ILEIA Newsletter Vol. 11 No. 3 p. 20

Weeds and trees

Rik Thijssen

From his experience in agroforestry development in Kenya, the author looks at benefits farmers can have from weeds and the positive effects of trees on weed growth.

Weeds in crop fields cannot be automatically judged damaging and in need of immediate control. At low density, weeds do not usually affect yields, and certain weeds can even stimulate crop growth. For example, in the rainfed areas of the Kenyan arid zone, some broad leaved and leguminous weeds increased the growth and yield of sorghum and millets. While other wild plants showed a promise in the control of insects and nematodes, especially on vegetables (Thijssen, 1991a). Before stressing the importance of weed control, it should be made clear whether or not a particular "weed" is harmful to a specific crop in a given area.

The best way to dispose of a weed is to make use of it, to promote it to the level of wanted plants. In fact, studies conducted in traditional farming systems in the high potential areas of the highlands in Kenya show that local farmers do not completely clear all weeds. This "relaxed" weeding might be mistaken for the consequence of labour shortage or low return for the extra work. However, a closer look at farmer attitudes towards weeds reveals that certain weeds are managed and even encouraged if they serve a useful purpose.

Useful weeds

Weeds are often used as fodder, especially in the dry season. Almost any weed in a field after harvest serves as food for animals. During this period, some weeds are more digestible than cultivated forages, have a higher crude protein content and are more widely available (Nuwanyakpa et al. 1983).

Many wild annuals found in cropland (Amaranthus, Brassica, Commelina, etc) can be valuable vegetables, especially during critical periods before the planted crops mature. It is of paramount importance to collect indigenous knowledge on edible species and to bring this information to other areas. The same is true for medicinal uses of weeds.

For example, the common knowledge in areas of western Kenya that Cassia didymobotrya has valuable properties for treatment of animal skin and intestinal problems has sofar not been shared with people in central Kenya where it is considered to be just a shrub that smells bad. Interestingly, Kokwaro (1976) indicates medicinal uses for most of the common weeds found in Kenya.

It has been reported that a weed "crop" in between cropping seasons can keep mycorrhizal fungi alive resulting in a higher availability of phosphorus (P) and zinc (Zn) for subsequent cropping (Thompson, 1991). More recently, experiments in western Kenya show enormous potential for some common weeds to increase soil fertility in cropland. Fast growing weeds that produce a lot of biomass are collected both on- and off-farm and used as green manure. Compared to commercial fertilisers, application of high amounts of Tithonia or Lantana green manure had very good results on growth and yield of maize. An additional benefit of the green manure was a significant reduction of the presence of the parasitic weed Striga (Table 1).

Apart from experimenting with weeds for soil fertility improvement, the Kenya Woodfuel and Agroforestry Programme (KWAP) has gained further experience on the profitable use of weeds. In the drier parts of Busia, in western Kenya, termite attacks on young tree seedlings are controlled using ashes of certain weeds. Close collaboration with farmers has learned that some wild plant species such as Tephrosia can effectively keep moles away from the cropland. Farmers leave and protect some "weeds" in the farm in order to harvest them later as firewood.

And in some areas of Migori, western Kenya, leaves of certain weeds are becoming more popular for smoke curing of tobacco. This could reduce the pressure on some valuable tree species considerably. Other important uses of wild plants on the farms include fibre from the bark of Triumfetta species for binding tomato plants to stakes and several flowering species as a source of nectar for honey producing bees.

Shade

In the control of weeds, trees and shrubs can have some direct and indirect effects. The canopy of trees reduces the amount of light reaching weeds on the ground. Generally, this results in a dramatic change in weed flora. The effect of Albizzia or Ficus shade trees in coffee plantations in Uganda is a typical example (Hoekstra et al. 1991). The combination of less light for germination and growth of weeds and mulch in the form of fallen leaves from coffee and trees reduced necessary labour inputs to a minimum.

In an intercropping system involving Gliricidia, Leucaena and Cassia species and maize in Machakos, Kenya, it was found that the weed flora in intercropped plots changed from grasses to broadleaved weeds, which were easier to control, and that reduction in weed biomass was a result of fast canopy closure by the trees (Thijssen, 1991a).

However, there is a need to properly manage the tree species for the purpose of providing shade at the right time. During the cropping season trees are lopped to provide mulch for weed control while in between seasons, trees are allowed to produce foliage to out shade weeds.

Mulch

The effectiveness of weed control through mulch varies considerably with the species used (Thijssen, 1991b). Generally, tree species with broad leaves that do not decompose fast are more suitable for this use. Good examples are Inga species from South and Central America, Cordia africana and Ficus species in East Africa (Thijssen et al. 1993), and Flemingia macrophylla from Asia (Budelman, 1988). Rows of Flemingia planted in maize in Machakos and in coffee in Embu, both Kenya, produced enough mulch every season to fully cover the soil surface of almost 1.5 m² per metre of alley.

At the begin of the next rains, about five months later, mulch coverage was still 42% in maize and 26% in coffee. One month after applying the mulch, weed biomass was, compared to fields without mulch, 78% and 92% lower for maize and coffee respectively.

Fallows

Fallow systems normally occur where crop production becomes more difficult because of problems with weeds, pests and diseases or declining soil fertility. A piece of land is left unused for a short or longer period until crop production is possible again. Improved fallows, with planted (often leguminous) trees, shrubs, and/or perennial herbs have shown potential for restoring soil fertility in a short time period. As an artificial copy of natural fallow fast growing species that produce a lot of biomass are allowed to grow for a few seasons on the fallow plot.

Before planting a crop again, the fallow bush is cleared and all green biomass is worked into the soil. Important agricultural weeds such as alang-alang (Imperata cylindrica) and other noxious grass weeds have been removed successfully from farmers fields using improved fallows of, for instance, Leucaena leucocephala. In western Kenya, KWAP found that after a two year fallow with the indigenous Sesbania sesban crop production could be up to 300% higher than before the fallow while the weed population was very low in the first season.

Allelopathy

This phenomenon has been defined as the chemical warfare between plants. Extracts or leachates from various plant parts of different species have shown to inhibit seed germination or growth of other plants. An exciting new finding by KWAP in western Kenya is that Calliandra calothyrsus, a popular agroforestry species for animal fodder and firewood production, can considerably reduce infestation of maize fields with the parasitic weed Striga. A pot experiment with Calliandra green manure showed a reduction in the presence of the, for this area, most important agricultural pest with almost 70% (Table 2). Field observations of this species interplanted with maize confirmed this finding.

Another allelopathic effect on Striga is caused by the indigenous species Croton megalocarpus. The presence of trees of this species can trigger germination of Striga seeds but it is not parasitised and in absence of a host the young Striga shoots will die.

Indirect effects

Especially in hot dry areas, the presence of shade from trees can stimulate people to spend the necessary time on weeding. Another observation was that where certain tree species become important in the farming system, the value of seedlings should not be underestimated.

In the Embu region of central Kenya, farmers are often prepared to hire labour for weeding because of an extra return they can get in the form of wildlings. Labourers are told to carefully uproot every wild seedling of certain tree species they find. The young trees are then planted by the farmer in a more appropriate part of the farm or in small containers to be sold at a later stage.

Tree-crop-weed interactions are extremely site specific and vary according to the plant species, environmental factors and management practices, which in turn vary according to the cultural traditions, social organisation and economic forces that influence agricultural systems. For this reason, weed ecology and management studies need a gender sensitive farming systems perspective. Weed management has to be more than removing undesirable vegetation because, although weeds do cause direct yield losses, they can have desirable effects as well.

Rik Thijssen, KWAP, PO Box 9050, Eldoret, Kenya.

References
– Budelman A. 1988. The performance of the leaf mulches of Leucaena leucocephala, Flemingia macrophylla, and Gliricidia sepium in weed control. In: Agroforestry Systems 6.
– Hoekstra D, Aluma J, Thijssen R. A diagnostic survey of trees in the banana based lake-shore system in Uganda. AFRENA reports, ICRAF, Nairobi.
– Kokwaro JO. 1976. Medicinal plants of East Africa. East African Literature Bureau, Nairobi.
– Nuwanyakpa MY, Bolsen KK. 1983. Nutritive value of seven tropical weed species during the dry season. In: Agronomy Journal Vol. 75.
– Thijssen R. 1991a. Alleycropping research in Kakuyuni, Kenya: an agronomist’s perspective. ICRAF, Nairobi.
– Thijssen R. 1991b. Agroforestry and soil organic matter; about green manure, mulch, decomposition and nutrient release. ICRAF, Nairobi.
– Thompson JP. 1991. Improving the mycorrhizal condition of the soil through cultural practices and effects on growth and phosphorus uptake by plants. In: Johansen C. et al (eds). Phosphorus nutrition of grain legumes in the semi-arid tropics. ICRISAT, India.

Table 1: Effect of inorganic and organic fertiliser on Striga

Table 2: Effect of green manure on Striga incidence in maize.

Back to Top