Rice, is the world most important staple food, with nearly 90% of the production being from Asian Countries (Trivedi et al., 2018). It is a crop with the highest production after sugarcane and maize (FAOSTAT, 2012). Flooding is the most predominant method of establishment with 30% of its production worldwide under rainfed system (Trivedi et al., 2018; Farook et al., 2011). However, there is a shift towards Direct seeding of rice due to advantages of lesser production cost, increased efficiency of irrigation water, labour and energy (Rao et al., 2017). It remains a suitable alternative to transplanting in situations where labor and irrigation water are scarce and can produce yields that are comparable to transplanting method (Signh et al., 2024). However, the emerging weeds are more competitive due to lack of a suppressing water layer, a risk eliminated by transplanting (Kaur and Singh, 2017). This calls for appropriate agronomic interventions since the resultant weed problem can reduce yields by up to 75% (Singh et al., 2024). Efforts to promote Direct seeding of rice have therefore focused on early maturing varieties, effective nutrient management and integrated weed management system (Joshi et al., 2013).

In Kenya, rice is the third most important crop with an annual demand increase of 12% and production that hardly meets 20% of the demand (NRDS 2, 2019). Its production is predominantly under conventional flood system, in small holdings, within irrigation schemes, managed by National Irrigation Authority (Alele et al., 2021). Transplanting of rice remain the predominant method of rice establishment in Kenya (NRDS 2). The method is intensive on irrigation water use and climate change has necessitated alternative water saving methods (Kumar et al., 2017).  Direct seeding has therefore been recommended as an alternative method that minimizes the negative effects of wet seeding of rice and saves on resources (Alam et al., 2020). The method is however labour intensive and prone to weed problems (Singh et al., 2016; Ajaykumar et al., 2022; Hashim et al., 2024).

In rice production, the nutrient N is important for growth and yield, its application and uptake is key to optimum yields (Gu and Yang, 2022). Nitrogen is an integral part of chlorophyll, serves to increase the photosynthetic area and LAI (Maurya et al., 2021). It enhances the dry matter production, improves rice growth rate, promotes elongation of internodes and activity of growth hormones like gibberellins (Gewaily et al., 2018). Therefore, increased application improves growth, yield, panicle length, number of filled grains, panicle and grain yields significantly components (Yoseftabar, 2013). This however is to optimum levels (Mrudhula and Suneetha, 2020; Murthy et al., 2012). The grain yield of rice can also be significantly influenced by different varieties and nitrogen levels (Maurya et al., 2021; Hossain et al., 2008). There are various N fertilization strategies that have been shown to improve use efficiency, growth and yield (Zou et al., 2023). Further, different varieties have varied responses to varied nutrient levels under varied soil conditions (Segdar et al., 2014). Split dose application has been recommended depending upon the nutrient status of the soil, crop demand and sources of nutrients (Shrestha et al., 2022).

Soil pH is known to have significant impact on numerous chemical reactions involving essential plant nutrients (Penn and Camberato, 2019). It affects sorption and uptake of essential nutrients (Barrow and Hartemink, 2023). Alkaline or high pH soils, negatively affect plant development and yield production (Saleem et al., 2023). Under high pH, N volatilization in the forms of NH₃ and N₂O is enhanced, thus leading to reduced soil N (Gao et al., 2024), Root growth and nutrient absorption is also inhibited (Munns and Tesler, 2008). According to McCouley et al. (2009), high soil pH is associated with low Cation availability, affecting essential plant nutrients solubility and organic matter decomposition in the soil. Consequently, growth, yield and grain quality indices are significantly affected (Chaudhry and Sidhu, 2022).

The objective of the experiment was to establish the effects of Varied N levels on growth and yield of direct seeded rice in Kenya.

Experimental site
 
The study was conducted at Kenya Agricultural and Livestock Research Organization (KALRO) research site in mwea, Ahero Irrigation Research Station (AIRS) and Tana irrigation scheme in hola, during 2024 and 2025 short and long rainy seasons. KALRO is located at an altitude of 1159 metres above the sea level within agro-ecological zones lower midland 3 (LM3) and lower midland 4 (LM4). Rainfall pattern is bimodal with an annual mean of about 930 mm and an average temperature is 22°C. The soils in the site are predominantly vertisols (black cotton soil). AIRS is located in Kisumu County, within Kano plains, at an altitude of about 1100 meters above the sea level, with an annual average rainfall of 900 mm. Soils in the site are vertisols, with a mean pH of 6.9. Tana River scheme is located within agro-ecological zone coastal lowland 3 (CL3), in the Tana delta flood plains. It has a bimodal rainfall pattern with 800-1000 mm of rainfall per year and an average temperature of 27°C. In both Ahero and Mwea sites, the soils were slightly acidic, with near optimum P levels but deficient in the nutrient K. Soils in hola site were alkaline, deficient in N and high in nutrient P (Table 1). Average temperatures for Ahero, Mwea and Hola experimental sites were 27°C, 23°C and 32° C, respectively. There was a gradual reduction in temperatures and an increase of up to 2°C towards crop maturity, for Mwea and Hola sites, respectively (Fig 1).




 
Experimental design
 
Treatments consisted of five nitrogen levels of 0, 30, 90, 120 and 150 kg N/ha, two varieties; 08FAN10 and Komboka) under direct seeding of rice. The treatments were laid out in a randomized complete block design with a split plot arrangement, in plot sizes measuring 5x5 m in three replications. The nutrients P and K were applied at standard rates of 50 kg P₂O₅ ha^-1 and 50 kg K₂O ha^-1, respectively, to all plots by broadcasting at 14 days after sowing. Each plot was separated by a 500-gauge polythene sheet, inserted one-foot-deep, to prevent lateral nutrient mobility. The seeds were sprouted by soaking in water for 48 hours, followed by incubation for 36 hours. Sprouted seeds were then hand drilled in lines at a spacing of 30 cm apart and later thinned to 1seedling each, 10 days after sowing. All other agronomic practices were as per Kenya agricultural research organization (KALRO) recomendations.
 
Data collection
 
Data was collected for growth and yield components. Growth data was collected for plant height and tiller numbers at 28, 45 and 65 days after sowing. Yield data was obtained at maturity for Panicle length, Panicle m^-2 and grain yield, after harvesting and threshing. Days to flowering and maturity were also recorded. Soils were sampled before the experiment and analyzed for N, P, K and pH, while weather data was recorded during the experimental period.
 
Data analysis
 
Data was summarized in excel package and subjected to analysis of variance using SAS Ver 9.4. Post hoc analysis (where there was significance) was carried out using the least significant difference test at p≤05.

Growth  and yield  response of direct seeded paddy rice varieties under varied Nitrogen levels treatments are shown in Table 2a and 2b.




 
Effects of direct seeding of rice (DSR) on plant height
 
The effects of nitrogen levels, varieties and site, were significant on plant height under DRS, for the test varieties. Treatment with 150 Kg N/ha gave significantly taller plants for 08FAN10 followed by 120 kg N/ha, while 120 kg N/ha gave significantly taller plants followed by 150 kg N/ha. For Komboka variety, treatment with 150 kg N/ha gave significantly (p≤0.05) taller plants followed by 120 kg N/ha (Table 2a). The test varieties were significantly taller in Hola and shortest in Mwea experimental sites (Table 2b). The average plant height for 08Fan10 test variety was 83.7 cm while for Komboka, it was 77.2 cm.
 
Effects of direct seeding of rice on tillers m-2
                 
The effects of nutrient N and experimental site, were significant (p≤0.05) on tillers m^-2 for the test varieties. The interaction effect of varieties and N levels, was also significant (p≤0.05) on tillers at 35 DAT (Table 2a).

Tillers m^-2 increased with increase in N levels for both test varieties. Treatment with 150 kg N/ha gave significantly high tillers m^-2 for both test varieties. The effect was however not different from that of 120 kg N/ha treatment, for 08FAN10 variety. For Komboka, 120 kg N/ha treatment, had the second highest tiller numbers but not different from the effect of 90 Kg N/ha treatment. Tillers m^-2 were highest in Hola, followed by Ahero and least in Mwea. The average tillers m^-2 for 08FAN10 and komboka varieties in ahero, Mwea and Hola sites were; 343, 317.5, 433.6 and 384.2, 349.5, 421.6, respectively. There were no significant (p≤0.05) differences in tillers m^-2, for the test varieties in hola experimental site.
 
Effects of direct seeding of rice on panicles m-2
 
The effects of nutrient N, varieties and site, were significant on panicles m^-2 (Table 2a, 2b). Treatment with 150 kg N ha^-1 gave the highest significant panicle m^-2 for Komboka variety.  However, for 08FAN10 variety, the effect was not different from that of 120 Kg N/ha treatment. In Ahero and Mwea trial sites, Komboka gave significantly more Panicles m^-2, but there were no significant (p≤0.05) difference panicles m^-2 for the two test varieties in Hola site. In both ahero and mwea sites, 0 kg N/ha and 30 kg N/ha treatments had the least panicle m^-2. There were no significant (p≤0.05) differences on Panicles m^-2, for the test varieties in Hola experimental site.
 
Effects of direct seeding of rice on days to flowering and maturity
 
The effects of varieties, nitrogen, experimental site and the interaction between Varieties and nitrogen, were significant (p≤0.05) on heading and maturity period. 

Komboka variety had significantly longer days to flowering and maturity compared to 08FAN10 variety, maturing 9 days later than 08FAN10. Treatment with 0 kg N/ha, gave the earliest heading and maturation period. On average, maturity period for 08FAN10 and Komboka was 105 days and 116 days, respectively. Mwea site, had significantly (p≤0.05) the longest heading and maturity period while Hola site was shortest, with a maturation period of 9 days earlier than Mwea and 4 days earlier than Ahero sites.
 
Effects of direct seeding of rice on grain yield
 
The effect of nitrogen, varieties and season, were significant (p£0.05) on grain yields for all varieties and all trial sites (Table 2a, 2b).

Treatment with 150 kg N/ha gave significantly (p≤0.05) higher grain yields for both test varieties. However, the effect was not different from that of 120 kg N/ha treatment for 08FAN10. In Ahero and Mwea, treatment with 150 kg N/ha gave significantly (p≤0.05) higher yields but not differences from that of 120 kg N/ha. In Hola, treatment with 120 kg N/ha gave significantly higher grain yields but not different from that of 150 kg N/ha. The average grain yields were 4.9 t/ha and 5.8 t/ha for 0FAN10 and komboka, respectively. Grain yields for 08FAN10 were significantly (p≤0.05) high in ahero (6.74 and 7.9 t/ha) and least in hola site (3.4 and 4.5 t/ha).

Under DSR, Increasing N levels significantly (p≤0.05) increased plant height and tiller numbers. Treatment with 150 KgN/ha gave significantly (p≤0.05) taller plants for Komboka rice variety in all trial sites, but this was not different from that of 120 and 90 kg N/ha treatments. For 08FAN10, treatment with 120 kg N/ha gave significantly (p≤0.05) taller plants but there were significant differences in plant height in hola site. 08FAN10 variety was significantly taller than komboka. The Increase in plant height with increase in N levels can be attributed to the role of N in enhancing plant growth, Leaf Area Index, photosynthetic area, being a constituent of chlorophyll and the role in hormonal elongation of internodes (Gewaily et al., 2018). Lack of significant differences between N level treatments is indicative of suitable optimum levels, as in previous findings by Mrudhula and Suneetha (2020). Plant height differences between 08FAN10 and Komboka can be associated with genetic differences and this is in concurrence with reported findings of Segdar et al., (2014). The lack of differences in height for the two varieties in Hola site can be associated with poor nutrient N use efficiency, arising from volatilization losses caused by the existing soil salinity phenomena.  Results of soil nutrient analysis has confirmed high soil salinity and pH conditions (Table 1). Previous findings have reported the effects of alkalinity on enhancing nutrient N loss thus hindering plant growth (Neina, 2019).

Tiller m^-2 and panicle m^-2 increased with increase in nutrient N levels for Ahero and Mwea test sites. Treatment with 150 kg N/ha gave significantly higher tillers m^-2 and Panicle m^-2 but there were no significant differences for the test parameters under 120 kg N/ha treatment for Komboka variety. For 08FAN10 variety, 150 kg N/ha gave significantly higher panicles m^-2 followed by 120 kgN/ha treatment. In Holla, the effect of nutrient N treatments was not significant (p≤0.05) on tiller m^-2 and panicles m^-2 for the test varieties. Tiller numbers and panicles m^-2 were highest in Hola, followed by Ahero and Mwea test sites. The increase in tiller numbers and panicles m^-2 with increase in nutrient N levels can be attributed to the role of the nutrient N in enhancing growth, tillering and panicles m^-2 (Yoseftabar, 2013). Lack of differences beyond 120 kg N/ha for komboka suggests an optimum N level requirement Mrudhula and Suneetha (2020). Further, lack of differences in N treatments in Hola site for the test varieties can be attributed to the high pH (Table 1) and the associated resultant nutrient N volatilization. The effect has previously been reported (Neina, 2019). The comparatively lower tiller numbers and panicles m^-2 for 08FAN10 in all sites may be associated with the varietal genetic differences (Segdar et al., 2014). Further, lack of differences in N treatments in Hola site may be attributed to the effect of high pH on volatilization and loss of the nutrient N, resulting into inadequacy thus affecting the test parameters (Neina, 2019).

Increasing N levels significantly (p≤0.05) delayed heading and maturity periods, which were shorter in Hola and Ahero, than in Mwea experimental sites. Variety 08Fan10 matured earlier than komboka by 9 days and there were 4 days’ differences each, in maturation periods between Ahero, Mwea and Hola sites. Earlier maturity in Hola, followed by Ahero can be attributed to the effect of high environmental temperatures on growth and yield (Fig 1). Previous findings have shown that high environmental temperatures have the effect of shortening the overall growth period through stress thus reducing accumulative photosynthesis (Stone, 2023). The comparative differences in maturity periods by the varieties, may have been contributed to by the genetic differences and the consequent physiological response to environmental conditions (Sitaresmi, 2016). 08FAN10 variety matured earlier than komboka variety.

Treatment with 150 kg N/ha gave significantly (p≤0.05) higher yields for both Komboka and 08FAN10 but the effects were not different from that of 120 kgN/ha treatment. In Hola site, treatment with 120 kg N/Ha gave significantly higher yields while there were no differences in yield for komboka under 90 kg N/ha, 120 kg N/ha and 150 kg N/ha. Komboka variety had significantly (p≤0.05) higher grain yields than 08FAN10 but the yields were highest in Ahero, followed by Mwea and least in Hola test sites. The higher yields from high nutrient N levels is attributed to the positive effect of N on growth and yield components (Yoseftabar, 2013). The lack of differences in 150 kg N/ha and 120 kg N/ha treatment suggests that treatment with 120 kg Nha is optimum and can be recommended for both test varieties. This is in concurrence with reported findings of optimum N levels beyond which there is no increase in yields (Mrudhula and Suneetha, 2020; Murthy et al., 2012). The comparatively low yields in Hola can be attributed to higher environmental temperatures and increased pollen abortion leading to incomplete pollination and reduced yields. High temperatures have also been reported to affect physiological processes like leaf abscission, scorching, senescence and growth (Parthasarathi et al., 2022). The observed consistent differences in yield between Komboka and 0FAN10 are partly varietal and genetic (Maurya et al., 2021; Hossain et al., 2008). Whereas the two test varieties showed good alkalinity tolerance, 0FAN10 variety seems comparatively better as indicated by the comparatively less effect on yield.

Application of 120 kg N/ha gave optimum growth and grain yield for 08FAN10 and while 150 kg N/ha is optimum for komboka under DSR. Further, the variety komboka showed better performance than 08FAN10 under different environments although 08FAN10 seemed less stressed by alkalinity.

The present study was Financed by the International Rice Research Institute- Nairobi Kenya.
 
Disclaimers
 
The views and conclusions expressed in this article are solely those of the authors and do not necessarily represent the views of their affiliated institutions. The authors are responsible for the accuracy and completeness of the information provided, but do not accept any liability for any direct or indirect losses resulting from the use of this content.

The authors declare that there are no conflicts of interest regarding the publication of this article. No funding or sponsorship influenced the design of the study, data collection, analysis, decision to publish, or preparation of the manuscript. Informed consent All animal procedures for experiments were approved by the Committee of Experimental Animal care and handling techniques were approved by the University of Animal Care Committee.