See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/325672594

Optimizing irrigation water levels to improve yield and water use efficiency of
vegetables: case study of tomato

Article  in  Sustainable Water Resources Management · June 2018

DOI: 10.1007/s40899-018-0250-8

CITATIONS READS

4 261

3 authors:

Kidane Welde Hintsa Libsekal Gebremariam

Tigray Agricultural Research Institute Tigray Agricultural Research Institute
8 PUBLICATIONS   102 CITATIONS    8 PUBLICATIONS   15 CITATIONS   

SEE PROFILE SEE PROFILE

Kiflom Degef Kahsay

Tigray Agricultural Research Institute
4 PUBLICATIONS   29 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Identification and prioritization of subwatersheds for land and water management in Tekeze dam watershed, Northern Ethiopia View project

Improving Irrigation Water Management View project

All content following this page was uploaded by Kidane Welde on 27 November 2018.

The user has requested enhancement of the downloaded file.

Sustainable Water Resources Management
https://doi.org/10.1007/s40899-018-0250-8

ORIGINAL ARTICLE

Optimizing irrigation water levels to improve yield and water use

efficiency of vegetables: case study of tomato
Kidane Welde1 · Hintsa Libsekal Gebremariam1 · Kiflom Degif Kahsay1

Received: 27 June 2017 / Accepted: 30 May 2018

© Springer International Publishing AG, part of Springer Nature 2018

Abstract
Optimizing irrigation water requirement is necessary to improve productivity of irrigated agriculture. Hence, adapting the
knowledge of irrigation scheduling for specific crop and location is important. Field experiment was then conducted in
Raya Alamata district of Tigray, Ethiopia to investigate the impact of different irrigation water levels on yield and irriga-
tion water use efficiency (IWUE) of tomato. Eight treatments (50, 35, 25% above estimated crop water requirement (CWR),
estimated CWR, 25, 35, 50% below the estimated CWR and farmers practice) were arranged in randomized complete block
design under three replications. Tomato water requirement was estimated using CROPWAT 8 software and it was estimated
to be 500 mm throughout the base period. The result indicates significant difference between marketable yield (MY), total
yield (TY) and IWUE of tomato by the level of irrigation. The application of estimated CWR of tomato gave higher MY
(36.37 ton/ha) and TY (38.58 ton/ha). Unmarketable fruit yield of tomato was unaffected by level of irrigation. Farmers’
practice gave considerable MY (36.32 ton/ha). But, the depth of water applied by farmers throughout the growing season
was 561 mm which was 12.2% above the estimated CWR. The yield obtained per unit of applied water (IWUE) ranges from
0.357 to 0.876 kg/m3 for the different irrigation water levels. This shows that IWUE increases with decreasing depth of water
application. In conclusion, deficit irrigation practice could be successful in saving irrigation water up to 35% of tomato CWR
without significant reduction in fruit yield of tomato in regions where water is a limiting factor for vegetable production.

Keywords  CWR​ · IWUE · Marketable yield · Unmarketable yield · Small-scale irrigation

Introduction Aboudrare 2004). Agricultural practices such as the use of

sprinkler and drip irrigation are in expansion in southern
Efficient utilization of irrigation water is becoming increas- zone of Tigray. But with this expansion, there remains a
ingly essential in developing countries to attain food security need for advancement in irrigation scheduling and CWR of
(Chun et al. 2010). The increasing worldwide shortage of vegetables.
water and costs of irrigation especially in arid and semiarid If the applied water is insufficient during the crop grow-
regions contributes to give emphasis to methods of irriga- ing season, the crop will not produce fully. This results in
tion that maximize water use efficiency (Youssef et al. 2007; low quality yield or loss of total yield (Yazar and Sezen
Cai and Rosegrant 2003). Hence, sustainable methods to 2006). Applying more water than required by crop evapo-
improve irrigation water productivity are getting emphasis transpiration consumptive use (ETc) can result in extra
in arid and semiarid regions like Ethiopia (Debaeke and pumping cost, wasted water and increased risk for leaching
of valuable chemicals below the rooting zone and possibly
in to the ground water (Cabello et al. 2009). Under these
Electronic supplementary material  The online version of this conditions, the root zone needs leaching to reduce salinity
article (https​://doi.org/10.1007/s4089​9-018-0250-8) contains and water logging (Sun et al. 2006). Effective irrigation is
supplementary material, which is available to authorized users.
possible only with regular monitoring of soil water and crop
* Kidane Welde development condition in the field and with the forecast-
kidanew2009@gmail.com ing of future crop water needs (Raes et al. 2009; Steduto
et al. 2009). Therefore, the irrigator needs to know effective
1
Tigray Agricultural Research Institute, Alamata Agricultural
Research Center, P.O.Box 56, Alamata, Ethiopia

13
Vol.:(0123456789)
 Sustainable Water Resources Management

irrigation management through different irrigation technolo- 14 °C, respectively. It was cross-checked (validated) with
gies (Karam et al. 2009). the New_LocClim 1.10 capabilities of weather generator.
Proper irrigation management requires estimation of Rainfall in Raya-Alamata is bimodal; short rainfall which
CWR as well as net and gross irrigation water needs (Chapa- occurs in March and May and a main rain that runs from end
gain and Orr 2008). Even though different crops and veg- June to beginning of September. The most intense rainfall
etables require different amounts of water, in southern zone occurs between July and August. Tumuga irrigation scheme
of Tigray regional state, CWR is not commensurate with is one of the low land parts of the district which is a bet-
the climate, soil and crop types. Small-scale farmers simply ter representative for the lowland part of southern zone of
irrigate their land based on the quota decided by the kush- Tigray Regional State. Dominantly, the main soil classes in
et’s1 water allocation committee. This leads, the study area the district are: clay, loam, sandy and saline (Girmay et al.
(Timuga irrigation scheme) starts to affect by water logging 2014). Onion, tomato and pepper are the most widely grown
and salinity problems. vegetable crops and are major cash crops for farmers during
The vegetable production at the study area is fully the irrigation season in the area.
dependent to irrigation water application (no significant The soil class of the experimental site was loam. The
rain at the growing period) for its water requirement under entire study area is characterized as alkaline soil in which
the limited source of irrigation water. As these vegetables traditional irrigation practices has been implemented for a
especially tomato are a source of cash for the farmers, there period of time using surface irrigation water sources. The
is an immense need for the improvement of the crop water total available soil moisture is 290 mm/m with a maximum
production function (CWP). The CWP expresses the relation infiltration rate of 40 mm/day. The pH of the soil was 8.69
between obtained marketable yield and the total amount of with electrical conductivity (ECe) of 0.64 mS/cm. The cat
water evapotranspired (Sam and Dirk 2008). Hence, besides ion exchange capacity of soil also accounts for 34.88 cmol/
the introduction of pressurized irrigation systems and its kg (Girmay et al. 2014).
expansion to the small-scale farmers of the study area, it
is necessary to adopt the practice of estimating CWR and Experimental design and descriptions
irrigation scheduling (Haregeweyn et al. 2011). Because,
previous studies have shown that the appropriate irrigation Thirty days later after sowing on 2 January 2013, tomato
scheduling is very important for water saving in irrigation seedlings were transplanted to an experimental field. Ferti-
agriculture and profitability improvement (Mofoke et al. lizers, DAP and urea were applied uniformly to each treat-
2006). Therefore, it is imperative to evaluate crop water ment (150 and 100 kg/ha, respectively). DAP was applied at
requirement, water use efficiency, and yield at different basal but urea was applied both during planting and 3 weeks
levels of irrigation water depth. The main objective of this after planting (1/3rd at basal and 2/3rd 3 weeks after plant-
experiment was to improve the production and productivity ing). The same amount of irrigation water was given to all
of tomato through optimizing the irrigation water manage- treatment plants till 16 January for the plants’ establishment
ment systems by mainly focusing on crop water requirement (seedlings grow uniformly), after then, application of defer-
and irrigation levels. ent depth of irrigation water began for the treatments based
on the preset treatment setting. Weed and insect control was
managed uniformly according to standard practices. Furrow
Materials and methods irrigation was the method of water application. The water
applied per each irrigation event was measured using 2 in.
Study area description Parshall flumes.
The experiment was conducted using field trials in a ran-
The experiment was carried out in Tumuga irrigation domized complete block design (RCBD) of the irrigation
scheme, Raya-Alamata district, Tigray regional state of treatments. The field was divided into three blocks (three
Ethiopia. Geographically, the experimental site is located replicates per treatment) with experimental plot area of
at 12.3°N latitude, 39.5°E longitude and altitude of 1436 m 4 × 4 m2. Each replication contained eight irrigation treat-
above sea level (Fig. 1). Long-term meteorological data ments randomly distributed in each block. The distance
(1997–2014) obtained from Raya-Alamata office of agri- between blocks and plots were 2 and 1  m, respectively.
culture and rural development indicates that the mean The row and plant spacing of tomato were 70 and 40 cm,
annual rainfall for the area ranges from 615 to 927 mm respectively. The experimental treatments were arranged
with mean maximum and minimum temperatures of 23 and under four conditions; one was different levels of deficit
imposed throughout the whole season, the second was a dif-
ferent level of over irrigation imposed throughout the whole
1
  The lowest administrative zone of Ethiopia.

13
Sustainable Water Resources Management

Fig. 1  Location of experimental site

Table 1  Treatment setting (level of irrigation water depth) equation (Luca et  al. 2003; FAO 2002) for the growing
No. Treatment naming Treat-
period from January to May 2013. Metrological parameters
ment like rainfall, temperature and sunshine hours were collected
code using New-LocClim software for a specific location. Crop
parameters (Kc values, rooting depth, yield response factor
1 50% above normal/estimated/CWR​ T1
to water stress and other related data) was taken from FAO
2 35% above normal/estimated/CWR​ T2
publications as used by Andreas and Karen (2002). The tex-
3 25% above normal/estimated/CWR​ T3
tural class of the soil was obtained from soil survey made by
4 Normal/estimated/CWR​ T4
Girmay et al. (2014).
5 25% below normal/estimated/CWR​ T5
6 35% below normal/estimated/CWR​ T6
Measurements and statistical analysis
7 50% below normal/estimated/CWR​ T7
8 Control/farmers’ practices T8
The amount of irrigation water applied to each experimental
plot was measured using 2 in. Parshall flumes. All treat-
ments had the same irrigation interval with different irriga-
season, third was normal irrigation water requirement and tion water levels. Rain fall occurrence during the growing
fourth was local farmer practice (Table 1). period was recorded. The fresh fruit yield was collected
from each experimental plot. Both marketable and unmar-
Crop water requirement and irrigation scheduling ketable tomato yield was measured for each plot. Irrigation
water use efficiency (IWUE) was calculated as the ratio of
The crop water requirement of tomato was determined obtained marketable yield and the total water consumption
by CROWAT 8 software using FAO Penman–Monteith (Stanhill 1986) for the whole season as:

13
 Sustainable Water Resources Management

Table 2  Depth of irrigation Treatment Description Amount of Total amount of water Irrigation

water applied for each treatment water (mm) applied ­(m3/ha) frequency

T1 50% above normal/estimated/CWR​ 750 7500 13

T2 35% above normal/estimated/CWR​ 675 6750 13
T3 25% above normal/estimated/CWR​ 625 6250 13
T4 Normal/estimated/CWR​ 500 5000 13
T5 25% below normal/estimated/CWR​ 375 3750 13
T6 35% below normal/estimated/CWR​ 326 3250 13
T7 50% below normal/estimated/CWR​ 250 2500 13
T8 Control/farmers’ practices 561 5610 13

GY Marketable and unmarketable fruit yield

IWUE = ,
Wap
Marketable fruit yield was significantly affected (p = 0.05)
where IWUE is the irrigation water use efficiency in fraction by the levels of irrigation water application (Table 3). A
(kg/m3), GY is the marketable fruit yield (kg/ha) and W­ ap is similar result was observed by Lincoln et al. (2008) dur-
the amount of water applied ­(m3/ha). ing their study on nitrogen rate and irrigation scheduling
All data were presented as means and statistically ana- on tomato fruit yield and yield components. There was a
lyzed using R 3.3.1 software (R Core Team 2016). The significant reduction in marketable yield of over 7.1 and
mean yields and IWUEs results of treatments were com- 6 ton/ha by adopting 50% below the normal/estimated crop
pared to each other for any significant differences using water requirement (250 mm) and 50% above the normal/
the least significant difference (LSD) method. LSD was estimated crop water requirement (750 mm), respectively, as
calculated from data, where the differences among means compared to yield obtained (36.37 ton/ha) by the estimated
were tested. Separate analyses were performed for each crop water requirement (500 mm). Under the conditions of
data at the probability level for determination of signifi- the experimental field, maximum fresh fruit yield (36.37 ton/
cance p = 0.05. ha) was obtained from the normal/estimated tomato crop
water requirement. But, minimum yield (29.27  ton/ha)
was scored at depth of water 50% below normal/estimated
CWR. Although the estimated crop water requirement versus
Result and discussion farmers’ practice (used as control) was not significant for
marketable yield, the farmers’ practice applied more water
Estimation of crop water requirement and irrigation than estimated crop water requirement. The depth of water
scheduling of tomato applied by farmers throughout the growing season was about

The estimated crop water requirement of tomato was

500 mm and a nearly similar result was estimated by Rich- Table 3  Mean marketable and unmarketable fruit yield for different
ard and Luis (2009) during their study estimating crop depths of water applications
coefficients from fraction of ground cover and height.
Treatment description Marketable Unmarketable
There was no rainfall event during the study period. yield (ton/ha) yield (ton/ha)
Hence, the specified water need of tomato was applied
through irrigation. The gross irrigation water was about Normal/estimated/CWR​ 36.37a 2.13
714 mm by taking the efficiency of furrow irrigation as Control/farmers’ practices 36.32ab 2.13
70%. The yield and irrigation water use efficiency analy- 25% above normal/estimated/CWR​ 34.87ab 2.27
sis starts from calculation of irrigation scheduling using 25% below normal/estimated/CWR​ 34.77ab 1.83
the estimated crop water requirement of tomato (500 mm). 35% below normal/estimated/CWR​ 32.57ab 2.47
Hence, the normal/net irrigation scheduling was set at 35% above normal/estimated/CWR​ 31.43ab 2.53
5 days irrigation interval. There were a total of 13 irriga- 50% above normal/estimated/CWR​ 30.37ab 2.50
tion events throughout the growing period. In addition to 50% below normal/estimated/CWR​ 29.27b 2.57
the estimated CWR six levels of water application was LSD 5.99 ns
developed as shown in Table 2 so as to determine the opti- CV (%) 10.28 22.57
mum irrigation depth at field level. Levels not connected with similar letter are significantly different

13
Sustainable Water Resources Management

561 mm which was 12.2% above the estimated water require- Table 4  Mean total fruit and IWUE for different depths of water
ment of tomato (500 mm). Based on the result, the applica- applications
tion of deficit irrigation up to 35% below tomato CWR can Depth of water applied Total yield (ton/ha) IWUE (kg/m3)
save a significant amount of irrigation water without signifi-
Normal/estimated/CWR​ 38.53a 7.28c
cantly affecting marketable yield reduction. Similarly, Sam
Control/farmers’ practices 38.50ab 6.47cd
and Dirk (2009) in their review on deficit irrigation as an
25% above normal/estimated/ 37.10ab 5.58de
on-farm strategy to maximize crop water productivity in dry
CWR​
areas also recommends up to 30% deficit for different crops.
25% below normal/estimated/ 36.60ab 9.27b
The analysis of variance result revealed that unmarketable CWR​
fruit yield per hectare did not differ significantly (p = 0.05) 35% below normal/estimated/ 35.03ab 10.02b
in response to the levels of irrigation water depth (Table 3). CWR​
Numerically, 25% below normal/estimated/CWR was better 35% above normal/estimated/ 33.87ab 4.65ef
in reducing the unmarketable yield (1.83 ton/ha) as com- CWR​
pared to unmarketable yield 2.5 and 2.57 ton/ha obtained 50% above normal/estimated/ 32.90ab 4.05f
CWR​
from 50% above and 50% below estimated water require-
50% below normal/estimated/ 31.87b 11.71a
ment application respectively.
CWR​
LSD 6.23 1.50
CV (%) 10.00 11.57
Total yield and irrigation water use efficiency
Levels not connected with similar letter are significantly different
Total yield per hectare was significantly affected by differ-
ent depth of water application (p = 0.05). The result indi-
cates that statistically highest total yield of fruits per hectare
(38.53 ton/ha) was obtained from normal/estimated/CWR.
Similar result also shows Vazquez et al. (2006) during their study result shows that, IWUE was reduced by 50% when
evaluation of drainage and nitrate leaching under process- irrigation water was applied at levels of 35% and 50% of
ing tomato. The minimum total yield (below 32.9 ton/ha) estimated tomato crop water requirement.
was obtained from the application of irrigation water at
50% below and 50% above the estimated evapotranspira-
tion of tomato. This reduction in fruit yield might be due
to unfavorable moisture conditions during tomato growth Conclusions
(extremely water stressed and extremely over irrigated). This
result supports the study by Mitchell and Shennan (1991) This study was conducted to improve yield and IWUE of
over tomato fruit yields and quality under water deficit tomato through optimizing irrigation water levels. Level
irrigation. of depth of irrigation water application in growing tomato
The use of different depths of water application arrange- significantly affects its marketable yield, total fruit yield
ments significantly affected (p = 0.05) the IWUE of tomato and IWUE, with surface furrow irrigation controlled by
(Table 4). This confirms the results obtained by Lincoln 2 in. Parshall flumes. Tomato yield was also increased by
et al. (2008) who found that IWUE of tomato was signifi- 15 and 17% for estimated tomato CWR compared to the
cantly affected by different levels of irrigation intervals 50% above and 50% below estimated CWR application
and depth of irrigation water application. The treatments’ respectively. It is concluded that under furrow irrigation
result for IWUE ranges from 4 kg/m3 obtained from 50% system, appropriate application of estimated tomato CWR
above estimated tomato water requirement to 12 kg/m 3 through flow measuring devices can allow tomato growers
obtained from 50% below estimated tomato water require- to obtain higher profitable yield. It can be also concluded
ment. This shows as the volume of water application that deficit irrigation practice could be successful in saving
decreases the IWUE increases proportionally even though irrigation water up to 35% of tomato CWR without sig-
significant yield reduction was obtained for the 50% above nificant reduction in fruit yield of tomato in regions where
and below estimated tomato water requirement applica- water is a limiting factor for crop production.
tion under the specific conditions. The high volume of
water which promotes excessive water percolation due to Acknowledgements  We kindly acknowledge Tigray Agricultural
Research Institute (TARI) as this study was financially supported by
the limited soil water holding capacity of the soil played
TARI. We would like to thank also the staff of Alamata agricultural
a significant role in the reduction of IWUE results. The research center for their facility and technical support during the field
work of this experiment.

13
 Sustainable Water Resources Management

Compliance with ethical standards  Karam F, Kabalan R, Breidi J, Rouphael Y, Oweis T (2009) Yield and
water production functions of two durum wheat cultivars grown
under different irrigation and nitrogen regimes. Agric Water
Conflict of interest  All authors have no any actual or potential conflict
Manag 96:603–615
of interest including any financial, personal or other relationships with
Lincoln Z, Johannes M, Michael D, Rafael M, Jason I (2008) Tomato
other people or organizations for the manuscript submitted to journal
yield, biomass accumulation, root distribution and irrigation water
of irrigation science.
use efficiency on a sandy soil, as affected by nitrogen rate and
irrigation scheduling. Agric Water Manag 96(2009):23–34
Luca E, Nagy Z, Berchez M (2003) Water requirements of the main
References field crops in Transylvania. J Cent Eur Agric 3(2):98–102
Mitchell JP, Shennan C (1991) Tomato fruit yields and quality under
water deficit and salinity. J Am Soc Hortic Sci 116(2):215–221
Abdel-Ati YY, El-Maziny MY, Abdalla S, Meleha ME, Hassan A,
Mofoke ALE, Adewumi JK, Babatunde FE, Mudiare OJ, Ramalan AA
Abdelraheem (2007) Effect of water stress and potassium fertili-
(2006) Yield of tomato grown under continuous-flow drip irriga-
zation on yield quantity and quality of potato. In: African Crop
tion in Bauchi state of Nigeria. Agric Water Manag 84:166–172
Science Conference Proceedings, vol 8, pp 445–455. Printed in
R Core Team (2016) R: a language and environment for statistical com-
El-Minia, Egypt. ISSN 1023-070X
puting. R Foundation for Statistical Computing, Vienna, Austria.
Andreas K, Karen F (2002) Crop water requirements and irrigation
http://www.R-proje​ct.org/. Accessed 15 July 2017
scheduling. Irrigation manual module 4. Water Resources Devel-
Raes D, Steduto P, Hsiao TC, Fereres E (2009) AquaCrop—the FAO
opment and Management Officers FAO. Sub-Regional Office for
crop model to predict yield response to water. II. Main algorithms
East and Southern Africa, Harare
and software description. Special issue on “Yield Response to
Cabello MJ, Castellanos MT, Romojaro F, Martinez-Madrid C, Ribas F
Water: Examination of the Role of Crop Models in Predicting
(2009) Yield and quality of melon grown under different irrigation
Water Use Efficiency”. Agron J 101:438–447
and nitrogen rates. Agric Water Manag 96:866–874
Richard G, Luis S (2009) Estimating crop coefficients from fraction of
Cai X, Rosegrant MW (2003) World water productivity: current situ-
ground cover and height. Irrig Sci 28:17–34
ation and future options. In: Kijne JW, Barker R, Molden D (eds)
Sam G, Dirk R (2009) Deficit irrigation as an on-farm strategy to maxi-
Water productivity in agriculture: limits and opportunities for
mize crop water productivity in dry areas. Agric Water Manag
improvement. International Water Management Institute (IWMI),
96:1275–1284
Colombo, pp 163–178
Stanhill G (1986) Water use efficiency. Adv Agron 39:53–85
Chapagain AK, Orr S (2008) An improved water footprint methodol-
Steduto P, Hsiao TC, Raes D, Fereres E (2009) AquaCrop the FAO
ogy linking global consumption to local water resources: a case
crop model to predict yield response to water. I Concepts. Spe-
of Spanish tomatoes. J Environ Manag 90:1219–1228
cial issue on “Yield Response to Water: Examination of the Role
Chun-Zhi Z, Zhi-Long B, Bao-Zhong Y, (2010) Determination of
of Crop Models in Predicting Water Use Efficiency”. Agron J
optimum irrigation water amount for drip-irrigated muskmelon
101:426–437
(Cucumis melo L.) in plastic greenhouse. Agric Water Manag
Sun HY, Liu CM, Zhang XY, Shen YJ, Zhang YQ (2006) Effects of
96(2009):595–602
irrigation on water balance, yield and WUE of winter wheat in the
Debaeke P, Aboudrare A (2004) Adaptation of crop management to
North China Plain. Agric Water Manag 85:211–218
water-limited environments. Eur J Agron 21:433–446
Vazquez N, Pardo A, Suso ML, Quemada M (2006) Drainage and
Food and Agriculture Organization (FAO) (2002) Crop evapotranspi-
nitrate leaching under processing tomato growth with drip irriga-
ration guidelines for computing crop water requirements: guide-
tion and plastic mulching. Agric Ecosyst Environ 112:313–323
lines for computing crop water requirements (FAO Irrigation and
Yazar A, Sezen SM (2006) Effects of full and deficit irrigation on yield
Drainage Paper No.56)
and water use efficiency of winter wheat in the arid southeast
Girmay T, Girmay G, Alem G, Abrhaley G, Yemane G, Tesfay H
Anatolia region of Turkey. In: Proceedings of the 1st international
(2014) Participatory rural appraisal report: Alaje Woreda, Tigray
symposium on land and water management for sustainable irri-
Region. Cascape Working Paper 2.6.1
gated agriculture, CD-rom
Haregeweyn N, Gebrekiros A, Tsunkeawa A, Tsubo M, Meshesha D,
Yazew E (2011) Performance assessment and adoption status of
family drip irrigation system in Tigray State, Northern Ethiopia.
In Water Conservation. InTech

13

View publication stats