Evaluation of Water Productivity under Furrow Irrigation for Onion (Allium cepa L.) Crop

Irrigation water management practices are the main strategies to improve water productivity. is research work was focused to study the performance of alternate and paired row furrow irrigation systems at three levels of irrigation (100%, 75%, and 50% of crop evapotranspiration) using dierent water productivity indicators for onion crops. e experiment had six treatments and replicated three times to evaluate the analysis of variance in SAS software. Water productivity indicators like crop water use eciency, eld water use eciency, and eld water expense eciency were determined through bulb yield and water which were used by the crop. e crop yield was expressed as the total yield of onion bulbs, and crop water use was expressed as crop evapotranspiration (ETc), gross depth of irrigation, and water expense. e estimated maximum values of crop water use efciency, eld water use eciency, and eld water expense eciency were 11.941, 16.152, and 9.361 kg m−3, respectively, for paired row furrow irrigation with 50% ETc. e performance of the paired row furrow irrigation system in crop yield and water use was better as compared to the alternate furrow irrigation system at all levels of irrigation.


Introduction
e water resource was limited by a lot of demand factors [1]. In line with this, agriculture is one of the consumer of this resource for agricultural crop production in the way of irrigation [2]. Irrigation is a source of water for agricultural production improvement to ful ll the growing food demands in the world [3]. e availability of water for irrigation is becoming limited from day to day because of the increasing consumption of water for di erent sectors such as home and industry. Agriculture is the largest water consumer, but overall irrigation e ciency in the case of surface irrigation at the farmers' elds is very low or insu cient [4,5]. is water-scarce is a major problem in many areas of the world; in this case, studying the alternative mechanisms to solve the problem is very important [6].
Furrow irrigation is the common surface irrigation method for water application to cropped elds [7]; however, furrow irrigation as practiced by farmers in Ethiopia results in large deep percolation losses and uneven water application [8]. ese not only result in large losses of limited water but also create problems of waterlogging and salinity [9]. erefore, the development of e cient furrow irrigation systems and irrigation water management practices are essential for higher water productivity.
Techniques such as partial irrigation and de cit irrigation can increase or enhance water productivity. When water productivity decreases but irrigation water increases [10], there is an increasing interest to study the crop water productivity of furrow irrigation systems. e study of the water use e ciency of furrow irrigation systems for onion crops is important using de cit irrigation [11].
ere are di erent possibilities of irrigation water applications in furrow irrigation systems. Conventional furrow irrigation (CFI) was the traditional method of furrow irrigation and was widely used by farmers in Ethiopia and any developing country [12][13][14]. e best water management techniques were alternate furrow irrigation (AFI), xed furrow irrigation (FFI), and also paired row furrow irrigation (PRFI). e crop planting in the case of PRFI is done at the top of the ridge in paired crop rows [15], and each crop row gets water from side furrows, but the furrow in between the crop rows is not constructed. e spacing between the furrows depends on crop type. e PRFI is similar to fixed furrow irrigation (FFI) in principle with the alternate irrigated furrow in which each irrigation is fixed. e difference is only the furrow spacing and construction of furrow in the case of FFI, but in the case of PRFI, the unirrigated furrow of FFI is not constructed. us, the cost of construction of the furrows is reduced; however, many studies have investigated and concluded that AFI is better as compared to CFI and FFI for water productivity [8,11,14]. Another option for AFI is there, which is PRFI, but AFI was never compared with PRFI under different irrigation water levels for water productivity throughout the world. erefore, this study focused on the performance evaluation of AFI and PRFI at different levels of irrigation for water productivity.

e Description of Study Area.
e field experiment was conducted at Arba Minch University demonstration farmland in the Gamo Zone, SNNPR National Regional State of Ethiopia.
e study area is geographically located at an altitude of 1203 m.a.s.l, latitude of 6°04′ N, and longitude of 37°33′ E. e location of the study area is shown in Figure 1.
Based on the data collected from the Arba Minch University Meteorological station, the mean monthly minimum and maximum air temperature in the study area vary from 17.4°C to 30.6°C, respectively. e average annual rainfall in the study area is 750 mm [16], although rainfall is erratic and uneven in distribution. e historical rainfall data show a bimodal behavior with an interval of February up to April and June up to September. e average relative humidity ranges from 39.4% (February) to 62.9% (May), and average annual daily sunshine duration varies from 6.3 hours to 9.1 hours. Based on climate properties of the study area, the agroecological zone of the study area was classified as dry low land [17].

Preexperimental Activities.
To conduct this experimental research, primary and secondary data were collected. Secondary data such as climatic and agronomic data of onion were collected from the Arba Minch University Meteorological station and FAO [18], respectively. e climatic data were maximum and minimum temperature, relative humidity, sunshine hours, wind speed, and rainfall. Other primary data such as soil physical and chemical characteristics were collected.

Experimental Treatments.
e experiment was conducted for alternate furrow irrigation (AFI) and paired row furrow irrigation (PRFI) systems with three levels of irrigation such as 100%, 75%, and 50%. e transplanting after 45 days of seedlings of the Red Bombay variety of onion was done in the ridge and furrow system following recommended agronomic practices. e treatment plot size for the two systems was 1.6 m × 3 m and the central rows for each treatment were considered experimental rows for the collection of field data. e side rows were nonexperimental (a buffer row) to minimize the border effects, and the plant-to-plant spacing in each row was 10 cm which has a plant density of 30 plants per row. ere were 6 treatments; each of the treatments was replicated three times; details of the treatments are given in Table 1. e location of different plots was decided by a randomized complete block design (RCBD). e spacing between each plot to plot and block to block was 1 m and 1.5 m, respectively. e total area required for this experiment was 240.7 m 2 (16.6 m × 14.5 m) ( Figure 2).

Crop Water Requirement Estimation.
e onion crop water requirement was estimated from reference crop evapotranspiration (ETo) and crop coefficient (Kc) using equation (1). e ETo was estimated using CROPWAT 8 software based on the Penman-Monteith method for 30year monthly average climatic data. (1) After the determination of crop evapotranspiration using the above relation, the net irrigation requirement (I) was estimated using where NIR represents the net irrigation requirement (mm), Pe represents the effective rainfall (mm), GW represents the groundwater contribution (mm), and SM represents the change in soil moisture (mm); the depth of the groundwater table during the crop season was more than 1.6 m; therefore, the groundwater contribution (GW) was negligible. e gross depth of irrigation water application was expressed as where NIR represents the net depth of irrigation estimated using equation (2) and Ea represents the overall irrigation efficiency measured in the field and found to be equal to 64%.

Crop Water Productivity.
e considered water productivity indicators were crop water use efficiency (CWUE), field water use efficiency (FWUE), and field water expense efficiency (FWEE) as expressed by the following equations [15,19].
where CWUE represents the crop water use efficiency (kg m −3 ), Y represents the crop yield (kg ha −1 ), and ETc represents the crop evapotranspiration in mm.
where FWUE represents the field water use efficiency (kg m −3 ), Y represents the crop yield (kg ha −1 ), and GIR represents the gross depth of irrigation water application (mm).
where X p represents the water expense (mm), estimated using the following equation where ω 1 represents the gravimetric soil moisture at the beginning of the crop growing season (transplanting) (fraction), ω 2 represents the gravimetric soil moisture at the end of crop season (harvesting) (fraction), Zr represents the crop root zone depth (mm), As represents the apparent specific gravity of soil in crop root zone depth, and, i represents the soil layer.

Data Analysis.
e results of onion yield and water productivity were analyzed and subjected to analysis of variance using SAS 9.0 program. e least significant difference (LSD) was used to compare the mean of each treatment. e general framework for this study is shown in Figure 3.

Results and Discussion
e soil texture of the soil in the experimental area was determined by using hydrometer analysis, and the field capacity of the soil was also measured by pounding water at the soil surface to saturate the soil column up to about 150 cm soil depth, covering the soil surface with a trace to prevent water evaporation from the soil surface and start measuring soil moisture content after 24 hours. e permanent wilting point of the soil was considered as the soil moisture content at 15 atmospheric tensions. e soil bulk density was calculated by taking undisturbed soil samples in the experimental area. Values of soil physical properties which were measured or determined in the laboratory are given in Table 2.
e important information during each irrigation on intervals of this research work is given in Table 3. e estimated values of seasonal crop evapotranspiration, net and the gross depth of irrigation, water, the expense, and collected value of the total bulb yield for each of the treatments are given in Table 4.

Statistical Analysis for Total Bulb Yield (TBY).
e ANOVA showed that the effect of the furrow irrigation system significantly affected TBY at P < 0.05. e maximum TBY (31.204 ton ha −1 ) was obtained for PRFI and significantly different from the lower TBY (29.445 ton ha −1 ) obtained in AFI, and in the same way, irrigation levels as the main effect had a significant effect on TBY. Water deficit is one of the essential factors for any crop production [20]. e maximum TBY (37.070 ton ha −1 ) was obtained at 100% ETc and significantly different for 75% ETc, and the lowest TBY (23.583 ton ha −1 ) was recorded at 50% ETc. e interaction of the furrow irrigation system and irrigation level significantly affected TBY at P < 0.05. As given in Table 5, the maximum TBY (37.863 ton ha −1 ) obtained for PRFI with 100% ETc (T4) was significantly different from all treatments, and a highly significant difference from the lowest value (23.078 ton ha −1 ) was obtained for AFI with 50% ETc (T3). e TBY was reduced by 2.64%, 16.38%, 23.46%, 36.38%, and 39.05% for treatments of T1, T5, T2, T6, and T3 when compared to the TBY obtained from T4, respectively. Here, 39.05% of the yield was reduced when 50% of water was saved. e reason for to maximum total bulb yield in T4 is due to a better-wetted root zone rather than deep percolation and soil evaporation loss.
Generally, this result revealed that TBY decreased in both furrow irrigation systems with decreasing irrigation levels. is argues against the results obtained by [21] which reported that the total bulb yield was reduced from 100% ETc to 50% ETc by 3.48 ton ha −1 . Similarly, [22] reported that the total bulb yield was reduced from 100% ETc to 50% ETc by 5.66 ton ha −1 .

Water Productivity Indicators.
e estimated values of different water productivity indicators using (4), (5), and (6) along with the results of statistical analysis are given in Table 6.

Statistical Analysis for Crop Water Use Efficiency (CWUE).
Analysis of variance showed that the furrow irrigation system and irrigation levels significantly affected CWUE at P < 0.05. e furrow irrigation system as the main effect, the maximum CWUE (10.596 kg m −3 ) obtained for PRFI, was significantly different from the lowest value (10.003 kg m −3 ) obtained for AFI. Similarly, with irrigation level as the main effect, the maximum CWUE (11.691 kg m −3 ) was obtained at 50% ETc, which was significantly different compared to CWUE recorded at 75% ETc (10.019 kg m −3 ) and the lowest value (9.188 kg m −3 ) obtained at 100% ETc. e interaction effect of the furrow irrigation system and irrigation water level on CWUE was significant at P < 0.05. It is observed from Table 6 that the maximum value of CWUE (11.941 kg m −3 ) was recorded for T6 (PRFI with 50 ETc). e maximum value of CWUE was significantly different compared to all other treatments. Contrary, the minimum CWUE (8.992 kg m −3 ) was recorded for treatment T1 (AFI with 100% ETc), which was significantly different from all other treatments. On the other hand, there was no significant difference between T2 (AFI with 75% ETc) and T4 (PRFI with 100% ETc).
e CWUE increased as the denominator (water) decreased or CWUE increased as the numerator (yield) increased [23,24]. is research work agrees with [25] which reported that CWUE decreased as the irrigation level increased. e maximum CWUE for treatment T6 was due to properly managed irrigation water. Contrary, the minimum CWUE for treatment T1 was due to the application of more Advances in Agriculture irrigation water that may result in deep percolation or soil water evaporation. erefore, the PRFI system with a 50% ETc level of irrigation can achieve the maximum CWUE. It results in 50% water-saving compared to full irrigation treatment (100% ETc). e irrigation water so saved may be used to bring more area under production. erefore, areas having water-scarce should use a paired row furrow irrigation system to maximize crop water productivity rather than maximizing crop yield per unit area.

Statistical Analysis of Field Water Use Efficiency (FWUE).
Analysis of variance showed that the furrow irrigation system and irrigation levels significantly affected FWUE at P < 0.05. e furrow irrigation system has the main effect; the maximum FWUE (12.121 kg m −3 ) obtained for PRFI was significantly different from the lowest value (11.452 kg m −3 ) obtained for AFI. Similarly, with irrigation level as the main effect, the maximum FWUE (15.813 kg m −3 ) obtained at 50% ETc was significantly different compared to FWUE obtained at 75% ETc (10.811 kg m −3 ) and the lowest value (8.736 kg m −3 ) obtained at 100% ETc. e interaction effect of the furrow irrigation systems and irrigation water levels on FWUE was significant at P < 0.05. It was observed from Table 5  On the other hand, there was no significant difference between T1 (8.549 kg m −3 ) and T4 (8.922 kg m −3 ) for both treatments with a 100% level of irrigation. However, FWUE was slightly higher for treatment T4 (PRFI) compared to T1 (AFI). It indicates that PRFI had higher crop productivity performance compared to the AFI system (Table 5).
e results revealed that FWUE increased as irrigation water decreased (Table 6). e reason to obtain the maximum FWUE for treatment (T6) was the lesser gross irrigation depth compared to other treatments except for treatment T3. e FWUE for treatment T6 (PRFI with 50% ETc) was higher compared to treatment T3 (AFI with 50% ETc). erefore, in areas of limited water resources, PRFI was preferable to AFI with 50% ETc to maximize FWUE. e saved water may be used to bring more area under irrigation    Advances in Agriculture 5 or for other commercial uses like domestic water supply for drinking or industrial use. Corresponding to this, [22] concluded that water productivity was maximum at a medium level of a deficit than higher soil moisture availability (full irrigation water availability); similarly, [26] reported that water use efficiency decreased with increased water supply.

Statistical Analysis for Field Water Expense Efficiency (FWEE).
e interaction effect of the furrow irrigation system and irrigation water levels on FWEE was significant at P < 0.05. As it is observed in Table 6, FWEE was slightly higher (9.361 kg m −3 ) for treatment T6 (PRFI with 50 ETc) compared to all other treatments. is maximum FWEE was significantly different compared to all other treatments. e minimum FWEE (7.619 kg m −3 ) was recorded for treatment T1 (AFI with 100% ETc) which was not a significant difference between T2 (7.68 kg m −3 ) and T4 (8.014 kg m −3 ). Furthermore, there was no significant difference between T4 and T5.
As shown in the result field, water expense efficiency was decreased with an increase in water expense. e differences in FWEE between treatments were very small due to soil moisture during transplanting being the same for all treatments, but during harvesting, different soil moisture for each treatment was collected. Practically, the amount of applied irrigation water for 100% ETc was different from 75% ETc and 50% ETc. Similarly, the soil moisture was also different for different treatments. at means the maximum soil moisture was observed for 100% ETc treatment less than 75% ETc and 50% ETc at the time of harvesting. is maximum soil moisture for 100% ETc was subtracted from soil moisture during transplanting (uniform value) to get a smaller value than 75% ETc and 50% ETc. erefore, the depth of water expense becomes counterbalancing with the decreased gross depth of irrigation. Generally, FWEE was increased when irrigation water decreased [19].
Generally, water productivity was affected by different factors. In line with this, [25] reported that water productivity was affected by the furrow irrigation system, irrigation water level, crop yield potential, and climatic conditions of the crop grown region. In the same way, [27] reported that water productivity was increased by improving yield or reducing irrigation water. Anyways, in the present research,

Conclusion
Water management in the furrow irrigation system was essential for the improvement of water productivity. Evaluation of water productivity under different furrow irrigation systems is important to identify the suitable furrow irrigation system with optimum levels of irrigation water.
In the present research, water productivity under alternate furrow irrigation and paired row furrow irrigation systems at three levels of irrigation water were studied. e crop water requirement was estimated using CROPWAT 8.0 software. e experiment was designed as a two-factorial and becomes six treatment and replicated three times. Statistically, the mean of collected data was separated using LSD at a 5% probability level in SAS software, and the effect of treatment was tested using bulb yield and water productivity indicators. e water productivity evaluation indicators were crop water use efficiency, field water use efficiency, and field water expense efficiency. e estimated values of crop water use efficiency, field water use efficiency, and field water expense efficiency varied 8.992-11.439 kg m −3 , 8.549-15.474 kg m −3 , and 7.619-8.914 kg m −3 for the alternate furrow irrigation system and 9.385-11.941 kg m −3 , 8.922-16.152 kg m −3 , and 8.014-9.361 kg m −3 for the paired row furrow irrigation system as irrigation level varied from 100% ETc to 50% ETc, respectively. erefore, the new outcome in this research was the paired row furrow irrigation system was identified with better performance than an alternate furrow irrigation system in water-saving, crop yield, and water productivity for onion crops. erefore, the use of the paired row furrow irrigation system is recommended for practicing irrigation. Further studies will be needed for spatial and temporal evaluation of water productivity under two methods with different crops.

Data Availability
All the available data can be accessed through request.

Conflicts of Interest
e authors declare that there are no conflicts of interest.