Rice (
Oryza sativa L.) is a major staple crop for a majority of the population worldwide (
FAOSTAT, 2016). More than 3.3 billion people in developing countries depend on rice for their calories (
Zeigler, 2017). In Kenya, rice is the third most important food crop after wheat and maize and production meets less than 15% of the domestic demand with a national consumption estimated at 983,000 metric tonnes compared to an annual production of 181,000 mt (
NRDS 2, 2019).
In Kenya, about 80% of the rice is grown under irrigated ecosystems, the largest of which is in Mwea Irrigation Schemes (
NRDS 1, 2008). Mwea Irrigation Scheme is in Kirinyaga county with an established 28,600 acres of paddy rice (
NIA, 2022). Annual mean temperature average is 23°C with a RH of 78%
(Kihoro et al., 2013). Soils are vertisols, black and cracking (
Sombroek, 1982). Basmati 370 paddy rice is the main variety grown with an estimated total annual production of 180,160,000 metric tons, which is 64% of the total rice produced in Kenya (
Economic Survey, 2025).
The main production season in Mwea is July to December and this is based on irrigation water availability (
NIA, 2022). On irrigation water availability to enable all the farmers to equitably share the inadequate irrigation water resource (
NIA, 2022). Sowing however is between July and August (
Ngige, 2004). Predominantly, Basmati 370 rice variety is grown in Mwea (
MRGM case report, 2020). Use of certified seeds and basal and top-dress inorganic fertilizers remain high at 61%, 82% and 93%, respectively (
Muhunyu, 2012).
In Mwea, Basmati 370 rice yields depends on the method of establishment, timing, agronomic practices and prevailing environmental conditions. The perod of June-July normaly experiences lower average temperatures (Fig 1). The effects of rice growth and yields as affected by environmental conditions have been reported
(Ionebala et al., 2023). Low temperatures have the effect of affecting pollen fertility (
Ashkra and Krishna, 2023). There also exist variations in agronomic practices including fertilizer application rates and regimes
(Muhunyu et al., 2012). Reported findings indicate that optimum fertilizer application is important for optimum growth, tillerimg and yields
(Peng et al., 2010).
According to
NIA (2022) the productivity of Basmati 370 is 3.95 t/ha, under continuous flooding and 21 Day old transplanting.
Ndiiri et al., (2017) reported average yields of 7.1 t/ha under System of Rice Intensification (SRI) (8-day old seedings with Alternate wetting and drying) and 5.3 t/ha under conventional flooding; 28-days old seedings, grown under continuous flooding. In a survey conducted within Mwea Irrigation Scheme in 2012,
Muhunyu (2012), reported average yields of 5.5 t/ha, under monoculture and conventional continuous flooding system. He however noted that biotic and abiotic factors limited the upward potential. According to
Matolo (2012), average grain yields of 14.85 t ha
-1 was obtained under SRI conditions of 16 plants/m
2 and 14 days seedling transplanting. This was 70% higher than 8.7 t/ha, from conventional method of 25 plants/m
2, transplanting done at 21 days after sowing. Using varied Nitrogen levels,
Njinju et al., (2018) recorded varied yields of between 4-5 -5.6 t/ha with N levels of 75 kg/ha recommended at optimum. According to
Samejima et al., (2020), Basmati rice yields are affected by the establishment period and the prevailing environmental condition at the time. Average yields of 8.8 t/ha, were obtained in October planting, while yields of less than 1.3 t/ha in March- April, transplanting period. The low results were attributed to the low temperatures during the June month leading to poor fertilization
(Guo et al., 2025). The reported research and farmers productivity data are varied and have not been systematically documented.
The objective of the study is to establish key farmers field practices and yields and compare with the research yields and document. This is based on the hypothesis that ‘there are no differences between the farmers yields and research yields’ in Mwea irrigation Scheme. In this study, farmers’ field practices were surveyed, yield established and compared with a five-monthly cropping yield trials, to determine Basmati rice yields in Mwea Irrigation Scheme, during the year 2021/2022.
Experimental site
The survey and trials were conducted in Mwea Irrigation Scheme (MIS) and Mwea Irrigation Agricultural Development Centre (MIAD), respectively. during the months of August 2021 to January 2022, which is the main rice growing period in the area. MIAD is within MIS. The main cropping season is August-December although some farmers practice double cropping and harvest in May, just before the June cold season sets in.
Mwea Scheme is located at an altitude of about 1100 meters above sea level, with an annual average rainfall of 900 mm at 0°37'S and 37°27'E. The climate is tropical within agro-ecological zones (AEZ) Lower Midland 3 (LM3) and Lower Midland 4 (LM4). The rainfall pattern is bimodal with an annual mean of about 930 mm with 66% reliability. The average temperature in Mwea is 22°C. The soils are predominantly vertisols (black cotton soils), imperfectly drained.
Weather for the period
The average temperatures and relative humidity for the period are shown in Fig 1. During the period, average, minimum and maximum temperatures were 23.4°C, 17.4°C and 29.5°C, respectively. March 2022 had the highest average temperatures while August 2021, had the lowest. Average relative humidity was 75.6% with December (80.2%) and March (70.3%) being the most and least humid months, respectively.
Soil nutrient status in mwea irrigation scheme
The soil pH was 6.4, slightly acidic and optimum for rice growth. The nutrients P and K were generally low with averages of 6.0 ppm and 81 ppm, respectively. Levels of P were however high in parts of Tebere. There were optimum levels of N and organic matter in the soil. Based on the soil nutrient status, the recommended nutrient N, P and K application rates for the Basmati370 rice variety are; 80 Kg N, 60 Kg P
20 and 60 Kg K
20 per hectare
(Wanjogu et al., 1995).
Experimental design and treatments
The scheme was stratified into the existing six main units namely; Mwea, Thiba, Wamumu, Karaba, Tebere and Mutithi/Curukia (out-growers). A list of farmers, maintained by IWUA, was obtained and 180 farmers, 30 each, from each unit selected in a sequence of 10; 1
st,10 and 300
th. Where a farmer was missing or land left to a leasee, the next was selected. Whereas the list indicated 5,576 farmers, Leasee farmers were eliminated, leaving 2,240 farmers. A structured questionnaire was administered to establish the common agronomic practices. The questionnaire was pre-tested and anomalies corrected, prior to final administration. In addition, three, 1 m
2 quadrant yield samples in a transect method, were taken from each of the farmers’ plots, threshed, weighed, moisture content recorded and yield determines at 14% moisture content, using the formula:
A concurrent field experiment was conducted at MIAD Centre during August to January. The period is the main cropping season, within the existing bimodal rainfall pattern and outside the cold season of June-July. January crop is harvested in May, at the onset land preparation for the main season. Basmati 370 rice variety seeds, sourced from MIAD Centre, was established in a monthly field trial, from August 2021 to January 2022. One (1) seedling each/hill, was planted at 21 days, at a spacing of 20 β × 20 cm, in plot sizes measuring 3β ×3 m, in RCB design, with three replications, every 20
th day of the month. The nutrients N, P and K were applied at MIAD standard rates of 80 Kg N ha
-1, 60 Kg P
2O
5 ha
-1 and 60 Kg K
2O ha
-1 (Wanjogu et al., 1995). Nutrient P and K were applied as basal during transplanting, while N was applied in splits, at 21 and 53 days after transplanting. Other agronomic practices were as per MIAD recommendations.
Data collection
A structured questionnaire administered provided information on crop establishment, fertilizers and herbicides used within the scheme. The total paddy harvest from each of the farmers’ one-acre plots, were weighed, moisture content recorded and yield determined at 14% moisture content. From the field experiments, plant height, tillers, grain weight, root biomass and grain yields were recorded, at 28, 45, 65, these being; active tillering, maximum tillering and heading stages of rice, respectively. Weather and soil data were also obtained from the MIAD research Centre.
Data analysis
Qualitative data analysis was done using SPSS version 10 and results were tabulated using frequency Tables. Quantitative data were subjected to analysis of variance using Statistical Analysis System, SAS (version 9.0) and Post hoc analysis, where there was significance, carried out using the Least Significant Difference test at p≤0.05.
Out of the targeted 180 farmers 167 were interviewed and crop samples obtained from their fields.
Paddy rice establishment
Paddy rice establishment is predominantly random (68%) by the farmers, majority 96.3% of whom transplant in the month of August (Table 1). Comparatively, low yields are experienced under random transplanting.
Herbicide use for weed control in mwea paddies
Nearly 100% of the farmers in MIS, were noted to be using herbicide for weed control. About 35% of the respondents reported using a combination of both early post emergent and late post emergent herbicide while 56% use only early post emergent herbicides (as 2,4-D). Use of Methsulfuron (as Topshot) was comparatively high at 35.3%. Correlation results indicate that a combination of Twiga+Dicopur gave higher yields, comparable to Pirata herbicide application (Table 2).
Fertilizer use among farmers
Fertilizer usage among farmers in MIS is shown in Table 3. About 85% of the farmers reported using basal fertilizers, containing the nutrient P and K, for crop establishment. The most predominant combination source of the nutrients P and K, was; DAP+MOP (40%), followed by TSP+MOP (24%). On sectional comparison, more farmers in Thiba (22%) followed by Wamumu (18%) and Karaba (16%), were using basal fertilizers. Mutithi section (out-growers) had the least percentage (12%) of basal fertilizer usage. The predominant fertilizer used (80%) for top dressing was Sulphate of Ammonia while use of Urea for top dressing was low (9.6%) among farmers (Table 3).
Grain yields in farmers’ fields
Farmers’ grain yields and corresponding fertilizers use is shown in Table 4,5,6 and 7. The average grain yield was 5.0 t/ha. There were significantly lower yields in Karaba (4.3t/ha) and Mutithi/out growers (4.54 t/ha). Other sections were indifferent with respect to grain yields. About 40% of the respondents reported average yields of 5.1 t/ha, 20% got average yields of 5.3 t/ha, while 16% had average 5.8 t/ha. Comparatively, more farmers in Tebere (27%) obtained higher yields (5.3 t/ha) followed by Thiba and Wamumu (20% each). Generally, farmers in Thiba section had higher grain yields, followed by Wamumu sections. Correlation analysis indicates that majority of the farmers use of DAP + MOP (70%) and SA for topdressing with optimum yields of 5.6 t/ha (Table 8).
Basmati 370 growth and yield in field trials
Plant height variation during the period
Growth and yield data for the experimental period of August to January 2021/2022 are shown in Table 9. There were significant differences in growth and yield parameters during the period. The average plant height was 119 cm with the tallest plant height attained is 125 cm from the September cropping. The shortest plant height attained was 108 cm during the November cropping.
Tillers numbers
November cropping gave the highest significant tiller numbers (31.9), followed by the August crop (23.6), while the least was in October (17.1). There were no Significant (p≤0.05) differences in the number of tillers for the months of September, October and December sowing. About 90% of the tillers were productive.
Root biomass (gm)
Root biomass Significantly (p≤0.05) varied for the period of study. The highest root biomass was for the crop sown in September while August cropping had the lowest biomass of 18.1 g/hill and 7.8 g/hill, respectively. The least Significant (p≤0.05) root biomass was from the month of August, but this was not different from that of September and December.
Thousand-grain weight (gm)
The thousand-grain weight averaged 21.9 grams for the period. The highest significant (p≤0.05) grain weight was obtained in October, but this was not different from the August and September cropping. The month of December had the least significant grain weight.
Grain yields (t/ha)
The effect of time of sowing on grain yield is shown in Table 9. December cropping had the least yields due to effect of bird pest thus considered an outlier. There were significant (p<0.05) differences in grain yield during the period. The average grain yield was 5.5 t/ha. November cropping had the highest significant grain yields (6.0 t/ha), but this was not different from August (5.9 t/ha) and September (5.6 t/ha) yields. The lowest significant yield was during the December off-season crop (1.3 t/ha). Grain yields had a sigmoid distribution with the highest proportion of farmers obtaining between 5.1-5.5 t/ha (Fig 2).
Majority of the farmers in Mwea (68%) establish rice through random (non-row) planting, with closer spacing and with a correspondingly lower yields. This may partly be attributed to the labour resource intensive requirement (
IRRI, 2007). Random planting is associated with closer spacing, nutrient and space competition and corresponding lower yields (
Banjade et al., 2023;
Goswami et al., 2020). Previous findings suggest that drudgery in rice crop establishment can be reduced by mechanization of transplanting
(Hossen et al., 2022) and this is currently not available in Mwea Irrigation scheme. Fertilizer usage was moderate (50%) and varied in different sections. There was predominant use of a combination of DAP+ MOP, followed by SA for top dressing. This combination provides N, P and K, at rates of 80 Kg N/ha, 60 Kg P
2O
5/ha and 60 Kg K
2O/ha, which is the rate recommended for Basmati paddy rice variety in Mwea
(Njinju et al., 2018). This is adequate for Basmati 370 whose yield potential is 5.5 t/ha
(Samejima et al., 2020; Muhunyu, 2012). The importance of proper nutrient management for optimum yields, previously been reported
(Zingore et al., 2022). In Tebere area, there was prevalent use of 23:23:0 as basal fertilizer. This may be associated with inherent high P and K levels in the soil (Table 10). Previous findings confirm the same (
Onderi and Danga, 2021).
Results also indicate that there was predominant (89.2%) use of herbicides among the farmers for weed control. This is an efficient weed control method and is less labour intensive
(Goswami et al., 2020). In Mwea, where weeding has previously been manual, herbicide use reduces drudgery especially to the women, who are also in reproductive roles (
JICA, 2015).
Onderi and Danga (2021), reported similar findings. The use of herbicides is cheap, efficient and has less drudgery in weed control
(Das et al., 2024).
Field trials indicated that
crop establishment in November gave higher tillers numbers, root biomass and grain yields, but not different from that of August and September cropping in the parameters measured. December cropping was least in performance. Whereas rice growth and yield are affected by weather parameters
(Ionebala et al., 2023), weather data imply no clear variation to possibly affect performance. The dip in RH during grain filling, could however not significantly affect the December crop yield. Other factors could have been the cause.
Farmers and field trial yields averaged 5.0 t/ha and 5.5 t/ha, respectively. The farmers’ yields were comparatively lower possibly due to differences in agronomic or cultural practices
(Ionebala et al., 2023; Muhunyu, 2012). Also, lower yields in sections of the schemes can be associated with inadequate irrigation water arising from lack of proper irrigation infrastructure
(Abdullahi et al., 2003). The average experimental yields of 5.5 t/ha is within reported previous findings
(Samejima et al., 2020; Onderi and Danga, 2021). The significantly high yields in the November crop can be attributed to higher tiller numbers, higher productive tillers and higher grain weight (Table 9). Lack of differences with August and September cropping indicates that August to November remain relatively good crop establishment period for optimum grain yields. It can be safely implied that the yield of Basmati 370 rice variety in Mwea Irrigation scheme is 5.5/ha.