This section focuses on manipulation of crop ecology, especially agronomy, to maximize returns on resources invested. Around the world, poor productivity of crops is often blamed on the genetic deterioration of seed as farmers save some of their crop for planting the next season rather than buying certified (or other quality-controlled) seed each year. This is true for groundnut in Malawi, where farmers’ yields remain low, averaging 600 kg/ha against yields more than 2,500 kg/ha on research stations. In 2017/18, the project evaluated the effect of seed generation (certified versus farmer-saved) and planting density (double rows versus single row) on groundnut biological nitrogen fixation and grain productivity. Certified seed had relatively larger pod yields; however, its productivity was reduced by poorer germination (compared with farmer-saved seed). This is likely the result of agro-dealers in Lilongwe using machinery for shelling; smallholders keep groundnut in the shell, and hand-shell close to planting time.

Across the three project regions of northern Ghana, six improved groundnut varieties were tested at four plant densities. Results were consistent between technology parks and farmers’ fields. The interaction of variety and plant density did not affect grain, fodder, or weed yields. However, groundnut varieties differed in grain and fodder yields, and plant density affected grain, fodder, and weed yields in all three regions. As plant spacing was increased, grain and fodder yields declined while weed biomass increased. The densest planting of groundnut (at 30 × 15 cm spacing) was preferred across the three regions because of its ability to suppress weeds, conserve soil moisture, decrease soil erosion (due to increased ground cover), and increase yield.

The maize–pigeon pea combination also contributes to nutrition security as maize provides carbohydrate and pigeon pea provides protein

In terms of the Sustainable Intensification Assessment Framework (SIAF), men and women farmers across the three regions scored the densest planting of groundnut higher than the traditional planting method for productivity, economic, environment, and human domains. However, the social domain score for dense planting was lower than that of the traditional method, possibly due to higher labor demands, especially at the time of establishment (Fig. A).

Combining sets of improved management practices (IMP) – typically including good-quality seed of improved varieties, healthy seedlings, and good agronomic practices – is a tried-and-tested route to improved productivity. The first season of IMP for vegetables in Karatu District, Tanzania (64 farmers) increased the yield of tomato by 48%, of African nightshade (Solanum villosum) by 30%, and of Ethiopian mustard (Brassica carinata) by 28%; and farmers’ respective incomes by 57%, 39%, and 40%. IMP also reduced postharvest losses by 86–98% for all three vegetables crops. Market participation increased by 14% for tomato, 36% for African nightshade, and 11% for Ethiopian mustard. Farmers said that IMP had a positive effect on productivity, profitability, and nutrition, but had less effect on the environment and social aspects of SIAF.

Among four cereal–legume intercrop and rotation schemes over three years in central Malawi, maize yield was greatest when it was monocropped after sole pigeon pea; however, groundnut–pigeon pea ‘doubled-up’ intercrop rotated with maize was the only one to perform, economically, as well as the farmer check. Sufficient economic and environmental returns are required to compensate for opportunity costs associated with maize production limitations due to small farm sizes.

In Mali, intercropping of soybean and sorghum increased sorghum growth and yield. As a monocrop, sorghum yielded 1.14 t/ha, but intercropped with soybean it was 2.325 t/ha. Meanwhile, application of organic manure increased cotton (Gossypium hirsutum) yield by 25% and biomass yield by 30%.

Growing-season drought in Babati District, Tanzania, suppressed any effects of novel intercropping technology mbili-mbili (no maize grain yield differences across treatments); however, early maturing common bean in the doubled-up legumes system outyielded that under mbili-mbili (0.5 t/ha compared with 0.3 t/ha).

During the first half of the grain growth period, maize variety Meru Hybrid 513, selected as best suited to the expected seasonal weather, had higher chlorophyll content than a similar system with ‘normal’ Meru Hybrid 515, suggesting that Meru Hybrid 513 has improved resistance to soil moisture stress, with consequent yield levels similar to those of sole maize. However, there may have been an effect of slow-release nitrogen (N) fertilizer, possibly increasing N availability after the onset of rain.

The amount of light intercepted by the maize canopy affects the proliferation of the understory legumes in intercrop. Photosynthetically active radiation (PAR) was recorded on pigeon pea after maize harvest. In most cases, doubled-up legumes had the highest light interception and best pigeon pea growth. This system had no maize planted, and common beans matured early, thus increasing light access for pigeon pea. The ability of pigeon pea to maximize use of PAR is associated with improved final yields and enhanced biological N fixation.

The maize–pigeon pea combination also contributes to nutrition security as maize provides carbohydrate and pigeon pea provides protein

Farmers evaluated legume–maize intercrop technologies based on sustainable intensification domains in Nsanama extension planning area, Malawi, where the program had introduced several technologies (fertilized maize, half fertilizer, no fertilizer, legume–maize rotations, doubled-up legumes technology, maize–pigeon pea intercrop, double-row planting of groundnut and soybean), via a structured questionnaire and focus group discussions.

The assessment focused on household food production, income generation, and labor requirements (Table 1). Nsanama has limited soil fertility, and household food security is a major concern. For household food production and income generation, most men farmers gave top ratings to fully fertilized maize, maize–pigeon pea intercrop, maize–groundnut intercrop, sole groundnut, sole pigeon pea, maize–groundnut rotation, and maize–pigeon pea rotation. Women gave similar ratings, except that none gave the top rating to sole pigeon pea, and they rated half-fertilized maize highly for income generation. Apart from yielding better, maize–pigeon pea intercrop contributes more pigeon-pea biomass, and maize and pigeon pea together make nsima (maize meal) and relish, a major meal in southern Malawi. The maize–pigeon pea combination also contributes to nutrition security as maize provides carbohydrate and pigeon pea provides protein. For input requirements, men rated unfertilized maize, maize–pigeon pea intercrop, sole groundnut, sole pigeon pea, maize–groundnut rotation, and maize–pigeon pea rotation highly; few women rated any of the options with the top score. On labor requirement, most men rated only sole pigeon pea as top, while more than half the women rated unfertilized maize as top.

Table 1. Men’s and women’s rating of sustainable intensification technologies by food security, income, and production input requirements

To make the best use of the biological N fixation of legumes, the legume may be cut at ground level (ratooned). Trials in Zambia revealed the highest maize yields were in full rotation with pigeon pea (no intercrop), and lowest in sole maize (no rotation) and intercrop uprooted at harvest (farmers’ practice). For pigeon pea productivity, ratooning two weeks after maize planting and again during maize harvest seemed to give best results. In terms of whole-system productivity, full rotation was again best, followed by maize with pigeon pea ratooned at harvest and three weeks after maize seeding. The lowest system yields were from sole pigeon pea and sole maize.