Tải bản đầy đủ - 0trang
VI. Nitrogen from Green Manure Crops
GREEN MANURING IN WETLAND RICE
Nitrogen Fertilizer Equivalence (NFE) of Green Manures
S. aculeatalsunn hemp1
S. rostratalsunn hemp
S . aculeata
Rinaudo et ul. (1983)
Bhardwaj and Dev (1985)
Crozat and Sangchyosawat
Jiao et al. (1986)
Morris et a/. (l986a)
Alazard and Becker (1987)
Ventura et al. (1987)
Ghai et a/. (1988)
Roy et al. (1988)
Williams et al. (1957)
Ten Have (1959)
Vachhani and Murty (1964)
Dargan et al. (1975)
Tiwari et al. (1980)
Bhardwaj et al. (1981)
Palm et ul. (1988)
Sharma and Mittra (1988)
O.P. Meelu (unpublished
97- I 50
Beri et a/. (1989a,b)
Morris et al. (1989)
Rabindra et al. (1989)
Shukla et al. (1989)
Y . Singh er a / . (1990)
F, flowering; GM, green matter.
falls between 25 and 50% in maize as compared to about 100% in rice
(Table V). Krishna Rao et al. (1961) and Jiao et al. (1986) have reported
that green manure N was as effective as fertilizer N when added on equal
N basis. Anti1 er 01. (1988) computed optimum N rates for wetland rice
as 152, 74, and 66 kg Niha after fallow, S . aculeata, and sunn hemp, re-
YADVINDER SINGH E T A L .
spectively. Like fertilizer N , application of green manure N in excess
of that needed for the highest rice yields could result in its poor efficiency
(Bhardwaj, 1982; Bhardwaj and Dev, 1985; Morris er al., 1989).
B. N TRANSFER
Similar to fertilizer N , the recovery of green manure N will vary considerably with soil type, management, and weather. Working on a clay soil
and using I5N labeled vetch green manure, Westcott and Mikkelsen (1987)
recorded 10% N recovery of added vetch N (60-120 kg N/ha). Recovery
was 16-25% with ammonium sulphate added on an equivalent N basis.
Apparent N recovery was also lower with vetch N (9-26%) than with
ammonium sulphate (19-53%). They observed that when conditions inhibited the early season mineralization of green manure N , recovery rates of
green manure N would not compare favorably with that of fertilizer N.
Contrarily, Williams and Finfrock (1962) working on different soils and
rice cultivars found apparent vetch N recovery rates as high as 103%.
Chapman and Myers ( 1987) indicated that inefficient management of green
manure resulted in its low efficiency in flooded rice. Apparent green
manure N recoveries have been reported to vary from 25 to 58% (Table
VI). Generally, recoveries of N from green manure were similar to those
A N D FERTILIZER
The amount of N accumulated by different green manures is not likely to
be able to provide the levels of N currently required by the high yielding
rice cultivars (Talley and Rains, 1982). Therefore, to achieve the yield
potential it often needs to be supplemented with inorganic fertilizers.
Supplementing basally applied green manure with top dressed fertilizer N
helps in maximizing both green manure and fertilizer N recovery (Morris
et al., 1986a,b). In situations where green manure N is sufficiently high, it
may be advisable to cover more area to green manuring so as to get
increased rice yields. This practice will also help avoid crop injury caused
by excess application of green manure.
Several workers (Beri and Meelu, 1981; Nagarajah and Nizar, 1982;
Rekhi and Meelu, 1983; Khind et al., 1985; Joseph, 1986; Rana et al., 1988;
Mahapatra and Sharma, 1989; Kalidurai and Kannaiyan, 1990) have reported that green manure plus 50% of the recommended fertilizer N resulted in higher rice yields than when recommended N rates were applied.
GREEN MANURING IN WETLAND RICE
Recovery of Green Manure N by Rice under Different Soil and Climatic Conditions
N recovery (95)
Appurent N recovery
A . nilotrca
S . rostrata
Morris el a / . (1986b)
Alazard and Becker (1987)
Ghai et al. (1988)
Y.Singh et al. (1988)
John et a/. ( 1 9 8 9 ~ )
Rabindra et a / . (1989)
Ladha rt al. (1989)
Shukla rt a / . (1989)
" N rrcouerv
S . uculeata
Gu and Wen (1981)
H u a n g e t a l . (1981)
Mo and Qian (1983)
Mean of 10 experiments.
Tiwari et ul. (1980) reported that green manure plus 40 kg fertilizer N/ha
produced rice yield comparable with 120 fertilizer N/ha alone. Ishikawa
(1988) reported that combining green manure and fertilizer N (56 : 38)
resulted in greater rice yields than from fertilizer N alone. Rabindra ef u l .
(1989) showed that application of a part of N (30%) through green manure
produced significantly more rice yield and N uptake than from 100 kg urea
N/ha alone. In most of the studies, rice yield potential was high when
green manure and an optimum quantity of fertilizer N were applied toget her.
Since a major fraction of green manure N is released and becomes
available to the rice plant within 2 to 3 weeks of its incorporation, the N
requirement of rice in the early growth period could be met by green
manure. Thus, Desai et al. (1957) observed that in rice green manured with
sunn hemp, top dressing of fertilizer N after first weeding (25-30 DT)
produced significantly higher yield (10%) than its application at transplant-
YADVINDER SINGH ET AL.
ing. Meelu and Morris (1988) showed that N applied with green manure in a
single dose at panicle initiation growth stage increased rice yield more than
the split-applied N . Results of field experiments conducted on a coarsetextured soil showed that Sesbania N was sufficient for the wetland rice at
the early growth stages, and application of fertilizer N at transplanting
could be delayed without any adverse effects on rice yield (Khind et af.,
1987b; Singh et al., 1987). Biswas (1988) reported that application of 30 kg
Nlha as urea 5-7 days before panicle initiation with basal incorporation of
Sesbania green manure improved grain yield, N uptake, and agronomic
efficiency. There is an obvious need to work out precise rate and time of
fertilizer N application to rice grown on fields amended with green manure
under different soil and climatic conditions.
VII. TRANSFORMATIONS OF GREEN MANURE NITROGEN IN
WETLAND RICE SOILS
The fate and availability of N in green manures are determined by the
rate and extent of manure decomposition and associated N mineralization.
The composition of green manure affects the overall mineralization rate.
The stem and roots of green manure are generally poorer in N content and
have wider CIN ratios than that of foliage portion (Palm et a f . , 1988).
These parts decompose slowly and could result in immobilization of N.
The net N release from green manure is the balance of all the N transformation processes occurring in the different parts of green manure. The
efficiency of green manure N should depend on the extent of N losses,
and the rate of N supply to the growing crops through mineralizationimmobilization turnover rate.
A. N MINERALIZATION
Release of mineral N from green manure is initially rapid, but slows
markedly within a fairly short time (P. K. Singh et al., 1981; Khind et af.,
1985; Y . Singh et al., 1988; Bhardwaj and Dev, 1985). This behavior is
typical of most organic amendments (Van Faassen and Smilde, 1985). In
the absence of rice plants, N release increased rapidly up to 2 weeks and
reached a plateau (Aspiras, 1966; Nagarajah, 1988; Nagarajah et af., 1989)
(Fig. 3). In the presence of rice plants, soil solution NHd+-N peaked at 2-4
weeks after green manure addition and then declined to low levels at 6-12
weeks (Fig. 3). The decline was attributed largely to plant uptake. The time
GREEN MANURING IN WETLAND RICE
DAYS AFTER TRANSPLANTING
FIG.3. Effect of Sesbania green manure on exchangeable NH4+ N in a field (a) without
rice crop and (b) with rice crop. (Adapted from Nagarajah, 1988.)
of occurrence of peak of mineral N released from green manure varied
greatly in different studies. It should not be surprising keeping in view the
effect of plant composition and environmental factors on rate of N release.
In some studies conducted in the absence of rice plant, a decline in
NH4+-Nafter a peak has been observed (P. K. Singh et al., 1981; Bhardwaj and Dev, 1985; Khind et al., 1987b; Beri et al., 1989b). It could be due
to losses of N via nitrification-denitrification and volatilization as NH3.
The simplest model representative of the N mineralization kinetics of
organic substrates added to soil is the mathematical formulation for the
first-order kinetics. Frankenberger and Abdelmagid (1985) found first order N mineralization rate constants for the legume residues incubated at
field capacity moisture regime to range from 0.045 to 0.325Iweek. Bouldin
(1988) proposed a simple two-component model to describe the N mineralization pattern of green manures. The organic material is treated as if there
are two distinct components-one decomposes rapidly, the other slowly.
It was proposed that 65% of the added N mineralizes during the first crop,
14% mineralizes during the second crop, and 3.3% mineralizes during each
succeeding crop. Recent work of Y. Singh rt al. (1988) indicated that N
mineralization kinetics of green manure could be described by two simultaneous first-order reactions: an initial fast reaction ( k = 2.12/week)
followed by a slow release of inorganic N ( k = 0.069Iweek). Gale and
Gilmour (1988) studied C and N mineralization from alfalfa as a threephase process. Under anaerobic conditions, the rate constant for slow
phase was near zero, and for rapid and intermediate phases these were
0.118 and 0.024/day, respectively. Gilmour et al. (1985)confirmed that the
amount of net N mineralized was related to net C mineralization.
There have been few studies on decomposition and N mineralization of
green manures under waterlogged conditions. And in some of the studies
inconclusive results have been obtained due to simultaneous loss of miner-