Use alley cropping

Supporting evidence from individual studies

1. A replicated, controlled trial in 1993 and 1995 in West Yorkshire, UK (Naeem et al. 1997) found winter barley Hordeum vulgare and winter wheat Triticum sp. had lower grain aphid Sitobion avenae (pest) densities in alley cropped plots than in control plots without tree rows. In 1993, alley cropped wheat had fewer grain aphids (average 8-23 aphids/wheat ear) than controls (22-39 aphids) in three of four plots, and alley cropped barley had fewer aphids in all four plots. Alley cropped plots had a lower ratio of wingless to winged grain aphids than controls (2.5-6.8 wingless to 1 winged aphid in alley cropped plots; 2.0-4.9 wingless to 1 winged aphid in controls). Wind speed was lower in alley cropped plots than in controls without trees. Alley cropped plots were 14 m wide containing a 10 m-wide crop area separated by 2 m-wide tree rows. Controls had only boundary hedges and no tree rows. Tree rows (established 1988) contained ash Fraxinus excelsior, cherry Prunus spp., sycamore Acer pseudoplatanus and walnut Juglans regia, planted in sets of five and spaced 4 m apart. Hazel Corylus avellana bushes were planted between the trees. There were four replicates. Aphids were sampled by suction sampling and direct counts.

   Study and other actions tested

   Referenced paper

   Naeem M., Compton S.G., Incoll L.D., Wright C. & Corry D.T. (1997) Responses of aphids to a silvoarable agroforestry landscape. Agroforestry Forum, 8, 18-20.

2. A paired, replicated, controlled study in 1991-1994 in West Yorkshire, UK (Griffiths et al. 1998) found more slugs in alley-cropped plots (averaging 14.3-20.6 slugs/pitfall trap) than in controls without trees (0.2-10.4 slugs) in 1992-1994. More roundback slugs Arion spp. occurred in alley-cropped plots (0.2-8.3 slugs/refuge trap) than in controls (0.0-0.3 slugs) in 1994. An average of 1.7-8.8 grey field slugs Derocerus reticulatum/refuge trap occurred in alley-cropped plots compared to 2.0-4.0 in controls in 1994. Within alley cropped plots, 5-8 roundback slugs and 6-9 grey field slugs/refuge trap were found in tree rows compared with 0-3 and 2-9 in crop alleys, respectively. The proportion of plants damaged by slugs was higher, and the number of emerging pea Pisum sativum plants was lower, next to tree rows than elsewhere in crop alleys or in the controls. Four plots of alley-cropped arable crops (using a wheat Triticum aestivum-barley Hordeum vulgare-pea rotation) were compared with paired, conventionally cropped controls. Rows of trees (containing five tree species) were established in 1987 and spaced 14 m apart. In 1994, slugs were sampled using 16 refuge traps (40 x 40 cm squares of roofing felt) in two pairs of alley-cropped plots and controls.

   Study and other actions tested

   Referenced paper

   Griffiths J., Phillips D.S., Compton S.G., Wright C. & Incoll L.D. (1998) Responses of slug numbers and slug damage to crops in a silvoarable agroforestry landscape. Journal of Applied Ecology, 35, 252-260.

3. A randomised, replicated, controlled trial in 1992-1995 at two sites in Kenya (Ogol et al. 1998) found that alley cropping had mixed effects on parasitism and predation of the pest maize stem borer Chilo partellus in maize Zea mays plots. Egg predation was lower in alley cropped plots (approximately 24-31% eggs predated/plot) than controls without hedges (44%) over two cropping seasons, but egg parasitism was similar over three seasons. Parasitism of larvae and pupae was higher in alley cropped plots than controls (4-6.4% vs. 3.1% parasitism/plot) at one site, but the second site showed the reverse (1.2-9.8% vs. 17.5%). Stem borer larvae mortality was slightly higher in alley cropped plots (averaging approximately 18-24% larval mortality/plot) than controls (17.5%) at one site. Hedgerow spacing (width of alleys) had mixed effects. Green lacewing Chrysopa spp. (natural enemy) egg abundance was similar between treatments. White leadtree Leucaena leucocephala hedgerows were established in 1992 and were 1.5 m (two plots), 2.25 m (six plots) or 3 m (two plots) apart with one, two or three maize rows between hedges, respectively. One plot per site was maize-only. Hedges were cut before cropping and the cuttings were mulched. Plots were 18 x 12 m (five replicates) or 12 x 10 m (four replicates).

   Study and other actions tested

   Referenced paper

   Ogol C.K.P.O., Spence J.R. & Keddie A. (1998) Natural enemy abundance and activity in a maize-leucaena agroforestry system in Kenya. Environmental Entomology, 27, 1444-1451.

4. A randomised, replicated, controlled trial in 1995-1996 in Machakos, Kenya (Girma et al. 2000) found more wasps (Hymenoptera) (65 vs. 45) and spiders (Araneae) (96 vs. 71) but fewer ladybirds (Coccinellidae) (14 vs. 23) in alley cropped plots compared to conventional plots. Alley cropped plots had fewer aphids (Aphidoidea) (520 vs. 895 individuals). Maize Zea mays had lower aphid Rhophalosiphum maidis and stalk borer (maize stalk borer Busseola fusca and Chilo spp.) infestations in alley cropped than conventional plots (21% vs. 32% and 17% vs. 30% infestation, respectively). However, alley cropped beans Phaseolus vulgaris had higher beanfly Ophiomyia spp. infestation than conventional beans (35% vs. 25% plants infested). The proportion of aphid Aphis fabae infestations in beans was similar between plots (14% vs. 13%). The type of hedge species affected the abundance of some but not all pest and natural enemies studied. Hedges in alley cropped plots were planted 8 m upslope of crops in 1993 (using nine tree species) and pruned to 0.5 m. Beans were grown in the short- and maize in the long-rain season. Alley cropping was replicated 36 times and conventional cropping four times in 10 x 10 m plots. See also 'Plant new hedges'.

   Study and other actions tested

   Referenced paper

   Girma H., Rao M.R. & Sithanantham S. (2000) Insect pests and beneficial arthropod populations under different hedgerow intercropping systems in semiarid Kenya. Agroforestry Systems, 50, 279-292.

5. A randomised, replicated, controlled trial in 1987-1988 at two sites in Mindinao, Philippines (MacLean et al. 2003) found that the weight of grass and broadleaved weeds averaged 3.4-86.1 g/m² and 0.7-51.3 g/m², respectively, in alley cropped plots of rice Oryza sativa compared to 1.2-16.4 g/m² and 2.6-35.6 g/m² in conventional plots. Grass weight was greater in alley cropped plots than controls at a site with low soil fertility and high erosion. Alley cropped plots had 0.8-25.8 rice seedling maggot Atherigona oryzae eggs/m crop row while conventional plots had 0.8-13.6 eggs. White grubs (Scarabaeidae) appeared less abundant in alley cropped than conventional plots in 1987 (8.5-11.5 larvae/5 m crop row vs. 29.8 larvae at one site, 0.3-0.6 vs. 2.0 larvae at a second site) but numbers were similar between these treatments in 1988. Stem borer damage resulted in 1.7-9.5 deadhearts (dried central rice shoots)/m of row in alley cropped plots compared with 0.78-16.3 deadhearts in conventional plots. Rice stover and grain yields averaged 0.66-6.27 t/ha and 0.09-1.48 (respectively) in alley cropped plots compared with 2.41-3.17 t/ha and 0.23-1.15 in conventional plots. Rice was planted in alleys between gliricidia Gliricidia sepium and cassia Cassia spectabilis hedgerows. Hedgerows followed contour lines and were spaced 3-6 m apart. Twenty-four alley crop plots (across two 0.6 ha sites) were compared with two plots receiving conventional farmers’ practice. Alley crop plots (grouped in this summary) comprised mulched, green manured, mulched and green manured, or non-amended treatments.

   Study and other actions tested

   Referenced paper

   MacLean R.H., Litsinger J.A., Moody K., Watson A.K. & Libetario E.M. (2003) Impact of Gliricidia sepium and Cassia spectabilis hedgerows on weeds and insect pests of upland rice. Agriculture, Ecosystems & Environment, 94, 275-288.

6. A replicated, randomised, controlled trial in 1999-2000 in Mtwapa, Kenya (Midega et al. 2004) (same study as Midega et al. 2005) found no significant differences in the number of parasitised stem borer (Lepidoptera) eggs, egg predation rates, larval and pupal parasitism and mortality rates between alley cropped maize Zea mays plots (with hedgerows of leucaena Leucaena leucocephala and/or gliricidia Gliricidia sepium trees) and plots without hedgerows. Hedges were planted in 1999 in plots of 16 x 13 m with 3.2 m alleys between hedges. Treatments included two plots with leucaena hedges, one plot with gliricidia hedges, two plots with alternating hedges of leucaena and gliricidia and four controls without hedges. This experimental design was replicated four times. Leucaena and gliricidia were pruned to 0.3 m before cropping and pruned foliage was applied as mulch. Four rows of maize were planted in 1999 and 2000 between hedges. Ten maize plants with stem borer egg batches were marked and inspected three days later for parasitism. Healthy and partially eaten eggs were analysed to identify parasitoids. Each week, stalks of 10 maize plants/plot were dissected to locate stem borers, which were then raised in the laboratory and assessed for parasitoids and mortality.

   Study and other actions tested

   Referenced paper

   Midega C.A.O., Ogol C.K.P.O. & Overholt W.A. (2004) Effect of agroecosystem diversity on natural enemies of maize stemborers in coastal Kenya. International Journal of Tropical Insect Science, 24, 280-286.

7. A replicated, randomised, controlled trial in 1999-2000 in Mtwapa, Kenya (Midega et al. 2005) (same study as Midega et al. 2004) found the average mortality of spotted maize stem borer Chilo partellus was similar between alley cropped maize Zea mays plots (with hedgerows of leucaena Leucaena leucocephala and/or gliricidia Gliricidia sepium trees) and control plots without hedgerows (91-97% vs. 94% mortality of larvae). When maize was intercropped with cowpea Vigna unguiculata, alley cropped and control plots had similar stem borer mortality (93.9% vs. 95.7%). Parasitism was not the major cause of stem borer mortality. Hedges were planted in 1999 in plots of 16 x 13 m with 3.2 m alleys between hedges. Treatments included two plots with leucaena hedges, one plot with gliricidia hedges, two plots with alternating hedges of leucaena and gliricidia and four controls without hedges. Trees were cut and the foliage was applied as mulch. Four maize rows were planted between hedges (0.8 m between each row). Maize and cowpea were planted in alternating rows in intercropped plots. Treatments were replicated 4-8 times. Ten maize plants/plot were collected on each sampling date and stem borer eggs, larvae and pupae were examined in the laboratory for parasitoids.

   Study and other actions tested

   Referenced paper

   Midega C.A.O., Ogol C.K.P.O. & Overholt W.A. (2005) Life tables, key factor analysis and density relations of natural populations of the spotted maize stemborer, Chilo partellus (Lepidoptera: Crambidae), under different cropping systems at the Kenyan coast. International Journal of Tropical Insect Science, 25, 86-95.

8. A controlled trial in 2004-2005 in Montana, USA (Stamps et al. 2009) found lower survival in pest alfalfa weevils Hypera postica from alley cropped plots than from a control without alleys on two of four sampling dates (19-58% vs. 41-73% larvae survived). Survival was similar with wide (24.4 m-wide) and narrow (12.2 m) alleys. Weevil fungal infection rates were higher in alley-cropped than control plots on two of four dates and parasitism was higher in narrow alleys than in wide alley and control plots on two dates. Alfalfa Medicago sativa yields were lower in plots with wide and narrow alleys (6,431-6,771 and 4,102-5,106 kg/ha, respectively) than in the control (8,800-9,223 kg/ha). Estimated costs were US$254-293/acre for wide and US$250-282/acre for narrow alley crops, compared with US$290-302/acre in the control. Predicted profit was only US$7-26/acre for wide alley crops and losses of US$-82 to US$-60 for narrow alley crops, compared with US$88-150 gains for the control. Alfalfa was established in a 2.5 ha plot within a black walnut Juglans nigra plantation using two distances between tree rows. A 2.5 ha control was planted with alfalfa monoculture. Weevil larvae were collected and reared in a laboratory to assess survival, fungal infection and parasitism.

   Study and other actions tested

   Referenced paper

   Stamps W.T., McGraw R.L. & Godsey L. (2009) The ecology and economics of insect pest management in nut tree alley cropping systems in the Midwestern United States. Agriculture, Ecosystems & Environment, 131, 4-8.