Nutr Cycl Agroecosyst (2006) 75:247–255 DOI 10.1007/s10705-006-9031-0

ORIGINAL PAPER

Wheat responses in semiarid Northern Ethiopia to N2 fixation by Pisum sativum treated with phosphorous fertilizers and inoculant K. Habtegebrial Æ B. R. Singh

Received: 31 January 2006 / Accepted: 31 May 2006 / Published online: 1 August 2006
Springer Science+Business Media B.V. 2006

Abstract Nitrogen fixation (N2) by leguminous crops is a relatively low-cost alternative to N fertilizers for smallholder farmers in Africa. Nitrogen fixation in pea (Pisum sativum L. cv. Markos) as affected by phosphorus (P) fertilization (0, 30 kg P ha–1) and inoculation (uninoculated and inoculated) in the semiarid conditions of Northern Ethiopia was studied using the 15N isotope dilution method and locally adapted barley (Hordeum vulgare L. cv. Bureguda) as reference crop. The effect of pea fixed nitrogen (N2) on yield of the subsequent wheat crop (Triticum aestivum L.) was also assessed. Phosphorus and inoculation significantly influenced nodulation at the late flowering stage and also significantly increased P and N concentrations in shoots, and P concentration in roots, while P and N concentrations in nodules were not affected. Biomass, pods m–2 and grain yield responded positively to P and inoculation, while seeds pod–1 and seed weights were not significantly affected by these treatments. Phosphorus and inoculation enhanced the

percentage of N derived from the atmosphere in the whole plant ranging from 53 to 70%, corresponding to the total amount of N2 fixed varying from 55 to 141 kg N ha–1. Soil N balance after pea ranged from – 9.2 to 19.3 kg N ha–1 relative to following barley, where barley extracted N on the average of 6.9 and 62.0 kg N ha–1 derived from fertilizer and soil, respectively. Beneficial effects of pea fixed N2 on yield of the following cereal crop were obtained, increasing the average grain and N yields of this crop by 1.06 Mg ha–1 and 33 kg ha–1, respectively, relative to the barley– wheat monocrop rotation. It can be concluded that pea can be grown as an alternative crop to fallow, benefiting farmers economically and increasing the soil fertility. Keywords Crop rotation Æ Inoculation Æ 15N dilution method Æ Phosphorus fertilization Æ Pisum sativum Æ Soil nitrogen balance

Introduction K. Habtegebrial Æ B. R. Singh Department of Plant and Environment Sciences, Norwegian University of Life Sciences, 5003, N-1432 ˚ s, Norway A K. Habtegebrial (&) Department of Land Resource Management, Mekelle University, 231, Mekelle, Ethiopia e-mail: [email protected]

Ethiopia is the largest producer of cool-season food legumes (CSFL) in Africa (Yemane and Skjelva˚g 2003). Among the CSFL, a pea variety, locally called Markos (Pisum sativum cv. Markos) is grown by subsistence farmers in Ethiopia during the cool season (June–September). Pea is the second most important food legume in Ethiopia,

123

248

next to fababean occupying about 19% (2,54,365 hectares) of the total cultivated land area under pulses (CSA 2005). The dry grain of this legume has been an important protein source in human diet for centuries, where animal protein is expensive or not sufficiently available. The straw of the crop is also used as animal fodder and as soil fertility restorer. Yields of pea are very low, mainly limited by poor soil fertility (Tsigie and Woldeab 1994), as it is cultivated in inherently poor soils often without fertilization. The potentially limiting nutrient in crop production in the high land plateau soils of Ethiopia is phosphorus (P) (Miressa and RoBarge 1999). Phosphorous is needed in relatively large amounts by legumes mainly for growth and for symbiotic N fixation. Phosphorus, in addition to promoting host plant growth, has specific effects on N2 fixation and nodule initiation, growth, development, and metabolic function (Jakobsen 1985; Leidi and Rodiguez-Navarro 2000). Nowadays, fallowing is a rare farming practice in Ethiopia due to high land pressure. Grain legumes could serve as alternative to fallow and in turn increase land use, weed control and reduce the need for fertilizers (Carlos and Minguez 2001). Most soils in Ethiopia are inherently low in organic matter and nitrogen (Pulschen 1987). Grain legumes could serve as possible alternatives to commercial N fertilizers, and legume– cereal rotation can be additional sources of income for the farmers and could provide a net input to soil N. Symbiotic N2 fixation is an important process for increasing the plant available N and is markedly increased by P fertilization and by inoculation using efficient, competitive and persistent strains of rhizobium (Giller 2001; Nuruzzaman et al. 2005). Grain legumes cause significant positive yield effects on the subsequent non-legumes when compared with rotations with non-legumes. Substantial dry matter and grain yield increases were obtained when cereals cropped after grain legumes (Amanuel and Tanner 1991; Chalk 1998; Jensen et al. 2004). The positive cereal response to N has been attributed to the transfer of biological fixed N, to N-sparing under the antecedent legume, and to less immobilization of nitrate during the decomposition of legume residue

123

Nutr Cycl Agroecosyst (2006) 75:247–255

(Chalk 1998). Legumes are also found to render phosphate more available to the subsequent crops and restrict phosphate fixation (Nuruzzaman et al. 2005). In Ethiopia, information on the biological N fixation capacity of grain legumes under rain fed field conditions is very scarce. Furthermore, with the abandonment of fallow practice and the limited access to mineral N fertilizers, the introduction of grain legumes by resource poor farmers as fallow crop is significantly advantageous. Thus, this study was aimed to evaluate the effect of Pfertilization and inoculation on yield, nodulation and N2 fixation capacity of Markos pea and the yield impact of fixed N2 on the next wheat crop under semi arid conditions of Northern Ethiopia.

Materials and methods Study site The experiment was conducted in the 2004 and 2005 growing seasons at the Mekelle University (MU) Campus, Tigray, Northern Ethiopia at 1314¢ N and 3932¢ E, and 2,100 meter above sea level. This site is located in a semiarid agro-ecological zone. Annual rainfall ranges between 200 and 700 mm. The rainfall distribution is bimodal with the short rainfall season (March–April) and the main rainy season (June–September). This pattern, however, is extremely variable with high probability of no rainfall during the short rainfall season. Monthly minimum and maximum temperatures (C) and rainfall (mm) were recorded for the growing seasons of 2004 and 2005. The annual rainfall of Mekelle for 2004 and 2005 were about 460 mm and 669 mm, respectively. Most of the rains (about 68–70%) came in the months of July and August (Fig. 1). The average annual rainfall for 2004 was below the long-term average of 600 mm. The average maximum and minimum temperatures for the growing seasons (June–October) in 2004 and 2005 were 24.7, 25.6, 13.3 and 13.8C respectively. Both minimum and maximum temperatures, in general did not differ significantly from the 35-year long-term average (1967–2003) for the area.

Nutr Cycl Agroecosyst (2006) 75:247–255 350

Rainfall (mm)

300

249

RF(2004) RF(2005)

250 200 150 100 50 0 Feb.

Mar.

Apr.

May

Jun

Jul.

Aug. Sept.

Months

Fig. 1 Monthly rainfalls (RF) for the 2004 and 2005 cropping seasons at Mekelle University

The land used for the experiment in 2004 was fallow for 3 years prior to planting. The plots used in the second year were adjacent to the plots used in the first year but was left fallow for the next experiment. The soil was clayey in texture and contained 18% sand, 30% silt and 52% clay. The soil had a pH of 7.2 and conductivity of 2 dS m–1, and contained 9.7 g kg–1 SOC, 0.8 g kg–1 total N, 6.99 mg kg–1 available P, 191 mg kg–1exchange–1 able potassium (K), 1.28 g kg exchangeable calcium (Ca) and < 1.5 mg kg–1 extractable molybdenum (Mo). Experimental design and treatments The experimental design was laid out in randomized complete blocks with three replications. The macro plots were 4 m by 4 m (16 m2) in size. Micro plot sizes were 1.5 m by 1.2 m (1.8 m2) for pea, and 1.2 m by 1.3 m (1.6 m2) for barley. Each micro plot was located in the center of individual macro plots. Two levels of inoculation (uninoculated and inoculated) and two levels of P fertilization (0 and 30 kg P ha–1) in factorial combinations were applied. Seeds of a pea variety called Markos ater (Pisum sativum cv. Markos) were obtained from the Ethiopian Seed Agency (ESA). Locally adapted barley (Hordeum vulgare L.cv. Bureguda) was used as a reference crop. The inoculants used were Rhizobium L.Vicia strain EAL-300 obtained from the National Soil Research Laboratories, Microbiology Unit (Ethiopia, Addis Ababa). The legume seeds were sown by hand with spacing

between plants and rows of 5 and 30 cm, respectively, as recommended by ESA. The final plant density was about 60 plants m–2. The reference crop was also sown by hand at 20 cm row spacing. The dates of sowing were July 10 and July 14 in 2004 and 2005, respectively. In the 2004 crop season, supplemental irrigation during mid- flowering and pod-filling phases was applied to reduce the effect of terminal drought. The legume seeds were inoculated at the time of sowing with a powder containing an equivalent of 108 viable bacterial cells g–1 of powder. Prior to inoculation, the seeds were surface sterilized. The seeds were washed in 70% (v/v) ethanol for 5 min, rinsed twice with sterilized water, shaken for 15 min in 30% (v/v) H2O2 (hydrogen peroxide), and rinsed with sterilized water four times. The seeds were soaked over night in distilled water and made ready for inoculation and sowing. Ten days after emergence of legume seeds, the micro plots for both the legume and barley crops were fertilized uniformly with 15 20 kg N ha–1 using 15NH15 N atomic 4 NO3 (5% excess) in a solution form (fertilizer dissolved in 450 ml H2O). The macro plots received also 20 kg N ha–1 as unlabeled urea. As basal dressing each plot was fertilized at sowing with 33 kg K ha–1 in the form of KCl. Hand weeding of pea plots was carried out twice before the plants got entangled with tendrils. Similar hand weeding was also applied to barley plots. The rotation crop wheat (HAR-1685) was sown on July 23 in the next cropping season (2005) after pea and barley plots. Crushed pea straw from the previous season was incorporated into the respective plots one month before sowing. Soil analyses Soil samples were analyzed for pH (1:2.5 soil: water ratio), electrical conductivity (ECe) of the saturated paste extract, soil organic carbon by Walkley and Black method (SOC) (Nelson and Sommers 1982), total N concentration as total Kjeldhal N. Available P was extracted using 0.5M NaHCO3 at pH 8.5 (Olsen et al. 1954), and extractable K and Ca were analyzed according to Knudsen et al. (1982) and Lanyon and Walter (1982). Extractable Mo was extracted with Tamm’s solution and determined by Perkin

123

250

Elmer Optima 3000 DV ICP- atomic emission spectroscopy (Haley and Melsted 1957). Plant sampling and analysis The effect of inoculation and P fertilization on nodulation and yield were assessed for both years. Nitrogen fixation potential using the 15N dilution method was carried out for the 2004 crop season only. Nodulation was assessed at late flowering stage of Markos pea. Ten randomly selected plants from each macro plot were uprooted. Soil adhering to the roots was removed by washing with tap water. The nodules remaining in the soil around the plant roots were picked by hand and washed with water. Nodules attached to each plant root were also removed and separately spread on a sieve for some minutes until the water has drained from the surface of the nodules. The total number, fresh weight and volume of the nodules were recorded. The shoots, roots and nodules of plants were separated and were oven dried at 70C for 72 h. The dry samples of these plant parts were ground to pass 0.5 mm sieve. Plant parts from the macro plots were analyzed for total N and P. Total plant N was analyzed as given in Bremer and Mulvaney (1982), using a total Leco CHN 1000 analyzer (Leco CHN 1000). Total plant P was analyzed using Thermo Jarrel Ash Polyscan 61EICPAtomic Emission Spectroscopy after the plant parts were prepared by multi wave microwave digester as described in Paar (1998). At physiological maturity, plants were harvested from 1 m2 micro plots, so that the outer rows were not used for both the legume and reference barley crops. They were divided into straw and pods or heads. These plant parts were oven dried at about 70C for 48 h. The plant material was ground to pass through a 0.5 mm sieve. Total N and % 15N atomic excess (a.e) of plant samples were analyzed at the Institute of Energy Technology (Norway) using a micromass optima isotope ratio mass spectrometer. The % 15N excess was calculated by the difference of the atomic % 15 N in the plant material (straw in legumes and barley, and pods in legumes and spikes in barley) and that of the natural abundance in the atmo-

123

Nutr Cycl Agroecosyst (2006) 75:247–255

sphere (0.3663%). The biological N2 fixation in Markos pea and the distribution of labeled N in the soil–plant system were calculated using the 15 N isotope dilution method according to Hardarson and Danso (1990), using Eqs. 1–4 below. The amount of N2 fixed was determined from the product % Ndfa and N yield for each replication, and the average was then estimated. Soil N balance (B) after pea production, was calculated by subtracting N output from N input using equation-5 (Amanuel et al. 2000). Roots were not removed from the soil, and the calculated potential N benefits are conservative, as they do not include root N. %Ndff pea=barley ¼ 100
½ð%15 N a.e.ðpea/barleyÞ= %15 N a.e.ðFertilizerÞ

ð1Þ

%Ndfa ¼ 100
½ð1  %15 N a.e.ðpeaÞ= %15 N a.e.ðbarleyÞ

ð2Þ

N2 fixedðkg ha1 Þ ¼ ½ð%Ndfa  total N in peaðkg ha1 Þ=100 ð3Þ

%Ndfspea ¼ 100  %Ndff pea  %Ndfa

ð4Þ

B ¼ ðNf þ N2 fixedÞ  Ng

ð5Þ

where: % Ndff is the percentage of nitrogen derived from fertilizer, % Ndfa is the percentage of nitrogen derived from the atmosphere and % Ndfs is the percentage of nitrogen derived from the soil, and B is soil N balance. The subscripts f and g denote applied fertilizer (kg ha–1) and the N removed by pea grain or by wheat straw and grains, respectively. Yield and yield components Plants harvested from 1 m2 areas of the macro plots were used for recording yield and yield component data. Grain yields and pods m–2 were obtained by collecting all pods from these plants.

Nutr Cycl Agroecosyst (2006) 75:247–255

251

Number of seeds pod–1 was counted by randomly picking 20 pods from each yield plot. Statistical analyses Field data were statistically analyzed using SAS (2002). Analyses of variance (ANOVA) were performed to evaluate the effects of treatments (inoculation and P rates), years and their interactions on plant characteristics and yield parameters. Comparisons among treatment means were made by the Duncan multiple range tests. Multiple stepwise regressions and correlation analyses between yield and yield components were also conducted.

Results Yield and yield components The dry matter and grain yields of 2005 were greater than those of 2004 but they did not differ significantly between the 2 years. However, N yield of grain for 2005 was significantly higher than that of 2004. Inoculation and P fertilization improved significantly the dry matter and grain yields, pods plant–1 and N yield of grains (Table 1). Grains pod–1 and single grain weight are not affected by inoculation and P fertilization. The dry matter yield, grain yield and pods plant–1 were improved by 44, 77 and 78%, respectively, relative to the control by the application of inoculation and P fertilization. Grain yield was strongly correlated with number of pods plant–1

(r = 0.85). Inoculation and P fertilization improved significantly the N yield of grain by increasing it to 113% relative to the control. No interaction between inoculation and P fertilization with respect to yield and yield components were observed. Nodulation and tissue nitrogen and phosphorus concentrations In both years and across all treatments, nodule color was invariably pink indicating rhizobium infection of the roots. But inoculation and P fertilization significantly increased the nodule number, nodule fresh weight and volume (Table 2). Inoculation increased the nodule number and nodule mass by 48 and 71%, and P fertilization by 35 and 60%, respectively. Nodules had the highest relative N and P concentrations, followed by shoot and root in that order (Table 2). Phosphorus fertilization did not affect N and P concentrations of nodules, while it enhanced N concentration of shoots and roots. Inoculation enhanced the N concentration of shoots only. A positive interaction effect between inoculation and P fertilization was observed in the N concentration of shoot tissues (Table 2). Nitrogen fixation and soil N balance The % 15N atomic enrichment in barley varied from 0.3551 to 0.5850% in the straw and from 0.4343 to 0.7169% in the spikes. The % 15N atomic enrichment in barley was always greater

Table 1 Effects of pea inoculation and P treatments on above ground yield and yield components taken from macro plots Treatments

UP0 UP1 IP0 IP1 I P I*P

Yield (Mg ha–1) DM

Grain

6.0 7.5 6.8 8.6 * ** ns

2.4 3.2 3.1 4.2 * * ns

Pods plant–1

3.5 4.2 4.7 6.2 ** * ns

Grains pod–1

3.9 3.8 3.6 3.9 ns ns ns

Seed weight

Grain N yield

(g)

(kg ha–1)

0.23 0.29 0.25 0.25 ns ns ns

89.7 142.6 134.2 191.0 * * ns

U = uninoculated, I = inoculated, 0 = 0 kg P ha–1 and P1 = 65 kg P2O5 ha–1 ns = not significant; *significant at P < 0.05; **significant at P < 0.01 Values are means of three replicates taken from 2-years’ yield

123

252

Nutr Cycl Agroecosyst (2006) 75:247–255

Table 2 Nodulation and N and P concentrations in different plant parts of pea Treatments

UP0 UP1 IP0 IP1 I P I*P

Nodule

%N

%P

Number

Weight (g)

Volume (ml)

Nodule

Shoot

Root

Nodule

Shoot

Root

23 36 39 44 * * ns

0.20 0.50 0.53 0.68 ** ** ns

0.20 0.49 0.52 0.68 ** ** ns

7.14 7.99 6.31 6.78 ns ns ns

1.66 3.44 2.95 3.50 ** ** *

2.10 2.34 2.15 2.33 ns ** ns

0.49 0.44 0.36 0.44 ns ns ns

0.17 0.25 0.17 0.29 ns ** ns

0.14 0.21 0.15 0.18 ns ** ns

U = uninoculated, I = inoculated, 0 = 0 kg P ha–1 and P1 = 65 kg P2O5 ha–1 ns = not significant; *significant at P < 0.05; **significant at P < 0.01 Values are means of three replicates taken from 2-years’ yield

160.0

than for pea, on a whole plant basis, a clear indication that the legume fixed N2 from the atmosphere. The dry matter and pod yields of pea for the micro plots were significantly improved by inoculation and P fertilization, increasing the yield by 73 and 48%, respectively, relative to the control (Table 3). The % Ndfa for pea varied from 52.7 to 70.4% on a whole plant basis, with an average value of 63%. The % Ndfa was significantly enhanced by inoculation and P fertilization in the pea straw but only by P fertilization in pods (Table 3). The enhancement varied from 59.5 to 77.7% in pods, and from 25.0 to 53.0% in the straw. These values indicate a different partition of fixed N2 between pods and straw (Table 3) and that a major fraction of fixed N2 was translocated to pods. The percent of N derived from fertilizer (% Ndff) and soil (% Ndfs) of pea varied also

Ndff

N-yield (kg/ha)

140.0

Ndfa

120.0

Ndfs

100.0 80.0 60.0 40.0 20.0 0.0 UP0

IP0 UP1 Treatments

IP1

Fig. 2 Contribution of N derived from fertilizer, atmosphere and soil (Ndff, Ndfa and Ndfs, respectively) to the total N-yield of pea. I = inoculated, U = uninoculated, P1 = phosphorus applied and P0 = with no phosphorus

from 4.0 to 3.3%, and from 43.3 to 26.2%, respectively, corresponding to average N yields values of 5.7 and 51.7 kg N ha–1 (Fig. 2). Similarly, the % Ndff and % Ndfs of the reference crop, barley varied from 9.0 to 12.0% and from

Table 3 Effect of inoculation and P fertilization on above ground yields, soil N balance (B) and the amount of N2 fixed by pea Treatment

UP0 UP1 IP0 IP1 I P I*P

DM yield(Mg ha–1)

N-yield (kg N ha–1)

Ndfa (%)

N2fixed (kg N ha–1)

B

Pod

Straw

Total

Pod

Straw

Total

Pod

Straw

Pod

Straw

Total

kg N ha–1

2.3 3.0 2.8 3.5 * ** ns

2.2 3.6 3.6 4.2 ** ** ns

4.5 6.7 6.4 7.7 ** ** ns

81.0 118.0 114.0 141.0 * * ns

20.7 52.2 65.4 57.2 ** * **

102.0 171.0 179.0 198.0 ** ** ns

59.5 76.8 66.0 77.7 ns ** ns

25.0 51.0 50.2 53.0 * * ns

50.0 90.4 75.3 110.4 * ** ns

5.3 26.9 32.0 30.2 ** * **

55.0 117.0 107.0 141.0 ** ** ns

– 9.2 18.9 13.2 19.3 * ** ns

U = uninoculated, I = inoculated, P0 = 0 kg P ha–1 and P1 = 65 kg P2O5ha–1 ns = not significant; *significant at P < 0.05; **significant at P < 0.01

123

Nutr Cycl Agroecosyst (2006) 75:247–255

253

Table 4 Grain and N-yields of wheat grown after pea –1

Discussion –1

Treatments

Grain yield (Mg ha ) N-Yield (kg N ha )

R0 UP0 UP1 IP0 IP1 LSD (Pr = 0.05)

2.7c 3.3bc 4.0ba 3.8ba 4.1a 7.2

32.9c 53.5b 73.9a 64.4a 72.2a 10.2

R = control, U = uninoculated, I = inoculated, 0 = 0 kg P ha–1 and P1 = 65 kg P2O5 ha–1 Pr = probability level of F-test Different letters represent significant difference at Pr = 0.05 level

92.0 to 88.0%, respectively, corresponding to average N yields of 6.9 and 62.0 kg N ha–1. The total N yield of pea and the amount of N2 fixed by the pea crop were also found to improve significantly by inoculation and P fertilization (Table 3). The total amount of N yield and N2 fixed by pea ranged from 102 to 198 and 55 to 141 kg N ha–1, respectively, corresponding to average increase of 95 and 154%, relative to the control. The amount of N2 fixed was highly correlated to the total dry matter (r = 0.94, P < 0.0001) and total N yield (r = 0.95, P < 0.0001) of pea. Total N yield was appropriated from 71.0 to 79.6% in the pods, and the remaining in the straw. Most of the total N2 fixed (79–90%) was also contained in the pods (Table 3). The N extracted by barley (spikes and straw included) from the soil varied from 65 to 73 kg N ha–1. The soils N balance for barley was in the range of – 45 to – 53 kg N ha–1. In the legume, the soil N balance varied from – 9.2 to 19.3 kg N ha–1 (Table 3). Legume wheat rotation The grain and N yields of wheat crop planted after the legume were improved significantly compared to the wheat crop planted after the cereal (Table 4). The incorporated legume straw and root residues may have played roles in improving the yield. The grain and N yields of wheat increased from 21.3 to 47.8% and from 62.4 to 119.5% respectively, relative to the cerealcereal rotation control.

The present work on Markos pea demonstrated the positive effects of inoculation and P fertilization on yield, nodulation and N2 fixation capacity of the legume. Nitrogen fixation has high sink strength for reduced carbon (C), whose supply is limited by the availability of N and P to the chloroplast (Jakobesen 1985). Phosphorus is found to enlarge the leaf area and improve the rate of assimilate production per unit leaf area (Yemane and Skjelva˚g 2003), as it is involved in photosynthetic energy transfer processes. Hence, the increase in dry matter yield (DM) with P supply could be attributed to both the size and efficiency of the assimilatory apparatus. Phosphorus fertilization increased grain yield as also reported by Yemane and Skjelva˚g (2003) from the experiment done in Ethiopia on Dekeko (Pisum sativum var. abyssinicum). According to these authors, the yield advantage of pea produced by the applied P could be attributed to an increase in the number of branches, which in turn increased the number of productive nodes and number of pods m–2. The number of pods m–2 was found to correlate strongly with grain yield in this experiment. The lack of significant P fertilizer effects in seed weight and number of seed pod–1 of Markos pea in this experiment was in agreement with findings of Yemane and Skjelva˚g (2003). Nitrogen and phosphorus deficiencies were found mainly to limit the grain yield of pea through reduction of tillers and number of flowering nodes directly affecting the number of pods per plant and not the number of seeds in a pod (Yemane and Skjelva˚g 2003). Phosphorus fertilization increased nodule number, nodule fresh weight and volume, similar to the findings of Jakobsen (1985) and Leidi and Rodiguez-Navarro (2000). These authors found that P fertilization affected nodule number, mass, relative growth and function in pea and other grain legumes. Nodules contained higher P than shoots and roots, indicating that they are strong sinks for P. Phosphorus concentration in nodules was not much affected by P fertilization the increase in nodule number may have a diluting

123

254

effect, as P is known to stimulate nodulation. The concentration of N and P significantly increased in shoot by a positive interaction of inoculation and P fertilization. According to Jakobsen (1985), increasing the P supply to pea plant increased considerably the N and P concentrations of shoots, enhancing the assimilation of reduced carbon, which in turn, increased the supply of photosynthetase to the nodules. Inoculation and phosphorus fertilization were also found to significantly improve the % Ndfa values of Markos pea, which. was greater in the pods than in the straw, appropriating 60–78% of the total N2 derived from the atmosphere. Carranca et al. (1999) reported up to 55% of the amount of N present in the pods at maturity was estimated to be derived from mobilization of N from the vegetative organs from the experiments done on pea in the Mediterranean conditions of Southern Portugal. The total N2 fixed was also significantly affected by inoculation and P fertilization, each contributing up to 44 and 60% to the total N2 fixed, respectively. Inoculation was found by Carranca et al. (1999), to improve the N2 fixation of pea to about 50%. Phosphorus fertilization was also reported to improve the N2 fixation in grain legumes through its specific roles in nitrogenase activities and host plant growth (Jakobsen 1985; Leidi and Rodiguez-Navarro 2000). The results for N2 fixation by pea were obtained by using barley as a single reference crop, which was also commonly used by other workers (Carranca et al. 1999; Amanuel et al. 2000). However, more reliable result could have been obtained if more than one-reference crop were used as reported by Boddey et al. (1990). The simplified N balance calculated in our study underestimates the legume contribution to soil N through turn over of fine roots and nodules, root N and N released from roots as rhizodeposition are not accounted for. From pea-wheat rotation, increase in average yields of 1.06 Mg ha–1 and 33 kg ha–1 of grain and N of wheat, respectively were obtained, accounting to 39 and 101% increase respectively, relative to the cereal-cereal monocrop rotation. Response in grain yield of cereals to previous crops of tropical grain legumes was reported by

123

Nutr Cycl Agroecosyst (2006) 75:247–255

People and Herridge (1990) to vary from 0.20 to 3.68 Mg ha–1 compared with cereal–cereal monocrop yields, with relative increase being in the range of 16–353%. Strong et al. (1986) reported up to 47% increase of wheat grain yield and 121% increase of N yield grown after pea relative to the wheat-oats monocrop rotation from experiments done in Australia. The positive cereal response to N has been partially attributed to the transfer of biological fixed N2, to N sparing under the antecedent legume, and less immobilization of nitrate during the decomposition of legume residues (Chalk 1998). The contribution from root and nodule residues and the N released from the roots as rhizodeposition adds also considerable N to the next cereal. These were estimated to constitute between 35 and 45% of the total residue N (Mayer et al. 2003). The other rotational benefits to cereals by legumes is the transfer of large amount of phosphate from the inorganic to the organic pool by enriching the SOM, as a result of high biomass production by the legume and thus, this process may render phosphate more available to the cereal by restricting phosphate fixation. The benefit is more when the legume is fertilized with enough supply of P prior to cereal rotation (Nuruzzaman et al. 2005). The results of this study show the beneficial effect on P mobilization in legume based rotation.

Conclusion Pea could serve as an alternative to fallowing, especially in areas affected by high farmland pressure, benefiting the farmers with additional income and improving the soil’s N and available P status. These, in turn, increase the yields of the subsequent cereals such as wheat, provided the legume was fertilized with P and its straw incorporated prior to the planting of the legume–cereal monocrop. Acknowledgements The authors duly acknowledge CND (Combating Nutrient Depletion) project funded by the Ministry of Foreign Affairs, Norway, and Mekelle University for partially funding this study. The first author also

Nutr Cycl Agroecosyst (2006) 75:247–255 appreciates the financial support received from the Loan Bank (La˚nekassen, Norway).

References Amanuel G, Kuhne RF, Tanner DG, Vlek PLG (2000). Biological nitrogen fixation in fababean (Vicia faba L.) in the Ethiopian highlands as affected by P fertilization and inoculation. Biol Fertil Soils 32:353–359 Amanuel G, Tanner DG (1991) The effect of crop rotation in two wheat production zones of southeastern Ethiopia. In: Saunders DA (ed) Wheat for the non-traditional warm areas. CIMMYT, Mexico, pp 486–490 Boddey RM, Urquiaga S, Neves MCP (1990) Quantification of the contribution of N2 fixation to field grown legumes – a strategy for the practical application of the 15N isotope dilution technique. Soil Biol Biochem 22:649–655 BremerJM, Mulvaney CS (1982) Total nitrogen. In: Page AL, Miller RH, Keeny DR (eds) Methods of soil analysis part 2. Chemical and Microbiological Properties, Agronomy Monograph No. 9, 2nd edn. SSSA, Madison, WI, pp 595–622 Carlos HDA, Minguez MI (2001). Cereal–legume rotations in a Mediterranean environment: biomass and yield production. Field Crops Res 70:139–151 Carranca C, deVarennes A, Rolston D (1999) Biological nitrogen fixation by fababean, pea and chickpea, under field conditions, estimated by the 15N isotope dilution technique. Eur J Agron 10:49–56 Chalk PM (1998) Dynamics of biological fixed N in legume–cereal rotation: a review. Aust J Agric Res 49:303–316 CSA (Central Statistics Authority) (2005) Agricultural Sampling Survey 2004/2005. Report on area and production for major crops. Vol. 1. CSA. Addis Ababa, Ethiopia Giller KE (2001) Nitrogen fixation in tropical cropping systems, 2nd edn. CABI Publishing, New York, pp 271–287 Haley LE, Melsted SW (1957) Preliminary studies of Molybdenum in Illinois soils. Soil Sci Soc Am Proc 21:316–319 Hardarson G, Danso SKA (1990) Use of 15N methodology to assess biological fixation. In: Hardarson G (ed) Training course series no. 2. IAEA, Vienna, pp 129–160 Jakobsen I (1985) The role of Phosphorus in nitrogen fixation by young plants (Pisum sativum). Physiol Plant 64:190–196 Jensen CR, Bjarne J, Mathias NA, Christiansen JL, Magensen VO, Poul F, Petersen CT (2004) The effects of lupins as compared with peas and oats on the yield of the subsequent winter barley crop. Eur J Agron 20:405–316 Knudsen D, Peterson GA, Pratt PE (1982) Lithium, sodium and potassium. In: Page AL, Miller RH, Keeny

255 DR (eds) Methods of soil analysis part 2. Chemical and Microbiological Properties, Agronomy Monograph No 9, 2nd edn. SSSA, Madison WI, pp 225–246 Lanyon LE, Walter RH (1982) Magnesium, calcium, strontium and barium. In: Page AL, Miller, RH, Keeny DR (eds) Methods of soil analysis part 2. Chemical and Microbiological Properties, Agronomy Monograph. No.9, 2nd ed. SSSA, Madison, WI, pp 247–260 Leidi EO, Rodiguez-Navarro DN (2000) Nitrogen and phosphorus availability limit N2 fixation in bean. New Phytol 147:337–346 Mayer J, Buegger F, Jensen ES, Schlote M, Jurgen H (2003) Estimating N rhizodeposition of grain legumes using a 15N in situ stem labeling method. Soil Biol Biochem 35:215–226 Miressa D, RoBarge WP (1999) Soil characteristics and management effects on phosphorus sorption by high land plateau soils of Ethiopia. Soil Sci Soc Am J 63:1455–1462 Nelson DW, Sommers LE (1982) Total carbon, organic carbon, and organic matter. In: Page AL, Miller RH, Keeny DR (eds) Methods of soil analysis part 2. Chemical and Microbiological Properties, Agronomy Monograph. No 9, 2edn. SSSA, Madison, WI, pp 539– 579 Nuruzzaman M, Lambers H, Mike DAB, Veneklaas EJ (2005) Phosphorus benefits of different legume crops to subsequent wheat grown in different soils of Western Australia. Plant Soil 175–187 Olsen SR, Cole CV, Watanabe FS, Dean LA (1954) Estimation of available phosphorus in soils by extraction with sodium bicarbonate. US Dep Of Agric Circ 939. Paar A (1998) Multiwave microwave sample preparation system instruction handbook. Gmbh, Graz, Austria People MB, Herridge DF (1990) Nitrogen fixation by legumes in tropics and subtropical agriculture. Adv Agron 44:155–223 Pulschen L (1987) Terminal report for the period May 1983–April 1987. Soils and Plant Nutrition Section, Agricultural Center, Debre Zeit, Ethiopia, 31 pp SAS Institute (2002) SAS user’s guide: Statistics. Statistical Analysis System Institute, Cary, NC Strong WM, Harbison J, Nielsen RGH, Hall BD, Best EK (1986). Nitrogen availability in Darling Downs soil following cereal, oilseed and grain legume crops. 2. Effects of residual soil nitrogen and fertilizer nitrogen on subsequent wheat crops. Aust J of Exp Agric 26:353–359 Tsigie A, Woldeab A (1994) Fertilizer response trails on food legumes. In: Asfaw T, Geletu B, Saxena MC, Solh MB (eds) Cool season food legumes of Ethiopia. AUN, IAR, ICARDA and Aleppo, Syria, pp 279–94 Yamane A, Skjelva˚g OA (2003) Effects of fertilizer phosphorus on yield traits of Dekeko (Pisum sativum var. abyssinicum) under field conditions. J Agron Crop Sci 189:14–20

123