African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 7, Num. 4, 1999, pp. 383-395

African Crop Science Journal, Vol. 7. No. 4,  pp. 383-395, 1999                                                             

Agronomic effectiveneness of poor quality manure supplemented with phosphate fertiliser on maize and groundnut in a maize-groundnut rotation

R. Chikowo, F. Tagwira and  M. Piha¹
Faculty of Agriculture and Natural Resources, Africa University, P. O. Box 1320, Mutare, Zimbabwe
¹Soil Science Department, University of Zimbabwe, P. O. Box 167, Mt. Pleasant, Harare, Zimbabwe

Code Number CS99029

ABSTRACT

A survey of groundnut and maize fields in the smallholder farms showed that soils sampled from maize fields had significantly (P<0.05) higher pH, calcium and phosphorus than those from groundnut fields. It was established that there was limited use of mineral and organic fertilisers to groundnut, and this could be linked to limited resources and the risk associated with fertilisation in an environment of low and uncertain rainfall.  The low fertility of groundnut soils has contributed to the decline in groundnut yields.  The agronomic effectiveness of cattle manure supplemented with phosphate fertiliser, on maize and groundnut, was evaluated in factorial experiments on five smallholder farms in Zimbabwe.  Increased groundnut and maize yields were obtained where manure had been applied at rates up to 20 t ha^-1 at the majority of the sites. The benefit of supplementing manure with phosphate fertiliser was not evident.  Factorial analysis indicated significant main effects of manure (P<0.01), for both direct and residual effects.  The main effect of phosphate was not significant at all but one site, while no manure x phosphate interactions were observed.  Regression analysis indicated linear effects of manure on maize yield and groundnut pod yield.  The different qualities of manure used at the different sites (based on total manure N content) may have contributed to the variations in response observed at the different sites.  Basic economic analysis revealed that it was more profitable to use direct and residual manure fertility on groundnut at Marange sites, while at a site in Wedza, farmers would get better returns with manure fertilisation on maize.

Key Words: Arachis hypogaea, cattle manure, soil fertility, Zimbabwe

RÉSUMÉ

Une enquéte conduite en champs de maïs et d’arachides chez les petits fermiers a montré que des échantillions de sol des champs de maïs avaient des niveaux de pH, calcium et de phosphate significativement superieurs à ceux des champs d’arachide.  Il a été trouvé qu’il y avait une utilization limitée d’engrais mineraux et organiques en champs d’arachide et ceci était lié aux ressources limitées et aux risques associés à la fertilisation dans un environement à faible et incertaine pluie.  La basse fertilité des sols a contribué à la reduction du rendement en arachide. L’efficacité agronomique de la fumure organique de vaches complementée d’engrais phosphaté sur le maïs et l’arachide, a été évaluée dans un essai factoriel chez 5 petits fermiers au Zimbabwe.  Des rendements élevés de maïs et d’arachides ont été obtenus où la fumure a été utilisée jusqu’ à des taux de 20 t ha^-1 dans la majorité des sites.  L’avantage d’ajouter l’engrais phosphaté sur la fumure n’a pas été évident.  L’analyse factorielle a montré des effects principaux significatifs de la fumure (P<0,01) ensemble pour les effets directs et résiduels. L’effect principal de la phosphate  n’a pas été du tout significatif alors que l’interaction entre la fumure et l’engais phosphaté a été signifactive.  L’analyse de la régression a montré des effets linéaires de la fumure sur le rendement du maïs et du rendement en gousses d’arachide.  Les differentes qualités de la fumure utilisée dans les different sites (basé sur la teneur en azote total du fumier) au raient contribué à la variation des réponses observées dans les differents sites. L’analyse économique préliminaire a révélé qu’il était profitable d’utiliser la fertilité directe et résiduelle du fumier à Murange, alors que les fermiers du site de Wedza pourraient recevoir de meilleurs revenus avec la fertilisation du maïs avec du fumier.

Mots Clés:  Arachis hypogaea, fumier de vache, fertilité du sol, Zimbabwe

Introduction

There has been a continuous decline in area planted to groundnut (Arachis hypogaea L.) in the smallholder farms in Zimbabwe (Dendere, 1987).  Groundnut yields in the smallholdings  are low (400-700 kg ha^-1) in marked contrast to yields of over 3000 kg ha^-1 obtained on research stations and commercial farms in southern Africa (Chiteka et al., 1992).  Yet groundnut is an important food crop and has high protein that could be exploited to arrest malnutrition problems in the communal areas.  Therefore, options have to be developed to give smallholder farmers alternative means to produce better groundnut yields.

Cattle manure is an integral component of soil fertility management in the communal areas in Zimbabwe.  The beneficial effects of cattle manure on maize yield are well documented.  Grant (1967) reported that the beneficial effect of manure on N availability for maize grown on the granitic sandy soils was due to the N released directly after application.  This N is most likely present as free mineral N in the manure as ammonia may be lost rapidly by volatilisation from manures (Murwira, 1995).  Hartley (1937) found that crop responses to manure on sandy soils are often due more to the contribution of P and the cation  such as calcium and magnesium, than the addition of N.  However, it has since been recognised that the majority of manures used in the communal areas in Zimbabwe have low P and N contents (Tanner  and Mugwira, 1984) and these P contents are insufficient to meet the requirements for sustained crop production (Palm, 1995).  Coupled with the widely P deficient soils in Zimbabwean smallholder farms (Grant, 1981), supplementation of cattle  manure with mineral P fertilisers to improve its agronomic effectiveness is a potentially useful strategy.  Additionally, the need for basal fertiliser N is widely recognised as it is readily observed that plants grown in soils with freshly-applied manure show N deficiency.  In fact, N is the most limiting nutrient to maize production in the smallholder sector in Zimbabwe, and soils in most areas of the smallholder agriculture are deficient in N (Mashiringwani, 1983).  The relatively high cost of mineral N is of a great economic concern in crop production.  As many legume species are known to fix N, there is need to increase the area under these crops and to include them in rotations and capitalise on the nitrogen they fix into the soil.  To date, little attention has been focused on the use of manure to fertilise crops other than maize under on-farm smallholder conditions.  The principal objectives of this study were (a) to quantify the contribution of different manures supplemented with P fertiliser at different sites to maize and groundnut productivity, and (b) assess the comparative profitability of using direct and residual manure on these two crops  when grown in sequence.

Materials and methods

Survey: nutrient status of maize and groundnut fields. A survey was conducted in June 1995 in two communal areas (Marange and Wedza) in eastern Zimbabwe.  A total of twelve villages and 7 farmers per village were involved in the survey.  Soils were sampled from the fields where the farmers had planted groundnut and maize during the 1994/95 season.  Soil sampling was limited to the plough layer (0-20 cm deep).  The soils were analysed for P, Ca, Mg and K using the Mehlich 3 universal extraction method (Mehlich, 1984).  Soil pH was determined in 0.01M CaCI₂. Two-tailed t-tests were used to compare the soil pH and nutrient concentrations for the maize and groundnut soils.

Field experiments. An experiment to assess the effect of manure supplemented with P on groundnut and maize yields in a maize/groundnut rotation was carried out at five on-farm sites in different agroclimatic regions of Zimbabwe.  Three rates of cattle manure (0, 10 and 20 t ha^-1) and three rates of P  as single super phosphate (0, 9 and 18 kg ha^-1 P) were combined in a factorial design with three complete replicates.

In the second season (1996/97)  maize was grown on the previous groundnut plots while the groundnut crop was grown on the plots previously grown to maize, to assess the benefits of maize following groundnut and the residual manure and P effects on both crops.  Plots measured 5 m long and 4.5 m wide, with rows 0.9 m apart and within row spacing of 30 cm for maize.  Maize cultivar R215, an early maturing variety, was used.  All maize plots received a top dressing of 68 kg ha^-1 N as ammonium nitrate. Plots for groundnut also measured 5 m long and  4.5 m  wide  with  rows 45 cm apart and within row spacing of 10 cm.  The groundnut variety Falcon was used.   The manure at each of the sites was first broadcast within the respective plots and  then ploughed under using mouldboard plough.  Factorial analysis of variance was done using Genstat, Version 5 (Genstat 5 Manual, 1989).  Regression analysis on the overall mean yields (maize grain, groundnut pod) due to manure application and on mean yields at different P rates were computed.

Soil and manure analysis.  Soil pH was determined in 0.01M CaCI₂. Exchangeable bases (Ca, Mg and K) were extracted from soil using acidified ammonium acetate and determined by atomic absorption spectrophotometry.  Soil available P was extracted using the Mehlich 3 extraction method (Mehlich, 1984), and the Murphy and Riley (1962) method was used for colour development and P was read on a UV spectrophotometer.  Total soil and manure N were determined by acid digestion of the soil or manure, followed by steam distillation in the presence of excess alkali, and then titration.  Manure P, and the rest of the elements were determined by complete oxidation of samples using Kjedahl procedures, followed by spectrometric analysis.

Economic analysis. Cost of manure was calculated using the fertiliser N equivalent value and assuming 50 percent was used during the first year and the other 50 percent used as residual fertility.  Linear regression slopes were used to determine the yield returns per unit increment in manure at the different sites.  In the preliminary economic analysis, the Grain Marketing Board’s maize and shelled groundnut buying prices (October 1996 and October 1997) of Z$1200 and Z$5000 were used in the calculations.  Overall economic analysis incorporated labour, at 150 person-days/ha for groundnut ($1800), and 80 person-days/ha ($960) for maize, fertiliser and seed costs.  The rationale of applying manure directly to groundnut or to maize was then evaluated.

Results

Survey. Mean P and Ca concentrations for soils sampled from maize fields were found to be significantly higher (P<0.05) than those from groundnut fields (Fig. 1).  Nine of the 12 villages had higher Ca concentrations in maize soils compared with groundnut soils, while P had higher values in maize soils for all the villages.  Soil pH was also found to be significantly higher in maize fields (P<0.05).  Statistically, there was no apparent difference between Mg and K concentrations of maize and groundnut soils, though maize soils had generally higher concentrations of these  elements (Table 1).  The t probability value of 0.054 for Mg was on the border line, and was evidence enough to suggest that there may be a real difference in Mg means.

Table 1. Chemical characteristics of soils taken at a depth 0-20 cm from farmers fields in Wedza and Marange, Zimbabwe

*Significantly different at P=0.05; ns-not significantly different

Soil and manure analysis.  For the field experiments, all soil pH values were below pH 5, indicating a moderately acid soil environment (Table 2).  The soil P concentrations were very low, particularly at Gonzo and Maekaeka sites.  The manures had varied nutrient contents (Table 3).

Table 2.  Some initial soil chemical properties of experimental sites

*TEB - total exchangable bases
**Mehlich method

Table 3. Nutrient contents of manures used at the different sites

Pooled data. Groundnut pod yield and maize grain yields was pooled together to determine if treatment comparisons were consistent enough over the farms for blanket recommendations to be made which would have general applicability.  The preliminary analysis of variance showed an overall significant farm x treatment interaction.  This may not be surprising in the light of different manure quality and the inherent differences in soil fertility between sites. 

Groundnut pod yield. An analysis of variance showed that there was no interaction between manure and P on groundnut pod yield for both seasons across all the sites (P>0.05).  The main effect of manure for both direct and residual fertility was highly significant across most of the sites (P<0.001), while the main effect of P was not significant at most sites (P>0.05) (Fig. 2).  Regression analysis on overall manure main effects showed that most of the variation in pod yield was accounted for by linear regressions (P<0.001).  At the Maekaeka site, during the 1996/97 season, the main effect of P was significant  on groundnut pod yield (P<0.05).  However, unlike at other sites, regression analysis showed that the linear model could not fully describe the yield trend due to manure application.  There was a significant deviation.  At Gonzo, there was little response to both manure and P and the yields were low (Fig. 2).

Maize grain yield. Factorial analysis of variance showed that there was no significant interaction between manure and P fertiliser on maize grain yield.  The main effect of P was also not significant at most of the sites except at Maekaeka where it had a significant effect (P<0.05) during the first season (Fig. 3).  Yields had a significant linear relationship to manure application.  At Gonzo, the maize grain yields show that the site had generally low yields, with a mean of 1.21 t ha^-1 during the first season. There was an average yield increment of 34 kg ha^-1 for every tonne of manure that was applied.  The direct and residual effects of manure were significant (P<0.010) on maize grain yield at all the sites (Fig. 3).

Soil pH and nutrient changes. Manure had a strong liming effect and significantly raised the soil pH across all the sites.  The soil nutrient levels (Mg, K) were substantially increased by application of manure.  The soil analysis data (Tables 4a and 4b) show that increasing the rate of manure application resulted in more nutrients being released into the soil.  Application of 20 t  ha^-1 cattle manure resulted in significantly larger nutrient concentrations than applying 10 t ha^-1 manure  at all sampling times.  At some sites the control had increased acidity at the end of the two seasons.

Table 4a. Mean pH, Mg, and K as a function of manure application rate at four sampling times at the Mashizha site

1 & 3 - six weeks after crop emergence for 1st and 2nd seasons, respectively
2 & 4 - grain filling stage for  1st and 2nd seasons, respectively

Table 4b. Effect of manure application on soil pH, Mg and K concentrations at the end of the second  season at Maekaeka, Chgamba and Mashizha sites

Economic analysis.  The preliminary economic analysis indicated that the highest returns were obtained at Maekaeka site on groundnut during the first year of manure application (Table 5).  Direct manure application was superior to residual fertility at Mazhizha site on both maize and groundnut.  More returns were obtained at Chakanetsa site using residual fertility on maize, while at Maekaeka site direct and residual cropping gave similar economic returns.  Responsiveness to manure was low at Gonzo for both crops.  The net economic returns over all costs at the three manure levels are shown in Table 6.  Maize net returns were greater than groundnut returns for both seasons at Mashizha site.  The net economic returns increased with increase in manure level.  At Chakanetsa and Maekaeka sites, the net economic returns from groundnut were superior to those from maize.  There were negative net economic returns to both crops at Gonzo site.  Growing groundnut was also uneconomic at Chigamba site (Tables 5 and 6).

Table 5. Regression slopes, manure value and net yield returns per tonne of manure for maize and groundnut using direct and residual manure fertility

Table 6. Net returns (ZW$) over all costs at the different manure rates for groundnut and maize

- = no data

Discussion and conclusions

Survey: Nutrient status of maize and groundnut soils. The lower fertility of the groundnut fields when compared with the maize fields is clear evidence of the inequitable distribution of resources between maize and groundnut.  The nutrient levels and soil pH were found to be lower in groundnut soils than maize soils.  The survey established that farmers apply manure and termitaria soils mainly to maize crop.  These soil amendments have a significant liming effect and at the same time release nutrients to the soil.  The difference in soil fertility of the maize and groundnut field soils was therefore a reflection of the different soil fertility management practices followed by farmers on different crops.  Maize is largely the staple food crop in Zimbabwe, hence farmers give preferential treatment to maize on either field allocation or fertilisation.  One reason often cited by farmers for such practices is the biological ability of groundnut to fix N, and thus produce something in a sandy and depleted field that will not necessarily produce a cereal crop.  The ability of groundnut to efficiently utilise this pathway under the critically deficient nutrient concentrations in most of the communal soils is seriously threatened.  Most villages had P and Ca concentrations  below  25 mg kg^-1 and  2  cmol+  kg^-1 , respectively, values considered in Zimbabwe to be the critical levels for deficiency for the Mehlich method (Jones and Piha, 1989; Piha, 1993).  Groundnut is being grown on the more depleted portions of the farmers’ fields, which contributes to the decline in groundnut yields.  Application of mineral P and Ca to such fields is largely expected to have a positive effects on productivity.

Field experiments. The addition of P to manure did not significantly improve the agronomic effectiveness of manure on both test crops.  Arnold (1932) also found out that mixing farm yard manure with P fertiliser gave only small increases in maize yield over manure applied alone.  Related studies on communal area manure (Mugwira, 1985) have also shown that the addition of K and S or their combinations with P did not influence the effectiveness of manure.  The effectiveness of manure was enhanced only when fertiliser N was added.  Response to the manure applied was remarkable at some sites, despite the fact that the total N in the manure was low.  Grant (1967) suggested that the benefits which acrue from fertilising with this type of manure are due more to the amounts of bases released than to the supply of N and P.  Murwira and Kirchmann (1993) also concluded that communal area manure gives little beneficial N fertiliser effects  in the short term.  Groundnut has a high Ca requirement for proper pod development (Cox et al., 1982), hence improved pod yield was a reflection of the contribution of manure to Ca supply and  manure liming effect.  A potential problem can arise though, as an antagonistic interaction between Ca and other cations supplied by manure can arise, and may result in need for higher soil Ca requirement ((Brady and Colwell, 1945; Hall, 1975).  This is especially so since manures have varied nutrient contents depending on the animal feeds, handling and storage systems.  These imbalances, together with mid season moisture stress that inhibit Ca uptake, probably contribute to increased pod development failure under manure fertilisation as perceived by many smallholder farmers (Chikowo, 1998).

The benefits of rotating maize with groundnut were not apparent.  Climatic and other environmental factors probably had a confounding effect.  Water logging probably impaired efficient N fixation by the groundnut and proper pod development during the extremely wet 1996/97 season.  The benefits of the rotation could also have been overshadowed by the relatively high rate of mineral N fertiliser (68 kg ha^-1) on the maize crop that followed groundnut and the fact that groundnut residues were not retained in the fields.  Nitrogen contribution from the legume to the soil is usually through decomposition of the N-rich crop residue. Where the residues are removed from the field, there is a net removal of nitrogen by the legume (Giller et al., 1994).

The magnitude of response to manure application was not consistent across the sites.  Soil fertility of all the sites were within the same range as indicated by the soil analysis (Table 1), but yields and responses at sites in the higher rainfall areas were significantly lower than those in the drier areas.  For maize, excessive rainfall resulted in water logging and loss of the applied N.  The resultant low yields at Gonzo, where the soil was very sandy, was probably a direct result of excessive leaching.  There was very poor nodulation at Gonzo site, and this could be linked to the very low P (5.9 mg kg-1) and a very acid solid environment, pH 4.3.  Consequently, the groundnut yield were less than 1 t ha^-1. Surprisingly, P application had no significant effect.  The significant response to P noted at Maekaeka was expected as soil P had been shown to be deficient (Table 2).

The much-simplified economic analysis from this work has shown that there are some areas where it would bring more returns by using manure on maize for at least the first two seasons.  In other areas groundnut showed remarkable response to manure using both direct and residual fertility (Table 5).  For example, at Maekaeka site, groundnut was superior to maize using both direct and residual fertility.  Farmers in this area could get more economic returns if they were to concentrate their inputs on groundnut or at least give equal opportunities to these crops in terms of manure use.  While the results of this study can be applied during wet seasons, it was difficult to predict the direct and residual effects of manure on pod filling during drier seasons.

Acknowledgements

We acknowledge with gratitude funding for this research work from the Rockefeller Foundation through the Forum for Agricultural Resource Husbandry.

References

Arnold, H.C. 1932.  Maize yield with farm yard manures versus farm yard manure supple-mented with phosphate fertilizers.  Rhodesia Agricultural Journal 29:26-28.

Brady, N.C. and Colwell, W.E. 1945. Yield and quality of large seeded type peanuts as affected by potassium and certain combinations of potassium, magnesium, and calcium.  American Society of Agronomy Journal 37:429-442.

Chikowo, R. 1998. Soil fertility management options for improved groundnut production in the smallholder sector.  M.Phil Thesis, University of Zimbabwe.

Chiteka, A., Cole, D.L., Mwenda, F., Rao, P. and  Hilderbrand, G.L. 1992.  Groundnut research in Southern Africa Development Community Region 1980-1990.  In: Groundnut: A Global Perspective.  Proceedings of an International Workshop, 25-29 November 1991.  Ngima, S.N. (Ed.). pp. 113-119. ICRISAT Centre, Patacheru, India. 

Cox, F.R., Adams, F. and Tucker, B.B. 1982.  Liming, fertilization, and mineral nutrition: In: Peanut Science and Technology.  Pattee, H. and Young, C. (Eds.), pp. 138-159. American Peanut Research and Educational Society.

Dendere, S. 1987. Constraints in groundnut proudction and research priorities for communal areas in Zimbabwe. In: Proceedings of the Second Regional Goundnut Workshop in Southern Africa.  ICRISAT. pp. 125-130.

Genstat 5 Committee of the Statistics, Department Rothamsted Experimental Station, 1989.  Genstat 5 Reference Manual.  Claredom Press, Oxford.

Giller, K.E., McDonagh, J.F. and Cadisch, G. 1994.  Can biological nitrogen fixation sustain agriculture in the tropics?  In: Soil Science and Sustainable Land Management in the Tropics.  Syers, J.K. and Rimer, D.L. (Eds.), pp. 173-191.  CAB. International, Wallington, UK.

Grant, P.M. 1967.  The fertility of sandveld soil under continuous cultivation.  The effect of manure and nitrogen fertilizers on the base status of the soil.  Rhodesia, Zambia and Malawi Journal of Agricultural Research 5:117-128.

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

Hall, M. 1975. Effects of phosphorus, potassium, and calcium on peanuts at Mauke, Cook Islands. Empire Journal of Experimental Agriculture 3:117-120.

Hartley, K.T. 1937. An explanation of the effect of farmyard manure in Nothern Nigeria.  Empire Journal of Experimental Agriculture 5:254-263.

Jones, U.S. and Piha, M. 1989.  Evaluation of four soil test extractants for Zimbabwean soils.  Communication in Soil Science and Plant Analysis 20:1857-1871.

Mashiringwani, N.A. 1983. The present nutrient status of the soils in the communal farming areas of Zimbabwe.  Zimbabwe Agricultural Journal 80:73-75.

Mehlich, A. 1984. Mehlich 3 soil test extractants: A modification of Mehlich 2 extractant.  Communication in Soil Science and Plant Analysis 15:1409-1416.

Mugwira, L.M. 1985. Effects of supplementing communal area manure with lime and fertilizer on plant growth and nutrient uptake.  Zimbabwe Agricultural Journal 82:153-159.

Murwira, H.K. 1995.  Ammonia losses from Zimbabwean cattle manure before and after incorporation into soil.  Tropical Agriculture (Trinidad) 72:269-276.

Murwira, H. and Kirchmann, H. 1993.  Carbon and nitrogen mineralization of cattle manures  subjected to different treatments in Zimbabwean and Swedish soils.  In: Soil Organic Matter Dynamics and Sustainability of Tropical Agriculture.   Mulungoy, K. and Merckx, R. (Eds.), pp. 189-198. John Wiley & Sons, UK.

Palm, C.A. 1995.  Contribution of agroforestry trees to nutrient requirements of intercropped plants.  Agroforestry Systems 30:105-124.

Piha, M. 1993.  Evaluation of Mehlich 3 extractant for estimating phosphorus deficiency and phosphorus sorption of Zimbabwean soils.  Communication in Soil Science and Plant Analysis 24:1397-1408.

Tanner, P.D. and Mugwira, L.M. 1984.  Effectiveness of communal area manure as sources of nutrients to young maize plants.  Zimbabwe Agriculture Journal 81:31-35.

©1999, African Crop Science Society

The following images related to this document are available:

Photo images