Geoderma 125 (2005) 343 – 354 www.elsevier.com/locate/geoderma

Leguminous fallows improve soil quality in south-central Cameroon as evidenced by the particulate organic matter status L-Stella Koutikaa,*, Christian Noltea, Martin Yemefackb, Rose Ndangoa, Daniel Folefoca, Stephan Weisea a

International Institute of Tropical Agriculture, Humid Forest Ecoregional Center, B.P. 2008 (Messa) Yaounde´, Cameroon b Institut Nationale de Recherche Agricole pour le De´veloppement, B.P. 2067 Yaounde´, Cameroon Received 27 January 2004; received in revised form 5 July 2004; accepted 9 September 2004 Available online 2 October 2004

Abstract Three experiments were conducted on three soil types in south-central Cameroon to evaluate the effects of leguminous fallows on soil quality as compared with nonleguminous fallows. Soil quality was assessed by analysing the status of particulate organic matter (POM) fractions (4000–53 Am): (i) at the end of fallow and after cropping of 3-year-old Chromolaena odorata, fallow with C. odorata removed by hand, and Pueraria phaseoloides fallow; (ii) in soil from 1-year-old C. odorata and P. phaseoloides fallow, before and after 6 weeks of growing maize in a pot experiment, which had two treatments: T1= +P N and T2= +N P; and (iii) at the end of a 2-year-old Calliandra calothyrsus and a 2- and 4-year-old C. odorata fallow. Both, the herbaceous (P. phaseoloides) and tree (C. calothyrsus) leguminous fallows improved soil quality of a nonacidic Typic Kandiudult and a Rhodic kandiudult. The N content of either the coarse (4000–2000 Am) or the medium (2000–250 Am) POM fraction was increased as compared to the nonleguminous C. odorata fallow. This trend was also found after cropping all fallows. C. odorata fallow is better adapted to improve soil quality in the acidic Typic Kandiudox, than both leguminous fallows. N addition to soil from C. odorata fallow increased maize growth in the pot experiment as well as the weight of coarse POM (cPOM). P addition to soil from P. phaseoloides fallow had the same effect in the Rhodic Kandiudult, while a more pronounced response to P addition was found in soil from C. odorata fallow in the Typic Kandiudult. A negative effect on cPOM weight after P and N addition was mainly found in soil from P. phaseoloides fallow in the Typic Kandiudox. D 2004 Elsevier B.V. All rights reserved. Keywords: P. phaseoloides; C. odorata; C. calothyrsus; Fallow management; Organic matter fractionation; Tropical soils

1. Introduction * Corresponding author. Present address: Universite´ Libre de Bruxelles, Laboratoire de Ge´ne´tique et Ecologie Ve´getales, Chausse´e de Wavre 1850, B-1160 Brussels-Belgium. Fax: +32 2 650 9170. E-mail address: [email protected] (L.-S. Koutika). 0016-7061/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.geoderma.2004.09.009

In shifting cultivation systems of the humid tropics, the soil productivity of the inherently infertile soils is maintained through a long fallow cycle in which soil nutrients and organic matter are recycled by the self-

344

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

regenerating natural vegetation. Planted leguminous fallows have been suggested to improve soil productivity and to shorten the fallow period (Sanchez et al., 1997). Calliandra calothyrsus (Meissner) has been reported as one of the best leguminous woody perennials for soil fertility management in the humid tropics due to its adaptation to both acid and nonacid soils (MacQueen, 1993; Duguma et al., 1994). However, Koutika et al. (2004) have shown a deterioration of soil properties under C. calothyrsus compared with Chromolaena odorata and Pueraria phaseoloides in a Typic Kandiudult located in the central province of Cameroon. By contrast, Tian et al. (1999) found that a fallow planted to P. phaseoloides could stabilise soil organic C. Nguimgo and Balasubramanian (1992) found that a natural fallow of C. odorata maintained soil fertility and crop yields if properly managed, while Kanmegne et al. (1999) had shown that C. odorata residues decomposed quickly and lead to an improvement of soil properties. Benefits of legumes were also found by Fisher et al. (1994), who observed positive effects of introduced legumes on soil C accumulation in Columbia. Oikeh et al. (1998) found an improvement in supplying mineral N and enhancing nutrient availability and crop yields in soil under legume–maize rotation in a system where biomass was removed from the field after harvest. Tian et al. (1993) argued that the effects of plant residues on soil and crops would differ, depending on the rates of residue decomposition and nutrient release. These authors also found that the rapidly decomposing plant residues provided crops with a large amount of nutrients in the early stages of crop growth. Particulate organic matter (POM) is an active SOM fraction, which supplies nutrients to the growing plant. It is more responsive to changes in agricultural management than total SOM (Paustian et al., 1993). As such, POM was considered as an indicator of soil quality (Gregorich and Ellert, 1993). Koutika et al. (2001) found an improvement of soil quality by an increase in N content of POM fractions under two legumes, P. phaseoloides and Mucuna pruriens, compared to natural regrowth, mainly composed of C. odorata. Another study conducted in the fallow systems of southern Cameroon has also shown a soil quality improvement by an increase in N content of POM fractions (POM quality), which was more

important for maize growth than the POM weight (POM quantity), while groundnut and cassava yields were more responsive to POM quantity (Koutika et al., 2002). The present paper reports on the evaluation of soil quality through POM status in three soil types under two leguminous fallows, planted to P. phaseoloides and C. calothyrsus, and two nonleguminous fallows (one dominated by C. odorata and the other in which C. odorata was removed by hand). The objectives were the following: (i) to evaluate POM status under P. phaseoloides and fallow with or without the removal of C. odorata at the end of fallow period compared with the period after cropping (experiment 1) and after maize growing period of 6 weeks (experiment 2); (ii) to compare maize growth in soil from P. phaseoloides fallow versus soil from C. odorata fallow (experiment 2); and (iii) to determine soil quality under C. calothyrsus as compared with C. odorata fallow, 2 and 4 years old (experiment 3).

2. Materials and Methods 2.1. Collection and preparation of soil samples All samples came from a benchmark area in the forest margins of southern Cameroon. This area has an increasing deforestation gradient from south to north (Thenkabail, 1999). Rainfall increases from about 1650 mm yr 1 in the northern part to 1900 mm yr 1 in the south (Santoir and Bopda, 1995). In the north, soils are developed from gneiss and are typically Rhodic Kandiudults. Soils of the central part, developed on granites, are Typic Kandiudults, while those in the south, developed on pyroxene granites, are mostly Typic Kandiudox (Champetier de Ribes and Aubague, 1956). The soil samples came from three different experiments. The experiment 1 is a fallow succession trial in three villages: Nkometou (3840VN, 11840VE), Mvoutessi (3810VN, 11850VE), and Mengomo (2820VN, 11830VE). This experiment was established in 1995 and has three fallow types: (i) bC. odorataQ natural regrowth vegetation, with C. odorata making up 68% of its plant composition; (ii) bwithout C. odorataQ, with the same natural regrowth as in (i) but with manual removal of C. odorata 3–4 times per year, the fallow plant composition is now

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

dominated to 28% by Sida spp. (L.) and to 19% by Stachytarpheta cayennensis (L.C. Rich) Schau; and (iii) bP. phaseoloidesQ, replacement of natural regrowth by sown P. phaseoloides at a rate of 12 kg ha 1. The soil bulk density at soil sampling in 1998 across villages was on average between 1.1 and 1.3, ranging from 0.95 to 1.50. Soil for the experiment 2 was taken after a period of 1-year-old fallows of C. odorata and P. phaseoloides plots of experiment 1. The experiment 3 is a tree fallow experiment with different planting patterns of C. calothyrsus trees (Nolte et al., 2003). This experiment is located in six villages, in the abovementioned three villages as well as in Nkolfoulou (3856VN, 11835VE), Awae (3856VN, 11837VE), and Akok (2835VN, 11814VE). The equidistantly (1.61.6 m) planted C. calothyrsus plot was sampled along with a 2-year- and a 4-year-old natural fallow, dominated by C. odorata. The soil bulk density at trial establishment in 1996 across villages was on average between 1.0 and 1.2, ranging from 0.75 to 1.43. For the experiment 1, soil samples were collected at the end of a 3-year-fallow period (1998) and after cropping (1999). In each location (village), fallow types were in three treatment plots. In each plot, soil was taken in an 88-m subplot in five places, i.e., 1 m from each corner and in the middle of the plots.

345

Sampling was done with an auger in the 0–10 cm layer, replicated five times. The individual sample was made up of approximately 10 independent cores. In experiment 2, soil was sampled in 2000 from three fields after a 1-year-old fallow of C. odorata and P. phaseoloides, respectively, with the same soil sampling procedure as in experiment 1. In experiment 3, soil was randomly sampled in a field of 238 m2 (1219 m) in 1999 at the end of a 2-year (C. calothyrsus and C. odorata)- and 4-year (C. odorata)-fallow period. All soil samples were also taken with an auger in 0–10 cm depth, air-dried, and ground to pass a 4-mm mesh. 2.2. Standard soil analyses The following analyses were made: (i) cations (Ca, Mg, K) and P were extracted by the Mehlich-3 procedure (Mehlich, 1984). Cations were determined by atomic adsorption spectrophotometry and P by the malachite green colorimetric procedure (Motomizu et al., 1983). Soil pH was determined in water and in 1M KCl at a 2:5 soil solution ratio. Extractable aluminum was extracted using 1M KCl and analyzed colorimetrically using pyrocathecol violet. Organic C was determined by chromic acid digestion and a spectrophotometric procedure (Heanes, 1984). Total N was determined using the Kjeldahl method for

Table 1 Chemical properties of the different soil types under C. odorata, without C. odorata, and P. phaseoloides fallow Soil and fallow types

Total N (%)

C/N

A

B

A

B

A

B

A

B

A

B

A

B

Rhodic Kandiudult Chromolaena Without Chromolaena Pueraria

2.32 2.24 2.25

2.22 2.25 2.11

0.14 0.14 0.14

0.14 0.14 0.13

16.34 16.58 16.32

16.12 16.28 16.37

5.84 5.77 5.96

5.96b 5.63a 5.91ab

4.92 4.75 4.99

4.87b 4.47a 4.98b

0.92a 1.01b 0.96ab

1.10b 1.15c 0.94a

Typic Kandiudult Chromolaena Without Chromolaena Pueraria

1.53a 2.23b 1.98a

1.18a 1.71b 1.36a

0.11a 0.14c 0.12c

0.09a 0.12b 0.13b

14.37a 15.88ab 16.09a

13.73b 13.91b 10.62a

6.04a 6.32b 5.96a

6.20ab 6.41b 6.08a

5.19a 5.44b 5.14a

5.21a 5.55b 5.27a

0.85ab 0.88a 0.82a

0.99c 0.86b 0.80a

2.41 2.38 2.31 (0.13)

2.42a 2.56b 2.66b (0.09)

0.17b 0.15a 0.15a (0.00)

0.16 0.17 0.17 (0.00)

14.67 15.72 15.57 (0.51)

15.61 14.83 15.30 (0.46)

4.86 4.97 4.86 (0.15)

4.93a 5.26b 5.18ab (0.13)

3.89a 4.14b 3.81a (0.14)

4.02a 4.32b 4.28b (0.14)

0.97b 0.84a 1.05c (0.03)

0.91a 0.94b 0.90a (0.02)

Typic Kandiudox Chromolaena Without Chromolaena Pueraria

pH–H2O

DpH

Organic C (%)

pH–KCl

Values in parentheses are standard errors. Different letters indicate differences within soil and fallow type (A) at the end of 3-year-fallow and (B) after cropping.

346

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

Table 2 Concentrations of nutrient elements of the different soil types under C. odorata, without C. odorata, and P. phaseoloides fallow Soil and fallow types

Mehlich P (ppm)

Ca (cmol kg 1)

Mg (cmol kg 1)

K (cmol kg 1)

ECEC (cmol kg 1)

Al saturation (%)

A

B

A

B

A

B

A

B

A

B

A

B

Rhodic Kandiudult Chromolaena Without Chromolaena Pueraria

4.78ab 3.46a 4.95b

5.15b 2.48a 8.33c

3.97ab 3.43a 4.72b

3.51ab 2.56a 4.17b

1.04ab 0.89a 1.14b

1.04ab 0.62a 1.18b

0.10b 0.05a 0.08b

0.06 b 0.04a 0.08c

5.12ab 4.41a 5.99b

4.62ab 3.22a 5.44b

0.17a 0.97a 0.75a

6.16a 13.34b 5.60a

Typic Kandiudult Chromolaena Without Chromolaena Pueraria

2.64 2.97 2.63

3.65a 4.39b 3.75ab

2.66a 5.12c 3.37b

2.31a 4.50b 3.00ab

0.75a 0.92a 0.94b

0.64a 0.81ab 0.97b

0.07 0.08 0.10

0.05a 0.07b 0.07b

3.49a 6.15b 3.52a

3.01a 5.39c 4.04b

0.17 0.33 0.63

4.74 2.06 4.00

3.98a 5.14b 3.99a (0.84)

5.29b 6.31c 4.92a 0.92)

1.90ab 2.07a 1.59a (0.69)

2.28a 3.03b 2.83ab (0.54)

0.47 0.45 0.49 (0.10)

0.52a 0.78b 0.91c (0.13)

0.08 0.07 0.09 (0.01)

0.08a 0.08a 0.09b (0.00)

4.37b 3.71ab 3.03a (0.68)

2.88a 3.90c 3.84b ns

16.39a 36.53b 30.92ab (7.99)

30.20b 18.76a 20.58a (4.20)

Typic Kandiudox Chromolaena Without Chromolaena Pueraria

Values in parentheses are standard errors. Different letters indicate differences within soil and fallow types (A) at the end of 3-year-fallow and (B) after cropping. ns, not significant.

digestion and ammonium electrode determination (Bremner and Tabatabai, 1972; Nelson and Sommers, 1972). 2.3. Greenhouse pot experiment The pot experiment was established according to an experiment carried out by Vanlauwe et al. (2000), with some modifications. There were 2 treatments: (i) without N ( N +P treatment) and without P (+N P treatment). For a soil sample, six pots were filled each with 3 kg of soil. There were three pots of dwithout NT

and three of dwithout PT treatments. Each pot of the dwithout NT treatment received 0.333 g triple superphosphate, 1.279 g CaSO4.2H2O, and 300 ml of a nutrient solution containing all other macro- and microelements [2.307 g l 1 K2SO4, 0.124 g l 1 Na2B4O7 10 H2O, 0.144 g l 1 ZnSO4.7H2O, 0.144 g l 1 CuSO4.5H2O, 0.288 g l 1 MgSO4.7H2O, 4.19 g l 1 CoCl2.6H2O, 4.95 g l 1 Na2MoO4 2H2O]. Each pot of dwithout PT treatment received 1.279 g CaSO4.2H2O and 300 ml of a nutrient solution containing all other macro- and oligoelements [15.7289 g l 1 KNO3, 2.307 g l 1 K2SO4, 0.124 g l 1 Na2B4O7 10 H2O, 0.144 g l 1

Table 3 Soil characteristics before the pot experiment of the different soil types from C. odorata and P. phaseoloides fallow Soil and fallow types

C (%)

N (%)

C/N

pH

P (cmol kg 1)

Ca (cmol kg 1)

Mg (cmol kg 1)

K (cmol kg 1)

ECEC (cmol kg 1)

Rhodic Kandiudult Chromolaena Pueraria

2.20 2.29

0.18 0.19

12.41 11.91

5.89 5.67

4.38 4.05

4.36 4.05

1.25 1.25

0.11 0.10

5.72 5.40

Typic Kandiudult Chromolaena Pueraria

1.92 2.08

0.19 0.19

10.37 10.86

6.32 6.42

3.01 4.69

5.07 5.99

1.15 1.03

0.20 0.29

6.42 7.32

2.09 2.13 (0.14)

0.18 0.19 (0.01)

11.47 11.03 (0.36)

4.71 4.82 (0.12)

2.79 2.45 (0.64)

1.80 1.83 (0.51)

0.48 0.60 (0.15)

0.09 0.08 (0.03)

2.37 2.51 (0.64)

Typic Kandiudox Chromolaena Pueraria

Values in parentheses are standard errors.

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

347

Table 4 Chemical characteristics of the different soil types under C. calothyrsus, 2-year C. odorata, and 4-year Chromoalena fallow Fallow types

C (%)

N (%)

C/N

pH–H2O pH–KCl DpH

P Ca Mg ECEC (ppm) (cmol kg 1) (cmol kg 1) (cmol kg 1)

Rhodic Kandiudult Calliandra 2-year Chromolaena 4-year Chromolaena Typic Kandiudult Calliandra 2-year Chromolaena 4-year Chromolaena Typic Kandiudox Calliandra 2-year Chromolaena 4-year Chromolaena

2.33 2.28

0.13 0.14

17.10 16.07

5.27 5.14

4.43 4.25

0.83 0.89

4.14 4.36

2.46 1.97

0.92 0.71

3.49 2.81

2.58 (0.20) 2.05 2.31

0.16 (0.01) 0.14 0.15

15.80 (0.96) 14.98 15.05

5.27 (0.47) 5.78 5.93

4.36 (0.54) 4.95 5.02

0.89 (0.12) 0.82 0.90

5.28 (1.73) 6.40 5.64

2.53 (1.18) 3.83 4.33

0.91 (0.24) 1.07 1.07

3.58 (1.40) 5.01 5.52

2.07 (0.19) 2.48 2.46

0.14 (0.01) 0.16 0.16

14.48 (0.89) 15.78 15.45

5.57 (0.66) 5.13 5.29

4.68 (0.53) 4.15 4.39

0.89 (0.01) 0.97 0.90

4.42 (1.48) 5.56 8.74

3.30 (1.08) 2.67 4.30

0.97 (0.22) 0.77 0.80

4.38 (0.01) 3.57 5.23

2.74 0.17 16.38 4.88 (0.20) (0.01) (0.91) (0.46)

3.94 (0.53)

0.93 5.50 1.98 (0.11) (1.59) (1.12)

0.61 (0.23)

2.73 (1.34)

Soil types

Values in parentheses are standard errors.

ZnSO4.7H2O, 0.144 g l 1 CuSO4.5H2O, 0.288 g l 1 MgSO4.7H2O, 4.19 g l 1 CoCl2.6H2O, 4.95 g l 1 Na2MoO4 2H2O]. The pots were arranged in the greenhouse according to a randomised complete block design with three replicates. The soils were watered before planting four seeds of treated maize (CMS 8704) in each pot. After germination, the maize plants per pot were thinned to two. The two removed plants were weighed, dried, and analysed for organic C and total N. The plants were watered every two days with deionized water. After 6 weeks of growth, the maize biomass (above and below ground) was collected, dried at 65 8C, and weighed. Total N, Ca, Mg, K, and P were analysed for the above and below ground dry mass of maize. The soil from the pot experiment was dried and analysed for C, N, pH, Ca, Mg, K, ECEC, available P, and POM status. 2.4. POM fractionation POM fractionation was made according to Cambardella and Elliot (1992) and Vanlauwe et al. (1999). The soil samples were dried at 65 8C for 12 h, and 100 g of dry soil was dispersed in 100 ml of Nahexametaphosphate-Na-carbonate solution and 400 ml distilled water by shaking for 16 h on an endover-end shaker at 140 rev min 1. After dispersion, the suspension was wet-sieved to separate the 4000– 2000-, 2000–250-, 250–53-, and 53–20-Am fractions.

The following POM fractions were obtained through separating them from mineral material by decantation: coarse (4000–2000 Am), medium (2000–250 Am), and fine (250–53 Am). The three fractions and organomineral fraction (53–20 Am) were dried at 65 8C and weighed. Organic matter C and total N in all fractions was determined, as described above. Table 5 Soil characteristic changes expressed as percentages after pot experiment of the different soil types from C. odorata (Chr) and (Pue) with T1 (+P N) and T2 (+N P) treatments Soil and C fallow types

N

C/N pH P

Ca

Mg

K

ECEC

Rhodic Kandiudult Chr-T1 9 22 Chr-T2 6 17 Pue-T1 11 21 Pue-T2 4 21

34 25 42 31

7 1660 6 46 4 1045 3 60

74 59 88 57

28 263 25 1581 16 290 34 1800

55 70 67 68

Typic Kandiudult Chr-T1 19 21 Chr-T2 12 15 Pue-T1 6 26 Pue-T2 1 21

41 32 43 28

7 1917 7 74 7 1507 4 56

67 48 54 32

22 23 24 26

52 65 44 47

Typic Kandiudox Chr-T1 14 Chr-T2 3 Pue-T1 1 Pue-T2 4

36 18 36 26

5 1411 116 3 105 70 1 1544 102 2 82 61

17 17 26 16

105 980 82 600

10 422 102 25 2111 127 26 487 83 21 2150 107

348

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

2.5. Statistical analyses The mixed model procedure of SAS (1989) was used to analyse all data of the experiments 1 and 3. Treatment means were compared using their least square means. The differences mentioned in the text are significant at pb0.05. In the experiment 2, statistical analysis was made using a mixed procedure for a 322 factorial design (SAS, 1989).

3. Results 3.1. General soil characteristics The lowest C and N contents were found in the Typic Kandiudult at the end of a 3-year fallow (Table 1). DpH under P. phaseoloides was higher in the Typic Kandiudox than in the Rhodic and Typic Kandiudult at the end of fallow. Table 2 shows that the highest p

value was found after cropping under P. phaseoloides fallow and the lowest under without C. odorata in the Rhodic Kandiudult. ECEC under P. phaseoloides in the Typic Kandiudox was lower than under the same fallow type in the Typic and Rhodic Kandiudult at the end of fallow and after cropping. The lowest Al saturation was found in the Typic Kandiudult and the highest in the Typic Kandiudox at the end of fallow. This pattern remained after cropping, but Al saturation increased. The highest C/N ratio was under C. odorata on Rhodic Kandiudult at the start of the pot experiment (Table 3). The lowest values of nutrients concentrations and ECEC were found in the Typic Kandiudox. In experiment 3, C under 4-year C. odorata (2.74) in the Typic Kandiudox was higher than under 4-year C. odorata (2.07%) in the Typic Kandiudult (Table 4). The C/N ratio was higher under C. calothyrsus than under 4-year C. odorata fallow on the Typic Kandiudult. Ca was higher under 2-year C.

Fig. 1. Weight of tPOM and its fractions in different soil and fallow types; C. odorata=1, without C. odorata=2, and P. phaseoloides fallow=3 (different letters denote significant differences at pb0.05).

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

349

Fig. 2. C of POM fractions as a percentage of soil C in different soil and fallow types; C. odorata=1, without C. odorata=2, and P. phaseoloides fallow=3 (different letters denote significant differences at pb0.05).

odorata (4.30 cmol.kg 1) than under 4-year C. odorata (1.98 cmol.kg 1) in the Typic Kandiudox (Table 4). In the pot experiment, total N decreased while organic C and the C/N ratio increased in both treatments (Table 5). Addition of superphosphate in T1 increased P by more than 1000%. Soil Ca

increased mainly in T1 with a higher increase in the Typic Kandiudox (superphosphate), whereas soil Mg decreased. Soil K increased by more than 500% in all T2 (due to a large addition of KNO3). This increase in soil exchangeable K was higher under P. phaseoloides than under C. odorata in the Rhodic Kandiudult.

Fig. 3. N of POM fractions as a percentage of soil N in different soil and fallow types; C. odorata=1, without C. odorata=2, and P. phaseoloides=3 (different letters denote significant differences at pb0.05).

350

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

3.2. POM status In the experiment 1, the weight of total POM (tPOM) and fine POM (fPOM) under C. odorata and P. phaseoloides decreased from cropping in the Rhodic Kandiudult (Fig 1a,b). In the Typic Kandiudult, the weight of tPOM sharply decreased under C. odorata and without C. odorata (Fig. 1c), while the decrease was less pronounced under the same fallow types for the fPOM (Fig. 1d). In the Typic Kandiudox, a significant increase in weight was only found for mPOM (Fig. 1e). The tPOM-C decreased under C. odorata, while an increase was noticed under P. phaseoloides in the Typic Kandiudult (Fig. 2a). The fPOM-C increased in C. odorata and P. phaseoloides fallows in the Typic Kandiudult (Fig. 2b). In the Typic Kandiudox, C increased in all POM fractions of C odorata and P. phaseoloides after cropping (Fig. 2c–f). In the Rhodic Kandiudult, only the coarse POM (cPOM)-N increased under P. phaseoloides (Fig.

3a). In the Typic Kandiudult, tPOM-N and cPOMN decreased under C. odorata and without C. odorata, while no significant change was noticed under P. phaseoloides (Fig. 3c,d). In the Typic Kandiudox, N contents of all POM fractions increased under C. odorata, whereas no change was noticed under without C. odorata (Fig. 3e–g). It was only fPOM-N which increased for the P. phaseoloides and without C. odorata fallow types (Fig. 3h). At the end of the pot experiment, there was an increase of cPOM weight in Chr-T2 and Pue-T1 of the Rhodic Kandiudult (Fig. 4a) and in Chr-T1 and Pue-T1 of the Typic Kandiudult (Fig. 4b). The weight of cPOM decreased in Chr-T2 and Pue-T1 of the Typic Kandiudox (Fig. 4c). Significant increase in C content was only in fPOM of both treatments from P. phaseoloides fallow of the Typic Kandiudox (Fig. 4d). In experiment 3, the mPOM-C was higher under the 4-year C. odorata than under the other fallow in

Fig. 4. Weight and C of POM fractions in different soil and fallow types, C. odorata (Chr) and P. phaseoloides (Pue) with T1 (+P (+N P) treatments (Chr-T1=1, Chr-T2=2, Pue-T1=3, and Pue-T2=4; different letters denote significant differences at pb0.05).

N) and T2

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

351

Fig. 5. C and N of POM fractions as a percentage of soil C and N in the Rhodic Kandiudult=1, Typic Kandiudult=2, and Typic Kandiudox=3 under C. calothyrsus, 2-year C. odorata, and 4-year Chromoalena fallow (different letters denote significant differences at pb0.05).

the Typic Kandiudox and lower in the Rhodic Kandiudult (Fig. 5a). The tPOM-N under the 4-year C. odorata was lower than those of the other fallow in

the Rhodic Kandiudult (Fig. 5b). The mPOM-N under C. calothyrsus was higher than under 4-year C. odorata fallow in the Typic Kandiudult (Fig. 5c).

Fig. 6. Dry matter (a), root weight (b), and Mg concentrations of maize plants(c), grown on Rhodic Kandiudult, Typic Kandiudult, and Typic Kandiudox soil after C. odorata (Chr) and P. phaseoloides (Pue) fallow with T1 (+P N) and T2 (+N P) treatments (different letters denote significant differences at pb0.05).

352

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

3.3. Characteristics of above- and belowground biomass of maize in the second experiment The dry matter of maize plants was higher in T1 treatments of the Rhodic Kandiudult and Typic Kandiudult (Fig. 6a). The dry matter of maize roots and Mg concentration was significantly higher in T1 from both fallow types of the Rhodic Kandiudult (Fig. 6b,c).

4. Discussion

fallow. The positive effect of C. calothyrsus on tPOM-N was greatest in the Rhodic Kandiudult, likely because these soils are more degraded. Population pressure is higher, and agriculture is more intensified in the area where Rhodic Kandiudults predominate. However, C. calothyrsus did not have a benefit on POM status compared to a 2-year C. odorata. A previous study has shown a deterioration of nutrient concentrations and POM status of C. calothyrsus compared with P. phaseoloides and C. odorata in a Typic Kandiudult (Koutika et al., 2004). No effect of C. calothyrsus on tPOM-N nor of mPOM-N was found in the acidic Typic Kandiudox.

4.1. POM assessment Cropping induced a decrease of tPOM-N and cPOM-N, mainly under fallow with or without removal of C. odorata, while no significant change was found under P. phaseoloides. Moreover, in the Typic Kandiudox, the tPOM-N and its fractions increased after cropping under C. odorata. This shows that, to improve soil quality, i.e., N contents of POM, P. phaseoloides is the adapted fallow on the nonacidic Rhodic and Typic Kandiudult, whereas C. odorata is the more adapted fallow type on the acidic Typic Kandiudox. These results are in agreement with previous findings (Koutika et al., 2002) showing that P. phaseoloides has the capability to increase the N concentration of POM fractions in soils with relatively low acidity. But in soils with chemical constraints (high acidity and Al saturation), C. odorata was more adapted than P. phaseoloides fallow (Koutika et al., 2002). Furthermore, according to the increase of cPOM weight in the pot experiment, the Rhodic Kandiudult showed a positive response to N addition under C. odorata in contrast to a P response after adding P to the soil from P. phaseoloides fallow. The Typic Kandiudult, however, responded positively to P addition under C. odorata, which confirms the beneficial effect of P. phaseoloides on soil quality of nonacidic soils. The negative response of both P and N addition on cPOM weight in the Typic Kandiudox was mainly found after P. phaseoloides fallow, which corroborates that C. odorata is the more adapted fallow type in acidic soils. C. calothyrsus increased the tPOM-N in the Rhodic Kandiudult and the mPOM-N in the Typic Kandiudult compared with the 4-year C. odorata

4.2. Effect of N/P fertiliser application, soil type, P. phaseoloides and C. odorata fallow on maize growth P-fertiliser application increased maize growth most on the Rhodic and Typic Kandiudult, the more degraded soils of the studied area. Concurrently, available P in the soil was increased by more than 1000%. Mg, K, and P concentrations in the maize plant increased notably in the aboveground biomass, mainly on the Rhodic and Typic Kandiudult. These results corroborate previous findings, which revealed that low nutrient concentrations, high soil acidity, and Al saturation are the limiting factors for crop production in Typic Kandiudox (Koutika et al., 2002). Only by alleviating chemical constraints or adopting varieties resistant to soil acidity could higher crop yields be attained on these soils. Maize responded more to fertiliser application than to soil and fallow types. 4.3. Changes in general soil characteristics Soil organic C and C/N ratio decreased in the Typic Kandiudult. This soil type probably lost its C at the end of the fallow and after cropping due to its coarse texture, which enhances C decomposition as found in a previous study (Koutika et al., 2002). Organic C decreased by 30% under natural fallow with and without C. odorata removal and by 46% under P. phaseoloides from the end of fallow to the period after the cropping. The decrease of total N was less pronounced and was only noted in natural fallows with (22%) or with lowered (17%) C. odorata in the fallow vegetation. These results confirm that soil

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

under P. phaseoloides fallow might provide more N to the growing plant during the cropping cycle than natural fallows composed of C. odorata. The beneficial effect of P. phaseoloides on soil quality was also found by a reduction of soil acidity. In fact, a decrease in DpH, which indicates the amount of functional acidity, confirmed the beneficial effect of P. phaseoloides on soil acidity even after cropping. The increase of P availability, C/N ratio, and pH value after cropping might be due to interactions between soil and roots of crops, such as groundnut that might enhance soil N content. However, after cropping, soil exchangeable K and ECEC decreased, while Al saturation increased especially in the Typic Kandiudult. The largest decrease was noticed with K, which is one of the limiting nutrient elements for starchy crops in the area. A study, comparing soils of the studied area with soils of the derived savannah in Nigeria, showed that soil K is much higher in the savannah soils (Vanlauwe et al., 2000). The increase of Al saturation was more pronounced in the Rhodic Kandiudult, probably due to soil degradation as a result of agricultural intensification. However, Al saturation remained highest in the Typic Kandiudox. The increase in pH values of Typic Kandiudox under C. odorata after the pot experiment might indicate that C. odorata fallow is more adapted in these acidic soils to correct soil acidity than P. phaseoloides.

5. Conclusions P. phaseoloides fallow did improve soil quality by increasing nutrient concentrations and N content in POM fractions in nonacidic soils at the end of the fallow period, which lasted through the cropping period. A 2-year C. calothyrsus fallow also showed a beneficial effect on POM-N, but only vis-a`-vis a 4year C. odorata. Both leguminous fallows, P. phaseoloides and C calothyrsus, seem to be more adapted for nonacidic soils. By contrast, C. odorata natural fallow seems to be the more adapted fallow type on acid soils. These results were confirmed by the pot experiment, which showed a positive response of maize growth in Rhodic and Typic Kandiudult after addition of P and N. This was

353

mainly found in soil from C. odorata. A negative response was found in the acidic Typic Kandiudox, mainly in soil from a P. phaseoloides fallow. Thus, longer fallows or mineral fertilisers, combined with an appropriate fallow period, would be necessary to improve crop growth and yield on Rhodic and Typic Kandiudult soils. An alleviation of chemical constraints or the adoption of acid-tolerant varieties seems to be necessary on acidic Typic Kandiudox soils. Further research should focus on the type of chemical reactions that optimise plant nutrition in acid soils of the study area.

Acknowledgments The authors thank the staff of the chemical and SOM laboratories at IITA-HFC in Yaounde´, Cameroon, for the analyses.

References Bremner, J.M., Tabatabai, M.A., 1972. Use of an ammonia electrode for determination of ammonium in Kjeldahl. Commun. Soil Sci. Plant Anal. 3, 159 – 165. Cambardella, C.A., Elliot, E.T., 1992. Particulate soil organic matter changes across a grassland cultivation sequence. Soil Sci. Soc. Am. J. 56, 777 – 783. Champetier de Ribes, G., Aubague, M., 1956. Carte geologique de reconnaissance a` l’e´chelle de 1/50000. Feuille de Yaounde´-Est avec notice explicant, Paris. Duguma, B., Tonye, J., Kanmegne, J., Manga, T., Enoch, T., 1994. Growth of ten multipurpose tree species on acid soils of Sangmelima, Cameroon. Agrofor. Syst. 27, 107 – 119. Fisher, M.J., Rao, I.M., Ayarza, M.A., Lascano, C.E., Sanz, J.I., Thomas, R.J., Vera, R.R., 1994. Carbon storage by introducing deep-rooted grasses in the South American savannas. Nature 371, 236 – 238. Gregorich, E.G., Ellert, B.H., 1993. Light fraction and macroorganic matter in mineral soils. In: Carter, M.R. (Ed.), Soil Sampling Methods of Analysis. Lewis Publisher, Boca Raton, pp. 175 – 203. Heanes, D.L., 1984. Determination of total organic C in soils by an improved chromic acid digestion and spectrometric procedure. Commun. Soil Sci. Plant Anal. 15, 1191 – 1213. Kanmegne, J., Duguma, B., Henrot, J., Isirimah, N.O., 1999. Soil fertility enhancement by planted tree-fallow species in the humid lowlands of Cameroon. Agrofor. Syst. 46, 239 – 249. Koutika, L.-S., Hauser, S., Henrot, J., 2001. Soil organic matter assessment in natural regrowth, P. phaseoloides phaseoloides and Mucuna pruriens fallow. Soil Biol. Biochem. 33, 1095 – 1101.

354

L.-S. Koutika et al. / Geoderma 125 (2005) 343–354

Koutika, L.-S., Sanginga, N., Vanlauwe, B., Weise, S., 2002. Chemical properties and soil organic matter assessment in fallow systems in the forest margins benchmark. Soil Biol. Biochem. 34 (6), 757 – 765. Koutika, L.-S., Meutem Kamga, J.G., Yerima, B., 2004. Comparative study of soil properties under Chromolaena odorata, Pueraria phaseoloides and Calliandra calothyrsus. Plant Soil (in press). MacQueen, J.D., 1993. C. calothyrsus series Racemosae: taxonomic information; OFI seed collections; trial design. Oxford Forestry Institute, Department of Plant Sciences. University of Oxford, Oxford. 157 pp. Mehlich, N.W., 1984. Mehlich 3 soil test extractant: a modification of the Mehlich 2 extractant. Commun. Soil Sci. Plant Anal. 15, 1409 – 1416. Motomizu, S., Wakimoto, P., Toei, K., 1983. Spectrophotometric determination of phosphate in river waters with molybdate and malachite green. Analyst (London) 108, 361 – 367. Nelson, D.W., Sommers, L.E., 1972. A simple digestion procedure for estimation of ammonium in Kjeldahl soils. J. Environ. Qual. 1, 423 – 425. Nguimgo, K.A.B., Balasubramanian, V., 1992. Effect of fallow and residue management practices on biomass production, weed suppression and soil productivity. In: Mulongoy, K., et al., (Eds.), Biological Nitrogen Fixation and Sustainability of Tropical Agriculture. Wiley, Chichester, pp. 463 – 473. Nolte, C., Tiki-Manga, T., Badjel-Badjel, S., Gockowski, J., Hauser, S., Weise, S.F., 2003. Effects of C. calothyrsus planting pattern on biomass production and nutrient accumulation in planted fallows of southern Cameroon. For. Ecol. Manag. 179, 535 – 545. Oikeh, S.O., Chude, V.O., Carsky, R.J., Weber, G.K., Horst, W.J., 1998. Legume rotation in the moist tropical savanna: managing soil nitrogen dynamics and cereal yields in Farmers’ fields. Exp. Agric. 34, 73 – 83.

Paustian, K.E., Elliot, T., Collins, H.P., Cole, C.V., Paul, E.A., 1993. Changes in active C fractions as a function of land management practices. Agronomy Abstracts. Madison, Wisconsin, pp. 257. Sanchez, P.A., Mokwunye, A.U., Kwesiga, F.R., Ndiritu, C.G., Woomer, P.L., 1997. Soil fertility replenishment in Africa: an investment in natural resource capital. Replenishment soil fertility in Africa, SSSA Special Publication, pp. 2 – 46. Santoir, C., Bopda, A., 1995. Atlas Regional Sud-Cameroun. ORSTOM Editions, Paris, pp. 3. SAS, 1989. (Fourth edition). SAS Stat User’s Guide, Version 68 vol. 2. SAS Institute, Cary, NC. 846 pp. Thenkabail, P.S., 1999. Characterization of the alternative-to-slashand-burn benchmark research area representing the Congolese rainforests of Africa using near-real-time SPOT HRV data. Int. J. Remote Sens. 20, 839 – 877. Tian, G., Kang, B.T., Brussaard, L., 1993. Mulching effect of plant residues with chemically contrasting compositions on maize growth and nutrient accumulation. Plant Soil 153, 179 – 187. Tian, G., Kolawole, G.O., Salako, F.K., Kang, B.T., 1999. An improved cover crop-fallow system for sustainable management of low activity clay soils of the tropics. Soil Sci. 164, 671 – 682. Vanlauwe, B., Aman, S., Aihou, K., Tossah, B.K., Adebeyi, V., Sanginga, N., Lyasse, O., Diels, J., Merckx, R., 1999. Alley cropping in the moist savanna of West Africa: III. Soil organic matter fractionation and soil productivity. Agrofor. Syst. 42, 245 – 264. Vanlauwe, B., Aihou, K., Aman, S., Tossah, B.K., Diels, J., Lyasse, O., Hauser, S., Sanginga, N., Merckx, R., 2000. Nitrogen and phosphorus uptake by maize as affected by particulate organic matter quality, soil characteristics, and land use history for soils from the West African moist savannah zone. Biol. Fertil. Soils 30, 440 – 449.