To make science and technology work for small farmers in sub-Saharan Africa I think it is imperative that account be taken of within-field soil variability.  This is especially true in semi-arid areas, but also in other parts of Africa.

The farmers themselves take variability into account: they often manage different parts of a field in a different manner.  They do this because it is more efficient, and because it reduces production risks. Technology developments and technology transfers that do not connect with this site-specific management by farmers, and with the underlying reasons, risk being ignored or turned down.  Farmer, extension and scientific knowledge about short-distance soil and crop growth variability must be combined for

– improved knowledge, understanding and communication by and for all parties concerned

– better design and analysis of agronomic experiments

– more relevant extension approaches

– better management options for the farmers.It is not difficult to achieve this.

It just needs a change in attitude from seeing variability as a problem to seeing it as an asset and an opportunity for subsistence farmers.  

One simple step is to use an easily assessed environmental parameter as a co-variable in the analysis of agronomic experiments.  A well-chosen parameter will explain a signifcant part of the spatial (and temporal) variability in experimental results.  That same parameter can then be used to extend the experimental results to farmers.  Simple parameters that a farmer can also use include microtopography (high spots and low spots in a field); degree of crusting of the soil; distance from certain termite mounds or from certain trees; wet parts an dry parts of a field; and the presence and absence of certain weeds.

A 12-page brochure on a variability project at ICRISAT Sahelian Center in Niger, that aimed to do just this, was financed by IUCN’s Commission on Ecosystem Management and is available from http://cmsdata.iucn.org/downloads/cem_csd_16_brochure_sahel_hq.pdf (2.6 Mb).  Its pictures tell the story and a number of references are included.  Its main conclusions are given below.  I will be happy to discuss this further with whomever may be interested.

The following practical results from the variability research in Niger can already be extended to farmers and/or researchers:·

• with-in field variability can play a yield stabilising role; this is especially important in times of uncertainty caused by climate change·
• over-manuring by many farmers in the Sahel can be reduced by spreading the available manure over a three times larger area;
• different aeolian sand deposits, often found side by side in a single field, should be managed differently;·
• nutrient use efficiency by farmers can be increased through microtopography-related site-specific management, such as not applying cattle manure and urine in wet and/or crusted areas;         this is important for the management of local fertility resources (manure, urine, compost, crop residues, domestic refuse), as well as for the management of mineral fertilisers: what can happen to cattle manure can happen to N, K and sometimes also P fertilisers;·
• with-in field rainfall infiltration may be increased by applying lime or gypsum, especially on older aeolian deposits that crust more easily;
• Macrotermes termites play an important role in local increases in soil fertility on sandy soils;
• Faidherbia albida seedlings in agro-forestry projects are much more likely to grow well, and survive to adulthood and full utility, if they are planted near an old Macrotermes mound.
• short-distance soil- and crop growth variability can be better incorporated into agricultural research through the use of covariables and other statistical techniques.