African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 4, Num. 2, 1996, pp. 197-206

Maize and sorghum yields under tied ridges of fertilised sandy soils in semi-arid south-east lowveld of Zimbabwe

E. Z. NYAKATAWA, M. BROWN^1 and D. MARINGA

Department of Research and Specialist Services, Chiredzi Research Station, P.O. Box 97, Chiredzi, Zimbabwe.
^1 Overseas Development Administration, Chiredzi Research Station, P.O. Box 97, Chiredzi, Zimbabwe

(Received 17 May, 1994; accepted 6 June, 1995)

Code Number: CS96057
Sizes of Files:
    Text:
    Graphics: Tables (gif) - 64.7K   

ABSTRACT

Growing maize and sorghum in furrows of 1.0 m and 1.5 m wide tied ridges with either 100 kg or 200 kg compound D ha^-1 (8:14:7 NPK, basal fertilizer) + 50 kg N ha^-1 top dressing were compared to the traditional farmers' practice of growing these crops on flat land without fertilizer. The crops were grown under rainfed conditions on sandy soils of Matibi 1 and Chivi communal areas in semi-arid agro-ecological regions IV and V in the south-east lowveld of Zimbabwe, from 1987/88 to 1989/90. Grain yield of maize in 1.0 m wide tied ridges was increased significantly by 22 to 85% over sowing on flat land and sorghum grain yield was increased by 18%. Application of inorganic fertilizer resulted in significant yield increases of 35 to 115% in maize grain, 59 to 200% in sorghum grain, 27 to 96% in maize stover and 63 to 161% in sorghum stover, compared to no fertilizer application. The benefits of growing crops in tied ridges and the incremental gross margins in Z$ ha^-1 of fertilizer application were more associated with monthly rainfall distribution than with the total summer rainfall.

Key Words: Grain yield, incremental gross margins, inorganic fertilizer, rainfed, sandy soils, soil fertility, stover yield, tied ridges

RESUME

La culture de maos et de sorgho en sillons espaces de 1.0 et 1.5 m avec des billons lies et une fertilisation de 100 kg ou 200 kg par ha du compose D (8:14:7 NPK, engrais de fond) avec 50 kg N/ha d'engrais de couverture a ete comparee avec la culture traditionelle des fermiers i.e une culture sur terre plate sans engrais. La culture etait etablie de 1987/1988 jusque 1989/1990 dans de conditions de precipitation naturelle et sur des sols sableux de Matibil et des aires de la commune Chivi dans des regions agro-ecologiques semi arides IV et V dans les plaines du sud est de Zimbabwe. Le rendement des grains de maos sur des billons de 1,0 m augmentait significativement de 22 jusque 85% par rapport ˆ la culture sur terre plate, alors que le rendement de grains de sorgho augmentait de 18%. Une application d'engrais inorganique donnait une augmentation significative du rendement de grains de maos de 35 jusque 115%, de grains de sorgho de 59 jusque 200%, de fourrage de maos de 27 jusque 96% et de fourrage de sorgho de 63 jusqu' e 161%, par rapport e la culture sans engrais. L'avantage des cultures en billons lies et les marges beneficaires en Z$ pour l'engrais Etaient plutot associes ˆ la distribution de la prEcipitation mensuelle qu' e la precipitation totale en saison pluvieuse.

Mots Cles: Rendement en grains, marges beneficaires, engrais inorganique, precipitation naturelle, sols sableux, fertilite du sol, rendement de fourrage, billons lies

INTRODUCTION

The south-east lowveld of Zimbabwe lies in agro-ecological regions IV and V (Vincent and Thomas, 1960) that are characterised by low, erratic and poorly distributed rainfall. The average annual rainfall in this region is 500 + 250 mm (Lovell, 1991). Most of the rain comes in heavy storms that last for short durations, between October and February, resulting in high runoff losses from the fields. Generally, soils of agro-ecological regions IV and V of Zimbabwe are naturally infertile sands derived from granite (Grant, 1981; Shumba, 1984) and are, therefore, unsuitable for crop production.

Communal farmers in the region grow maize, sorghum and pearl millet on flat land with little use of inorganic fertilizers and, because of insufficient soil moisture and poor soil fertility, the yields of these crops have remained low. Techniques to ensure that as much of the rain-water received is made available for crop use and to improve soil fertility need to be developed to increase crop yields in the south-east lowveld of Zimbabwe.

Growing crops in tied ridges was introduced in West Africa in the 1950s (Anon., 1987). Reports by Mataruka (1985), Prestt (1986), Jones et al. (1987), Nyamudeza (1987), and Jones and Nyamudeza (1991) suggested that growing crops in tied furrows to conserve and concentrate rain-water into the root zone of crops gave significant yield increases over the traditional farmers' practice of planting on flat land on heavy soils such as the red paragneiss and the basalt-derived vertisols. This study was undertaken to investigate the effect of tied ridges and inorganic fertilizer application on grain and stover yields of maize and sorghum under rainfed conditions on sandy soils in the south-east lowveld of Zimbabwe.

MATERIALS AND METHODS

Maize and sorghum were grown under three landforms: on flat land (traditional farmers' practice), in furrows of 1.0 m and 1.5 m wide tied ridges and at three inorganic fertilizer levels: zero, 100 kg compound D ha^-1 (8:14:7 NPK) basal fertilizer + 50 kg N ha^-1 top dressing and 200 kg compound D ha^- 1 basal fertilizer + 50 kg N ha^-1 top dressing. The experiment was conducted on farmers' fields in Matibi 1 and Chivi communal areas in semi-arid agro-ecological regions IV and V in the south-east lowveld of Zimbabwe, from 1987/88 to 1989/90. The Matibil sites were Runde, Negari, Neshuro, Kuwirirana and Shazhaume, while the Chivi sites were Chivi and Mushava. The ridges were constructed using a tractor drawn, duck-foot shaped ridger. Raised soil bunds (cross ties) were erected across the ridges at about 2 m intervals to prevent water from flowing out of the furrows between the ridges. Sowing stations were marked in the furrows and 100 and 200 kg ha^-1 basal compound D fertilizer treatments were applied by uniformly incorporating the fertilizer into the soil about 10 cm from the planting rows. The farmers were given seeds of maize, variety R201, and sorghum, variety SV2. The maize variety is an early maturing hybrid while the sorghum is open pollinated and both varieties were developed for marginal rainfall areas. The crops were sown when sufficient soil moisture had been achieved, which was in the first week of November in 1987/88 season and in the last week of October in 1988/89 and 1989/90 seasons. Three maize seeds were hand sown into holes that were 45 and 30 cm apart in 1.0 m and 1.5 m wide tied ridges, respectively. Sorghum seed was drilled by hand into the planting rows. The crops were thinned to one plant per station, three to four weeks after sowing, giving plant densities of 44 000 plants ha^-1 for sorghum and 22 000 plants ha^-1 for maize. Within row spacings for sorghum were 22 and 15 cm in 1.0 m and 1.5 m wide tied ridges, respectively. The experiment was laid as a 3 x 3 factorial (three landforms and three fertilizer treatment levels) in a randomised complete block design with three replications at each site. The gross plot size was 6 x 6 m = 36 m^2. The compound D fertilizer treatments were given as 50 kg N ha^-1 top dressing in split applications of 25 kg N ha^-1 in form of ammonium nitrate (34.5% N) at thinning and at flowering. A rain gauge was installed at each site to enable farmers to take daily rainfall records throughout the duration of the experiment.

The following procedure was used to classify the rainfall distribution at each site into a category of good, mediocre or bad. Based on irrigation schedules, maize and sorghum crops generally require a total of about 400 and 350 mm, respectively, from December to January in natural regions IV and V. Since the crops were sown late October to early November, the critical months of crop growth were late December and January when the crops were flowering and filling the grain. On the basis of the above information, the three rainfall distribution categories were as follows. Good: 200-400 mm total from December to January for maize, 150-300mm total from December to January for sorghum; Mediocre: 160-200 mm total from December to January for maize, 120-150mm total from December to January for sorghum; and Bad: < 160 mm total from December to January for maize and <120 mm total from December to January for sorghum. Any year which had very little (<20 mm), or no rainfall in January was classified as bad. Farmers were encouraged to keep the experimental plots weed free throughout the season. At maturity, crops within the central 12 m2 were manually harvested and grain yield at 12.5% moisture content and stover dry weight after oven drying at 70 C for 72 hours were measured. Analysis of variance was performed on individual site data using the MSTAT-C (Michigan State University) statistical package. Tests of homogeneity of residual error variances were done to determine if results of the same crop from the individual sites and years could be pooled into combined analyses. Grain yield data for maize and sorghum were subjected to an economic analysis to determine the incremental gross margins of the fertilizer treatments in Zimbabwean dollars per hectare.

RESULTS

Monthly rainfall data for each site from October to February, which was the critical period of crop growth, from 1987/88 to 1989/90 is presented in Table 1. The rainfall for Sarahuru and Negari in 1987/88 was above the long term summer average of 400 mm (Nyamudeza et al., 1991) and was well distributed, whereas that for Mushava, Runde and Neshuro was below the long term summer average and was either mediocre or poorly distributed. In 1988/89, rainfall for all the sites was below the long term summer average and was poorly distributed (Table 1) with the exception of the distributions for Negari and Shazhaume which were good and mediocre, respectively. In 1989/90, rainfall for all the sites were below the long term summer average (Table 1) but the distributions were either good or mediocre.

Table 1

Maize. Grain yield in 1.0m wide tied ridges (Table 2) was significantly greater than that on the flat at Mushava in 1987/88, Negari in 1988/89 and Shazhaume in 1988/89 and 1989/90. Grain yield in 1.5 m wide tied ridges was significantly greater than that on the flat only at Shazhaume in 1988/89. Compared to 1.5m wide tied ridges, grain yield in 1.0m wide tied ridges was significantly greater at Mushava in 1987/88, Negari in 1988/89, and at Shazhaume and Runde in 1989/90. Grain yield in plots with the medium fertilizer treatment level (100 kg compound D ha^-1 + 50 kg N ha^-1 top dressing) was significantly greater than that in plots with the zero fertilizer treatment level at Mushava in 1987/88 and at Runde, Negari and Shazhaume in 1989/90 (Table 2).

In plots with the high fertilizer treatment level (200 kg compound D ha^-1 + 50 kg N ha^-1 top dressing), grain yield was significantly greater than that in plots with the zero fertilizer treatment level at Mushava in 1987/88, Runde in 1987/88 and 1989/90, Negari in 1988/89 and 1989/90 and Shazhaume in 1989/90. Grain yield in plots with the high fertilizer treatment level was significantly greater than that in plots with the medium fertilizer treatment level at Negari in 1988/89 and at Shazhaume in 1989/90.

Table 2

Sorghum. Grain yield in 1.0 m wide tied ridges (Table 4) was significantly greater than that on the flat at Sarahuru in 1987/88, Neshuro in 1989/90 and Kuwirirana in 1989/90, whereas that in 1.5 m wide tied ridges was significantly greater than that on the flat only at Kuwirirana in 1989/90. There were no significant differences in grain yield between 1.0 m and 1.5 m wide tied ridges at any site in any year. Grain yield in plots with the medium fertilizer treatment level (Table 4) was significantly greater than that in plots with the zero fertilizer treatment level at Negari in 1987/88, Neshuro in 1989/90 and Kuwirirana in 1989/90. In plots with the high fertilizer treatment level, grain yield was significantly greater than that in plots with the zero fertilizer treatment level at Sarahuru and Negari in 1987/88 and at Neshuro and Kuwirirana in 1989/90. Grain yield in plots with the high fertilizer treatment level was significantly greater than that in plots with the medium fertilizer treatment level at Sarahuru and Negari in 1987/88 and at Kuwirirana in 1989/90.

Table 3

Table 4

Table 5

Table 6

DISCUSSION

Growing maize in furrows of tied ridges designed to conserve and concentrate rain water into the root zone under rainfed conditions generally gave significant grain yield increases over the traditional farmers' practice of planting crops on flat land. Grain yield of maize in tied ridges was increased by 22 to 85% compared to sowing on flat land. Jones et al. (1989) reported a maize yield increase of 45% due to tied ridges on the vertisols. Grain yield of sorghum and stover yields of both maize and sorghum were, however, less affected by sowing in tied ridges. Grain yield of sorghum in tied ridges was on average 18% greater than that on flat land. Similar studies by Saleem et al. (1987) reported a 67% increase in sorghum yield in tied ridges compared to sowing on flat land. On the Vertisols, Jones et al. (1989) found a 26% increase in sorghum yield in tied ridges compared to the flat. The differences in response of sorghum to tied ridges could be attributed to soil type. Vertisols which have a higher clay content than sandy soils can conserve the soil moisture in the furrows between the tied ridges thereby giving the crop a longer period of utilising the soil moisture. On the other hand, sandy soils quickly lose their soil moisture due to poor water holding capacity thus reducing the effectiveness of the tied ridges. The difference between maize and sorghum in response to tied ridges on sandy soils could be explained by the fact that sorghum is more drought tolerant than maize hence its response to soil moisture in tied ridges would be less than that of maize.

Inorganic fertilizer increased grain yield of maize by 35 to 115% and that of sorghum by 59 to 200%. Stover yields for maize and sorghum were increased by 27 to 96% and 63 to 161%, respectively, in response to inorganic fertilizer application. Similar results were obtained by Ngambeki et al. (1991) who reported grain yield increases of up to 183% in maize due to application of inorganic and organic fertilizers. Application of inorganic fertilizers gave significant increases in maize grain and sorghum stover yields on more occasions than sorghum grain and maize stover yields. Also, the increases in maize stover yields from sowing in tied ridges was less significant than that of grain yield, while in sorghum these were about the same. These results could be attributed to the fact that maize which does not tiller responded to more soil moisture in tied ridges by allocating more assimilates towards grain, whereas sorghum in tied ridges initially produced more tillers compared to that on the flat, which, however, produced less grain due to soil moisture competition. Grain and stover yields of maize in the medium and high fertilizer treatment levels were significantly greater than that in the low fertilizer treatment level on more occasions than those of sorghum. However, the percentage increases in grain and stover yields of sorghum from inorganic fertilizer application were greater than those of maize.

The responses of maize and sorghum crops to tied ridges and to inorganic fertilizer were more affected by monthly rainfall distribution than by the total summer rainfall. On average, significant grain and stover yield increases of both crops due to tied ridges or application of inorganic fertilizer were realised in seasons with mediocre or good rainfall distribution. Poorly distributed rainfall at Runde and Neshuro in 1987/88 and Sarahuru, Chivi and Neshuro in 1988/89 resulted in no response to tied ridges and inorganic fertilizer by both maize and sorghum. In years with very low and poorly distributed rainfall, there was not enough rainfall to cause significant concentration of water in the furrows at the times it was most needed by the crops. As a result crop responses to tied ridges were less and inorganic fertilizer had little benefits on crop yields as reflected by the negative incremental gross margins (Table 6). Therefore, due to inadequate soil moisture, resulting from low and poorly distributed rainfall, fertilizer applications under rainfed conditions in the south-east lowveld of Zimbabwe require extremely good timing in order to realise benefits in yield and economic returns. The risks, such as Òfertilizer burnÓ, and accelerated crop failure associated with bad timing of fertilizer application have reduced farmers' confidence in fertilizer usage in the region. As a result, opportunities for fertilizer application are often ignored even if they could potentially increase crop yields and economic returns.

Tied ridges may not have any yield benefits in very wet years. Nyamudeza et al. (1992) reported that tied ridges did not significantly affect sorghum yields in years with adequate rainfall. Grain yields of maize in 1.0 m wide tied ridges were generally, greater than those on the flat probably because of higher soil moisture availability in tied ridges compared to the flat. However, maize grain yield in 1.5 m wide tied ridges was generally less than that in 1.0 m wide tied ridges and more or less equal to that on the flat. This could probably be due to higher within-row competition for soil moisture in 1.5 m wide tied ridges since the plants were much closer (within row spacing of 30 cm) than in 1.0 m wide tied ridges (within row spacing of 45 cm).

Roth et al. (1986) reported that the use of the tied ridge technology in combination with inorganic fertilizer gave the highest returns per unit of land and labour invested among all the technologies that were evaluated in Burkina Faso. Mazhangara (1993) found positive economic returns to cash and labour from using tied ridges in communal areas of the south-east lowveld of Zimbabwe. Results of the study in this paper suggests that if rainfall is adequate and well distributed, there is a potential of increasing sorghum yields with fertilizer application, more than those of maize on the sandy soils of Matibi 1 and Chivi communal areas.

ACKNOWLEDGEMENTS

The authors would like to thank Mr I. Gadzikwa for his part in trial management and data collection. Our thanks also go to Mr. P. Nyamudeza, Dr. M.D.S Nzima and Mr. E. Mwashayenyi for their comments and suggestions during the preparation of this paper. Finally, we would like to acknowledge and thank Katharine Morse of NRI for her encouragement and assistance in reviewing this paper.

REFERENCES

Anon., 1987. Cereal technology development -West African semi-arid tropics: A farming systems perspectives. International Programs in Agriculture, Purdue University,West Lafayette, Indiana. pp 37-51.

Grant, P.M. 1981. The fertilization of sandy soils in peasant agriculture. Zimbabwe Agricultural Journal 78:169-175.

Jones, E., Nyamudeza, P. and Nyati, C. 1987. Rainfed cropping in Natural Region V. In: Cropping in Natural Region V. Proceedings of a workshop held in Harare by Agritex and DR&SS. pp 531-542.

Jones, E., Naymudeza, P. and Busangavanye, T. 1989. Rainfed cropping and water conservation on vertisols in the S.E. lowveld of Zimbabwe. In: Vertisol Management in Africa. IBSRAM Proceedings no. 9. Bangok. pp 133-143. Jones, E. and Nyamudeza, P. 1991. The relationship between rainfall and the yield of cotton, sorghum and maize grown in tied furrows and on the flat in the South East Lowveld of Zimbabwe. In: Proceedings of the Second Annual Scientific Conference. Mbabane, Swaziland. October 7-9, 1991. pp. 100-107.

Lovell, C.J. 1991. Measurement of evaporation, transpiration and soil moisture depletion under maize during a hot rainy season in Zimbabwe. In: Proceedings of the Second Annual Scientific Conference, SADC-L&WMRP, October 7-9, 1991. Mbabane, Swaziland. pp 127-141.

Mataruka, D.F. 1985. Review of the constraints to maize production in the communal areas in natural regions III, IV and V. Zimbabwe Agricultural Journal 82:171-175.

Mazhangara, E.P. 1993. Policy consideration to enhance tied furrow technique adoption in the south-eastern lowveld vertisol of Zimbabwe. In: Cereal Grain Policy Analysis in the National Agricultural Research Systems of Eastern and Southern Africa. pp. 14-35.

Ngambeki, D.S., Loehman Edna, Yu Ziyou and Deuson, R. 1991. New agricultural technologies in northern Cameroon. Documentation and Analysis in a Risk Context. pp. 23.

Nyamudeza, P. 1987. Rain water management on vertisols for crop production in semi-arid regions. In: Management of Vertisols in Sub-Saharan Africa. Proceedings of a conference held at ILCA, Ethiopia.

Nyamudeza, P., Nyakatawa, E.Z., Jones, E. and Maringa, D. 1991. Water and soil fertility management for increased crop production on the light soils of the south-eastern lowveld of Zimbabwe. In: Proceedings of the Second Scientific Conference, SADCC-L&WMRP. Mbabane, Swaziland. October 7 - 9. pp 232-240.

Nyamudeza, P., Mazhangara, E.P. and Kamba, E. 1992. Adoption of the tied-furrow technique, and the effects of the technique and the previous crop on residual moisture and yields of sorghum and maize on vertisols. In: Africaland, Management of Vertisols in Africa. pp. 69-81.

Prestt, A.J. 1986. Basin tillage. A review. Zimbabwe Agricultural Journal 83:11.

Roth, M., Abbot, P., Sanders, J. and McKinzie, L. 1986. An application of whole - farm modelling to new technology evaluation, Central Mossi Plateau, Burkina Faso. International Programs in Agriculture, Purdue University. pp. 31-32.

Saleem, M.A., Mohamed Suleiman, H. and Kaufmann, R. von. 1987. Tillage methods for the cropping of sorghum in legume fodder banks in the sub-humid zone of Nigeria. Field Crops Research 16:85-94.

Shumba, E.M. 1984. Yields of maize in the semi-arid regions of Zimbabwe. Zimbabwe Agricultural Journal 81:91-94.

Vincent, V. and Thomas, R.G. 1960. An Agricultural Survey of Southern Rhodesia, Part 1. Agro-ecological survey, Government Printers, Salisbury.

Copyright 1996 The African Crop Science Society

The following images related to this document are available:

Line drawing images