Improving onion productivity and producer income through nitrogen management

.


Introduction
Intensive nutrient management for onions (Allium cepa L.) involves using fertilizer as efficiently as possible.The principle behind nutrient management is balancing soil nutrient inputs with crop requirements (Fekadu and Dandena, 2006).Nitrogen (N) is an essential nutrient whose deficiency limits crop productivity.Compared to other vegetables, onions require high amounts of nitrogen and are applied at different times dur ing the growth (Geisseler et al., 2022).The recovery of fertilizer can be low, 30 to 40% (Halvorson et al., 2002;Sharma et al., 2012), due to the shallow onion root system.Low N fertilizer recovery in onions is, in part, due to the variable amounts of mineral N present in the soil before sow ing (Brewster, 2008).Appropriate N fertilizer management requires knowledge of crop demand and the time it is needed (IPNI, 2012).
A low rate of N causes a low yield of onions due to a shortage of N required for the chlorophyll pigment that is responsible for photosynthe sis (Jilani et al., 2004;Khan et al., 2021).Excessive N is hazardous to the environment, weaken the foliage and predispose the plants to pathogenic diseases (Geary et al., 2015).These factors decreased the potential of processing photosynthesis that resulted in yield reduction due to a decrease in assimilates in plant leaves (Reay et al., 2012).
The frequency of N application affects the produc tivity of onions (Grant et al., 2012).When applying N fertilizer all at once, more N is susceptible to denitri fication, leaching, or volatilization.When N fertilizer is applied faster than plants can use it, soil bacteria convert it to nitrate and decrease N use efficiency (Zhang et al., 2013).
The majority of smallscale farmers use less nitro gen and apply it all at once, which is suboptimal (Shura et al., 2022).There is often excessive applica tion of N fertilizer by largescale commercial farmers near harvest.It is assumed that synchronization of crop demand for N fertilizer and time of application with sufficient amounts improves yield of onions and reduces waste of N, reducing production costs and environmental pollution due to unused excess N leaching in different forms (Geisseler et al., 2022).However, the optimum amount of N that matches the crop demand during the growth period is little explored.We hypothesized that the low N use effi ciency of the onion due to its shallow and sparse root system can be improved by controlled application of N fertilizer during the growth period.Therefore, this work was conducted to determine the optimum N rate and proper application frequency during the growth period for economically feasible onion pro duction.

Site description
The study was conducted at Alage Agricultural Technical Vocational Educational and Training College, near Bulbula Town, Ethiopia, under supple mental irrigation (when precipitation is lacking) from June to October 2021.The area is situated between 7°65" N latitude and 38°56" E longitude at 1600 m above sea level in the dry plateau of the southern part of the Ethiopian rift valley system.The area is characterized by a bimodal rainfall pattern where a short rainy season occurs during March and April and the main rain starts in June and extends to September, with high rainfall in July and August.The mean annual rainfall is 800 mm, and the annual mean minimum and maximum temperatures are 11 and 29°C, respectively.The soil of the area ranges from sandy loam to sandy clay loam, with some clay loam and a few clay soils, and is slightly alkaline, pH 7.8 (Alemayehu and Bewket, 2016).

Physico-chemical properties of the experimental field soil
Before planting, five soil samples were randomly taken from the field at a depth of 0 to 20 cm in a zigzag pattern using an auger.Samples were mixed to produce a representative composite sample of 1 kg.The soil sample was airdried and ground to pass 2 and 0.5 mm (for total N) sieves and analyzed for total P, total N, pH, organic carbon (OC), exchangeable cations, and physical properties at the Batu Agricultural Research Centre Soil Laboratory.
The soil was a silty clay loam with 18.0% sand, 50.5% silt, and 31.5% clay.The soil was slightly alka line in reaction, with a pH (H 2 O 1:2.5) of 7.82, which is within the range of ideal soil pH for onion bulb pro duction (Graham et al., 2004).Total N, available P, organic carbon (OC), and CEC of the soil before plant ing were 0.12%, 11.48 mg kg 1 , 1.39%, and 31.74cmol (+) kg 1 , respectively.The total N content of the soil was within the range of low, according to Havlin et al. (1999).The cation exchange capacity (CEC, 31.74 meq/100 g) of the soil was high according to the rat ing of Jackson (1975).The carbontonitrogen ratio (C:N) was 11.5%.

Planting materials
The onion, cv.Bombay Red, was used.It is adapt ed to areas of 700 to 2000 m above sea level.The size of the bulb of this variety ranges from 85 to 90 g, with a yield potential of 25 to 30 t ha 1 under research conditions (Lemma and Shimeles, 2003).Sources of fertilizers were urea (46% N) and Triple Super Phosphate (TSP) (46% P 2 O 5 ) for N and phos phorus, respectively.The nitrogen fertilizer was applied immediately after weeding.Onion seeds were sown 56 days before transplanting on a 1 m wide seed bed 10 m in length (area 10 m 2 ).Seedlings were grown under suitable conditions of fertilization, weeding, and pest control.Seven days before trans planting, seedlings were gradually exposed to field conditions and withheld from the water supply (hard ened off) in the nursery, and then manually trans planted at 56 days old.The TSP was applied during transplantation.All cultural practices and crop pro tection measures (diseases and insect control) were carried out uniformly for all plots (EARO, 2004).

Treatments, design and experimental procedure
Treatments were arranged in factorial combina tions in a randomized complete block design with three replications.While transplants were being developed, the soil was manually pulverized twice with an oxendriven plough.After the soil was pulver ized, levelling and ridge preparation were done with hoes and spades.Rates of N were: 50, 100, 150, 200, or 250 kg ha 1 in band application according to appli cation frequencies, which were: NAF1 (all at once 1 week after transplanting), NAF2 (half of the N at 1 week after transplanting, another half of the N at 21 days after transplanting), NAF3 (onethird of the N at 1 week after transplanting, onethird of the N at 21 days after transplanting, and onethird of the N at 42 days after transplanting), or NAF4 (onefourth at 1 week after transplanting, onefourth of the N at 21 days, onefourth of the N at 42 days after transplanti ng, and onefourth of the N at 63 days after trans planting).Treatment combinations were assigned randomly to experimental units within each block.The national blanket recommendation of N fertilizer for onion production is 100 kg ha 1 , which could be considered a control treatment.Double row planting was done by hand on ridges about 20 cm high at a spacing of 40 cm for water furrows, 20 cm between rows on raised beds, and 5 cm between plants within rows.There were 60 plots corresponding to the 20 treatment combinations with three replications.The unit plot size of the experiment was 2 x 2.5 m (5 m 2 ).Blocks were separated by 1.5 m, and the space between each plot within a block was 1 m.In each plot, 10 rows were prepared, and in each row, 50 seedlings were manually planted.Generally, 500 onion seedlings were planted per plot.The outer 2 rows on both sides of the plot and the 2 plants at both ends of the rows were border plants.The plants in the six central rows were used for measurements.

Data collection
Ten plants were randomly selected from each plot's central six rows, and data on growth perfor mance, quality indicators, and yield components were recorded for each of them.For the data on bulb yield, all of the plants in each plot were harvested.Thus, the following information was collected:

Maturity and growth parameters
Days to maturity were recorded as the number of days from seedling transplanting to a day at which more than 80% of the plants in each plot showed yel lowing of leaves or attained physiological maturity.Plant height was measured from the ground to the tip of the leaves on 10 randomly selected plants from the central rows in each plot at maturity.Leaf length was recorded at physiological maturity from the sheath to the tip of the leaf from the third youngest leaves of ten representative plants, which was used to count the number of leaves per plant using a ruler.Leaf diameter was measured from the thirdyoungest leaves at the bottom, middle, and tip parts of the leaves from ten randomly selected plants using a veneer caliper.Leaf number per plant was counted as the total number of leaves from 10 randomly selected plants at maturity, and the average of the ten plants was taken.The aboveground biomass was harvested by cutting the plant at the crown part, dry ing it in an oven at 650°C until a constant weight was attained, and the shoot dry matter was determined and expressed in grammes at harvest.Additionally, the total dry biomass was determined by summing the shoot and bulb dry weights of the sample.

Yield and yield related parameters
The mean bulb diameters of ten sample bulbs were measured at the maximum wider portion of matured bulbs using calipers.The bulb length of ten sample bulbs was measured along the length of the bulb from the basal end to the top end, at which the bulb neck was removed from matured bulbs using calipers after harvest.The average fresh weight of ten randomly taken mature bulbs was measured using a sensitive balance and finally expressed in grammes.Fir bulb dry weight Ten bulbs were ran domly taken from each plot and chopped, mixed thoroughly, placed in an aluminum paper bag, and put in the oven to dry at 650°C until a constant dry weight was attained.Then each sample was immedi ately recorded as a bulb's dry weight.The bulb dry matter concentration (%) was determined by ran domly selecting ten bulbs from each plot and chop ping them, mixing them thoroughly, then weighting them and recording the fresh weight.Then each sub sample was placed in an aluminum paper bag and put in an oven at 650°C until constant dry matter was attained.Each subsample was then immediately weighed and recorded as a dry matter yield.The dry matter concentration was determined using the loss weight, and the fresh sample was weighed to the nearest gramme using the formula set by Ruck (1969) and Dantata (2014): Marketable bulb yield (t ha 1 ) was determined from the weight of healthy and marketable bulbs that range from 20 g to 160 g in weight (Lemma and Shimeles, 2003).The marketable yield was deter mined from the net plot at the final harvest.Unmarketable bulb yield (t ha 1 ) was measured as the total weight of unmarketable bulbs that are under sized (<20 g), diseased, decayed, and bulbs from plants with physiological disorders such as thick neck and split were measured from a net plot at final har vest.
Undersized bulb yield (t ha 1 ) was determined by taking under sized bulbs (<20 g) as unmarketable bulbs per net plot and converted to t ha 1 as deter mined.Total bulb yield (t ha 1 ): The total bulb yield was measured from the total harvest of net plot as a sum weight of marketable and unmarketable yields that was measured in kg per plot and finally convert ed into t ha 1 .Harvest index (%) was expressed as the ratio of total bulb dry weight to the total biomass dry weight and expressed in percentage.
The TSS was determined at harvesting time from ten randomly selected bulbs per plot using the proce dures described by Waskar et al. (1999).Aliquot juice was extracted using a juice extractor, and 50 ml of the slurry was centrifuged for 15 minutes.The TSS was determined by a hand refractometer (ATAGO TC 1E) with a range of 0 to 32 Brix and resolutions of 0.20 Brix by placing 1 to 2 drops of clear juice on the prism, washing it with distilled water, and drying it with tissue paper.

Data analysis
Data were subjected to analysis of variance using Stat8 software (ver.8.1.1,Analytical Software, Tallahassee, FL).Assumptions of ANOVA were tested, and no violation was observed.The data analysis was done with a generalized linear model.The N rate and application frequency were fixed effects, and the block was random.If the interaction was significant, it was used to explain the results.If the interaction was not significant, the main effects were separated using LSD (Walter and Duncan, 1969).

Economic analysis
A partial budget analysis was made to determine the exact rate of return that producers gain on their investment by changing existing cultural practices to alternative ones.The potential response of added fertilizer corresponding to labor costs and the price of fertilizers (variable costs for urea fertilizer at 16.2 birr) throughout the crop growing season was evalu ated.The birr is the local currency.1 birr is equal to 0.031 US during the growth period.The economic outcome was analyzed using agronomic indices for N and its frequencies.The economic analysis was com puted using accepted procedures (CIMMYT, 1988), where: Avy (gross average bulb yield, average yield of each treatment); Ajy (adjusted yield, the average yield adjusted downward by 10% to reflect the differ ence between experimental yield and yield of the farmers); GFB (gross field benefit determined by mul tiplying field price that farmers receive for the crop by adjusted yield); NFB (net field benefit calculated by subtracting total costs from gross field benefit; GFB for each treatment); and MRR (marginal rate of return %) calculated by dividing change in net benefit by the marginal cost reflecting change in cost (CIM MYT, 1988).

Results and Discussion
The main effects of N rate and application fre quency influenced plant height, leaf length, leaf width, and number of leaves.The interaction effect of two factors was not significant for these parame ters.Plant height increased with increasing N rate from 50 to 250 kg ha 1 N.The tallest and shortest onion plants were at 250 kg ha 1 and 50 kg ha 1 , respectively.Onions treated with N at 250 kg ha 1 exceeded their mean height by 34.23% compared with the 50 kg ha 1 Ntreated onions (Fig. 1A).In response to An N application frequency, the tallest plants were for 3 time's N application (Table 1).Application of N at all frequencies (once, twice, and four times) was not different for plant height.The increase in plant height with the addition of higher N fertilizer could be attributed to the increased avail ability of N for growth as a result of protein synthesis and the accumulation of carbohydrates (Rizk, 2012).The tallest plants recorded at three applications of N may be because of increased N use efficiency, decreasing N loss by leaching and volatilization (Rizk, 2012).Application of N three times better matched the availability of N with crop N demand compared to application all at once, or 2 or 4 applications of N (Brewster, 2008).This result is consistent with the findings of Morsy et al. (2012) and Nasreen et al. (2007), who reported onion plant height increased as N fertilizer rate increased.
Shorter leaves were from the 50 kg ha 1 N that decreased by about 31.56% and 33.95% compared to values obtained from 200 and 250 kg ha 1 N, respec tively (Fig. 1).The effects of 100 and 150 kg ha 1 N were similar.The influence of N rate at 200 and 250 kg ha 1 was not significant for leaf length.The longest and shortest leaves were when N fertilizer was applied three times and all at once, respectively (Fig. 1B).These values agree with Rao et al. (2013), who reported that higher N fertilization increased onion leaf length, and Khan et al. (2002), where the longest leaves were when N was applied three times.Increased leaf length at three times N applications could be due to increased recovery of N by onions and decreased N loss (Ali and Ceyhan, 2001).
The narrowest leaves were from plants treated with 50 kg ha 1 N. Leaf diameter increased with increased N, from 50 to 150 kg ha 1 .The N rates of 150, 200, and 250 kg ha 1 had similar leaf diameters (Fig. 1A).Leaf diameter was influenced by the fre quency of N applications.Three times N application increased leaf diameter by 38.99% compared to applying N all at once (Fig. 1B).The increase in leaf diameter with an increase in N rate from 50150 kg Means followed by the same letters within a column are not significantly different at P<0.05.ha 1 and 3 applications of N could be associated with a better supply of N and better N use efficiency.Application of N with three or more applications and a higher supply of N could increase leaf thickness, capture resources for photosynthesis, and promote better growth and development.The lowest leaf diameter was due to the low N rate applied to onions (Abdissa et al., 2011;Woldeyohannes et al., 2013;Seid et al., 2014).
Increasing N from 50 to 150 kg ha 1 increased leaf number.Beyond 150 kg ha 1 , the number of leaves decreased.Onion leaf number increased as the N fer tilizer rate increased from 50 to 150 kg N ha 1 (Fig. 1A).Increasing N frequency from one to three appli cations increased leaf number.At 4 N applications, leaf number decreased (Fig. 1B).This may be due to a lack of enough nitrogen at the early growth stage at which leaf formation was initiated.When N is applied four times during the growth period, a quarter of it will not be used by the onion plants for the initiation of leaf formation due to late application (Geisseler et al., 2022).Increases in the number of leaves with a further increase in the rate of N could be attributed to enhanced photoassimilate production, cell divi sion, and vegetative growth (Suthar, 2009).The N plays a role in leaf production and vegetative growth (Nasreen et al., 2007).Increases in the number of leaves per plant with up to 3 applications of N may be attributed to increasing N use efficiency (Geisseler et al., 2022).This indicates that onethird of the N fertilizer in the first application was enough for leaf number production and vigorous vegetative growth, which agrees with Mengel et al. (2006).
Bulb diameter and bulb length were influenced by N rate and N application frequency, but the interac tion of N rate and frequency was not significant.Increasing the N rate from 50 to 150 kg ha 1 increased bulb diameter by 43.66%.Increasing the N rate beyond 150 kg ha 1 did not increase bulb diameter (Fig. 2B).The widest bulbs were for the 3times N application (Fig. 2B), which agrees with Nasreen et al. (2007).Development of wider bulbs with increasing frequency and rate of N fertilizer could be associated with the availability of more growth resources due to efficient N use as bulbs develop.Increasing N rate and split application are associated with promoting cell elongation, aboveground vegetative growth, and the synthesis of chlorophyll, resulting in dark green leaves (Geisseler et al., 2022).These results agree with Soleymani and Shahrajabian (2012) and Ghaffoor et al. (2003).
Increasing the rate of N application from 50 to 150 kg ha 1 increased bulb length.At N rates of 150, 200, or 250 kg ha 1 bulb lengths were similar (Fig. 2A).Increasing application frequency up to three times increases bulb length.When all N fertilizer was applied at once, bulb length was reduced (Fig. 2B).The reason for the longest bulbs at 150 kg ha 1 N could be the supply of optimum N (Fageria and Baligar, 2005).The increase in bulb length at 3 N applications might be due to the recovery of N by onions and its subsequent use for growth (Singh and Chaure, 1999;Bahadur and Singh, 2005;Mengel et al., 2006).
Bulb neck thickness was only affected by the N rate.The widest bulb necks were due to the applica tion of 250 kg ha 1 N, and the narrowest bulb neck diameter was 50 kg ha 1 N (Fig. 2A).The reason for the widest neck diameter at 250 kg ha 1 N rate might be due to high N that resulted in excessive vegetative growth and delayed maturity, resulting in a large neck size (Grant et al., 2012;Morsy et al., 2012).The increase in neck diameter might show its involve ment in the synthesis of amino acids, as they link together to form proteins and make up metabolic processes required for plant growth, including neck thickening (Jilani, 2004).
Fig. 2 The effect of nitrogen rate (A) and application frequency (B) on bulb diameter, bulb length and bulb neck of onion.
The main effects of N rate and N application fre quency influenced the shoot dry matter weight and harvest index of onion plants.The interaction effect of an N application and its application frequency was not significant.The unmarketable yield was only affected by the N rate.Bulb dry matter, dry total bio mass weight, bulb fresh weight, marketable bulb yield, total soluble solids, and total bulb yield were affected by the interaction effect of N rate and appli cation frequency in addition to the main effects.
The highest shoot dry matter weight was 250 kg ha 1 N.As the N rate increased from 50 to 250 kg ha 1 , the shoot dry matter of onions increased by 50.51% (Fig. 3A).This might be due to excessive vegetative growth that resulted from the application of excess N (Kandil et al., 2013).The lowest shoot dry weight yield on 50 kg ha 1 N might also be because of N defi ciency, which limits cell division and expansion, chloroplast development, chlorophyll concentration, and enzyme activity (Soleymani and Shahrajabian, 2012).The highest shoot dry matter weight per plant was at 3 N applications.The lowest shoot dry matter weight was from treatment at N applied once (A).Shoot dry matter yield from 3 applications of N increased by 17.42% compared to shoot dry matter produced when N was applied at one time (Fig. 3B).Three applications of N fertilizer increased the N use efficiency of onions (Sharma, 1992).The present find ing agrees with Nasreen et al. (2007), who indicated the shoot dry matter weight increased with an incre ment of N fertilizer rate applied three times.
The harvest index was increased from 63.74 to 71.51% as N was increased from 50 to 150 kg ha 1 .Beyond 150 kg ha 1 N, the harvest index decreased (62.2% at 250 kg ha 1 N) due to increase above ground dry biomass compared to the bulb weight of onions.The harvest index recorded at 50 kg ha 1 N was similar to that recorded at 200 and 250 kg ha 1 N (Fig. 3A).The average harvest index across all N rates was 66.2%.The highest harvest index at 150 kg ha 1 N might be due to increased bulb weight due to enough N sufficient for photosynthesis and assimilate pro duction that increased bulb dry weight with optimum above ground biomass resulting in a higher harvest index (Geisseler et al., 2022).The low harvest index from treatment with 50 kg ha 1 was due to low bulb dry matter weight due to N deficiency (Nasreen et al., 2007).The low results obtained from 200 and 250 kg ha 1 might be because of etiolated growth of above ground biomass and low bulb growth performance resulting from the application of excess N (Negash et al., 2009;Abdissa et al., 2011).
With increasing frequency of N up to 3 applica tions, the unmarketable bulb yield of onions decreased.The highest value of unmarketable bulb yield was all N applied once.This was followed by the 4N applications.The lowest unmarketable bulb yield was 3 N applications (Fig. 3B).The unmarketable yield from the application of all N at once was exceeded by the unmarketable yield from three applications of N.This may be due to the loss of N because of volatilization and leaching of N, which decrease N use efficiency if N supply is sufficient.Sustainable application of N during the active growth period can decrease unmarketable bulb yield (Balemi et al., 2007;Biesiada and Kołota, 2009;Soleymani and Shahrajabian, 2012).
Increasing the rate of nitrogen application up to 150 kg ha 1 increased the average bulb weight by increasing the frequency of nitrogen application up to three times.The highest average fresh bulb weight was recorded at an N rate of 150 kg ha 1 applied three times (Table 1).This treatment approximated the expected bulb weight for the cultivar.The lowest bulb size was for the 50 kg ha 1 N all applied at once (Table 1).Application of 150 kg ha 1 N three times exceeded the lowest fresh bulb weight recorded at 50 kg ha 1 N all applied at once by 66.4%.
The increase in fresh bulb weight with an increase Fig. 3 The effect of nitrogen rate (A) and application frequency (B) on shoot dry matter, harvest index and unmarketable yield of onion.
in nitrogen fertilizer rate might be due to a sufficient supply of nitrogen that enhances cell division and expansion, chloroplast development, chlorophyll concentration, and enzyme activity.The increase in fresh bulb weight with increasing nitrogen applica tion frequency up to 3 times might be due to the increasing matching of the crop demand for N with the plant requirement for N (Khan et al., 2002;Yadav et al., 2003).When all of the nitrogen is supplied ahead of crop growth, more of that nitrogen is sus ceptible to denitrification, leaching, or volatilization (Brady, 1985).
Increasing the N from 50 to 150 kg ha 1 increased the production of marketable bulbs across the board by increasing the N application frequency up to three times.Beyond that, marketable bulb yield was not increased.The highest marketable bulb yield was from onions, which provided 150 kg ha 1 N in 3 appli cations.The lowest marketable bulb yield was in response to 50 kg ha 1 N applied at once (Table 1).This might be attributed to an optimum rate of N fer tilizer and sustained application of N in relation to the demand of the crop, with reduced loss of N. Split applications can be timed to match the N available with crop demand.This reduces the residence time of fertilizer N in the soil and the risk of N being lost.The marketable bulb yield of onions per unit area is a function of the N dose supplied and the application frequency of N fertilizer (Naik and Hosamani, 2003;Latif et al., 2010).A higher marketable bulb yield was achieved at a 150 kg ha 1 rate of N fertilization applied three times (Balemi et al., 2007;Soleymani and Shahrajabian, 2012).
Total bulb yield increased in response to increas ing N rates up to 150 kg ha 1 across increasing fre quency of N application up to 3 times.The highest total bulb yield was obtained from onion plants at 150 kg ha 1 N and 3 applications (Table 1).The lowest total bulb yield was in response to the application of N at 50 kg ha 1 all at once.The increased total bulb yield in response to 150 kg ha 1 N and 3 applications might be due to plants receiving enough N. High N use efficiency and crop recovery occurred when N was applied 3 times during active growth (Tsai et al., 2012).The decrease in total bulb yield with applica tion of N beyond 150 kg ha 1 might be due to luxury consumption of N that affects onion plant metabo lism by decreasing the ability of the root surface to absorb phosphorus and decreasing the assimilate preparation (Mahdieh et al., 2012).The highest bulb yield due to the application of 150 kg ha 1 N applied three times agrees with Gebremedhin et al. (2018).Enhanced leaf number and length may lead to increased assimilate production and increased bulb yield (Geisseler et al., 2022).The dose of N up to 120 kg ha 1 increased total bulb yield, but below this rate, total bulb yield decreased (Jilani et al., 2004).A low rate of N applied all at once produced lower total yields compared to higher N doses applied in 3 splits (Balemi et al., 2007;Soleymani and Shahrajabian, 2012).
Plants treated with 150 kg ha 1 N applied three times produced the highest total dry biomass.Plants treated with 50 kg ha 1 N all at once produced the lowest dry total biomass weight (Table 1).The total dry biomass from plants treated with 150 kg ha 1 N applied three times was about 64.09% higher than the lowest total dry biomass weight produced by onion plants treated with 50 kg ha 1 N applied all at once (Table 1).An increase in total dry biomass in response to an increasing rate of N may be associat ed with sufficient supply and efficient use, which enhance vegetative growth and contribute to an improved rate of photosynthesis and assimilate pro duction (Nasreen et al., 2007;Sikder et al., 2010;Daniel et al., 2021).
Total soluble solids increased with increasing N across the frequency of application.The highest total soluble solids were for plants grown at 250 kg ha 1 N with 3 applications.The lowest total soluble solids were for plants grown at 50 kg ha 1 N applied all at once (Table 1).
Total soluble solids from treatment with 250 kg ha 1 N applied three times exceeded total soluble solids obtained at 50 kg ha 1 N applied all at once by about 29.4% (Table 1).The possible reason for increasing total soluble solids with a higher applica tion rate of N along with increasing N application fre quency might be increased chlorophyll content and dry weight per plant (Naik and Hosamani, 2003;Mengel et al., 2006;Moursy et al., 2007;Morsy et al., 2012).
The minimum acceptable MRR is 100% (CIMMYT, 1988).The labor cost for applying N fertilizer was increased depending on the labor required for each N fertilizer application frequency.The cost of labor for N fertilizer application was 100, 200, 300, and 400 birr ha 1 for 1, 2, 3, or 4 applications of N fertilizer, respectively.The field price of onions during harvest ing was 15 birr per kg.All total variable costs were subtracted from the gross benefit to obtain the net benefit (Table 2).Partial budget analysis indicated the highest MRR of 7.295% was from an application of 50 kg ha 1 N rate in 2 applications.For every 1 birr invested in 100 kg ha 1 N applied three times, growers can expect to recover the 1 birr and obtain an additional 7.295 birr.The higher net benefit with an acceptable MRR of 240.50% was from 150 kg ha 1 N applied three times (Table 2).The most attractive combinations for farm ers were in response to an application of 150 kg ha 1 N, which provided three times the highest mar ketable yield and net benefit.Onion producers may maximize their net benefit by using 150 kg ha 1 of N applied three times.

Conclusions
The interaction of N rate and application frequen cy resulted in the highest total bulb yield (48.84 g plant 1 ), marketable bulb yield (46.95 t ha 1 ), total dry biomass weight (13.5 g plant 1 ), and bulb dry biomass yield (9.7 g plant 1 ), when the combination of N rate at 150 kg ha 1 and three times application frequencies of N was realized.The highest value of total soluble solids (13.13°Brix) was recorded at the treatment combination of 250 N kg ha 1 with three times the application frequency.The partial budget analysis revealed that the highest net benefit of Birr 628317 with an acceptable MRR of 240.5% was obtained from the application of N at 150 kg ha 1 and three fre quencies of N application.Generally, the treatment combination of an N rate of 150 kg ha 1 with three times the N application frequency can be recom mended to achieve a high bulb yield of onions with the highest net benefit.charge of the Alage Teachers Vocational Education Technical College, which partly supported this research by providing field space.

Table 1
Interaction effects of N rate and application frequency on bulb size, marketable bulb yield, total bulb yield, bulb dry weight, total dry biomass, and total soluble solids of onions Fig.1The effect of nitrogen rate (A) and application frequency (B) on plant height, leaf length, leaf diameter and leaf number of onion.