Blackwell Science, LtdOxford, UKWBMWeed Biology and Management1444-61622003 Weed Science Society of JapanDecember 2003304213221Research PaperCompetition between weeds and rice in LaosT. Inamura
et al.
Weed Biology and Management 3, 213–221 (2003)
RESEARCH PAPER
Competition between weeds and wet season
transplanted paddy rice for nitrogen use, growth and
yield in the central and northern regions of Laos
TATSUYA INAMURA,1* SHUICHI MIYAGAWA,2 OLAYVANH SINGVILAY,3 NIVONG SIPASEAUTH3
and YASUYUKI KONO4
1
Graduate School of Agriculture, Kyoto University, Kyoto, 2Faculty of Agriculture, Gifu University, Gifu, Japan,
3
National Agriculture and Forestry Research Institute, Vientiane, Laos and 4Center for South-east Asian Studies, Kyoto
University, Kyoto, Japan
To quantify the effect of weeds on the nitrogen (N) use, growth and yield of wet season paddy
rice in the central and northern regions of Laos, we surveyed the paddy fields in these regions
in October 1999 and November 2000. We found 13 weed species in total, but there were few
major weeds abundant at the survey sites. In the infertile soils under rainfed conditions, weed
growth was poor. Rough rice yield, the number of panicles, the number of seeds per square
meter, the above-ground biomass of paddy rice and the amount of N accumulated in the
above-ground biomass of paddy rice (amount of N in rice) were suppressed by competition
with weeds. However, harvest index (HI) and nitrogen use efficiency (NUE) of paddy rice
were not suppressed by competition with weeds. The amount of N in rice was suppressed by
competition with weeds, the number of panicles decreased as the amount of N in rice
decreased, and the number of seeds per square meter decreased as the number of panicles
decreased. As a result, rough rice yield was suppressed by competition with weeds. The weeds
competed with paddy rice for N uptake during the productive tillering stage. However, the
ability of paddy rice to compete for N uptake with weeds was not reduced under rainfed lowland conditions. When the weeds were completely removed, the amount of N in rice
increased. Rough rice yield may be increased by 10% under rainfed lowland conditions and
by 17–19% under irrigated conditions.
Keywords: competition, nitrogen, paddy rice, rainfed lowland, weed.
INTRODUCTION
Rice is the most important crop in Laos. In 1999, the
area planted with rice was approximately 717 600 ha,
occupying more than 80% of the total cropped land area.
Wet season paddy rice accounts for about 66% of the
total rice in area and 71% in production. In the period
from 1980 to 1998, total annual rice production ranged
from 1.0 to 1.7 million tons (Schiller et al. 2001), and
in normal rainfall years, the country is scarcely self*Correspondence to: Tatsuya Inamura, Graduate School of
Agriculture, Kyoto University, Kitashirakawa, Sakyo-ku, Kyoto
606–8502, Japan.
Email: [email protected]
Received 28 April 2003; accepted 10 July 2003
sufficient in rice. In a rainfed environment, an unusual
dry spell may cause substantial national rice deficits
(Khotsimuang et al. 1995).
The national policy for improving rice self-sufficiency in
Laos is to raise total rice production to approximately 2.2
million tons per year. Most of this increase is proposed to
come from a 27% expansion of the cultivation area and
a 29% increase in average yield of the wet season lowland
rice (Khotsimuang et al. 1995).
The major production constraints in the rainfed lowland
paddy fields (the main rice-producing environment of
the wet season lowland rice) are drought and poor soil
fertility in Laos (Schiller et al. 2001). To achieve the targeted yield in the rainfed lowland environment of Laos,
improved rice varieties (Fukai et al. 1999), soil fertility
214
T. Inamura et al.
management (Lathrilayvong et al. 1996) and integrated
pest management are needed. Recent work on the rainfed lowland rice in Laos and Thailand shows the importance of genotypic variation in nitrogen (N) uptake and
the efficiency of N use in determining genotypic variation in grain yield (Inthapanya et al. 2000).
of Laos (Table 1, Fig. 1) were surveyed in October 1999
and November 2000. Seventy paddy fields were chosen
for the survey at the rice maturity stage, considering the
average yield at the survey site.The altitude of the survey
paddy field was measured by the Global Positioning System (GPS12; Garmin International, Olathe, KS, USA).
In the rainfed lowland environment, local farmers
acknowledge weeds as one of the top three production
constraints (drought, pests and weeds). The major weed
species reported by farmers are the broad-leaved weed,
Ludwigia octovalvis (Jacq.) Raven, and the Cyperaceous
weed, Fimbristylis miliacea (L) Vahl. Marsilea minuta Presl.
is also a common problem weed in southern provinces,
while Xyris indica L. is prevalent in the central region
(Khotsimuang et al. 1995).
Major weed species in the paddy fields were identified
according to Harada et al. (1996) and Noda et al. (1994).
Weeding frequencies, irrigation methods and frequencies of the use of chemical fertilizer were recorded based
on interviews with the owners of the paddy fields.
Despite the perceived significance of weeds as a potential
yield constraint, the time committed to weed control
rarely exceeds 10 days ha-1 per cropping season. In infertile soils under rainfed lowland rice cultivation, weed
growth is poor (Ampong-Nyarko & De Datta 1993)
and subsequent labor input for weed control is saved
(Khotsimuang et al. 1995). In paddy fields with
improved soil fertility, labor input for weed control
increases are associated with increased semidwarf rice
variety and chemical fertilizer inputs. This potential significance has been reported previously (Kawano et al.
1974; Khotsimuang et al. 1995).
Previous studies have attempted to explain the nature of
competition between weeds and rainfed lowland rice in
other areas of tropical Asia (Ahamed & Moody 1982;
Moody 1982; Moody et al. 1986). Generally there are
only three or four weed species that are economically
important for rice farmers in the tropics, and a 10–75%
reduction in rice yield by weeds has been reported.
The limited information from the research on rice under
irrigated and rainfed lowland conditions suggests that
uptake of N and its utilization efficiency are important
to increasing grain yield in a low soil-fertility environment (Fukai et al. 1999). Prior studies have been confined to quantifying the competition of weeds with rice
under irrigated and rainfed lowland conditions in Laos.
The objective of the present study was to analyze the
competition between weeds and paddy rice for growth,
yield and N use in the main rice-producing environment of Laos.
Weeds and paddy rice plants within a radius of 1 m were
sampled from two spots in each paddy field. The aboveground biomass of weeds and paddy rice, rough rice
yield and yield components were determined after ovendrying at 70∞C to constant weight.The N concentration
of the above-ground part of plants in weeds and rice
were analyzed by the Kjeldahl method. Two yield
parameters of paddy rice, harvest index (HI) and nitrogen use efficiency (NUE) were calculated as follows:
HI = rough rice yield (gm–2)/above-ground part of
paddy rice (gm–2) and NUE = number of seeds (m–2)/
amount of N in rice (gm–2).
Soil samples from two spots in each paddy field were
collected as a composite of three subsamples taken from
the surface soil (0–10 cm deep). All soil samples were
air-dried and ground to pass through a 1 mm sieve
before analysis. Soil pH was measured with 1 : 2.5
(w/w) soil water suspensions with a pH meter. Total N
concentration and total carbon (C) concentration were
analyzed with a trace mass spectrophotometer (Tracer
MAT, ThermoQuest, Tokyo, Japan). Mineralizable N of
soil was obtained as the difference between the amount
of N extractable with 2 M KCl solution before and after
incubation at 30∞C for 4 weeks under waterlogged
conditions. In the analyses, the concentration of NH+4
was determined by the indophenol method.
Results were expressed as the means calculated from the
mean values of 4–15 paddy fields in the survey sites.
Means were compared by the Duncan’s multiple range
test at 5% level of significance. Data obtained were analyzed statistically using analysis of covariance to determine the levels of significance between the regression
lines.
RESULTS AND DISCUSSION
MATERIALS AND METHODS
Characteristics of the paddy fields
Paddy fields in 41 villages located in the Vientiane
municipality and three provinces in the northern region
Survey site locations are shown in Figure 1. The
Vientiane municipality (Vientiane Plain) is an ancient
Luang Namtha
Oudomxay
Vientiane
Municipality
Luangprabang
1–1
1–2
2–1
2–2
2–3
3–1
3–2
4–1
Region A
Region B
Region C
Region A consisted of survey site 1–1 and 1–2. Region B consisted of survey site 2–1 and 2–2. Region C consisted of survey site 2–3, 3–1, 3–2 and 4–1. Means followed by the same letter are
not significantly different at the 5% level based on Duncan’s multiple range test. *N = 46; **N = 50. ACF, applied chemical fertilizer; C, carbon; N, nitrogen; PF, paddy field; PFI, PF irrigated;
PFW, PF weeded.
9.32b
8.94b
21.40a
11.08b
13.83b
10.37b
8.73b
22.40a
9.13c
18.45a
13.76b
1.163b
0.983b
2.459a
2.601a
2.327a
1.352ab
1.241b
1.805ab
1.073c
2.450a
1.752b
0.097b
0.090b
0.225a
0.190a
0.188a
0.082b
0.081b
0.195a
0.093c
0.215a
0.143b
5.61b
5.32b
7.77a
5.64b
7.12a
5.74b
5.30b
5.28b
5.47c
7.16a
5.96b
18.9ab
24.0ab
27.6a
17.1b
17.4b
21.6ab
25.6ab
23.4ab
21.4a
24.6a
21.3a
15
15
10
4
8
8
4
6
30
14
26
Xaythany
Naxaithong
Xiang Ngeun
Nambak
Luangprabang
Beng
Xay
Namtha
10
11
5
2
4
4
2
3
21
7
13
0.00
0.00
1.00
0.50
1.00
1.00
1.00
1.00
0.00
0.86
1.00
0.20
0.57
1.00
1.00
0.75
0.50
0.75
0.33
0.42
1.00
0.58
0.80
0.57
0.80
0.00
0.25
0.00
0.00
0.00
0.67
0.67
0.08
Total soil
N (%)
Planting density
PR (m-2)
No.
PF
No.
villages
District
Province
Site/region
Table 1. Research paddy fields covered by the survey
Ratio
PFI
Ratio
PFW*
Ratio
ACF**
Soil pH
Total soil
C (%)
Miner.
soil N (mg
100 g-1)
Competition between weeds and rice in Laos
215
Fig. 1. Survey sites in the central and northern regions of
Laos. 1-1, Xaythany district; 1–2, Naxaithong; 2–1, Xiang
Ngeun; 2–2, Nambak; 2–3, Luangprabang; 3–1, Beng;
3–2, Xay; 4–1, Namtha.
low-level terrace approximately 150 m above sea level.
In the northern provinces, the surveyed paddy fields
occupied rather narrow river valleys, elevated 300–
700 m above sea level.
As indices of soil fertility, total N concentration of soil,
total C concentration of soil and mineralizable soil N are
shown in Table 1. In the paddy fields of the Xaythany,
Naxaithong, Beng and Xay districts, soil fertility was
low, but mineralizable soil nitrogen was higher than that
in northern Thailand (Kawaguchi & Kyuma 1969).
The soil at survey sites, except for Xiang Ngeun and
Luangprabang districts, was acidic.
Production methods
In the Vientiane municipality, none of the paddy fields
surveyed were irrigated. In the Nambak district, half of
the fields were irrigated. All other paddy fields surveyed
were irrigated from ponds or rivers (Table 1).
Chemical fertilizers were used in the survey sites of
Xaythany, Naxaithong, Xiang Ngeun and Luangprabang districts (Table 1), at a rate less than 50 kg ha-1 N
equivalent.
Hand transplanting was practiced in all paddy fields surveyed. Planting density was higher in the Xiang Ngeun
district, but lower in the Nambak and Luangprabang districts than in others (Table 1).
Ratios of paddy fields weeded to total paddy fields surveyed were lower in the Xaythany and Namtha districts
216
T. Inamura et al.
than in other districts (Table 1). None of the farmers
surveyed used herbicides, and hand weeding was the
only method for weed control. Most of the paddy fields
surveyed were weeded 40 days after transplanting. However, some paddy fields were weeded at the heading
stage, 90 days after transplanting.
Major weed species in survey sites
Thirteen weed species were recorded in total
(Table 2). However, there were few species of major
weeds (the broad-leaved weed and Cyperaceous
weed), with some regional differences. The major
weed species in our survey were the same as those
reported by Khotsimuang et al. (1995). The broadleaved weed, Ludwigia octovalvis (Jacq.) Raven, and the
Cyperaceous weed, Fimbristylis littoralis Gaudich, were
generally the most consistently cited abundant weeds,
although Xyris indica L. was common at the survey
sites in the Xaythany and Luangprabang districts.
Gramineous weeds, except for Ischaemun rugosum
Salisb, have not been reported.
Above-ground biomass, nitrogen concentration
and the amount of nitrogen in the above-ground
part of weeds
Above-ground biomass of weeds at the rice maturity
stage was greater at the survey sites of the Xiang Ngeun,
Luangprabang and Beng districts than at other survey
sites, but not statistically different among survey sites
(Table 3). The N concentration of the above-ground
part of weeds (referred to as N concentration of weeds,
hereafter) at the rice maturity stage was significantly
lower in the Nambak district, and significantly higher
in the Naxaithong district than in other districts. The
amount of N accumulated in the above-ground part of
weeds (amount of N in weeds, hereafter) at the rice
maturity stage was greater in the Xiang Ngeun, Luangprabang and Xay districts and lower in the Naxaithong
and Namtha districts than in other districts.
Classification of the survey sites
The relationship between the above-ground biomass of
weeds and crop yield varies with the crop variety, species of weed, fertilizer level and crop density (Kawano
et al. 1974; Smith 1988). To quantify the nature of
competition, the survey sites were classified into
regions A, B and C based on the growth and yield traits
of paddy rice.
The number of seeds per panicle and rough rice weight
at the survey sites of the Xaythany and Naxaithong districts were lower than those in other survey sites. In the
Xiang Ngeun and Nambak districts, HI was higher than
that in the Xaythany, Xay and Namtha districts. In the
Luangprabang, Beng, Xay and Namtha districts, NUE
was lower than that in other districts (Table 4). Based on
these results we classified the survey sites into three
regions: region A (Xaythany and Naxaithong districts),
region B (Xiang Ngeun and Nambak districts) and
region C (Luangprabang, Beng, Xay and Namtha
districts).
Table 2. Major weeds grown in research sites
Weed type
Gramineous (few)
Axonopus compressus (Sw.) Beauv
Echinochloa colonum (L) Link
Ischaemun rugosum Salisb.
Paspalum distichum L.
Cyperaceous (medium)
Cyperus difformis L.
Fimbristylis littoralis Gaudich
Scirpus juncoides Roxb.
Broad-leaved (abundant)
Ageratum conyzoides L.
Ludwigia octovalvis (Jacq.) Raven
Ludwigia hyssopifolia (C. Don) Exell
Marsilea crenata Presl
Sagittaria trifolia L.
Xyris indica L.
Plant length (cm)
15–20
20–80
60–120
20–40
10–70
25–50
20–60
30–120
75–150
20–150
5–10
60–100
30–45
1–1
1–2
2–1
2–2
2–3
1
1
3
2
2
2
1
3
3
1
1
1
1
1
3
3–2
4–1
1
1
1
1
1
2
1
3–1
2
1
1
2
2
1
2
2
1
3
2
1
1
1
2
2
1
3
1
3
1
3
1
1
1
1
Competition between weeds and rice in Laos
217
Table 3. Comparison of above-ground biomass, the amount of nitrogen (N) accumulated and N content in above-ground
parts of paddy rice and weeds at rice maturity stage among survey sites and regions
Surevey site/region
Paddy rice
Weeds
Above-ground
biomass (gm-2)
Amount N
(gm-2)
N conc.
(%)
Above-ground
biomass (gm-2)
Amount N
(gm-2)
626a
884a
823a
725a
637a
710a
739a
961a
755a
795a
750a
3.25a
4.45a
5.08ab
4.10a
4.39ab
5.89ab
3.34a
6.32b
4.02b
4.90ab
6.01a
0.56b
0.58b
0.64ab
0.58b
0.87a
0.82ab
0.63ab
0.85a
0.57b
0.62b
0.81a
50.3a
11.0a
96.9a
58.2a
97.6a
117.3a
80.2a
26.4a
30.6a
85.9a
84.5a
0.49a
0.12a
0.87a
0.44a
0.87a
0.79a
0.82a
0.25a
0.30a
0.75a
0.69a
1–1
1–2
2–1
2–2
2–3
3–1
3–2
4–1
Region A
Region B
Region C
N conc.
(%)
1.11ab
1.25a
0.84ab
0.71b
0.79ab
0.89ab
1.11ab
0.94ab
1.17a
0.79b
0.90b
Means followed by the same letter are not significantly different at the 5% level by Duncan’s multiple range test.
Table 4. Comparison of some characteristics related to rough rice yield among survey sites and region
Survey
site/region
Rough rice
yield (gm-2)
No. panicles
(gm-2)
No. seeds
per panicle
No. seeds
per m2
(¥100 m-2)
1000-seeds
weight (g)
HI
NUE
Plant length
(cm)
1–1
1–2
2–1
2–2
2–3
3–1
3–2
4–1
Region A
Region B
Region C
193a
292ab
361b
300ab
217ab
259ab
214ab
318ab
242b
344a
253b
96.4b
129.0ab
159.2a
79.5b
69.9b
73.9b
69.5b
74.6b
113.2a
136.4a
72.1b
100.5b
109.1b
112.5ab
158.2a
127.5ab
133.7ab
125.9ab
160.9a
104.9b
125.5a
136.9a
93b
138ab
167a
124ab
87b
122ab
77b
109ab
115b
155a
95b
21.2c
21.3c
22.6bc
24.3bc
24.7bc
26.6ab
24.9bc
29.0a
21.3c
23.1b
26.3a
0.309c
0.349abc
0.443a
0.417a
0.343abc
0.375abc
0.295c
0.340bc
0.329b
0.436a
0.345b
2668ab
3157a
3162a
3220a
1618b
1808b
1882b
1499c
2913a
3179a
1690b
139.3ab
138.4ab
113.1b
143.9ab
170.5a
150.1ab
156.0ab
168.2a
138.9b
121.9b
161.5a
Means followed by the same letter are not significantly different at the 5% level based on Duncan’s multiple range test.
Characteristics of the three regions
In region A, all paddy fields surveyed were rainfed and
soil was moderately acidic and infertile. In region B,
almost all paddy fields surveyed were irrigated and the
soil was fertile but not acidic. In region C, all paddy
fields surveyed were irrigated and the pH and fertility of
soil were between those in regions A and B. The percentage of paddy fields weeded was higher in region B
than in regions A and C. The percentage of paddy fields
with chemical fertilizer applied was higher in regions A
and B than in region C (Table 1).
Major and common weeds were the broad-leaved weed,
Ludwigia octovalvis (Jacq.) Raven, and the Cyperaceous
weed, Fimbristylis littoralis Gaudich, in regions A and C,
respectively. In region B, the major and most common
weed was the Cyperaceous weed, Fimbristylis littoralis
Gaudich (Table 2).
The amount of N in weeds and above-ground biomass
of weeds was lower in region A than in regions B and C,
but not significantly different among the three regions.
The N concentration of weeds was significantly higher
in region A than in other regions. The amount of N in
-0.161†
-0.149†
0.101†
-0.013†
0.354†
0.046†
*, ** Significant at the 5% and 1% level, respectively; † not significant.
Amount of
N in weeds
A
B
C
A
B
C
-0.457*
-0.579*
-0.432*
-0.506**
-0.579*
-0.434*
-0.411*
-0.586*
-0.469*
-0.476**
-0.619*
-0.469*
-0.462**
-0.554*
-0.440*
-0.518**
-0.536*
-0.442*
-0.371*
-0.605*
-0.396*
-0.440*
-0.560*
-0.401*
-0.293†
-0.453†
-0.003†
-0.292†
0.339†
0.003†
-0.447*
-0.573*
-0.478*
-0.505**
-0.566*
-0.472*
0.018†
0.456†
-0.134†
-0.026†
0.487†
-0.165†
0.002†
0.260†
0.061†
NUE
HI
1000-seeds
weight
No. seeds
per m2
No. seeds
per panicle
AGB of weeds
The regression coefficient for the regression line in
Figure 2 shows the ability of paddy rice to compete with
weeds for light. The coefficients for regions B and C
were significantly higher than that for region A. Characteristics related to a tall growth habit are positively
related to the competitive ability for light of paddy rice
(Kawano et al. 1974). Plant length of paddy rice in
region A was not shorter than that in region B. However,
in region A, paddy rice grew with tall-type weeds, Ludwigia octovalvis (Jacq.) Raven and Paspalum distichum L
(Table 2). Plant length of paddy rice relative to that of
weeds has been suggested to be important for the evaluation of the competitive ability of rice. C-4 plants are
No. panicle
A statistically significant and negative correlation was
observed between the above-ground biomass of weeds
and paddy rice in all regions (Fig. 2). The regression
coefficients and recurrence constants of the regressions
were not statistically significantly different between
regions B and C. However, the regression coefficient for
region A was significantly different from those of regions
B and C. The two regression lines for regions B and C
were not significantly different, and the regression
coefficient was -0.944. The regression line for region
A was significantly different from that for regions B and
C. The regression lines y = -2.744x + 839 and y =
-0.944x + 830 were obtained in region A and regions
B + C, respectively.
Rough rice
yield
Competition between weeds and paddy rice
for light
Accumulated
N of rice
Both the above-ground biomass of weeds and the
amount of N in weeds were significantly and negatively
correlated with above-ground biomass of rice, the
amount of N in rice, rough rice yield, number of panicles and number of seeds per square meter in all regions
(Table 5). No significant effect of the weeds on the
number of seeds per panicle, 1000-seeds weight, HI and
NUE was observed (Table 5).
AGB of rice
Effect of weeds on the growth and yield of
paddy rice
Region
Paddy rice grown in regions A and B was the paniclenumber type and its NUE was higher than that in region
C. Paddy rice grown in region C was the panicle weight
type and had heavier 1000-seeds weight, longer plant
length and lower NUE than that in regions A and B
(Table 4).
Parameter
rice and N concentration of rice grown in the research
sites were significantly higher in region C than in
regions A and B (Table 3).
-0.186†
-0.200†
0.054†
T. Inamura et al.
Table 5. Correlation coefficient between characteristics related to rough rice yield and above-ground biomass (AGB) of weeds or the amount of nitrogen (N)
accumulated in above-ground part of weeds
218
Competition between weeds and rice in Laos
14
Amount of N in paddy rice gm–2
2000
Aboveground biomass of paddy rice
219
1500
1000
500
0
0
100
200
300
400
500
12
10
8
6
4
2
600
Aboveground biomass of weeds
Fig. 2. Relationship between the above-ground biomass
of paddy rice and weeds at rice maturity in different regions
() Region A, y = 2.74ax + 839c, r = 0.457; () region B,
y = -0.70bx + 876d, r = 0.579; () region C, y = 1.14bx +
859d, r = 0.432*. Regression coefficients or regression constants followed by different letters were significantly different at the 5% level. * Significant at the 5% level.
well-adapted to a climate with high temperatures, high
light intensity and limited water supply. In rainfed and
upland conditions, C-4 weeds dominate over C-3
weeds. It is widely recognized that rice growing under
submerged soil conditions competes better with weeds
than that growing in upland conditions (Matsunaka &
Saka 1977). In region A, paddy rice was grown under
rainfed conditions (Table 1) with C-4 weeds such as
Paspalum distichum L and Fimbristylis littoralis Gaudich
(Table 2). The competitive ability for light of paddy rice
depends not only on weed plant length, but also on N
concentration of weeds growing with paddy rice
(Chisaka 1966).The N concentration of weeds in region
A was statistically higher than that in regions B and C
(Table 3). In region A, paddy rice competed with C-4
weeds and relatively tall weeds with high N concentration under rainfed conditions.
Competition between weeds and paddy rice
for nitrogen
The amount of N in weeds negatively and significantly
correlated with that in rice in all regions (Fig. 3). The
regression coefficients were not significantly different
among the three regressions, but the recurrence constants of the regressions in regions B and C significantly
differed from that for region A.The three regression lines
0
0
2
4
6
–2
Amount of N in weeds gm
Fig. 3. Relationship between the amount of nitrogen (N)
accumulated in above-ground parts of paddy rice and
weeds at rice maturity in different regions. () Region A,
y = 1.33ax + 4.42a, r = 0.476; () region B, y =
–0.53ax + 5.40b, r = 0.579; () region C, y = 1.68ax
+ 7.00b, r = 0.477*. Regression coefficients or regression
constants followed by different letters were significantly different at the 5% level. * Significant at the 5% level.
in Figure 3 were statistically parallel, and the regression
coefficient was -0.934. The regression line in region B
was significantly equal to the regression line in region C,
but regression lines for regions B and C were significantly different from the regression line for region A.
The regression lines y = -0.934x + 4.300 and y =
-0.934x + 6.173 were obtained for region A and
regions B + C, respectively.
The regression coefficient of the regression line in
Figure 3 shows the ability of paddy rice to compete with
weeds for N uptake. This was not significantly different
among the three regions. The competitive ability of rice
varies with growth stage of paddy rice at the time of
competition with weeds. No significant impact of weeds
was observed on the number of seeds per panicle and
1000-seeds weight, but a significant effect on the number of panicles and the number of seeds per square meter
was observed (Table 5). Under a low N supply, plants
primarily compete for N in the soil first (Kawano et al.
1974), and the amount of N in rice controls the number
of stems in the vegetative stage. Therefore, it is clear that
the weeds presently studied competed with paddy rice
for N uptake during the productive tillering stage.
220
T. Inamura et al.
Table 6. Correlation coefficient among rough rice yield, yield components and
the amount of nitrogen (N) accumulated in above-ground part of paddy rice
Parameter
Region A
Region B
Region C
0.974**
-0.042†
0.851**
-0.213†
0.971**
0.673**
0.777**
0.546**
0.884**
-0.412†
0.826**
0.044†
0.765**
0.681*
0.717**
Rough rice yield
No. seeds per m2
1000-seeds weight
No. seeds per m2
No. panicles
No. seeds per panicle
No. panicles
N accumulated in paddy rice
*, ** Significant at the 5% and 1% level, respectively; † not significant.
In region A, the N concentration of weeds was higher,
but that of paddy rice was lower than in other regions
(Table 3). Therefore, in region A, the competitive ability
of paddy rice for N (the regression coefficient of regression line in Fig. 3) was higher than that of paddy rice for
light (the regression coefficient of regression line in
Fig. 2). This relationship has been reported previously
(Chisaka 1966).
The amount of N in paddy rice in region A was lower
than that in regions B and C (Fig. 3), probably due to
lower soil fertility in region A (Table 1).
Mechanisms of the competition with weeds
Rough rice yield was significantly correlated with the
number of seeds per square meter in all regions, but not
with rough rice weight in regions A and B. The number
of seeds per square meter was significantly correlated
with the number of panicles in all regions, but not with
the number of seeds per panicle in regions B and C. The
number of panicles was significantly correlated with the
amount of N in rice in all regions (Table 6), and the
amount of N in rice was significantly correlated with
that of weeds in all regions (Fig. 3).
The amount of N in paddy rice was decreased by competition with the weeds. Reduction in the amount of N
in paddy rice decreased the number of panicles, which
in turn decreased the number of seeds per square meter.
Thus, the result was a decrease in rough rice yield.
than that in regions B and C. The ability of paddy rice
to compete with weeds for N uptake was not significantly different among the three regions. When the
weeds were completely removed, the amount of N in
paddy rice increased to 4.29, 5.58 and 6.64 g N m-2 in
regions A, B and C, respectively. If we assume that NUE
and 1000-seeds weight are not influenced by the amount
of N in paddy rice, rough rice yield in regions A, B and
C may be increased by 10, 19 and 17%, respectively, by
perfect weed control.
In infertile soils in areas under rainfed lowland rice cultivation, weed growth is poor and the effect of weeding
on rice yield is low. However, in fertile soils under irrigated rice cultivation, weed growth is vigorous and the
effect of weeding on rice yield is high.
The present results will be useful for setting the economic
thresholds for weed control, not only in rainfed lowland
rice cultivation but also in paddy fields with improved soil
fertility. Moreover, they suggest that weed control practices completed up until rice tillering could lead to
important yield increases in Laotian rice production.
ACKNOWLEDGMENTS
We wish to thank Dr Y.Yamasue of the Graduate School
of Agriculture, Kyoto University, for his advice on the
identification of weed species. The present study was
supported by a Grant-in-Aid for Science Research from
the Ministry of Education, Culture, Sports, Science and
Technology of Japan (No. 11309006).
CONCLUSION
We quantified the impact of weeds on the N use, growth
and yield of wet season paddy rice in the central and
northern regions of Laos. In region A, all paddy fields
surveyed were rainfed and their soil fertility was lower
REFERENCES
Ahamed N.V. and Moody K. 1982. Weeds in cropping systems as affected
by landscape cells position and weeding regime land with a low
ponding potential. Philippine Agriculturist 159–168.
Competition between weeds and rice in Laos
Ampong-Nyarko K. and De Datta S.K. 1993. Effect of light and nitrogen
and their interaction on the dynamics of rice-weed competition. Weed
Res. 33, 1–8.
Chisaka H. 1966. [Competition between rice plants and weeds]. Weed
Res. Jpn. 5, 16–22 (in Japanese).
Fukai S., Inthapanya P., Blamey F.P.C. and Khunthasuvon S. 1999.
Genotypic variation in rice grown in low fertile soils and droughtprone, rainfed lowland environments. Field Crops Res. 64, 121–130.
Harada J., Shibayama H. and Morita H. 1996. Weeds in the Tropics.
Association for International Cooperation of Agriculture and Forestry,
Tokyo, 1–304.
Inthapanya P., Sipaseuth Sihavong P., Sihathep V., Chanphengsay M.,
Fukai S. and Basnayake J. 2000. Genotype differences in nutrient
uptake and utilization for grain yield production of rainfed lowland
rice under fertilized and non-fertilized conditions. Field Crops Res. 65,
57–68.
Kawaguchi K. and Kyuma K. 1969. Lowland Rice Soils in Thailand. The
Center for Southeast Asian Studies, Kyoto University, Kyoto, 270.
Kawano K., Gonzalez H. and Lucena M. 1974. Intraspecific competition,
competition with weeds, and spacing response in rice. Crop Sci. 14,
841–845.
Khotsimuang S., Schiller J.M. and Moody K. 1995. Weeds as a production
constraint in the rainfed lowland rice environment of the Lao PDR.
In: Proceedings of the 15th Asian-Pacific Weed Science Society Conference
221
(Tsukuba, Japan, 24–28 July 1995). Asian-Pacific Weed Science Society,
Tsukuba, 444–454.
Lathrilayvong P., Schiller J.M. and Phommasack T. 1996. Soil limitations
for rainfed lowland rice in Laos. In: Breeding Strategies for Rainfed
Lowland Rice in Drought Prone Environments. ACIAR, Ubon
Ratchathani, 74–90.
Matsunaka S. and Saka H. 1977. [C3 and C4 plants classification and weed
control]. Weed Res. Jpn. 22, 177–183 (in Japanese).
Moody K. 1982. Weed control in sequential cropping in rainfed lowland
rice growing area in tropical Asia. Weed control in small farms. AsianPacific Weed Science Society, Honolulu, 49–62.
Moody K., De Datta S.K., Bhan V.M. and Mamna G.B. 1986. Weed
control in rainfed lowland rice. In: Progress in Rainfed Lowland Rice,
International Rice Research Institute, Los Banos, 359–370.
Noda K., Teerawatsakul M., Prakongvongs C. and Chaiwiratnukul L.
1994. Major Weeds in Thailand. Botany and Weed Science Division,
Department of Agriculture, Bangkok.
Schiller J.M., Linquist B., Douangsila K., Inthapanya P., Douang
Boupha B., Inthavong S. and Sengxua P. 2001. Constraints to rice
production systems in Laos. In: Increased Lowland Rice Production in the
Mekong Region (ed. by Fukai S. and Basnayake J.). ACIAR, Tokyo, 3–
19.
Smith R.J. 1988. Weed thresholds in southern USA rice, Oryza sativa.
Weed Technol. 2, 232–241.