Effect of increasing area of cassava (Manihot esculenta Crantz) relative to Flemingia (Flemingia macrophylla) on biomass yield, soil fertility and soil erosion

Ngo Tien Dzung and T R Preston^*

Goat and Rabbit Research Center, Sontay, Hatay, Vietnam
dzungvbdp@yahoo.com
* University of Tropical Agriculture, TOSOLY, AA #48 Socorro, Colombia

Abstract

An experiment was carried out at the Goat and Rabbit Research Center, Sontay, Hatay, Vietnam, from February 2003 to November 2004 to measure effects on biomass yield, soil fertility and soil erosion in response to increasing ratio of rows of cassava rows to rows of the leguminous shrub Flemingia macrophylla in an inter-cropping system. Two rows of Flemingia were associated with 1, 2, 3, 4 and 5 rows of cassava. Flemingia was planted first and cassava 14 days later. Fertilizer was applied in the form of organic manure (from mixed cattle and buffalo excreta) before planting at the rate of 2 kg/m² fresh weight in the first year and the same amount again in March the second year. No other fertilizers were applied during the experimental period. The first harvest was made when the cassava reached 100 cm in height and subsequent harvests of the re-growth from 56 to 75 days. All the foliages were removed at 30 cm above ground level.

Biomass yield of cassava and Flemingia in an inter-cropping system was improved by increasing the number of cassava rows relative to cassava in the first year, but decreased slightly in the second year. The maximum biomass yield was obtained in the system with 2 rows of cassava and two rows of Flemingia. The crude protein contentof the cassava foliage deceased as the area of cassava rlative to Flemingia increased. Soil fertility over a 24 month period increased in the plots with the highest ratio of Flemingia and decreased as the ratio of cassava to Flemingia was increased. Protein content in the cassava foliage decreased linearly with decreased in soil fertility. Soil erosion was high in all plots in the first year and was directly related to the relatve area planted with Flemingia. In the second year soil eroson was much less and was inversely related to the proportion of Flemingia n the inter-cropping system.

It is concluded that the optimum ratio of Cassava to Flemingia, in terms of biomass yield and maintenance of soil fertility, is two rows of each planted alternately.

Keywords: Biomass, bio-test, cassava, Flemingia, foliage, inter-cropping, soil fertility, soil erosion

Introduction

Cassava or tapioca (Manihot esculenta, Crantz) is an annual root crop grown widely in tropical and sub-tropical areas. It is a cash crop cultivated by smallholder farmers within the existing farming systems in many countries. Growing and using cassava as a perennial forage was first proposed by Moore (1976) based on observations at CIAT in Colombia. High yields of foliage were obtained when cassava was managed as a semi-perennial crop with repeated harvesting of the foliage at 2-3 month intervals. This idea was taken up in the Dominican Republic by Ffoulkes and Preston (1978) who showed that the fresh foliage could be used as the sole source of protein and fibre for supplementing a liquid diet of molasses-urea for fattening cattle. Growth rates were over 800 g/day and were not improved when 400 g/day of additional soya bean meal was given. However, although successful at the level of the animal the system could not be sustained agronomically. Yields of foliage fell rapidly with successive harvests and were negligible by the fourth harvest, due to a lack of appreciation of the need to return to the soil the considerable amounts of nitrogen and other nutrients removed by repeated harvesting (T R Preston, unpublished observations).

In studies on the use of cassava hay as cattle feed in Thailand Wanapat et al (1992, 1997, 2000) described a combined forage/root management system for cassava with two or more harvests of the foliage prior to letting the root develop to maturity. More recent research, first in Vietnam (Preston et al 2000, 2001) then in Cambodia (San Thy and Preston 2000), has demonstrated that the cassava plant can be maintained as a semi-perennial forage crop for at least 2 years provided there is heavy fertilization either with goat manure or with the effluent from biodigesters charged with pig manure.

Another approach to maintaining soil fertility is to plant the cassava in association with a forage legume such as Gliricidia sepium and Desmanthus virgatum (Preston et al 2000), Flemingia macrophylla (Ngo Tien Dzung et al 2003; Nguyen Phuc Tien et al 2003) and cow pea (Polthanee et al 2001).

Flemingia macrophylla is a leguminous shrub with high biomass yields that grows well in acid soils (Dinh Van Binh et al 1998). Flemingia improves soil fertility though return to the soil of organic matter from fallen dead leaves and by nitrogen fixation (Dinh Van Binh et al 1998; Andersson 2002). The foliage of Flemingia is relatively high in crude protein (about 17% in dry matter) but it can only replace about 15% of the diet of growing goats because of low palatability and low digestibility (Nguyen Thi Mui et al 2002). Recent research indicated that planting Flemingia and cassava in alternate double rows was better than alternate single rows in terms of total biomass yield, maintenance of soil fertility and reduced soil erosion, and that both these associations were superior to growing of cassava in monoculture (Ngo Tien Dzung 2003).

The overall objective of the present study was to develop a sustainable system for maintaining cassava as a semi-perennial forage crop for feeding to live stock. The hypotheses to be tested were:

Biomass yield in an inter-cropping system of cassava and Flemingia would be improved by increasing the number of cassava rows from 2 to 3, 4 or 5 rows, for every 2 rows of Flemingia.
Soil fertility would be maintained even with the highest ratio of cassava in the association.

Material and methods

Location and climate

The experiment was carried out from February 2003 to December 2004 on sloping land at 105^o25' E and 21^o06' N, 220m above sea level in the Bavi region, Hatay province, North Vietnam. The soil characteristics and composition at the experimental site were: pH_KCl 4.24, organic matter (OM) 4.64%, N_total: 0.17%, K₂O_total 0.48% and P₂O_{5 total} 0.052%. In general, the soil is acid with low fertility and the OM has been reduced by erosion. Drought occurs frequently in the area.

The climate in the area is tropical- monsoon, with a wet season between April and November and dry season from December to March. Annual rainfall was 1874mm in 2003 and 1756mm in 2004 with a mean daily temperature from 19.7 to 31.5^oC.

Treatments and experimental design

There were five treatments:

2FM-1CF: Two rows of Flemingia associated with 1 row of cassava
2FM-2CF: Two rows of Flemingia associated with 2 rows of cassava
2FM-3CF: Two rows of Flemingia associated with 3 rows of cassava
2FM-4CF: Two rows of Flemingia associated with 4 rows of cassava
2FM-5CF: Two rows of Flemingia associated with 5 rows of cassava

The plots were each 10*20m (200 m²) arranged as a randomized complete block design with 4 replications on 5000 m² of sloping land. The slope of the plots ranged from 17.3 to 17.7%.

Crop establishment, management and measurements

Planting

The Flemingia was planted with 50 cm between the two rows and 5 cm between seeds. The cassava was planted with 50 cm between rows and 15 cm between stem cuttings (length of stem cutting was 20 to 25 cm). The cassava was planted first followed 14 days later by the Flemingia. Organic manure (from pigs, cattle and buffalo) was applied before planting and after 6 months of planting at the rate of 2 kg/m². No other fertilizer was applied during the experimental period, which lasted for 2 years. Manure application was the same in all plots.

Harvesting

The first harvest was made when the cassava or Flemingia reached 100 cm in height (about 3 months after planting), all the foliage being removed at 30 cm above ground level. The same pattern was followed for the re-growth (about every 75 days). In each plot, 5 randomized sub-plots (4*4 m = 16m²) were chosen to estimate biomass yield, which was calculated as an average of the 5 sub-plots. The sub-plots were chosen without a zone around them. The roots from the cassava were harvested at the end of the two years of the experiment.

Measurements

Samples of foliage (Flemingia and cassava) and of cassava roots were taken from each of the 5 sub-plots and mixed prior to analysis for dry matter (DM) and nitrogen (N).

Soil sampling and biological test of soil fertility:

Biological test for soil fertility and soil analysis

At the beginning, and at 6-month intervals, samples of soil were taken from the plots for biological test of soil fertility (Nguyen Thi Mui et al 1996; Boonchan Chantaprasarn and Preston 2004) and for soil analysis. The soil samples were taken from each of the 5 sub-plots which were then bulked for each treatment/replicate. The growth periods of the maize in the biological test were 35 days in the summer season and 42 days in the winter season.

Soil erosion

At the lower limit of each plot, a canal system was established following the contour. The canals were 60 cm deep, 60 cm wide and 10 m long, and were lined by plastic sheet to catch the soil that was washed away. The soil loss was weighed after each rainfall.

Chemical analysis

The samples of the foliages, and the cassava roots, were analysed according to AOAC (1990) for DM and crude protein. Samples of soil were analysed for pH_KCl (KCl 0.1M), OM, N, P and K, according to AOAC (1990).

Statistical analysis

The data were analysed statistically using the GLM procedure of the Minitab Software, version 13.1 (Minitab 2000). The treatment means which showed significant differences at the probability level of P<0.05 were compared with each other using Tukey's pair-wise comparison procedure (Minitab 2000). The model used in the analysis was:

Yij = µ + Ti +Bj + (TB)ij + eij

Where: Yij is dependent variable, µ the overall mean, Ti: treatment (i= 1,...5), Bj: block (j = 1; …4), (TB)ij: interaction between treatment and block and eij the experimental error

Results and discussion

Chemical composition of cassava and Flemingia foliages

There were contrasting results for composition of Flemingia and cassava foliage in the two years (Table 1).

Table 1: Mean values for content of DM and crude protein (CP) in the foliage of cassava and Flemingia in each of the two years
	Flemingia		Cassava
	DM, g/kg	CP, g/kg DM	DM, g/kg	CP, g/kg DM
Year 2003
2FM-1CF	250^a	148	239	171
2FM-2CF	249^a	154	242	165
2FM-3CF	240^ab	164	246	161
2FM-4CF	223^ab	172	252	160
2FM-5CF	212^b	175	253	159
SE mean	6.0	4.5	7.2	3.7
Year 2004
2FM-1CF	237^a	177^a	226	186
2FM-2CF	246^ab	177^a	231	186
2FM-3CF	250^ab	174^a	242	179
2FM-4CF	256^b	163^b	252	175
2FM-5CF	256^b	163^b	258	173
SE mean	3.9	2.1	6.7	3.4
^{a, b}Means within columns within years, without common superscripts differ at P<0.05

As the ratio of cassava to Flemingia increased, there were increases in crude protein and decreases in DM content of the Flemingia foliage in the first year (2004) while the opposite trends were observed in the second year (Figures 1 and 2). By contrast, the DM content of cassava foliage increased and the crude protein decreased in both years as the ratio of cassava to Flemingia was increased (Figures 3 and 4).


Figure 1: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on DM content of foliage of Flemingia in each of the two years	Figure 2: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on CP content of foliage of Flemingia in each of the two years


Figure 3: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on DM content of foliage of cassava in each of the two years	Figure 4: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on CP content of foliage of cassava in each of the two years

Biomass yield

In the first year, biomass yield of Flemingia was low but increased in the second year (Table 2; Figures 5 and 6). In both years the maximum yield of biomass DM and crude protein was recorded for the treatment of 2 rows Flemingia; 2 rows cassava (Figures 7 and 8). In the second year, yield of DM and crude protein decreased linearly as the ratio of cassava to Flemingia increased (Figures 7 and 8). In the optimum treatment (2 rows Flemingia: 2 rows cassava) total biomass yield was higher in the second year. A similar finding was reported by Nguyen Phuc Tien et al (2003).

Table 2: Mean values for yield (tonnes/ha) of DM and crude protein in foliages of cassava and Flemingia in each of the two years
	2FM-1CF	2FM-2CF	2FM-3CF	2FM-4CF	2FM-5CF	SEM/P
Year 2003
DM	8.14^a	10.31^b	11.03^c	11.16^cd	11.72^d	0.15/
Flemingia	3.50^a	3.40^a	2.85^b	2.21^c	1.67^d	0.075/
Cassava	4.63^a	6.91^b	8.18^c	8.95^d	10.06^e	0.113/
Crude protein	1.31^a	1.66^b	1.78^c	1.81^cd	1.89^d	0.025
Flemingia	0.52^a	0.52^a	0.47^b	0.38^c	0.29^d	0.012
Cassava	0.79^a	1.14^b	1.31^c	1.43^d	1.60^e	0.018
Year 2004
DM	11.32^a	11.99^b	11.06^a	10.15^c	10.00^c	0.164
Flemingia	7.62^a	6.72^b	5.30^c	4.12^d	3.43^e	0.115
Cassava	3.69^a	5.26^b	5.76^c	6.03^c	6.56^d	0.090
Crude protein	2.04^a	2.17^b	1.95^a	1.72^c	1.69^c	0.029
Flemingia	1.34^a	1.18^b	0.92^c	0.67^d	0.56^e	0.019
Cassava	0.69^a	0.98^b	1.03^b	1.05^b	1.13^c	0.016
Two years
DM	19.46^a	22.31^b	22.10^b	21.32^b	21.72^b	0.308
Crude protein	3.35^a	3.83^b	3.73^bc	3.54^ac	3.58^c	0.052
^{a bc} Means within rows, within years, without common superscripts differ at P<0.05


Figure 5: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on DM yield of foliage of Flemingia and cassava in 2003	Figure 6: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on DM yield of foliage of Flemingia and cassava in 2004


Figure 7: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on DM yield of foliage (combined Flemingia and cassava) in 2003, 2004 and for both years	Figure 8: Effect of treatment (rows of cassava relative to 2 rows of Flemingia) on crude protein yield of foliage (combined Flemingia and cassava) in 2003, 2004 and both years

Soil fertility

Soil fertility as measured by the maize "bio-test" increased with time over 24 months for treatments 2FM:1CF and 2FM:2CF but declined for the treatments with 4 or 5 rows of cassava per 2 rows of Flemingia (Table 3 and Figure 9). Other measures of soil fertility (organic matter and N content) also showed a linear decline as the ratio of cassava to Flemingia increased (Figures 10 and 11).

Table 3: Least square means for weight of root and green biomass of maize plants grown in soil from the experimental plots (g/pot)
	2 FM-1CF	2FM-2CF	2FM-3CF	2FM-4CF	2FM-5CF	SE
Beginning	32.8	34.7	33.7	32.9	33.2	2.20
6 months	36.5	35.1	34.9	33.4	33.6	1.50
12 months	37.25^a	36.81^a	34.0^ab	31.7^b	29.6^b	1.32
18 months	39.9^a	37.1^ab	33.9^b	30.7^b	30.6^b	1.56
24 months	40.1^a	38.1^a	34.8^ab	31.6^b	28.7^b	1.47
^{a, b} Means within rows without common superscript differ at P<0.05

Figure 9: Changes in soil fertility with time according to ratio of Flemingia to cassava


Figure 10. Changes in organic matter in soil according to ratio of Flemingia to cassava	Figure 11. Changes in N content of soil according to ratio of Flemingia to cassava

The protein content of the cassava foliage was positively related with soil fertility (Figure 12).

Figure 12: Relationship between relative soil fertility (according to maize biotest)
and crude protein content of foliage of cassava (end year 2004)

In the first year, soil erosion varied inversely with the ratio of Flemingia to cassava as growth of the Flemingia was slower than that of the cassava leading to reduced soil cover in plots with highest proportion of Flemingia (Table 4 and Figure 13). The opposite effect was observed in the second year when soil loss was lower in all plots compared with the first year and increased linearly as the proportion of cassava in the plots increased.

Table 4: Least square means for parameters of soil fertility
	2FM-1CF	2FM-2CF	2FM-3CF	2FM-4CF	2FM-5CF	SE
Beginning
pH_KCl	4.30	4.26	4.22	4.30	4.24	0.032
OM, %	4.69	4.10	4.76	4.39	4.75	0.259
N %	0.16	0.17	0.165	0.155	0.162	0.015
P₂O₅, %	0.055	0.041	0.051	0.036	0.090	0.015
K₂O, %	0.58	0.47	0.45	0.40	0.48	0.039
After 2 years
pH_KCl	4.37	4.22	4.28	4.30	4.28	0.05
OM, %	5.10	4.42	4.56	4.21	4.08	0.121
N %	0.24	0.187	0.15	0.142	0.132	0.015
P₂O₅, %	0.105	0.081	0.104	0.078	0.085	0.007
K₂O, %	0.63	0.515	0.565	0.30	0.28	0.055
Beginning compared to after 2 years
pH_KCl	NS	NS	NS	NS	NS
OM, %	*	*	NS	NS	*
N %	*	NS	NS	NS	**
P₂O₅, %	**	**	**	NS	NS
K₂O, %	NS	NS	*	*	*
Soil erosion (tonnes/ha)#
Year 1	50.0	46.7	43.0	42.4	39.1	2.25
Year 2	12.8^a	18.7^a	22.1^ab	23.9^b	29.5^b	2.25
2 years	63.2	65.4	65.1	66.3	68.7	6.7
^{a, b} Means within rows without common superscript differ at P<0.05 # The slope of the plots ranged from 17.3 to 17.7%

Figure 13: Soil erosion in plots with varying ratios of Flemingia to cassava in the two years

Soil cover is an important factor in control of erosion by water through interception and absorbing the kinetic energy in the rain. The soil cover reduces the direct impact of the rain drops and prevents the surface from becoming sealed as well as preserving the soil structure. Andersson (2002) compared plots with monoculture of cassava or Flemingia and reported that in the second year, soil loss in the former was 34 tonnes compared with only 5.4 tonnes/ha for the Flemingia. Nguyen Phuc Tien et al (2003) also found that soil fertility was maintained over two years in plots inter-cropped with Flemingia and cassava in equal proportions.

Conclusions

Biomass yield of cassava and Flemingia in an inter-cropping system was improved by increasing the number of cassava rows relative to cassava in the first year, but decreased slightly in the second year. The maximum biomass yield was obtained in the system with 2 rows of cassava and two rows of Flemingia.
The crude protein contentof the cassava foliage deceased as the area of cassava relative to Flemingia increased.
Soil fertility over a 24 month period increased in the plots with the highest ratio of Flemingia and decreased as the ratio of cassava to Flemingia was increased.
Protein content in the cassava foliage decreased linearly with decrease in soil fertility.
Soil erosion was high in all plots in the first year and was directly related to the relative area planted with Flemingia. In the second year soil erosion was much less and was inversely related to the proportion of Flemingia in the inter-cropping system.
The optimum ratio of Cassava to Flemingia, in terms of biomass yield and maintenance of soil fertility, is two rows of each planted alternately.

Acknowledgements

The Senior Author acknowledges the support of the International Foundation of Science (IFS) for financing this research (Grant No. B/3085-1).

References

Andersson J 2002 Possible Strategies for Sustainable Land Use in the Hilly Area of Northern Vietnam. Swedish University of Agricultural Sciences, Department of Soil Sciences, Division of Agricultural Hydrotechnics, S-750 07 Uppsala, Sweden

AOAC 1990 Official methods of Analysis. 15^th Edition. Association of Analytical Chemist, Washington D C

Boonchan Chantaprasarn and Preston T R 2004 Measuring fertility of soils by the bio-test method. Livestock Research for Rural Development. Volume 16, Art. #78. Retrieved , from http://www.cipav.org.co/lrrd/lrrd16/10/chan16078.htm

Dinh Van Binh, Nguyen Phuc Tien and Nguyen Thi Mui 1998 Study on biomass yield and quality of Flemingia macrophylla and on soil fertility. Proceeding on Workshop of Animal Nutrition Science, Ministry of Agriculture and Rural Development, Vietnam, pp.137

Ffoulkes D and Preston T R 1978 Cassava or sweet potato forage as combined sources of protein and roughage in molasses based diets: effect of supplementation with soybean meal. Tropical Animal Production (3) :186-192 http://www.utafoundation.org/TAP/TAP33/3_3_1.pdf

Moore C P 1976 El uso de forraje de yuca en la alimentacion de rumiantes Seminario Internacional de Ganaderia Tropical, Acapulco, Mexico

Ngo Tien Dzung, Dinh Van Binh, Nguyen Thi Mui and Preston T R 2003 Improving biomass yield of cassava (Manihot esculenta Crantz) and soil fertility through mixed culture with Flemingia(Flemingia macrophylla). Cassava hay as protein source for lactating goats: IFS Research grant report, No. B/3085-1.

Nguyen Thi Mui, Preston T R, Dinh Van Binh, Le Viet Ly and Ngo Tien Dzung 1996 Effect of planting season and type of fertilizer on biomass yield and quality of sugar cane. Livestock Research for Rural Development, Volume 8, Number 4, November 1996 http://cipav.org.co/lrrd/lrrd8/4/mui841.htm

Nguyen Thi Mui, Ledin I, Udén P and Ding Van Binh 2002 Effect of replacing a rice bran-soya bean concentrate with Jackfruit (Artocarpus hererophyllus) or Flemingia (Flemingia macrophylla) foliage on the performance of growing goats. Livestock Production Science 72 (2001), 253-262

Nguyen Phuc Tien, Ngo Tien Dung, Nguyen Thi Mui, Dinh Van Binh and Preston T R 2003 Improving biomass yield and soil fertility by associations of Flemingia (Flemingia macrophylla) with Mulberry (Morus alba) and cassava (Manihot esculenta) on sloping land in Bavi area. In: Proceedings of Final National Seminar-Workshop on Sustainable Livestock Production on Local Feed Resources (Editors: Reg Preston and Brian Ogle). HUAF-SAREC, Hue City, 25 - 28 March, 2003. Retrieved , from http://www.mekarn.org/sarec03/tienbavi.htm

Polthanee A, Wanapat A, Wanapat M and Wachirapokor C 2001 Cassava-Legumes intercropping: A potential food -feed system for dairy farmers. International Workshop Current Research and Development on Use of Cassava as Animal Feed. Khon Kaen University, Thailand, July 23-25, 2001 http://www.mekarn.org/procKK/polt.htm

Preston T R, Rodriguez L and Khieu Borin 2000 Associations of cassava and legume trees as perennial forage crops for livestock. Workshop-seminar "Making better use of local feed resources" January 2000. SAREC-UAF http://www.mekarn.org/sarpro/trplrkb.htm

Preston T R 2001 Potential of cassava in integrated farming systems. International Workshop Current Research and Development on Use of Cassava as Animal Feed. Khon Kean University. Thailand, July 23-24^th, 2001 http://www.mekarn.org/procKK/pres.htm

San Thy and Preston T R 2000 Cultivating cassava (Manihot esculenta) as perennial forage for goats, effects on foliage yield and composition of variety, associations with shrub legumes and recycled biodigester effluent. Livestock Research for Rural Development. (13) 2, http://www.cipav.com.org/lrrd/lrrd13/2/sant132

Wanapat M, Wachirapakorn C, Chanthai C and Sommart K 1992 Utilization of cassava leaf (Manihot esculenta Crantz) in concentrates mixtures for swamp buffaloes in Thailand. Proc. Feeding Strategies for Improving Ruminant Productivity in Area of Fluctuation Nutrient Supply FAO/IAEA, Vienna, Austria

Wanapat M, Pimpa O, Petlum A and Boontao U 1997 Cassava hay: A new strategic feed for ruminants during the dry season. Livestock Research for Rural Development (9) 2: Http://www.cipav.org.co/lrrd/lrrd9/2/metha92.htm

Wanapat M, Puramongkorn T and Siphuak W 2000 Feeding of cassava hay for lactating dairy cows. Asian-Australasian Journal of Animal Science 13, 478-482

Received 1 June 2006; Accepted 12 December 2006; Published 8 February 2007

Go to top