| dc.description.abstract | Food shortage is a major problem in sub Saharan Africa as population increases. Over 89% of
Kenyans are food poor and are malnourished especially in rural areas. Water deficit is a major
problem 'in rice grown under rainfed conditions in Kenya. It affects plant growth and
development and ultimately leads to a considerable yield reduction or crop failure. NERICA
(New Rice for Africa) are high yielding rainfed rice varieties with early maturity and have shown
high potential to revolutionize rice farming even in Africa's stress afflicted ecology. However,
the effect of drought in NERICA rice varies with the variety and its coincidence with different
growth stages. It is therefore necessary to establish the effect of water deficit at different stages
of crop growth in order to develop crop management strategies to minimise risk in rice
production. The objective of this research was to investigate the effects of water deficit on
growth and development of NERICA rainfed rice with a view of identifying the most drought
tolerant variety that may be grown under rainfed conditions in Kenya. This study was carried out
in the University botanic garden, Maseno during 2008-2010. Plants were subjected to water
deficit treatments in the green house and in the field. The experiment layout was factorial 2 in a
Randomized Complete Block Design (RCBD) with three replicates. The seeds were sown in 20
litres Polyvinyl chloride pots (PYC pots) in the green house and in 2 m x 2 m plots with a
spacing of 0.5 m between the plots in the field. The treatment combinations in the greenhouse
consisted of three levels of water deficit, viz T1-well watered throughout the lifecycle; T2-water
deficit at vegetative stage; T3-water deficit at reproductive stage; and five varieties of NERICA
rice. The field experiment was based on rainfall precipitation with supplemental irrigation during
long drought durations. The parameters measured included soil moisture content, physiological
parameters which included germination rate, transpiration rate, stomatal conductance, net
photosynthesis, chloronhvll fluorescence, electron transport rate and leaf relative water content.
Biochemical parameters included chlorophyll and protein content. Morphological parameters
included plant biomass, plant height and root length. Yield and yield component parameters
included tiller number, panicle length, days to heading, harvesting and maturity, field grain ratio
and yield. The collected data was subjected to analysis of variance (ANOY A) and means
separated using least significance difference (LSD). Correlation analysis was used to examine
the nature and strength of the relationship among parameters. The parameters recorded a similar
trend in the green house and in the field. Soil water content, physiological, biochemical and yield
parameters of the five NERICA rice varieties were affected more by reproductive stage water
deficit as compared to vegetative stage water deficit while the morphological parameters were
more sensitive to water deficit occurring at vegetative compared to water deficit at reproductive
stage. The most sensitive growth stage to water deficit was the reproductive stage with yield
reduction from 44% to 58% depending on variety. The overall results indicate that there is
genetic variability in the NERICA rice varieties studied and NERICA 2 and 4 were tolerant to
water deficit occurring at vegetative or reproductive stage as compared to NERICA 1, 3 and 5
because their physiological, biochemical and yield parameters were less affected by water
deficit. NERICA 2 and 4 could play an important role in improving rice production in Kenya and
therefore their production in various agroecological areas under rainfed condition should be
cncDUl-c.;;?j ~;-~~ thf3)' qr~ W?t~,. rl~f;,:it toleraT't Tl-.prp is MPPrl t(\ rlptf",..,..,i1"~ I"rm<:llmntivp,
moisture level for the NERICA varieties that will not affect the yield severely hence denying a
farmer chances of breaking even in their production. Two peer reviewed publications have been
published from the results of this work and are attached at appendix 2, | en_US |
