Improving nitrogen and phosphorus use efficiencies through inclusion of forage cowpea in the rice–wheat systems in the Indo-Gangetic Plains of India

Erratum

Improving nitrogen and phosphorus use efficiencies throughinclusion of forage cowpea in the rice–wheat systems

in the Indo-Gangetic Plains of India$

B.S. Dwivedi*, Arvind K. Shukla, V.K. Singh, R.L. YadavProject Directorate for Cropping Systems Research, Modipuram, Meerut 250 110, India

Abstract

In high productivity zones of Indo-Gangetic Plains in south Asia, the rice–wheat system is stressed due to production fatigue

as evidenced by declining soil organic matter content, low efficiency of fertilizer use and diminishing rates of factor productivity.

We, therefore, conducted field experiments at Modipuram, India, to conserve soil organic carbon, improve N and P use

efficiency, and increase yields of rice–wheat system through inclusion of forage cowpea during the summer before cultivating

the rice–wheat system. Cowpea forage harvested at 50 days removed greater amounts of N and P through aboveground biomass

than those recycled through belowground roots and nodules. The NO3-N in soil profile below 45 cm depth after wheat harvest

was greater under fallow during summer than under cowpea, suggesting that cowpea minimized NO3-N leaching beyond 45 cm

depth. Similarly, in the treatments receiving both 120 kg N and 26 kg P ha�1, NO3-N in soil below 45 cm depth was lower

compared to those receiving N or P alone. After three crop cycles, soil OC content in 0–15 and 15–30 cm depths was greater

compared to initial OC in plots having cowpea. P applied at 26 kg ha�1 increased available P content over initial P content, and

also over P content of soil under no P treatments. The available P content was, however, invariably low under summer cowpea

plots as compared to that under no cowpea ones. With continuous rice–wheat cropping, the bulk density (BD) of soil increased

over the initial BD at different profile-depths, more so at 30–45 cm depth in no cowpea plots, but inclusion of summer cowpea

helped decreasing the BD in the surface (0–15 cm) and sub-surface (15–30 and 30–45 cm) soil layers. Summer cowpea grown on

residual fertility after wheat harvest did not influence rice yield, but increased wheat grain yield (P < 0:05 during the terminal

year), when both the crops received fertilizer N and P at recommended rates. Skipping of N or P or both, however, resulted in

consistently low yield of these crops under summer cowpea treatments than those under no cowpea treatments, although the

differences were not necessarily significant every year. The use efficiency of applied N and P fertilizer in rice and wheat,

measured as agronomic efficiency and apparent recovery, was increased with the use of fertilizer N and P at recommended rates,

and also with inclusion of summer cowpea.

# 2003 Elsevier Science B.V. All rights reserved.

Keywords: Bulk density; Summer cowpea; Fertilizer N and P; Nutrient use efficiency; Typic Ustochrepts

1. Introduction

The rice–wheat cropping system (RWCS), managed

over 10.5 million ha in the Indo-Gangetic Plain region

(IGPR) of India, is the most widely practised annual

rotation providing food, income and employment to

Field Crops Research 84 (2003) 399–418

$ doi of original article 10.1016/S0378-4290(02)00169-7.* Corresponding author. Tel.: þ91-121-570708;

fax: þ91-121-571548.

E-mail address: [email protected] (B.S. Dwivedi).

0378-4290/$ – see front matter # 2003 Elsevier Science B.V. All rights reserved.

doi:10.1016/S0378-4290(03)00060-1

millions of rural and urban producers and consumers

(Paroda et al., 1994; HobbsandMorris, 1996). Adoption

of Green Revolution Technologies (GRTs, i.e., high

yielding varieties, massive use of chemical fertilizers

and expansion in irrigated areas) since mid-1960s

increased the yield of both the component crops of

RWCS, raising the contribution of RWCS to the extent

of 23% in the total food grain production of India. It is,

however, since 1990s that RWCS is showing signs of

stress with production fatigue, particularly in the high

productivity zones of the Trans-Gangetic Plains (TGP)

representing the states of Punjab and Haryana, and the

Upper Gangetic Plains (UGP) representing western

Uttar Pradesh (Yadav et al., 1998b). Since both rice

and wheat are exhaustive feeders of plant nutrients,

a high annual productivity results in removal of

nutrients in substantial amounts that often exceed

replenishments through fertilizers and manures, lead-

ing ultimately to deterioration in soil fertility (Hegde

and Dwivedi, 1992). In many such areas of TGP and

UGP, yield decline and decreasing factor productivity

have been noted in RWCS (Yadav, 1998). Farmers

have started to use greater than recommended rates of

chemical fertilizers, particularly N fertilizers, to main-

tain the yield levels previously attained with relatively

less fertilizer. Recent diagnostic surveys on nutrient

management practices prevailing in RWCS dominated

areas of western Uttar Pradesh (UGP) revealed that

nearly one-third of rice–wheat growing farmers apply

as much as 180 kg fertilizer N ha�1 to each rice and

wheat crop as against the local recommendation of

120 kg N ha�1 (Dwivedi et al., 2001).

Such emerging trends of indiscriminate use of ferti-

lizer N have to be curbed because of two major con-

cerns. Firstly, the TGP and UGP are the areas where

fertilizer use has already reached very high levels, and

further increase is unlikely to give economic returns

(Yadav et al., 2000). Secondly, as the fertilizer N use

efficiency in these coarse-textured permeable soils is

quite low due to excessive N losses (Katyal et al., 1985;

Singh et al., 1991a), increase in fertilizer N application

rates may also enhance the extent of nitrate-leaching

and thereby pollution of groundwater (Singh et al.,

1995), which is used for drinking by the rural masses.

Hence the use efficiency of applied fertilizers, and not

the fertilization rate, needs to be increased in order to

sustain the productivity of RWCS, as well as to mini-

mize the environmental hazards.

Balanced fertilizer use, i.e., use of fertilizer nutri-

ents in right proportion and in adequate amount, and

integrated plant nutrient supply (IPNS) are considered

as promising agro-techniques to sustain yield, increase

fertilizer use efficiency (FUE), and to restore soil

fertility under intensive cropping (Yadav et al.,

1998c). In this context, the advantage of green manure

legumes and short duration grain legumes in RWCS

have been widely investigated and documented

(Tiwari et al., 1980; Singh et al., 1991b; Mann and

Garrity, 1994; Ahlawat et al., 1998). The possibilities

of inclusion of fast-growing forage legumes as a catch

crop (i.e., a crop planted and harvested between two

regular crops in consecutive seasons) during summer

after the wheat harvest did not, however, receive due

attention, although great scope exists for this practice

(Yadav et al., 1998a). In the present study, we mea-

sured the effect of summer cowpea raised as a forage

crop, on important soil fertility parameters, and on the

yield and nutrient use efficiency in rice and wheat

crops grown with or without adequate N and P ferti-

lizer inputs.

2. Materials and methods

2.1. The experimental site

A field experiment commencing in 1997–1998 was

continued for three consecutive years, i.e., up to 1999–

2000 on a Typic Ustochrept soil of the Research Farm

of the Project Directorate for Cropping Systems

Research, Modipuram, Meerut, located at 29840Nlatitude, 778460E longitude and 237 m above mean

sea level, in western Uttar Pradesh representing UGP

zone of the IGPR of India. The UGP is an intensively

cultivated zone with 150% cropping intensity.

Rice–wheat is cultivated on about 2.66 million ha in

UGP (Yadav et al., 1998b). The climate of Meerut is

semi-arid subtropical, with dry hot summers and cold

winters. The average annual rainfall is 810 mm,

and nearly 80% of the total rainfall is received

through northwest monsoons during July–September.

The average monthly minimum and maximum tem-

peratures fluctuate from 6.7 to 7.5 and 17.9 to 21.7 8Cin January (the coolest month), and from 23.4 to 25.2

and 38.1 to 40.9 8C in May (the hottest month),

respectively.

400 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

The soil of the experimental site was a sandy loam

(15% clay, 18.5% silt, 64.5% sand) of Gangetic alluvial

origin, very deep (>2 m), well drained, flat (about 1%

slope), and represented one of the most extensive soil

series, i.e., Sobhapur series of northwest India.

2.2. Treatments and crop cultivation

The treatments comprising two levels each of fertili-

zer N (0 and 120 kg N ha�1) and P (0 and 26 kg

P ha�1) were applied to both rice and wheat, grown

in sequence with summer cowpea (forage) or summer

fallow, i.e., no cowpea. Eight treatment combinations,

i.e., 2 N rates � 2 P rates � 2 summer crop options

were randomized within a block, and four such blocks

were maintained. The plot size was 5 m � 6 m. Both

rice and wheat received N and P applications as per

treatment through urea (46.4% N) and single super-

phosphate (6.99% P), respectively. K was applied

at a uniform rate of 33 kg ha�1 through muriate of

potash (49.8% K). In addition, 5 kg Zn ha�1 was also

applied uniformly as zinc sulphate (21% Zn) to

each rice crop. One-third dose of N, and entire P, K

and Zn were applied before rice transplanting/wheat

sowing, and the remaining N top-dressed in two

equal splits, 30 and 55 days after transplanting/sowing.

Summer cowpea was grown without applying any

fertilizer.

Cowpea cv. ‘Russian Giant 2’, was grown as per

treatment in rows 45 cm apart using 30 kg seed ha�1

during the second week of May in all 3 years, and

was harvested 50 days after sowing (DAS) for forage

purposes. Thereafter, 25-day-old seedlings of rice cv.

‘Saket 4’ were transplanted at 20 cm � 15 cm spacing

during the first week of July in all the plots. After rice

harvesting in second week of October, wheat cv. ‘UP

2338’ was sown in rows 20 cm apart in the same layout

using 100 kg seed ha�1. Wheat was harvested in the

third week of April during all years.

All the crops were grown under assured irrigated

conditions. Three irrigations, one pre-sowing and the

other two at 15-day interval after sowing, were applied

to cowpea. In rice, 12 irrigations each during 1997–

1998 and 1998–1999, and 11 irrigations during 1999–

2000 were applied. At each irrigation 5 cm standing

water was maintained, and the interval between two

irrigations depended on the disappearance of standing

water. Wheat received five irrigations at the critical

stages for water stress, viz., crown root initiation (21

DAS), maximum tilling (55 DAS), jointing (75 DAS),

ear emergence (100 DAS) and milking (135 DAS). At

maturity, rice and wheat were harvested manually

from ground level using sickles, and the harvested

aboveground biomass was removed from the plots.

2.3. Soil and plant analysis

Before commencement of the experiment in 1997–

1998, soil samples were collected from 0 to 105 cm

profile-depth at every 15 cm interval, at four sites in

the experimental field using a core sampler of 8 cm

diameter. The sub-samples were mixed and bulked,

and representative soil samples for each depth were

drawn for chemical analysis. The post-harvest soil

samples (0–105 cm profile-depth at every 15 cm inter-

val) were also drawn following the same procedure

every year after completion of rice–wheat cycle, i.e.,

after wheat harvest. The initial and post-harvest soil

samples were pulverized using a wooden pestle-mor-

tar and sieved through a 100 mesh sieve. The pro-

cessed samples were analysed for organic carbon (OC)

(Walkley and Black’s method), mineral N (Bremner

and Keeney, 1965) and available (0.5 M NaHCO3, pH

8.5 extractable) P content (Olsen and Sommers, 1982).

The bulk density (BD) of soil at different profile-

depths in the initial and the terminal (post-wheat

1999–2000) soil samples was measured using alumi-

nium cores. The initial samples were also analysed for

pH (1:2 soil water suspension) and mechanical com-

position (international pipette method), following

standard analytical procedures (Page et al., 1982).

Important physico-chemical characteristics of the soil

before commencement of the experiment are given in

Table 1.

The dry matter (DM) samples of cowpea, and grain

and straw samples of rice and wheat collected from

each plot were dried at 70 8C in a hot-air oven. The

dried samples were ground in a stainless steel Wiley

mill, and wet-digested in concentrated H2SO4 for

determination of total N and in di-acid mixture

(HNO3 and HClO4 mixed in 4:1 ratio) for determina-

tion of total P content. The N content was determined

by the Kjeldahl method using a Kjeltec auto-analyser,

and P content by the vanadomolybdate yellow colour

method (Piper, 1966) using an UV-VIS spectrophot-

ometer.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 401

2.4. Statistical analysis and computations

For treatment comparisons in the field experiments,

the ‘F-test’ was used, following the procedures of

factorial randomized block design (Cochran and Cox,

1957). The CD (critical difference), defined as the

least significant difference beyond which all treatment

differences are statistically significant, was computed

to determine statistically significant treatment differ-

ences in Tables 3–6 as

CD ¼ ðp2VEr�1Þt5% (1)

where VE is the error variance, r the number of

replications of the factor for which CD is calculated,

t5% the table value of t at 5% level of significance at

error degrees of freedom.

In Tables 2 and 7 and Figs. 1–5, standard error (S.E.)

of the treatment means was computed as

S:E: ¼ S:D:ðpNÞ�1(2)

where S.D. is the standard deviation of mean, and N

the number of observations on which the mean is

based.

In order to quantify the effect of fertilizer N and P

input and summer cowpea on the nutrient (N or P) use

efficiencies in rice and wheat, computations were

made using following equations:

AEN or P ¼ DY F�1n (3)

ARN or P ¼ DU F�1n � 100 (4)

where AEN or P is the agronomic efficiency, often

termed as incremental efficiency, of applied N or P

fertilizer, DY the incremental yield due to fertilizer N

or P input, Fn the amount of fertilizer N or P applied,

ARN or P the apparent recovery of fertilizer nutrients,

and DU the incremental uptake of N or P due to

fertilizer application. TheDY,DU and Fn are expressed

as kg ha�1.

3. Results

3.1. DM yield, nodulation and NP accumulation in

forage cowpea

The forage cowpea grown during the summer sea-

son produced a DM yield of 3.3 t ha�1 in 1997–1998,

but in subsequent years DM yield ranged between 2.18

and 3.54 t ha�1 in different plots having variable appli-

cations of N and P in the preceding wheat and rice

crops (Table 2). The DM yield of cowpea in no P plots

as compared with treatments receiving 26 kg P ha�1 in

both rice and wheat was lower by 21% in 1998–1999

and 27% in 1999–2000. N application to preceding

cereals, on the other hand, had a negligible effect on

DM yield of cowpea. The effective nodule count,

which ranged between 242 and 443 m�2, also revealed

the advantage of residual P left in the soil after harvest

of P fertilized rice and wheat crops. Averaged over N

rates, effective nodules in no P plots were 268 m�2 in

1998–1999 and 256 m�2 in 1999–2000, which

increased to 388 and 410 m�2, respectively, in the

plots receiving fertilizer P during the previous cereal

crops. Data pertaining to nutrient (N and P) removal by

forage cowpea and nutrient recycling through roots and

nodules are presented in Table 2.

3.2. Changes in physico-chemical characteristics

of the soil

3.2.1. Soil OC content

Soil OC content measured up to a profile-depth

of 0–105 cm after 3 years of continuous cropping

Table 1

Physico-chemical characteristics of soil measured at commencement of field experiment during 1997–1998

Soil depth (cm) BD (Mg m�3) pH OC (%) NO3-N (mg kg�1) NH4-N (mg kg�1) Available P (mg kg�1)

0–15 1.46 8.13 0.43 4.8 14.3 8.3

15–30 1.55 8.02 0.38 5.0 17.1 6.7

30–45 1.48 8.10 0.32 4.9 19.5 6.2

45–60 1.52 7.84 0.35 4.7 18.4 6.3

60–75 1.54 7.95 0.26 4.4 18.1 5.8

75–90 1.59 7.62 0.19 3.6 17.0 5.8

90–105 1.58 7.74 0.14 3.6 14.2 5.9

402 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

Table 2

DM yield, nodulation, and N and P accumulation in forage cowpea grown in RWCS during pre-rice summer seasons of 1997–1998, 1998–1999 and 1999–2000

Particulars 1997–1998 1998–1999 1999–2000

N0P0a N0P26 N120P0 N120P26 N0P0 N0P26 N120P0 N120P26

Aboveground DM yield (t ha�1) 3.30 0.05b 2.45 0.07 3.03 0.09 2.78 0.08 3.54 0.11 2.18 0.06 2.91 0.08 2.30 0.09 3.21 0.09

N accumulation in

aboveground DM (kg ha�1)

79.20 1.31 54.45 1.58 69.75 2.28 60.15 2.12 79.65 2.29 43.80 1.08 63.75 1.87 48.15 1.47 69.00 1.96

P accumulation in

aboveground DM (kg ha�1)

11.88 0.17 8.25 0.24 10.31 0.26 8.33 0.21 12.03 0.46 6.53 0.25 9.32 0.20 5.97 0.31 9.63 0.38

Root dry weight (kg ha�1) 570 8.4 362 16.8 486 12.9 392 16.8 528 14.7 345 9.8 492 20.9 353 8.4 504 18.1

N recycled through roots

(kg ha�1)

5.30 0.07 3.65 0.12 5.15 0.13 3.65 0.10 5.24 0.14 3.48 0.09 4.82 0.13 3.35 0.09 5.15 0.17

P recycled through roots

(kg ha�1)

1.08 0.01 0.54 0.02 0.83 0.03 0.59 0.02 0.95 0.03 0.48 0.02 0.89 0.02 0.42 0.02 0.75 0.02

Effective nodules (m�2) 443 5.8 287 9.1 372 13.5 248 8.4 404 9.1 269 7.9 395 12.4 242 6.4 425 11.9

Nodule dry weight (kg ha�1) 12.23 0.18 8.12 0.17 10.83 0.35 8.01 0.34 11.76 0.26 7.23 0.15 11.34 0.35 7.40 0.18 12.03 0.38

N recycled through nodules

(kg ha�1)

0.57 0.01 0.35 0.01 0.50 0.03 0.36 0.01 0.57 0.02 0.33 0.01 0.51 0.03 0.33 0.01 0.53 0.02

P recycled through nodules

(kg ha�1)

0.024 0.001 0.015 0.001 0.021 0.001 0.012 0.001 0.024 0.001 0.012 0.001 0.021 0.001 0.011 0.001 0.023 0.001

a Fertilizer N and P rates (N0P0, N0P26, N120P0 and N120P26) refer to N and/or P application to preceding cereal (rice and wheat) crops. For details, see Section 2.b Mean and S.E.M.; n ¼ 16 for 1997–1998, and n ¼ 4 for 1998–1999 and 1999–2000.

B.S

.D

wived

iet

al./F

ieldC

rop

sR

esearch

84

(20

03

)3

99

–4

18

40

3

revealed that the plots treated with recommended N

and P fertilizer, i.e., 120 kg N and 26 kg P ha�1 in rice

and wheat had greater OC content in the 0–15, 15–30

and 30–45 cm soil layers under summer fallow as well

as forage cowpea situations, compared with no ferti-

lizer N and P (control) or the treatments kept devoid of

either of these nutrients (Fig. 1). Inclusion of a sum-

mer legume (forage cowpea) increased soil OC over

the initial content, as well as over the summer fallow

treatments; the extent of increase was greater with N

and P fertilization at recommended levels under forage

cowpea treatments. Use of 120 kg N and 26 kg P ha�1

increased soil OC over the initial values by 11.6% in

the 0–15 cm, 10.5% in the 15–30 cm and 6.3% in the

30–45 cm soil depths. The corresponding increase in

OC in summer fallow plots was only 2.3, 2.6 and

3.1%, respectively. A depletion in soil OC up to the

45 cm soil profile as compared to the initial content

was noticed in control plots, irrespective of cowpea

treatments. The treatment effects were, however, not

evident in soil below 45 cm depth. The OC content

showed a decreasing trend with soil depth. At 90–

105 cm soil depth, the values were lower by 62–70%

compared to those in 0–15 cm layer.

Fig. 1. Soil OC content at different profile-depths after wheat harvest in the terminal year as influenced by fertilizer N and P, and inclusion of

summer cowpea in RWCS. Bars indicate S.E.M., n ¼ 4.

404 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

3.2.2. Bulk density

The BD of the surface soil (0–15 cm) was

1.51 Mg m�3 at the beginning of experiment. It, how-

ever, increased with increasing soil profile-depth,

measuring 1.70 Mg m�3 in the 90–105 cm soil layer

(Fig. 2). At the termination of the field experiment,

i.e., after wheat 1999–2000, BD was not influenced

by N and P treatments. With continuous rice–wheat

cropping for 3 years without inclusion of cowpea,

the BD values at the 30–45 cm soil depth were greater

(1:63 0:02 Mg m�3) than the initial BD (1:560:01 Mg m�3) at this depth, indicating a tendency

toward sub-surface soil compaction. In the plots

with summer cowpea, no such soil compaction was

observed at the 30–45 cm depth (Fig. 2). When aver-

aged across fertilizer N and P treatments, the BD of

soil at different profile-depths up to 45 cm was smaller

(1:51 0:01 to 1:56 0:01 Mg m�3) under summer

cowpea plots, compared to BD (1:53 0:01 to 1:630:02 Mg m�3) under summer fallow plots.

3.2.3. Nitrate-N content

The nitrate-N (NO3-N) content of surface soil

(0–15 cm) was greater in fertilizer N treated plots

as compared to that in no N, as well as greater in

cowpea treatments than that in the summer fallow, the

differences being prominent during second and term-

inal years of cropping (Fig. 3). In the summer fallow

treatments after wheat harvest, the NO3-N content in

the surface soil layer in plots having 120 kg N treat-

ments was greater than in plots having no N, by 7%

in the first year, 71% in the second year and 88% in

the terminal year. With inclusion of cowpea during

summer, the increase in NO3-N content was 21% in

the first year, 105% in the second year and 92% in the

terminal year.

Fig. 2. Changes in BD of soil at different profile-depths after three crop cycles as influenced by inclusion of summer cowpea in RWCS. Bars

indicate S.E.M., n ¼ 4.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 405

Fig. 3. Effect of fertilizer N and P, and summer cowpea on the distribution of nitrate-N in soil profile after wheat harvest in different years.

Bars indicate S.E.M., n ¼ 4.

406 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

When compared with the initial content of surface

soil layer, the NO3-N content in 120 kg N ha�1 treated

plots was greater in the terminal year by 1.1 mg kg�1

under summer fallow and by 3.9 mg kg�1 under

summer cowpea treatments. On the other hand, con-

tinuous skipping of N to both rice and wheat under the

summer fallow treatment led to NO3-N depletion by

1.6 mg kg�1 over the initial status.

Fertilizer N and P as well as cowpea treatments

markedly influenced the distribution of NO3-N in the

soil profile. Whereas NO3-N content was increased

over the initial content up to a profile-depth of 75 cm

in the summer fallow plots receiving N alone, a

constant decrease in NO3-N content was observed

when 120 kg N and 26 kg P ha�1 were applied

together (Fig. 3). Cowpea in the system not only

favoured greater NO3-N content in the surface layer

under different N and P treatments compared to sum-

mer fallow, but also resulted in lower NO3-N content

in deeper soil layers, as evident from the decreasing

values of NO3-N with increasing profile-depth.

This advantage of the summer cowpea in minimizing

NO3-N leaching was, however, more spectacular

during the second and the terminal years of the experi-

ment.

3.2.4. Ammonium-N content

Application of N alone or in combination with P

increased the ammonium-N (NH4-N) content over

N0P0 or N0P26 treatments, with no specific effect

of summer crop (Fig. 4). When averaged over P

treatments, NH4-N content of the surface soil layer

(0–15 cm depth) after one crop cycle in 120 kg

N ha�1 treated plots was greater compared to that

under N-skipped plots by 47% in summer fallow and

by 54% in summer cowpea. A similar pattern of

increase in NH4-N content consequent to fertilizer

N application was observed after second and final

crop cycles.

Compared with the initial NH4-N content, which

was almost uniformly distributed throughout the soil

profile (0–105 cm), the values of NH4-N in different

soil layers up to 90 cm soil depth were greater under

fertilizer N treated plots in all the crop cycles, irre-

spective of the summer crop treatment. Under N-

skipped plots, the NH4-N content in the upper three

soil depths (i.e., up to 45 cm) was less than the initial

content, particularly after the terminal crop cycle.

The NH4-N content of soil below 45 cm was, however,

maintained at the initial level even under the N0P0

(control) plots.

3.2.5. Available P content

Available P content of the soil (0–15 cm profile-

depth) increased over the initial content, consequent

to P fertilization at 26 kg ha�1 to rice and wheat. The

magnitude of increase was greater in the third year of

experimentation (Fig. 5). To the contrary, a conco-

mitant depletion of available P content compared with

the initial value was observed under no fertilizer P

treatments, particularly under those receiving fertili-

zer N alone. When averaged across the summer crop

treatments, available P content of surface soil (0–

15 cm) under N0P0 (control) was depleted compared

to the initial content (8.3 mg kg�1) by 1.8 mg kg�1

after first rice–wheat cycle (1997–1998), 2.5 mg kg�1

after second crop cycle (1998–1999) and by

2.8 mg kg�1 after third crop cycle (1999–2000).

The extent of depletion over the initial P content

was relatively greater under the plots treated with

120 kg N kg�1 alone, i.e., 2.7 mg kg�1 in 1997–1998,

2.9 mg kg�1 in 1998–1999, and 3.5 mg kg�1 in 1999–

2000.

Compared with summer fallow, forage cowpea

grown during summer led to greater depletion of

available P under no P treatments (Fig. 5). With the

use of fertilizer P, the magnitude of increase in avail-

able P content of soil over no P was also smaller under

forage cowpea plots than under summer fallow plots.

The average available P content of soil at 0–15 cm

depth across fertilizer N and P treatments in summer

fallow plots was 8.6, 9.4 and 9.7 mg P kg�1 after

the wheat harvest in 1997–1998, 1998–1999 and

1999–2000, respectively. The corresponding values

of available P under summer cowpea were 8.1, 7.7

and 7.8 mg kg�1, respectively, thus indicating a gen-

eral decline in available P content of the soil conse-

quent to the inclusion of summer cowpea in RWCS.

Data on the distribution of available P in the soil

profile revealed a decrease in P content with increasing

soil depth, irrespective of the treatments imposed

(Fig. 5). Fertilizer N and P, and summer crop treat-

ments influenced available P content of soil at 15–30

and 30–45 cm depth, the effect being similar to that at

0–15 cm. The P content below 45 cm hardly exhibited

any treatment effect.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 407

Fig. 4. Effect of fertilizer N and P, and summer cowpea on the distribution of ammonium-N in soil profile after wheat harvest in different

years. Bars indicate S.E.M., n ¼ 4.

408 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

Fig. 5. Effect of fertilizer N and P, and summer cowpea on the distribution of available P in soil profile after wheat harvest in different years.

Bars indicate S.E.M., n ¼ 4.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 409

3.3. Effect on rice and wheat yield

3.3.1. Rice yield

In rice, application of 120 kg N ha�1 out-yielded

no N treatment significantly (P < 0:01), by 2.52–

3.23 t ha�1 in different years (Table 3). This yield

response to N, across P and summer crop treatments,

was 86–94% over no N plots (2.82–3.44 t grains ha�1)

from 1997–1998 to 1999–2000. Crop response to P

increased under continuous cropping, and was always

greater when rice followed summer cowpea rather

than summer fallow. The magnitude of increase in

rice grain yield due to application of 26 kg P over no P

under summer fallow plots was 4% in 1997–1998,

9% in 1998–1999 and 12% in 1999–2000. The yield

increases under summer cowpea plots were 7, 14 and

34%, respectively, in different years. When averaged

over fertilizer N and summer crop treatments, grain

yield with P fertilization was greater over no P by

0.27 t ha�1 during the initial year, 0.46 t ha�1 during

the second year and 0.80 t ha�1 during the terminal

year. This is 6, 11 and 22% response over no P, res-

pectively. During the terminal year, yield differences

due to fertilizer P, and summer crop � fertilizer P

interactions were statistically significant (P < 0:05).

In P-skipped plots, rice yields tended to be low when

rice was preceded by cowpea instead of summer

fallow.

3.3.2. Wheat yield

Application of fertilizer N alone or in conjunction

with P enhanced the wheat yield significantly over

no N and P (control) under summer fallow as well

as summer cowpea treatments during all 3 years

(Table 4). The grain yield under no N treatments

ranged between 1.63 and 1.87 t ha�1 in different

years, whereas N fertilization at 120 kg ha�1 did

produce an additional yield of 2.76–3.03 t ha�1. Simi-

larly, the yield response to 26 kg fertilizer P ha�1 over

no P was computed at 0.57–0.86 t ha�1 (i.e., 20–31%)

in different years, the magnitude of the response was

of course greater in the terminal year. The N � P

interaction was statistically significant during all

years, as the N fertilized crop accrued a greater

advantage (in terms of yield gain) due to P application

compared with the crop receiving no fertilizer N.

Table 3

Effect of fertilizer N and P applied to rice and wheat on the grain yield of rice (t ha�1) as influenced by summer cowpea (forage) in RWCS

Fertilizer NP

rate (kg ha�1)

1997–1998 1998–1999 1999–2000

Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean

N0P0 3.44 3.27 3.35 3.08 2.78 2.93 2.71 2.36 2.54

N0P26 3.50 3.52 3.51 3.20 3.06 3.13 3.01 3.17 3.09

Mean 3.47 3.40 – 3.14 2.92 – 2.86 2.77 –

N120P0 6.62 6.37 6.49 5.39 5.12 5.28 5.27 4.36 4.82

N120P26 6.95 6.80 6.87 6.05 5.92 5.98 5.88 5.82 5.72

Mean 6.79 6.59 – 5.72 5.52 – 5.58 5.09 –

Mean of P rates

P0 5.03 4.82 4.93 4.24 3.95 4.10 3.99 3.36 3.68

P26 5.23 5.16 5.20 4.63 4.49 4.56 4.45 4.50 4.48

Overall mean 5.13 4.99 – 4.43 4.22 – 4.22 3.80 –

CD (P < 0:05)

Summer

crop (C)

Fertilizer

N (N)

C � N Fertilizer

P (P)

C � P N � P C � N � P

1997–1998 NSa 0.32 NS NS NS NS NS

1998–1999 NS 0.36 NS NS NS NS NS

1999–2000 0.29 0.29 NS 0.29 0.40 NS NS

a Not significant at P < 0:05.

410 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

When averaged over fertilizer N and P rates, the effect

of the summer crop on wheat yields was not signifi-

cant. Nonetheless, the grain yield improvement due to

fertilizer P was greater under the summer cowpea

treatments, compared with summer fallow treatments,

but the interaction of summer crop � fertilizer P was

statistically significant during 1999–2000 only.

3.4. Effect on total N and P uptake by rice

and wheat

3.4.1. Total N uptake

In rice, total N uptake, i.e., N uptake by grainþstraw was significantly greater (P < 0:05) under

N fertilized plots compared with those devoid of N

fertilizer during all 3 years (Table 5). The mean N

uptake by rice under no N plots, across fertilizer P and

summer crop treatments, was 58.6 kg ha�1 in 1997–

1998, 44.8 kg ha�1 in 1998–1999 and 44.7 kg ha�1 in

1999–2000, which was increased to 113.6, 92.5 and

89.8 kg ha�1, respectively, consequent to N applica-

tion at 120 kg ha�1. Compared with no P treatments,

total N uptake by the crop was invariably greater in the

treatments receiving 26 kg P ha�1 (P < 0:05 during

1999–2000). The effect of the summer crop followed a

trend similar to grain yield, though the differences in

N uptake by rice and wheat due to the summer crop

(cowpea or no cowpea) were generally not significant.

The total N uptake by wheat was influenced sig-

nificantly (P < 0:05) by fertilizer N and P input

(Table 5). Whereas application of 120 kg N accounted

for a two-and-a-half-fold increase in total N uptake

over no N plots, irrespective of the year of experi-

mentation, the magnitude of increase in N uptake due

to use of P fertilizer ranged between 17.3% in the

initial year and 30.2% in the terminal year. Introduc-

tion of summer cowpea resulted in smaller N uptake

by wheat as compared to the crop raised in summer

fallow plots; the extent of decrease was more specta-

cular in the treatments receiving N alone or no N and

P fertilizer.

3.4.2. Total P uptake

The mean P uptake by rice in different years varied

from 9.0 to 15.3 kg ha�1 under N0P0 (control) and

from 22.7 to 31.8 kg ha�1 under N120P26 treatments

Table 4

Effect of fertilizer N and P applied to rice and wheat on the grain yield of wheat (t ha�1) as influenced by summer cowpea (forage) in RWCS

Fertilizer NP

rate (kg ha�1)

1997–1998 1998–1999 1999–2000

Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean

N0P0 1.67 1.50 1.58 1.72 1.48 1.60 1.86 1.58 1.72

N0P26 1.75 1.61 1.68 2.01 1.69 1.85 2.08 1.95 2.02

Mean 1.71 1.56 – 1.87 1.59 – 1.97 1.77 –

N120P0 4.23 4.05 4.14 4.28 3.87 4.08 4.05 3.77 3.91

N120P26 5.09 5.28 5.18 4.91 5.35 5.13 5.17 5.51 5.34

Mean 4.66 4.67 – 4.60 4.61 – 4.61 4.64 –

Mean of P rates

P0 2.95 2.78 2.87 3.00 2.68 2.84 2.96 2.68 2.82

P26 3.42 3.45 3.44 3.46 3.52 3.49 3.63 3.73 3.68

Overall mean 3.18 3.11 – 3.23 3.10 – 3.29 3.20 –

CD (P < 0:05)

Summer

crop (C)

Fertilizer

N (N)

C � N Fertilizer

P (P)

C � P N � P C � N � P

1997–1998 NSa 0.24 0.34 0.24 NS 0.34 NS

1998–1999 NS 0.25 NS 0.25 NS 0.36 NS

1999–2000 NS 0.30 NS 0.30 0.42 0.42 NS

a Not significant at P < 0:05.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 411

under summer fallow conditions (Table 6). The cor-

responding P uptake values under summer cowpea

treatments were 6.3–13.9 and 17.6–31.1 kg ha�1,

respectively. When averaged across fertilizer N and

summer crop treatments, total P uptake by the P

fertilized rice crop was greater than a crop receiving

no P, by 22% in the initial year, 29% in the second year

and 46% in the terminal year. Fertilizer N application

not only increased P uptake as compared to that under

N-skipped plots, but also augmented the magnitude of

the response to P fertilization. The changes in P uptake

due to N or P fertilizers, and due to N � P interaction

were significantly different from zero (P < 0:05) dur-

ing all the years.

Table 5

Total N uptake in rice and wheat as influenced by fertilizer N and P application to rice and wheat, and inclusion of summer cowpea (forage) in

RWCS

Fertilizer NP

rate (kg ha�1)

1997–1998 1998–1999 1999–2000

Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean

Total N uptake (kg ha�1) by rice

N0P0 56.7 54.8 55.8 46.9 40.9 43.9 43.1 36.1 39.6

N0P26 63.8 58.7 61.3 49.3 42.1 45.7 48.8 50.7 49.8

Mean 60.3 56.8 – 48.1 41.5 – 46.0 43.4 –

N120P0 111.3 106.0 108.7 92.9 84.7 88.8 84.9 78.4 81.7

N120P26 120.5 116.2 118.4 99.4 92.9 96.2 95.5 100.1 97.8

Mean 115.9 111.1 – 96.2 88.8 – 90.2 89.3 –

Mean of P rates

P0 84.0 80.4 82.2 69.9 62.8 66.4 64.0 57.3 60.7

P26 92.2 87.5 89.9 74.4 67.5 71.0 72.2 75.4 73.8

Overall mean 88.1 84.0 – 72.2 65.2 – 68.1 66.4 –

Total N uptake (kg ha�1) by wheat

N0P0 39.8 36.9 38.4 37.7 32.1 34.9 39.3 32.6 36.0

N0P26 40.1 39.2 39.7 41.7 36.5 39.1 42.7 44.2 43.5

Mean 40.0 38.1 – 39.7 34.3 – 41.0 38.4 –

N120P0 95.0 91.6 93.3 95.4 84.3 89.9 90.0 78.5 84.3

N120P26 107.9 110.9 109.4 107.4 107.4 107.4 108.1 118.3 113.2

Mean 101.5 101.3 – 101.4 95.9 – 99.1 98.4 –

Mean of P rates

P0 67.4 64.3 65.9 66.6 58.2 62.4 64.7 55.6 60.2

P26 74.0 75.1 74.6 74.6 72.0 73.3 75.4 81.3 78.4

Overall mean 70.7 69.7 – 70.6 65.1 – 70.1 68.5 –

CD ðP < 0:05Þ

1997–1998 1998–1999 1999–2000

Rice Wheat Rice Wheat Rice Wheat

Summer crop (C) NSa NS 7.04 NS NS NS

Fertilizer N (N) 5.58 5.34 7.04 4.89 4.02 4.08

C � N NS NS NS 4.89 NS NS

Fertilizer P (P) NS 5.34 NS 4.89 4.02 4.08

C � P NS NS NS NS NS 5.76

N � P NS 7.55 NS 6.92 NS 5.76

C � N � P NS NS NS NS NS 8.17

a Not significant at P < 0:05.

412 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

Compared with rice, subsequent wheat crops

removed less P from the soil, irrespective of treatments

(Table 6). Nonetheless, total P uptake by wheat, as

influenced by fertilizer N and P or summer cowpea,

exhibited a pattern similar to rice. In this case, summer

crop � fertilizer N interaction was also significant in

two of the 3 years.

4. Discussion

In our study, highest yields of rice and wheat were

recorded in the treatments fertilized with 120 kg

N ha�1 and 26 kg P ha�1, and skipping of either nutri-

ent resulted in significant yield loss, irrespective of

growing a legume (cowpea for forage) during summer

Table 6

Total P uptake of rice and wheat as influenced by fertilizer N and P application to rice and wheat, and inclusion of summer cowpea (forage) in

RWCS

Fertilizer NP

rate (kg ha�1)

1997–1998 1998–1999 1999–2000

Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean Summer

fallow

Summer

cowpea

Mean

Total P uptake (kg ha�1) by rice

N0P0 15.3 13.9 14.6 12.0 9.2 10.6 9.0 6.3 7.7

N0P26 17.0 18.1 17.6 14.2 11.8 13.0 12.0 10.3 11.2

Mean 16.2 16.0 – 13.1 10.5 – 10.5 8.3 –

N120P0 26.3 24.7 25.5 20.2 16.7 18.5 16.8 10.6 13.7

N120P26 31.8 31.1 31.5 26.3 23.1 24.7 22.7 17.1 19.9

Mean 29.1 27.9 – 23.3 19.9 – 19.8 13.9 –

Mean of P rates

P0 20.8 19.3 20.1 16.1 13.0 14.6 12.9 8.4 10.7

P26 24.4 24.6 24.5 20.3 17.5 18.9 17.4 13.7 15.6

Overall mean 22.6 21.0 – 18.2 15.1 – 15.2 10.9 –

Total P uptake (kg ha�1) by wheat

N0P0 10.2 8.4 9.3 9.8 7.4 8.6 8.6 6.1 7.4

N0P26 11.8 9.7 10.8 13.1 9.9 11.5 11.5 9.4 10.5

Mean 11.0 9.1 – 11.5 8.7 – 10.1 7.8 –

N120P0 21.2 21.6 21.4 22.6 18.7 20.7 18.3 12.8 15.6

N120P26 28.0 29.0 28.5 29.6 26.5 28.1 25.5 20.7 23.1

Mean 24.6 25.3 – 26.1 22.6 – 21.9 16.8 –

Mean of P rates

P0 15.7 15.1 15.4 16.2 13.1 14.7 13.5 9.5 11.5

P26 19.9 19.4 19.7 21.4 18.2 19.8 18.5 15.1 16.8

Overall mean 17.8 17.2 – 18.8 15.7 – 16.0 12.3 –

CD (P < 0:05)

1997–1998 1998–1999 1999–2000

Rice Wheat Rice Wheat Rice Wheat

Summer crop (C) 1.16 NS 0.94 1.13 NS 1.04

Fertilizer N (N) 1.16 1.12 0.94 1.13 0.98 1.04

C � N NSa 1.58 NS NS 0.98 1.47

Fertilizer P (P) 1.16 1.12 1.33 1.60 0.98 1.04

C � P NS NS NS NS NS NS

N � P 1.63 1.58 1.33 1.60 1.38 1.47

C � N � P NS NS NS NS NS NS

a Not significant at P < 0:05.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 413

or maintaining the field as summer fallow. It may,

therefore, be inferred that these soils having low organic

matter and total N contents (Singh, 1963) can support

highly productive rice–wheat systems only under

assured and adequate supply of nutrients. With ade-

quate fertilization, anannual (rice þ wheat) yieldof11–

12 t ha�1 was obtained in different years of the study,

which supports the view that with proper management,

the coarse-textured (sandy loam and loamy sand) soils

managed under the rice–wheat system in the IGPR are

amongst the most productive soils of the world, despite

their inherent low fertility (Aulakh and Singh, 1997).

We observed a consistent effect of fertilizer N on the

yield of both crops, which remained largely unaffected

by the inclusion of summer cowpea in the system, but

the yield responses of rice and wheat to P fertilizer

varied in accordance with the summer crop (cowpea

or fallow), as well as with the year of experimentation.

Greater yield response to P, in terms of percent

increase in grain yield over the no P treatment, in

wheat (20–31%) compared with that in rice (6–22%)

in different years is explainable. The electrochemical

changes that the soil undergoes during the submerged

rice-culture leads to a flush of available P (Ponnam-

peruma, 1985; Kirk et al., 1990), and as a consequence

response to fertilizer P in rice is either smaller than in

wheat (Dwivedi, 1994) or even absent in some cases

(Gill and Meelu, 1983; Rekhi et al., 2000). Although

flooding increases the supply of native P to rice,

subsequent drying promotes P fixation and depresses

its availability to post-rice (wheat) crop (Willet and

Higgens, 1978; Willet, 1979; Sah and Mikkelsen,

1989). Hence, wheat grown after transplanted sub-

merged rice exhibits tremendous response to P, and the

use of fertilizer P in adequate amounts along with N

becomes inevitable to meet fixation as well as crop P

requirements (Timsina and Connor, 2001). These

arguments supporting judicious fertilization in RWCS

are further substantiated in the present case, when the

results are explained in terms of FUE of N and P

fertilizers. The AEN, AEP, ARN and ARP computed

as a measure of FUE of applied N and P in rice and

wheat (Table 7) revealed that both AEN and ARN

increased when 120 kg N ha�1 was supplemented

with 26 kg P ha�1, and not applied alone. For instance,

in summer fallow treatments, the AEN ranged between

34.8 and 45.5 in rice, and between 42.3 and 48.1 in

wheat with the application of 120 kg N alone during

different years. Use of 26 kg P ha�1 along with N

increased the corresponding AEN values, which ran-

ged between 36.4 and 47.2 in rice, and 54.5 and 56.5 in

wheat. As wheat is more responsive to fertilizer P, the

advantage of balanced NP fertilization was more

apparent in wheat. The results thus support the find-

ings of long-term experiments (LTEs) conducted in

India, where NPK fertilization at recommended rates

improved N use efficiency over the use of fertilizer N

alone (Nambiar, 1994; Swarup and Wanjari, 2000).

In another series of LTEs, the yield response of rice

and wheat to fertilizer N declined over years at several

locations in IGPR of India, when P fertilization was

ignored continuously (Hegde and Dwivedi, 1992).

Many reports from India and elsewhere as reviewed

recently (Ladha et al., 1988; Singh et al., 1991b;

Timsina and Connor, 2001) amply show the advan-

tages that green manure and grain legumes have in

improving the productivity of RWCS and in regener-

ating the soil fertility, mainly by way of biological N

fixation (BNF) and nutrient recycling from deeper soil

layers. The possibilities of inclusion of forage legumes

in RWCS have, however, been less explored, and their

effects on soil or crop parameters rarely documented

(Lauren et al., 1998; Yadav et al., 1998a). In the

present case, inclusion of summer cowpea as a forage

legume did rather deplete the soil of native N and P,

because: (i) all aboveground parts of cowpea grown on

post-wheat residual fertility were harvested for forage

purpose, which resulted in annual removal of 43.8–

79.2 kg N ha�1 and 6–12 kg P ha�1 under different

treatments, and (ii) cowpea roots and nodules incor-

porated after cowpea harvesting could recycle meagre

quantities of N and P, compared with the removal of

these nutrients by aboveground parts (Table 2).

Whereas entire P removed by cowpea apparently came

from the soil, the share of soil N is also likely to be

substantial if only 54–70% of the N accumulated by

cowpea DM is assumed to be derived from BNF

(Awonaike et al., 1990). In cases where a large portion

of legume biomass is removed from the field, legume

crops may even deplete soil N fertility (Peoples and

Herridge, 1990). Nevertheless, the mineral N content,

i.e., NO3-N and NH4-N, analysed after the wheat

harvest (Figs. 3 and 4) did not decline in the surface

soil layer of summer cowpea treatments during two

out of 3 years. The depletion in available P content of

the soil owing to inclusion of summer cowpea in the

414 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

Table 7

N and P use efficiency in rice and wheat as influenced by inclusion of summer cowpea (forage) in RWCS

Fertilizer NP

rate (kg ha�1)

Agronomic efficiency of N (AEN) Apparent recovery of N (ARN, %) Agronomic efficiency of P (AEP) Apparent recovery of P (ARP, %)

1997–1998 1998–1999 1999–2000 1997–1998 1998–1999 1999–2000 1997–1998 1998–1999 1999–2000 1997–1998 1998–1999 1999–2000

Rice

Summer fallow

N0P0 – – – – – – – – – – – –

N0P26 – – – – – – 2.3 0.09 4.6 0.20 11.5 0.42 6.52 0.24 8.76 0.32 11.61 0.26

N120P0 26.50 0.84a 19.25 0.92 21.33 0.55 45.51 1.35 38.3 1.14 34.8 0.91 – – – – – –

N120P26 28.75 0.76 23.75 0.72 23.92 0.82 47.21 1.19 41.80 1.29 36.4 1.00 12.7 0.46 25.4 0.99 23.50 0.65 20.90 0.78 23.50 0.78 22.73 0.63

Summer cowpea

N0P0 – – – – – – – – – – – –

N0P26 – – – – – – 9.60 0.27 10.80 0.25 31.20 0.85 16.29 0.60 9.79 0.20 15.57 0.60

N120P0 25.83 0.71 19.50 0.82 16.67 0.61 42.63 1.47 36.50 1.17 35.30 1.35 – – – – – –

N120P26 27.33 0.91 23.83 0.66 22.08 0.80 47.88 1.08 42.30 1.11 41.20 1.47 16.50 0.62 30.80 0.60 56.2 1.58 24.39 0.95 24.65 0.84 25.04 0.79

Wheat

Summer fallow

N0P0 – – – – – – – – – – – –

N0P26 – – – – – – 3.10 0.09 11.20 0.46 8.50 0.22 6.17 0.24 12.73 0.58 11.17 0.32

N120P0 21.33 0.82 21.33 0.46 18.25 0.76 46.01 1.08 48.1 1.37 42.3 1.15 – – – – – –

N120P26 27.83 0.74 24.17 0.65 25.75 0.74 56.46 1.68 54.80 1.39 54.50 1.65 33.10 1.10 24.20 0.65 43.10 1.23 26.31 0.58 26.80 0.72 27.95 0.77

Summer cowpea

N0P0 – – – – – – – – – – – –

N0P26 – – – – – – 4.20 0.17 8.10 0.18 14.20 0.41 5.06 0.16 9.41 0.38 12.60 0.43

N120P0 21.25 0.64 19.92 0.53 18.25 0.81 45.62 1.29 43.50 1.43 38.30 0.95 – – – – – –

N120P26 30.58 0.82 30.50 1.10 29.67 0.79 59.73 1.75 59.10 1.55 61.70 2.02 47.30 0.91 56.90 1.88 66.90 1.95 28.13 0.89 30.28 0.77 30.35 0.85

aMean and S.E.M.; n ¼ 4.

B.S

.D

wived

iet

al./F

ieldC

rop

sR

esearch

84

(20

03

)3

99

–4

18

41

5

rice–wheat system was, however, apparent in 0–15

and 15–30 cm soil layers under no P as well as

fertilizer P-treated plots during all the years of experi-

mentation (Fig. 5). In fact, soil sampling and analysis

after cowpea harvesting could have helped to explain

the mineral N and available P contents of the soil vis-

a-vis nutrient withdrawals by cowpea.

Despite the removal of N and P in sizeable amounts

from the soil by cowpea, its inclusion in the rice–

wheat system resulted in increased wheat yields as

well as FUE in the plots treated with 120 kg N and

26 kg P ha�1, the magnitude of yield increase over the

summer fallow treatment being greater in the terminal

year. Two hypotheses may be given to support the

advantage of summer cowpea. Firstly, incorporation of

cowpea roots and their subsequent decomposition has

increased the soil OC content over the initial OC

content by 11.6% in the 0–15 cm layer, 10.5% in

the 15–30 cm layer and by 6.3% in the 30–45 cm

layer (Fig. 1). This could have favoured better germi-

nation and establishment of the wheat crop. Secondly,

decreased soil compaction as indicated by reduction in

BD (Fig. 2) due to growing a tap rooted legume crop

(cowpea), and decomposition of its root residues

favoured wheat root growth and root penetration to

deeper soil layers. The wheat roots may have other-

wise been restricted in the post-rice environment

due to sub-surface compaction caused by puddling

(Oussible et al., 1992; Aggarwal et al., 1995). Rice

yields remained unaffected in adequately fertilized

plots, as improvement in soil physical parameters

owing to inclusion of summer cowpea in the system

is unlikely to influence the growth and yield of rice

planted after puddling. In the plots not receiving either

N or P fertilizer, yields of both rice and wheat were

invariably smaller under summer cowpea than those

under summer fallow, though the differences were not

necessarily significant every year.

After the wheat harvest, the NO3-N content below

30 cm soil depth was smaller in N and P fertilized

plots compared with those receiving N alone, and also

smaller in summer cowpea plots compared with sum-

mer fallow ones during all 3 years (Fig. 3). As the

NO3-N content in the soil profile was not measured

after the cowpea harvest, it is difficult to explain the

relatively low NO3-N content at greater depth under

the summer cowpea treatment in terms of N removal

by cowpea. It is, however, possible that a better-

established wheat crop in summer cowpea treatments

absorbed NO3-N from deeper profile-depths, which

ultimately resulted in low NO3-N content at these

depths, as compared to the summer fallow treatments.

If NO3-N content in lower profile-depths is considered

to be an indicator of N leaching, the results of this

study inferred that the extent of N leaching can be

minimized with N and P fertilization at recommended

rates to both rice and wheat, as well as with inclusion

of summer cowpea. In the intensively cultivated areas

of northwestern India, particularly those managed

under the irrigated rice–wheat system with heavy

fertilizer N dressings, leaching of NO3-N and resultant

groundwater pollution is a serious concern amongst

agricultural and environmental scientists (Aulakh and

Singh, 1997). Improvement in N use efficiency is

considered one of the most promising agro-techniques

to curb NO3-N leaching. Research efforts seeking to

increase N use efficiency are mainly focused on

balanced use of chemical fertilizers, integrated nutri-

ent management, green manuring, deep placement

of N fertilizers and use of modified urea materials

(Timsina and Connor, 2001). In the present study, AEN

and ARN increased with combined use of N and P at

recommended rates, and with inclusion of forage

cowpea in RWCS during post-wheat summer season.

5. Conclusion

The results of this study show that in UGP the

60–70-day period between wheat harvesting and rice

transplanting can be successfully utilized for raising

cowpea as a forage legume in RWCS. Cowpea, besides

providing green forage during the otherwise forage-

scarce summer season, may also help to improve

annual productivity and nutrient use efficiency in

RWCS, provided both rice and wheat crops receive

recommended fertilizer input. There is, however, a need

to clarify the role of forage cowpea in improving the

physical properties of the soil, and in using the NO3-N

left after an adequately fertilized wheat crop.

References

Aggarwal, G.C., Sidhu, A.S., Sekhon, N.K., Sandhu, K.S., Sur, H.S.,

1995. Puddling and N management effects on crop response in a

rice–wheat cropping system. Soil Till. Res. 36, 129–139.

416 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418

Ahlawat, I.P.S., Ali, M., Yadav, R.L., Rao, J.D.V.K., Rego, T.J.,

Singh, R.P., 1998. Biological nitrogen fixation and residual

effect of summer and rainy season grain legumes in rice and

wheat cropping systems of the Indo-Gangetic Plain. In: Rao,

J.D.V.K., Johansen, C., Rego, T.J. (Eds.), Residual Effects of

Legumes in Rice and Wheat Cropping Systems of the Indo-

Gangetic Plains. Oxford/IBH Publishing Co., New Delhi, India,

pp. 31–54.

Aulakh, M.S., Singh, B., 1997. Nitrogen losses and fertilizer N use

efficiency in irrigated porous soils. Nutr. Cyc. Agroecosyst. 47,

197–212.

Awonaike, K.O., Kumarasinghe, K.S., Danso, S.K.A., 1990.

Nitrogen fixation and yield of cowpea as influenced by cultivar

and Bradyrhizobium strain. Field Crops Res. 24, 163–171.

Bremner, J.M., Keeney, D.R., 1965. Steam distillation methods for

determination of ammonium, nitrate and nitrite. Anal. Chim.

Acta 32, 485–495.

Cochran, W.G., Cox, G.M., 1957. Experimental Designs. Wiley,

New York.

Dwivedi, B.S., 1994. Effect of farmyard manure on the rate and

frequency of P application in rice–wheat system. Annual

Report (1992–1994). PDCSR, Modipuram, India, pp. 10–12.

Dwivedi, B.S., Shukla, A.K., Singh, V.K., Yadav, R.L., 2001.

Results of participatory diagnosis of constraints and opportu-

nities (PDCO) based trials from the state of Uttar Pradesh. In:

Subba Rao, A., Srivastava, S. (Eds.), Development of Farmers’

Resource-based Integrated Plant Nutrient Supply Systems:

Experience of an FAO–ICAR–IFFCO Collaborative Project and

AICRP on Soil Test Crop Response Correlation. IISS, Bhopal,

India, pp. 50–75.

Gill, H.S., Meelu, O.P., 1983. Studies on the utilization of

phosphorus and causes for its differential response in rice–

wheat rotation. Plant Soil 74, 211–222.

Hegde, D.M., Dwivedi, B.S., 1992. Nutrient management in rice–

wheat cropping system in India. Fert. News 37 (2), 27–41.

Hobbs, P.R., Morris, M., 1996. Meeting south Asia’s future food

requirement from rice–wheat cropping systems: priority issues

facing researchers in the post-green revolution era. NRG Paper

96-01. CIMMYT, Mexico.

Katyal, J.C., Singh, B., Vlek, P.L.G., Craswell, E.T., 1985. Fate and

efficiency of nitrogen fertilizers applied to wetland rice. II.

Punjab, India. Fert. Res. 6, 279–290.

Kirk, G.J.D., Yu, T.R., Choudhary, F.A., 1990. Phosphorus

chemistry in relation to winter regime. In: Phosphorus

Requirements for Sustainable Agriculture in Asia and Oceania.

IRRI, Los Banos, Philippines, pp. 211–223.

Ladha, J.K., Watanabe, I., Soan, S., 1988. Nitrogen fixation by

leguminous green manure and practices for its enhancement in

tropical lowland rice. In: Sustainable Agriculture: Green Manure

in Rice Farming. IRRI, Los Banos, Philippines, pp. 165–183.

Lauren, J.G., Duxbury, J.M., Beri, V., Razzaque II, M.A., Sattar,

M.A., Pandey, S.P., Bhattarai, S., Mann, R.A., Ladha, J.K.,

1998. Direct and residual effects from forage and green manure

legumes in rice-based cropping systems. In: Rao, J.D.V.K.,

Johansen, C., Rego, T.J. (Eds.), Residual Effects of Legumes in

Rice and Wheat Cropping Systems of the Indo-Gangetic Plain.

Oxford/IBH Publishing Co., New Delhi, India, pp. 55–81.

Mann, R.A., Garrity, D.P., 1994. Green manures in rice–wheat

cropping systems in Asia. In: Ladha, J.K., Garrity, D.P. (Eds.),

Green Manure Production Systems for Asian Ricelands. IRRI,

Los Banos, Philippines, pp. 27–42.

Nambiar, K.K.M., 1994. Soil Fertility and Crop Production Under

Long-term Fertilizer Use in India. ICAR, New Delhi, India.

Olsen, S.R., Sommers, L.E., 1982. Phosphorus. In: Page, A.L.,

Millar, R.H., Keeney, D.R. (Eds.), Methods of Soil Analysis.

Part 2. American Society of Agronomy and Soil Science

Society of America, Madison, WI, pp. 403–448.

Oussible, M., Crookstan, R.K., Larson, W.E., 1992. Subsurface

compaction reduces the root and shoot growth and grain yield

of wheat. Agron. J. 84, 34–38.

Page, A.L., Millar, R.H., Keeney, D.R., 1982. Methods of Soil

Analysis. Part 2. American Society of Agronomy and Soil

Science Society of America, Madison, WI.

Paroda, R.S., Woodhead, T., Singh, R.B., 1994. Sustainability of

Rice–Wheat Production Systems in Asia, vol. II. RAPA

Publication, Bangkok, Thailand.

Peoples, M.B., Herridge, D.F., 1990. Nitrogen fixation by legumes

in tropical and subtropical agriculture. Adv. Agron. 44,

154–223.

Piper, C.S., 1966. Soil and Plant Analysis. University of Adelaide,

Australia.

Ponnamperuma, F.N., 1985. Chemical kinetics of wetland rice soils

relative to soil fertility. In: Wetland Soils: Characterization,

Classification and Utilization. IRRI, Los Banos, Philippines,

pp. 71–90.

Rekhi, R.S., Benbi, D.K., Singh, B., 2000. Effect of fertilizers and

organic manures on crop yields and soil properties in rice–

wheat cropping system. In: Abrol, I.P., Bronson, K.F., Duxbury,

J.M., Gupta, R.K. (Eds.), Long-term Soil Fertility Experiments

in Rice–Wheat Cropping Systems. Rice–Wheat Consortium

Paper Series 6. Rice–Wheat Consortium for the Indo-Gangetic

Plains, New Delhi, India, pp. 1–6.

Sah, R.N., Mikkelsen, D.S., 1989. Phosphorus behaviour in

flooded-drained soils. I. Effects on phosphorus sorption. Soil

Sci. Soc. Am. J. 53, 1718–1722.

Singh, A., 1963. A critical evaluation of green manuring

experiments of sugarcane in north India. Empire J.

Exp. Agric. 31, 205–210.

Singh, B., Singh, Y., Khind, C.S., Meelu, O.P., 1991a. Leaching

losses of urea-N applied to permeable soils under lowland rice.

Fert. Res. 28, 179–184.

Singh, Y., Khind, C.S., Singh, B., 1991b. Efficient management of

leguminous green manures in wet land rice. Adv. Agron. 45,

135–187.

Singh, B., Singh, Y., Sekhon, G.S., 1995. Fertilizer N use efficiency

and nitrate pollution of groundwater in developing countries. J.

Contam. Hydrol. 20, 167–184.

Swarup, A., Wanjari, R.H., 2000. Three Decades of All India

Coordinated Research Project on Long-term Fertilizer Experi-

ments to Study Changes in Soil Quality, Crop Productivity and

Sustainability. IISS, Bhopal, India.

Timsina, J., Connor, D.J., 2001. Productivity and management of

rice–wheat cropping systems: issues and challenges. Field

Crops Res. 69, 93–132.

B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418 417

Tiwari, K.N., Pathak, A.N., Tiwari, S.P., 1980. Fertilizer manage-

ment in cropping system for increased efficiency. Fert. News 22

(3), 3–20.

Willet, I.R., 1979. The effects of flooding for rice culture on soil

chemical properties and subsequent maize growth. Plant Soil

52, 373–383.

Willet, I.R., Higgens, M.L., 1978. Phosphate sorption by reduced

and reoxidised rice soils. Aust. J. Soil Res. 16, 319–326.

Yadav, R.L., 1998. Factor productivity trends in a rice–wheat

cropping system under long-term use of chemical fertilizers.

Exp. Agric. 34, 1–18.

Yadav, R.L., Dwivedi, B.S., Prasad, K., Gangwar, K.S., 1998a.

Overview and prospects for enhancing residual benefits of

legumes in rice and wheat cropping systems in India. In: Rao,

J.D.V.K., Johansen, C., Rego, T.J. (Eds.), Residual Effects of

Legumes in Rice and Wheat Cropping Systems of the Indo-

Gangetic Plain. Oxford/IBH Publishing Co., New Delhi, India,

pp. 207–225.

Yadav, R.L., Prasad, K., Gangwar, K.S., 1998b. Analysis of

eco-regional production constraints in rice–wheat cropping

system. PDCSR Bulletin No. 98-2. PDCSR, Modipuram, India,

68 pp.

Yadav, R.L., Prasad, K., Gangwar, K.S., Dwivedi, B.S., 1998c.

Cropping systems and resource use efficiency. Indian J. Agric.

Sci. 68, 548–558.

Yadav, R.L., Dwivedi, B.S., Prasad, K., Tomar, O.K., Shurpali,

N.J., Pandey, P.S., 2000. Yield trends, and changes in soil

organic-C and available NPK in a long-term rice–wheat system

under integrated use of manures and fertilizers. Field Crops

Res. 68, 219–246.

418 B.S. Dwivedi et al. / Field Crops Research 84 (2003) 399–418