| Abstract | Water scarcity is a key consideration in irrigated agriculture. Innovations in water saving
technologies are dire needs in today’s rice culture to produce staple food for Asian
people. A series of studies was conducted to investigate the effect of different cycles of
inundation and suspension of irrigation and its timing and durations on growth and yield
of rice under conventional rice production (CRP) and system of rice intensification (SRI)
for increasing grain yield and water productivity. The studies were conducted in the
Agricultural Science Research Farm of the Asian Institute of Technology, Thailand
during October 2007 to August 2009.
The study comprised three different experiments. The first experiment tested two
methods of establishment [viz. direct seeding (DSR) and transplanting with 8-day old
seedlings (TPR)] under three water management schedules [viz. conventional water
management (CWM), two-week inundation followed by two-week suspension of
irrigation (I2S2) and one-week inundation followed by three-week suspension of
irrigation (I1S3] in a 2 x 3 factorial experiment arranged in a randomized complete block
design with four replicates. The selected growth and yield parameters were recorded and
also simulated using CERES-Rice model in DSSAT 4.0.2.0 with concurrent weather data
and compared with the observed data.
Soil moisture contents (SMCs) at 0-15 cm and 15-30 cm soil layers reached near the
lower level of the readily available water (RAW) during drying period of 14 days in I2S2,
21 days in I1S3, but never reached below RAW. There were interactions between
establishment method and water management for leaf area index (LAI), root parameters,
yield and yield components, harvest index and water productivity. Highest LAI was
recorded by I1S3 in DSR. The same combination reported 11% more grain yield than
continuous inundation. There was no significant difference in yield between I1S3 of DSR
and I2S2 of TPR. In addition, higher productive tiller number per hill and filled grain
number per panicle, and lower grain sterility were also found in the same treatment
combination. Highest water productivity (WP), root length density and rooting depth
were found in the I1S3 water management with DSR. CERES-Rice model showed
excellent results in predicting days to anthesis and time to maturity, biomass and harvest
index at maturity. However, simulations of leaf area index, panicles per hill, soil
moisture deviated from observed values. Overall results showed that there are greater
benefits of water saving and income from direct seeding over transplanting, and in the
modified water management of 1-week inundation followed by 3-week suspension of
irrigation (I1S3) over 2-week inundation followed by 2-week suspension of irrigation
(I2S2) in SRI.
The second experiment investigated plant growth, yield and yield components of rice
when transplanted with varying ages under alternating inundation followed by
suspension of irrigation water management. In a greenhouse pot study, nine ages of
seedlings (viz. 4- to 12-day old) were transplanted and performance was compared with
DSR in a completely randomized design with three replicates.
Soil moisture content in the 0-15, 15-25 and 25-50 cm soil layers reached near the lowest
level of RAW during suspension of irrigation in all the treatments, but never reached
below RAW. In the 15-25 cm soil layer, SMC reached PWP at the second period of
suspension of irrigation. The number of tillers, leaf area per hill, plant height and above
iv
ground non-grain biomass did not differ between DSR and transplanted rice (TPR) with
4- to 12-day old seedlings. In all the treatments, maximum tillering occurred between 5
and 6 weeks after establishment. The root depth, total root length, root dry weight and
shoot to root ratio were significantly greater in DSR than TPR.
There was a significant effect of seedling age at transplanting on number of productive
tillers per hill and filled grains per panicle, panicle length, 1000-grain weight, grain yield
and panicle setting rate. The harvest index was not significantly affected. Establishment
of rice by direct seeding and transplanting up to 12 days had shown no difference in
handling the transplanting shock in a short time and resuming growth phase rapidly.
There was no difference in the tillering potential and the growth dynamics of seedlings
of all ages used, but 9-day old seedlings had higher yield under alternative inundation
and suspension of irrigation (AIS) water management of rice.
In the third experiment, four different water management schedules [viz. conventional
water management (CWM), two-week inundation followed by two-week suspension of
irrigation (I2S2), one-week inundation followed by three-week suspension of irrigation
(I1S3) and one-week inundation followed by four-week suspension of irrigation (I1S4)]
were evaluated in greenhouse pots with transplanted rice to determine the most
appropriate method for increasing or maintaining grain yield and water productivity.
Experimental design was completely randomized design with six replicates.
The commencement of flowering and physiological maturity of rice occurred in the
shortest period with CWM, and delayed with increasing the period of suspension of
irrigation during vegetative phase. The I1S4 treatment had the longest period in both
flowering and physiological maturity. The number of tillers at flowering and harvesting,
plant height at panicle initiation and flowering, leaf area per hill at flowering, and shoot
biomass, root depth, root length and root diameter, and shoot to root ratio at harvest were
significantly affected by water management practices. Free proline accumulated in
leaves was lowest in CWM, increased with increasing the period of suspension of
irrigation, and the highest was in I1S4 treatment. The grain yield and WP were also
significantly affected by water management method: CWM had the highest grain yield
while I1S3 had the highest WP. CWM used 0.74 m3/pot (equivalent to 26,000 m3/ha) of
water during the growth of rice. Compared to CWM, water management treatments I2S2,
I1S3 and I1S4 used 79, 72, and 77 % water, respectively during the lifespan. Therefore,
I1S3 water management could provide both conservation of water and increased grain
yields. |