture". For those of you not steeped in agriculture, I'll give a brief summary:

1. "Conventional agriculture" is the ages-old practice of turning over the soil. The reasons for this is that a) the soil is loosened up for planting, and b) turning over the soil buries weeds. In Zambia, the turning over of the soil is accomplished either by hand (typical of the northern half of Zambia where tsetse flies are endemic), or by animal traction (e.g., oxen) pulling a single-share Ransome Victory plow [a post about the R.V. will follow in the future]. Anyway, the long and short is that you turn over the soil and plant right away; this gives your maize, sorghum, groundnuts, etc. time to get ahead of the weeds that will eventually resprout. You then weed the field once or twice and (hopefully) harvest.
2. Conservation agriculture is a possibly older practice (in some sense) that is coming back into vogue with recent concerns regarding soil erosion as a result of conventional agriculture. You see, when you turn over the soil surface and expose the bare soil to wind and rain, a surprising number of tons of soil are lost through erosion. This fact was starkly pointed out during the Dust Bowl years in the U.S., and through a number of publications, most notably "The Ploughman's Folly", written by Richard H. Faulkner in 1945. Conservation agriculture (hence, CF) puts emphasis on less disturbance of the soil profile, and therefore, less instances of soil erosion and the associated loss in soil fertility.

I think I'll post the manual online once I get back to my laptop (the Conservation Farming Unit (CFU) website is down these days), but basic tenants are:

• Digging planting basins or furrows that will be utilized year after year (preventing the need of turning over the entire soil surface).
• Applying lime (to reduce soil acidity) and fertilizer precisely (in basins or furrows) according to the crop being planted.
• Planting crops on time (for maize in Zambia, that means the first big rain after Nov. 15th).
• Weeding regularly to reduce the overall weed population.
• Retaining crop residues on the soil surface after harvesting in order to cover the soil during the long dry season.
• Annual rotation of crops (i.e. planting maize, then groundnuts, then cowpeas, etc.)

There is little doubt that CF is numerically a "better" system ... there are a number of glowing reports about how CF is sweeping Zambia. That may be the case given the amount of promotion of the practice being carried out by CFU, but my eyes and my gut tells me that actual, spontateous adoption (i.e. what Stephen Carr refers to as "osmosis between farmers") is slow. Though the technology isn't way out of reach, it is a technical system that needs to be diligently followed in order to for it to work; i.e., the chances for it to fail are quite high.

WARNING: I'm blithely skipping through tomes and years' worth of socio-agricultural research with the following statement: Simply presenting an agricultural methodology that is "better" in terms of lower ecological impact, higher agronomic returns, and so forth does not automatically guarantee adoption by farmers. Anyway, assuming people will adopt a system that is "better" disregards the reasons or history behind their existing systems and how those fit into broader socio-economic-political-cultural-ecological landscapes. When adopting a new system, farming families have to take into account literally hundreds of factors, particularly in the case of risk-averse, subsistence farmers. Put simply, what you and your ancestors have done since time out of mind is far more reliable and trustworthy than a brand-new system about which you have no clue.

That said, CF does work; when done correctly, yield gains, particularly in maize (the mainstay of Zambian agriculture and its staple food) can be huge, i.e. 300% increases. I've done it before in Kasempa during the drought season of 2004-05 and it worked alright, given the lack of rain and my lack of anything relating to farming knowledge; I did it again in successive years through 2007. What I've learned is that it takes a lot of trial-and-error to fine-tune the system; differences in soils, weed populations, rain, etc. mean that the same system doesn't work exactly the same in different places, such as Serenje and Kasempa. You desperately need to go through those iterations in order to have something you know will work when you roll it out to the farmers, because failure in the first season can stain the reputation of a methodology, making it less likely to be tried out by other farmers in successive seasons. Furthermore, to quote Roland Bunch, you have to "Start slowly and start small," working closely with just a few farmers.

So, with this convoluted portrait of CF, we come to our field visits to farmers earlier this week. Our agency set up six lead farmers as Field Extension Officers (FEOs) in spatially disparate villages throughout Senanga district; each lead farmer works with 10 farmers to monitor their CF plots. Earlier in the year, our agency set up the parameters for the plots, which are 1 lima (0.25 hectares or 2,500 sq. meters) in size and composed of maize (50%), cowpeas (25%), and groundnuts [peanuts] (25%). The FEOs were educated in CF techniques in Kaoma; later, they and all 60 farmers were given seeds, fertilizer, and lime.

On our monitoring visit, we found that for the most part, CF was having a very rocky start for a number of reasons:

1. Site selection: In two of the villages of the four villages I visited, the fields were quite distant from each other, making regular visitations and monitoring a difficult task, especially given that the FEOs work on foot.
2. Soil variability: Senanga is divided into two very distinct soil/water regimes: a) the "East Bank" [of the Zambezi River], which features a drift of Kalahari sand that occupies much of Western Province, b) the Barotse Floodplains [Upper Zambezi Floodplains] which features a significant amount of silt deposition from annual flooding. What we found was that on the sandy uplands, leaching was a significant issue; even in cases where the farmers had applied fertilizer as recommended and planted early, the crops showed evidence of nutrient deprivation (particularly phosphorous). Conversely, in the floodplain, nutrient loss wasn't as big an issue, but the constant moisture and rich soil meant the weed pressure was much greater; in roughly half the plots, it was difficult to discern where the crops were amongst the weeds.
3. Plant population and spacing: It appeared that most of the basins had been planted with half or less the recommended number of seeds. This was not surprising, as the traditional way of planting is to simply drop a pinch of seeds in a hole or furrow; also, I've found over the years Zambians prefer to have plenty of space for their plants to grow in the belief that bigger cobs or beans will result. Also, Zambians tend to be sparing of seeds; putting four maize seeds or eight groundnuts may seem rather profligate.
4. Weeding: Though the floodplains do favor faster weed growth, even in upland fields we noted that weeding was being performed on the traditional schedule (twice); this means weeds are allowed to grow in thickly prior to being chopped back, which leads to nutrient and moisture competition with the crops. From my experience, the problem of weed interference is considered by farmers as one of shading moreso than anything else; also, weeding is hard labor that I think many Zambian farmers put off until it is absolutely necessary. Another point is that many of these weeds are the Stygian, nuclear-powered quack grass I've encountered before; I now think it was the inspiration for the many-headed Hydra in the ways that it just keeps growing back.
5. Pests: Spring hares (linyuka), moles (lipeba), and grasshoppers have in turn hammered the maize, groundnuts, and cowpeas in many places. Though this is not directly related to the CF practices, it does affect crop performance.
6. Monitoring: This is the hardest one to point out, but our lead farmers and their charges are all doing CF for the first time. No one really has any institutional knowledge regarding the practice, and therefore has no basis from which to conduct monitoring. Therefore, many of our farmers are committing sins of omission rather than sins of commission; they don't really know what they're doing right or wrong.

This is not to say that CF is foundering; a number of farmers, particularly widows, pointed out merits to the system, esp. the fact that their crops were ahead of many other farmers and that they didn't have to wait to rent plows for their field. Also, a few fields with good conditions and management were doing quite well.

However, it was a good lesson in agricultural development which bears enumerating:

1. Sites need to be spatially close together so that monitoring can be done quickly and frequently. Also, fewer sites should be selected so that experts can also easily monitor and check in with lead farmers to ensure basin spacing, fertilizer application, planting, and weeding are accomplished on time and correctly.
2. Work on practical measures to counteract the tendency of nutrient leaching in sandy soils needs to be carried out. Similarly, work on alternative methods of weed suppression, particularly for the ubiquitous couch [Elytrigia repens] and kapinga grasses, needs to be carried out for farmers working in the floodplains.
3. Work on practical measures to keep pests off the crops, i.e. living barriers, sprays, noise makers, etc.
4. Last, but not least, we (as ag developers) need to be invested in ensuring what we are promoting can and will work for farmers. That means constant monitoring, test plots, sharing experiences with farmers, etc. ... i.e. significant time and work.

Erghh ... I've got tired fingers, enough of the Two Ears of Corn for today.

Ah, the waning days of 2010 ... here's to a better, more prosperous, well-fed world in 2011.