This study was conducted (2010–2012) to analyse the efficiency of irrigation scheduling in maize production based on soil moisture measurements (Watermark soil moisture sensors) in years with extreme weather events at the research site of the Agricultural Institute in Osijek, eastern Croatia. Three irrigation treatments and four maize hybrids were studied. In the extremely rainy 2010, the highest yield of maize grain was obtained in rainfed plots (control = 9.24 t ha−1). A significantly (P < 0.01) lower yield (−8%) was obtained in fully irrigated plots (a3 = 8.59 t ha−1). This was opposite to the results obtained from the extremely warm 2011 and very dry 2012, when grain yield was higher as the amount of irrigation water was increased. Maize grain yield in the fully irrigated plot was 25% (2011) and 40% (2012) higher compared with the control plots (dryland). According to our results, the main factor for irrigation efficiency in extreme weather conditions is to properly determine the optimum level for soil moisture sensors and ground water level in relation to root depth.

INTRODUCTION

Agriculture plays an important role in the economy and life of eastern Croatia. Over the last decade agricultural production has been affected by numerous extreme weather events, droughts and excessive rainfall. As a result, yields of summer crops have been low and unstable. Some previous research results have shown that droughts in eastern Croatia have become more frequent and more intense. For instance, Šoštarić et al. (2012) have analyzed weather conditions for Osijek area (eastern Croatia) by hydrothermal coefficient by Seljaninov (HTC) and concluded that in the 1973–2011 period 14 years were dry, yet in the 1994–2011 period droughts occurred every 6 years and in the same period an extremely wet year occurs every 8 years. Furthermore, during the 2000–2011 period 5 years (2000, 2003, 2007, 2009 and 2011) were extremely dry, highlighting the need for irrigation to stabilize crop production. Several studies in Central Europe have shown significant yield losses in maize production during periods of drought. Kovačević et al. (2013) analyzed on-farm maize yield during extremely wet (2010) and dry years (2012) in Croatia, Serbia, Hungary and Bosnia and Herzegovina and concluded that maize yields were 53, 38, 38 and 40%, respectively, lower in the unfavorable dry year of 2012 compared with the wet year 2010. They suggested the use of irrigation in critical stages of maize growth and growing more drought tolerant genotypes. The positive effect of irrigation on maize grain yield and quality has been the subject of previous studies (Ibrahim & Kandil 2007; Hussein & Pibars 2012; Jampatong et al. 2012; Tariq & Usman 2009; Oktem 2008). Regardless of irrigation system and scheduling method used in research, maize grain yields under irrigation increased. Although the Republic of Croatia is a country rich in clean and fresh water it has the smallest irrigated acreage in the European Union (EU) (Pejdo & Šiljković 2007). According to AQUASTAT (FAO 2012) only 1.1% of agricultural land in Croatia is irrigated which is substantially less than in other EU countries. Total area equipped for irrigation amounts to 1,390 ha in eastern Croatia and 9,275 ha nationwide. The total agricultural acreage in eastern Croatia amounts to 212,013 ha highlighting the potential for further irrigation development. For instance, intense drought damage in the year 2012 were estimated at over 105 million euros (Croatian Bureau of Statistics 2013). In the last few years damage was not only caused by drought but excessive rainfall as well. The growing season of 2010 was extremely wet in the areas of eastern Croatia and damage was estimated at over 67 million euros (Croatian Bureau of Statistics 2011). All data are expressed for farm conditions. The negative impact of excessive rainfall on crops has been the subject of numerous studies (Bancy 2000; Dat et al. 2004; Ifabiyi & Omoyosoye 2011; Parent et al. 2008; Rosenweig et al. 2002). For instance, Rashid & Rasul (2011) analyzed the influence of climate variability on maize production. They found that the correlation between grain yield and amount of rainfall was up to 61% and that grain yield increased with the rainfall up to 450 mm, while it decreased with the rainfall above 450 mm because of runoff water effects on soil fertility. Furthermore, they claimed that an excessive rainfall during a vegetation stage (third to ninth leaf) causes yield to decrease. The most water sensitive stage of maize crop is the reproductive one and excessive rainfall during tasseling, silking and grain formation causes yield reduction. Reduction of maize grain yield as well as lower grain quality caused by water stress have been the subjects of numerous studies (Blum 2005; Cakir 2004; Carena et al. 2009; Hu & Buyanovski 2003; Ibrahim & Kandil 2007; Mohammadkhani & Heidari 2008; Scott et al. 2006), all of which have reported that yield reduction due to drought can be as high as 50% and total yield loss depends on maize growth stage as well as the duration and the severity of the stress. The objective of this paper was to analyse the efficiency of irrigation scheduling in maize production for an experimental site over 3 years with different weather conditions (extremely wet, average and extremely dry years).

MATERIALS AND METHODS

This study was conducted in 2010–2012 at the research site of the Agricultural Institute Osijek, eastern Croatia (45 °32′N and 18°44′E, altitude 90 m). The experimental design was a randomized complete block with three replications. Irrigation was the main factor (a) and maize hybrids were the sub-factor (b). Three irrigation treatments (Table 1) were studied and included a dryland treatment (a1) which received no irrigation (control) and two irrigation management scenarios.

Table 1

Irrigation treatments applied to maize at Osijek (2010–2012)

Year 2010 2011 2012 
Irrigation treatment a2 a3 a2 a3 a2 a3 
Sensors buried in soil – date 12 May 17 May 8 May 
Number of readings 40 36 45 
Soil water content (% of FWC) 60–100 80–100 60–100 80–100 60–100 80–100 
Year 2010 2011 2012 
Irrigation treatment a2 a3 a2 a3 a2 a3 
Sensors buried in soil – date 12 May 17 May 8 May 
Number of readings 40 36 45 
Soil water content (% of FWC) 60–100 80–100 60–100 80–100 60–100 80–100 

Irrigation treatments were based on measuring soil moisture content with the use of Watermark Soil Moisture Sensors (model 200SS). The treatments were designed to irrigate at 60–100% (a2) of field water capacity (FWC) and fully irrigated plots at 80–100% FWC (a3). The Watermark Soil Moisture Sensors were buried in the soil after sowing of the maize crop. The sensors were buried at two depths (20 and 30 cm), two sensors in each irrigation plot and three replications (18 sensors in total). The Watermark sensors were calibrated to the soil on a trial plot by comparing gravimetric measurements and sensor readings. The calibration curve is presented in Figure 1 (Marković 2013).

Figure 1

Calibration curve of Watermark Sensors for the study sites.

Figure 1

Calibration curve of Watermark Sensors for the study sites.

The readings ranged from 0 to 199 cbar (0 denotes the maximum amount of water that soil can retain (100% FWC), while 199 cbar denotes dry soil). In the a3 irrigation plots (full irrigation) the irrigation set point has a value of 40 cbar, which according to the manufacturer general guidelines represents the usual range for irrigation (30–60 cbar). In the a2 plots the irrigation timing has a value of 60–80 cbar. Soil water measurements were made every 2–3 days, depending on the amount of rainfall and irrigation intervals. The plots were irrigated using a traveling sprinkler system. Water for the system was pumped from a 37 m deep well at a 5–7 l s−1 flow rate using an electric pump (5.5 kW). The total size of the irrigation plot was 235 m2 plus side borders which prevented the overlapping of water from different irrigation treatments. The amount of water added in one irrigation event was 35 mm for both irrigation treatments. Amount of water was calculated using the following model: 
formula
1
where the AAW = amount of added water in one irrigation interval (m3 ha−1), h = the depth of irrigation water penetration (m), Bd = bulk density (g cm3), FWC = field water capacity (%), RAW = readily available water (%). The soil at the trial site of the Agricultural Institute in Osijek is hypogley (hydro-meliorated) with its main characteristics presented in Table 2.
Table 2

Properties of the soil

Cm WC (%) AC (%) BD (g cm−3Porosity (%) CC % Humus (%) pH (H2O) N (%) 
0–32 36.57 5.25 2.58 41.82 32.5 1.56 7.5 0.13 
32–50 35.59 6.24 2.65 41.83 31.3    
Cm WC (%) AC (%) BD (g cm−3Porosity (%) CC % Humus (%) pH (H2O) N (%) 
0–32 36.57 5.25 2.58 41.82 32.5 1.56 7.5 0.13 
32–50 35.59 6.24 2.65 41.83 31.3    

WC = water capacity; AC = air capacity; BD = bulk density; CC = clay content.

The maize hybrids OSSK596 (b1), OSSK617 (b2), OSSK602 (b3) and OSSK552 (b4) were planted 0.70 m between rows. The size of the hybrid plots was 19.6 m2. Maize (two rows in each plot) was harvested using a plot combine. Sowing, harvesting time and maize production technology are presented in Table 3. Irrigation efficiency (IE) was calculated as 
formula
2
where Yi is yield in irrigated plots while Yd is yield at dry farming (a1) (Takac et al. 2008). Boss (1979) proposed that the irrigation water use efficiency (IWUE) could be quantified as 
formula
3
where Yi is yield at irrigated plots, Yd represents yield at rainfed treatment (a1) while I (mm) represents the total amount of water delivered through the irrigation system on each irrigation plot. Grass reference evapotranspiration (ETo) was calculated using the FAO56 procedure (Allen et al. 1998). The weather data were obtained from a meteorological and hydrological service and included minimum and maximum air temperatures, wind speed, rainfall, soil temperature, relative humidity and solar radiation on a daily basis. The automatic weather station was located 1.5 km from the plot location. At the experimental site the groundwater levels were measured by using an observation well near to the study site.
Table 3

Maize production technology 2010–2012

Year 2010 2011 2012 
Urea (46%) fertilization 20 April 18 April 12 April 
PK fertilization (1/2 PK) 22 April 19 April 12 April 
0:20:30–250 kg ha−1 
P2O5 (45%) – 50 kg ha−1 
K2O (60%) – 75 kg ha−1 
Soil tillage with rotary harrow 19 April 
Preparation of research plots 24 April 19 April 27 April 
Sowing 6 May 3 May 28 April 
Weed control – Radazin T50 2 l ha−1 7 May 4 May 3 May 
+ Dual Gold 960 EC 1.4 l ha−1 
Sensors setup 12 May 17 May 8 May 
Fertilization CAN (27%) 11 June 29 June 30 June 
Plant density reduction 18 June 30 May and 31 May 29–31 May 
Inter row cultivation and side dressing CAN (27%) 18 June 7 June 8 June 
Harvesting 12 November 3 November 5 November 
Basic NPK fertilization in autumn (1/3 N; ½ PK) 19 November 
Year 2010 2011 2012 
Urea (46%) fertilization 20 April 18 April 12 April 
PK fertilization (1/2 PK) 22 April 19 April 12 April 
0:20:30–250 kg ha−1 
P2O5 (45%) – 50 kg ha−1 
K2O (60%) – 75 kg ha−1 
Soil tillage with rotary harrow 19 April 
Preparation of research plots 24 April 19 April 27 April 
Sowing 6 May 3 May 28 April 
Weed control – Radazin T50 2 l ha−1 7 May 4 May 3 May 
+ Dual Gold 960 EC 1.4 l ha−1 
Sensors setup 12 May 17 May 8 May 
Fertilization CAN (27%) 11 June 29 June 30 June 
Plant density reduction 18 June 30 May and 31 May 29–31 May 
Inter row cultivation and side dressing CAN (27%) 18 June 7 June 8 June 
Harvesting 12 November 3 November 5 November 
Basic NPK fertilization in autumn (1/3 N; ½ PK) 19 November 

The statistical analyses of yield data which included analysis of variance and Fisher's least significant differences test (LSD), where conducted using the SAS statistical software (SAS Institute, Inc., Cary, NC, USA) for Windows. Calibration curve of Watermark Sensors and regression analysis was conducted using STATISTICA 7 (StatSoft, Inc. Tulsa, OK, USA) statistics and analytics software package.

RESULTS AND DISCUSSION

Weather conditions

The average weather conditions in Osijek during the months April to September from 2010 to 2012 are shown in Table 4. The first year of the experiment was warm and extremely wet. At the global scale, the year 2010 (along with 2005) was the warmest year since weather records began in the nineteenth century. Total amount of year rainfall was 1,038.2 mm, 60% above the long-term average for this region. The amount of rainfall during the growing season was 676, 308.6 mm above the long-term average. The excessive amount of rainfall in 2010 caused a variety of problems for maize production. Sowing was postponed (6 May) because of a rainy April (71.1 mm) after which the Osijek area experienced an exceptionally wet May with 120.8 mm of rainfall. The highest rainfall in 2010 was registered in July when 107.2 mm of rainfall was recorded (Figure 2) during 1 day (22 July).

Table 4

Daily value averages of minimum (Tmin) and maximum air temperatures (Tmax), relative humidity (%), wind speed (km day−1), sunshine hours, solar radiation (MJ m−2 day−1) and grass reference evapotranspiration (mm day−1)

   Tmin Tmax Humidity Wind Sun RAD ETo 
Year Month °C °C km day−1 MJ m−2 day−1 mm day−1 
2010 April 7.2 17.9 74 1.8 6.9 16.9 2.15 
May 11.8 21.9 76 2.1 5.5 17.1 2.69 
June 15.5 25.5 78 1.8 7.3 20.4 3.53 
July 17.6 28.9 74 1.7 9.5 22.8 4.17 
August 15.8 27.8 75 1.5 9.9 21.5 3.81 
September 11.3 20.7 82 1.5 4.8 12.4 2.03 
Average 13.2 23.8 76.5 1.4 7.3 16.9 3.1 
2011 April 7.5 19.7 68 1.8 6.8 17.3 2.27 
May 10.1 22.9 70 1.6 8.5 21.6 3.25 
June 14.7 27.1 72 1.7 9.9 24.2 4.13 
July 15.8 28.2 69 1.7 8.5 21.8 3.87 
August 15.7 30.4 65 1.5 10.2 22.4 3.88 
September 13.2 28.4 65 1.5 8.7 17.5 2.87 
Average 12.8 26.1 68 1.5 8.8 20.8 3.38 
2012 April 6.5 19.0 72 1.9 6.1 16.1 2.18 
May 11.0 23.1 73 1.6 8.1 20.9 3.19 
June 15.3 28.5 69 1.6 10.6 25.2 4.34 
July 17.5 31.7 59 1.7 10.7 24.9 4.47 
August 15.3 32.2 53 1.6 11.4 24.0 4.10 
September 12.4 26.3 67 1.7 6.5 14.9 2.48 
Average 13.0 26.8 66 1.7 8.9 21.0 3.46 
   Tmin Tmax Humidity Wind Sun RAD ETo 
Year Month °C °C km day−1 MJ m−2 day−1 mm day−1 
2010 April 7.2 17.9 74 1.8 6.9 16.9 2.15 
May 11.8 21.9 76 2.1 5.5 17.1 2.69 
June 15.5 25.5 78 1.8 7.3 20.4 3.53 
July 17.6 28.9 74 1.7 9.5 22.8 4.17 
August 15.8 27.8 75 1.5 9.9 21.5 3.81 
September 11.3 20.7 82 1.5 4.8 12.4 2.03 
Average 13.2 23.8 76.5 1.4 7.3 16.9 3.1 
2011 April 7.5 19.7 68 1.8 6.8 17.3 2.27 
May 10.1 22.9 70 1.6 8.5 21.6 3.25 
June 14.7 27.1 72 1.7 9.9 24.2 4.13 
July 15.8 28.2 69 1.7 8.5 21.8 3.87 
August 15.7 30.4 65 1.5 10.2 22.4 3.88 
September 13.2 28.4 65 1.5 8.7 17.5 2.87 
Average 12.8 26.1 68 1.5 8.8 20.8 3.38 
2012 April 6.5 19.0 72 1.9 6.1 16.1 2.18 
May 11.0 23.1 73 1.6 8.1 20.9 3.19 
June 15.3 28.5 69 1.6 10.6 25.2 4.34 
July 17.5 31.7 59 1.7 10.7 24.9 4.47 
August 15.3 32.2 53 1.6 11.4 24.0 4.10 
September 12.4 26.3 67 1.7 6.5 14.9 2.48 
Average 13.0 26.8 66 1.7 8.9 21.0 3.46 
Figure 2

Daily rain (mm) during maize growing season at Osijek (2010–2012).

Figure 2

Daily rain (mm) during maize growing season at Osijek (2010–2012).

The local government authorities declared a natural disaster due to flooding and excessive rainfall on 28 May and 9 June. The damage was estimated at 6.7 million euros. The total amount of rainfall in June was 234 mm which is 166% greater than the long-term average for the region. Although there was an excessive amount of rainfall and water logging in our case there was no yield reduction caused by the periods of floods during the 2010 growing season. Surface water was removed from the given area by temporary drainage ditches. Water was collected and transported by gravity to a drainage network (canal). The damage for young maize plants was minimal due to the short period of water logging and there was no decreased accessibility to oxygen.

Furthermore, wet soils are cold so corn growth was slow. Yet, with the drainage we managed to lower the moisture content of the upper soil layers. However, the groundwater level was very high (Figure 3) and had a negative impact on irrigation scheduling and soil moisture monitoring discussed later. Most of the summer crops in farm conditions of eastern Croatia (Osijek area) were completely destroyed by flooding. Yet in areas without water logging and with usual groundwater levels the yields of maize grain (7.5 t ha−1) under farm conditions were 16% higher in comparison to those in the extremely warm and dry 2011 and 50% higher in comparison to the extremely warm and very dry 2012. The significantly (P < 0.01) lower yields in the years 2011 and 2012 were the result of insufficient and poorly distributed rainfall and inadequate soil moisture during June, July and August (Figure 2). For instance, during August of 2011 and 2012 in the Osijek area there was only 4.6 and 4.0 mm of rainfall, respectively, while air temperatures were considerably above long-term average. The monthly average air temperatures (Table 5) in the June–September period were higher by 1.3, 1.1, 2.7 and 3.7 °C in the year 2011, and 3, 3.7, 3.8 and 2.3 °C in the year 2012.

Table 5

Average air temperatures (°C) for Osijek region in 2010–2012 period

  Month
 
    
Year II III IV VI VII VIII IX XI XII  GP 
2010 −0.8 1.4 6.8 12.4 16.5 20.4 23.2 21.7 15.6 9.1 8.9 0.2 11.3 18.3 
2011 1.1 0.7 6.4 13.2 16.7 20.8 22.2 23.0 20.3 10.6 2.3 3.4 11.7 19.4 
2012 2.2 -4.1 8.7 12.5 16.9 22.5 24.8 24.1 18.9 12.1 9.0 2.3 12.5 19.9 
LTA 1.2 1.6 6.1 11.3 16.5 19.5 21.1 20.3 16.6 11.2 5.4 0.9 10.8 17.6 
  Month
 
    
Year II III IV VI VII VIII IX XI XII  GP 
2010 −0.8 1.4 6.8 12.4 16.5 20.4 23.2 21.7 15.6 9.1 8.9 0.2 11.3 18.3 
2011 1.1 0.7 6.4 13.2 16.7 20.8 22.2 23.0 20.3 10.6 2.3 3.4 11.7 19.4 
2012 2.2 -4.1 8.7 12.5 16.9 22.5 24.8 24.1 18.9 12.1 9.0 2.3 12.5 19.9 
LTA 1.2 1.6 6.1 11.3 16.5 19.5 21.1 20.3 16.6 11.2 5.4 0.9 10.8 17.6 

LTA = long-term average (1961–1990); = year average; GP = growing period average.

Figure 3

Groundwater levels during the growing season (2010–2012).

Figure 3

Groundwater levels during the growing season (2010–2012).

The 2011 and 2012 drought and heat stresses caused more than 64 and 105 million euros, respectively, in direct losses to agriculture. The local authorities declared a natural disaster due to drought on 1 August, 8 September (2011) and 27 July (2012). Although there was excessive rainfall during the 2010 growing season, lack of rainfall (33.5 mm) occurred in July (Figure 2) and represents the difference between amount of rainfall in July 2010 and long-term average (1961–1990) for this region. In the same period, the average air temperatures were 2.1 °C higher than normal and maximum ETo was 4.17 mm day−1 (Table 4).

Soil moisture conditions at experimental plot

Watermark sensor readings in 2010 (Figure 4) were very low in the April–June period due to excessive amount of rainfall in the spring period. The sensor readings show that there was no need to irrigate until July since the soil moisture was very high, 80–100% FWC (Figure 4). Owing to the lack of rainfall in 2010 and high air temperatures, soil moisture depletion occurred in July which made it necessary to irrigate. In the growing season of 2010 only one irrigation event was applied in the a2 plots (35 mm) and three irrigations on the fully irrigated a3 plots (105 mm) due to the frequent and high amounts of rainfall (Figure 2). This was considerably less than in 2011 and 2012, when 105 mm and 175 mm, respectively, of irrigation water were applied to treatment a2 and 245 mm was applied in both years to treatment a3 (Table 6). The amount of irrigation was based on soil moisture content, amount of rainfall and also on the distribution of rainfall during the growing season. The severe lack of rainfall and extremely high temperatures during the 2011–2012 period caused significant soil moisture depletion (Figure 4). Irrigation events are presented with the × mark in Figure 4. During 2011 and 2012, drought occurred at the most sensitive stages for maize growth. It was very difficult to keep soil moisture values at a high level not only due to the lack of rainfall but also due to the extremely high temperatures (Table 5) and ground water levels below 350 cm (Figure 3). For instance, ETo in June, July, August 2012 was 4.34, 4.47 and 4.1 mm day−1, respectively, so it was difficult to compensate for water losses.

Table 6

Number of irrigation events at both irrigation treatments

Irrigation regimes a2 a3 
Year mm mm mm 
2010 35 35 105 
2011 35 105 245 
2012 35 175 245 
Irrigation regimes a2 a3 
Year mm mm mm 
2010 35 35 105 
2011 35 105 245 
2012 35 175 245 

a2 = 60–100% FWC; A3 = 80–100% FWC; mm = amount of irrigation water added in one irrigation regime; n = number of irrigation regimes.

Figure 4

Soil moisture sensors readings (2010–2012).

Figure 4

Soil moisture sensors readings (2010–2012).

Yield of maize hybrids and water efficiency under different irrigation regimes

The average yields of maize grain for the different irrigation treatments are shown in Table 7. The statistical analysis of maize yields revealed that there was significant (P < 0.01) difference between irrigation treatments in all 3 years. Although the 2010 growing season for Osijek area was extremely wet and warmer than normal (1961–1990) with an average air temperatures of 11.3 °C which was almost 4.7% above normal (10.8 °C), and the amount of rainfall in July at 31.5 mm which was 51% below average, our results are opposite to what we expected since the highest yield of maize grain was in the rainfed treatment (a1). According to the results of our study, although there was a need to irrigate since the sensor readings were indicating soil moisture depletion during July (2010), the observation of groundwater level revealed that water in July was 100 cm from the soil surface and readily available for maize growth (Figure 3). Installation of soil moisture sensors at 30 cm depth was adequate for average climatic years, but in extreme weather conditions such as in 2010 it was necessary to take into account the groundwater level which was extremely high (Figure 3). Consequently, maize yield was significantly reduced (P < 0.01) in fully irrigated treatments in comparison to the control in 2010. Irrigation irrespective of ground water level may have decreased accessibility to oxygen in the plant root zone and induced stress so the yield of the fully irrigated treatment (a3) was almost 8% lower compare to the rainfed treatment (a1). IE in 2010 was −7% for the a2 treatment and −65% for the a3 treatment. Furthermore, IWUE in 2010 was −3.5 (a2) and −6.19 (a3) kg ha mm−1. Yield trend was reversed in 2011 and 2012 when maize grain yield was higher as the amount of irrigation water increased. For instance, in 2011 the maize grain yield from the a2 treatment was 18% higher compared to the rainfed treatment (a1). In the same year, in fully irrigated treatment (a3) yield was 25% higher than that of the control plots (a1) and 11% higher than that of the a2 treatment. IE was 118% (a2) and 125% (a3), and IWUE 12.9 (a2) and 7.68 (a3) kg ha mm−1 in 2011. In 2012, reduced rainfall and higher air temperatures resulted in greater yield in the irrigated treatment. Maximum yield was obtained in fully irrigated treatment (a3), 40% greater compared to control (a1) and 13% greater compared to the a2 treatments. In the extremely warm and very dry 2012 IE was 167% (a2) and 291% (a3), while IWUE was 9.53 (a2) and 11.87 kg ha mm−1 (a3). Averages across hybrids yield of maize grain ranged from 7.62 (b4) to 10.55 t ha−1 (b3) in 2010, from 8.40 (b1) to 8.76 t ha−1 (b3) in 2011 and from 8.27 (b1) to 9.28 t ha−1 (b2, b3) in 2012. Hybrid OSSK602 (b3) had a higher yield compared with the other hybrids and proved to have good tolerance to drought and adaptability to different weather conditions (extremes). For instance, hybrid b3 had a significantly higher yield in the extremely wet 2010, almost 28% higher compared to b4 in the fully irrigated treatment where stress due to excessive amount of water occurred. Furthermore, in the dry years of 2011 and 2012, in the rainfed treatment when the soil moisture level was very low (Figure 3) maximum yield was also measured for the b3 hybrid. The result is similar to previous research results (Marković et al. 2012) when in the dry season of 2007 hybrid OSSK602 achieved the highest yield (the same hybrids were included in the research).

Table 7

Analysis of variance for grain yield of four maize hybrids grown under different irrigation regimes

  Yield (t ha−1)
 
 
 b1 b2 b3 b4  Analysis of variance 
2010 
 a1 8.42 9.89 11.14 7.50 9.24  LSD0.05 LSD0.01 F value 
 a2 9.19 9.25 10.50 7.52 9.17 0.25 0.33 14.74** 
 a3 8.13 8.37 10.01 7.83 8.59 0.18 0.26 351.75** 
 8.58 9.17 10.55 7.62  a × b 0.36 0.52 20.06** 
2011 
 a1 7.27 7.31 7.72 7.57 7.47  LSD0.05 LSD0.01 F value 
 a2 8.81 8.90 9.21 8.37 8.82 0.24 0.32 122.30** 
 a3 9.13 9.44 9.35 9.47 9.35 0.17 0.24 6.39** 
 8.40 8.55 8.76 8.47  a × b 0.34 0.49 5.24** 
2012 
 a1 6.70 6.92 7.97 7.90 7.37  LSD0.05 LSD0.01 F value 
 a2 8.21 8.94 9.51 9.51 9.04 0.53 0.70 57.87** 
 a3 9.90 10.44 10.35 10.35 10.28 0.16 0.22 75.02** 
 8.27 9.28 9.28 8.90  a × b 0.31 0.45 9.30** 
  Yield (t ha−1)
 
 
 b1 b2 b3 b4  Analysis of variance 
2010 
 a1 8.42 9.89 11.14 7.50 9.24  LSD0.05 LSD0.01 F value 
 a2 9.19 9.25 10.50 7.52 9.17 0.25 0.33 14.74** 
 a3 8.13 8.37 10.01 7.83 8.59 0.18 0.26 351.75** 
 8.58 9.17 10.55 7.62  a × b 0.36 0.52 20.06** 
2011 
 a1 7.27 7.31 7.72 7.57 7.47  LSD0.05 LSD0.01 F value 
 a2 8.81 8.90 9.21 8.37 8.82 0.24 0.32 122.30** 
 a3 9.13 9.44 9.35 9.47 9.35 0.17 0.24 6.39** 
 8.40 8.55 8.76 8.47  a × b 0.34 0.49 5.24** 
2012 
 a1 6.70 6.92 7.97 7.90 7.37  LSD0.05 LSD0.01 F value 
 a2 8.21 8.94 9.51 9.51 9.04 0.53 0.70 57.87** 
 a3 9.90 10.44 10.35 10.35 10.28 0.16 0.22 75.02** 
 8.27 9.28 9.28 8.90  a × b 0.31 0.45 9.30** 

a1 = rainfed; a2 = 60–100% FWC; a3 = 80–100%; b1 = OSSK596; b2 = OSSK617; b3 = OSSK602; b4 = OSSK552.

** = P < 0.01.

CONCLUSION

Maize grain yields in rainfed treatments across several years were highest in the extremely wet year (2010) when water was not a limiting factor. If rainfall comes in the form of an intense single-day event, excess surface water must be removed from the upper soil layer to limit water logging and sustain accessibility to oxygen. Irrigation scheduling based on monitoring of soil moisture should be adjusted to extreme conditions. In such conditions, the main factor for irrigation efficiency is to properly determine the optimum level for soil moisture sensors installation and ground water level in relation to root depth of maize plant. Otherwise, there is a possibility that irrigation will reduce accessibility to oxygen in the upper soil level so the plants are induced to stress. In our study, the yield in fully irrigated treatment was almost 8% lower compared to rainfed treatment under excessive rainfall. During extremely dry years, irrigation linearly increased (P < 0.01) the yield of maize grain and maximum yields were obtained in fully irrigated treatment. In extreme weather events and variability of climate, it is important to choose the hybrids with good tolerance to drought and good adaptation to different growing conditions. Based on our study results, maize hybrid OSSK602 has demonstrated good adaptation to environmental variability and showed maximum yield potential in extreme weather events.

REFERENCES

REFERENCES
Allen
R. G.
Pereira
L. S.
Raes
D.
Smith
M.
1998
Crop Evapotranspiration: Guidelines for Computing Crop Water Requirements
.
Irrigation and Drainage Paper No. 56
.
FAO
,
Rome
,
Italy
.
Boss
M. G.
1979
Standards for irrigation efficiencies of ICID
.
Journal of Irrigation and Drainage
105
,
37
43
.
Carena
M. J.
Bergman
G.
Riveland
N.
Eriksmoen
E.
Halvorson
M.
2009
Breeding maize for higher yield and quality under drought stress
.
Maydica
54
,
287
296
.
Croatian Bureau of Statistics
2011
Statistical Yearbook of the Plant Production in Osijek Baranya County, 2010 (in Croatian)
.
Croatian Bureau of Statistics
2013
Statistical Yearbook of the Plant Production in Osijek Baranya County, 2012 (in Croatian)
.
Dat
J.
Capelli
N.
Folzer
H.
Bourgeade
P.
Badot
P. M.
2004
Sensing and signalling during plant flooding
.
Plant Physiology and Biochemistry
42
,
273
282
.
FAO
2012
AQUASTAT Global Map of Irrigation Areas
.
Food and Agriculture Organisation of the United Nations
,
Rome
,
Italy
.
Hu
Q.
Buyanovski
G.
2003
Climate effects on corn yield in Missouri
.
Journal of Applied Meteorology
42
,
1626
1635
.
Hussein
M. M.
Pibars
S. K.
2012
Maize response to irrigation system, irrigation regimes and nitrogen levels in sandy soil
.
Journal of Applied Sciences Research
8
,
4733
4743
.
Ibrahim
S. A.
Kandil
H.
2007
Growth, yield and chemical constituents of corn (Zea mays L.) as affected by nitrogen and phosphorus fertilization under deficit irrigation intervals
.
Journal of Applied Sciences Research
10
,
1112
1120
.
Ifabiyi
I. P.
Omoyosoye
O.
2011
Rainfall characteristic and maize yield in Kwara state, Nigeria
.
Indian Journal of Fundamental and Applied Life Sciences
3
,
60
65
.
Jampatong
S.
Wongpila
W.
Stamp
P.
Messmer
R.
2012
Yield of tropical Asian maize (Zea mays L.) at alternating row irrigation and at severe drought
.
Maydica
57
,
249
251
.
Kovačević
V.
Kovačević
D.
Pepo
P.
Marković
M.
2013
Climate change in Croatia, Serbia, Hungary and Bosnia and Herzegovina: comparison the 2010 and 2012 maize growing season
.
Agriculture
19
,
16
22
.
Marković
M.
2013
Influence of irrigation and nitrogen fertilization on yield and quality of maize grain (Zea mays L.)
.
PhD Thesis. Faculty of Agriculture
,
University of J.J. Strossmayer in Osijek
,
Osijek
,
Croatia
.
Marković
M.
Péter
P.
Sárvári
M.
Kovačević
V.
Šoštarić
J.
Josipović
M.
2012
Irrigation water use efficiency in maize (Zea mays L.) produced with different irrigation intervals
.
Acta Agronomica Hungarica
60
,
21
27
.
Mohammadkhani
N.
Heidari
R.
2008
Effects of drought stress and soluble proteins in two maize varieties
.
Turkish Journal of Biology
32
,
23
30
.
Parent
C.
Capelli
N.
Berger
A.
Crevecoeur
M.
Dat
J. F.
2008
An overview of plant response to soil waterlogging
.
Plant Stress
2
,
20
27
.
Pejdo
A.
Šiljković
Ž.
2007
Possibilities for irrigation in Croatia by using groundwater reserves
.
Geodaria
12
,
111
129
.
Rashid
K.
Rasul
G.
2011
Rainfall variability and maize production over the Potohar plateau of Pakistan
.
Pakistan Journal of Meteorology
8
,
63
74
.
Rosenweig
C.
Tubiello
F. N.
Goldberg
R.
Mills
E.
Bloomfield
J.
2002
Increased crop damage in US from excess precipitation under climate change
.
Global Environmental Change
12
,
197
202
.
Scott
M. P.
Edwards
J. W.
Bell
C. P.
Schussler
J. R.
Smith
J. S.
2006
Grain composition and amino acid content in maize cultivars representing 80 years of commercial maize varieties
.
Maydica
51
,
417
423
.
Šoštarić
J.
Marković
M.
Šimunić
I.
2012
Irrigation – wish or necessity? In
:
Proceedings of 4th International Scientific and Expert Conference of the International TEAM Society. 17–19 October
,
Slavonski Brod
,
Croatia
1
,
17
20
.
Takac
J.
Nejedlik
P.
Siska
B.
2008
Irrigation water use efficiency
.
Cost734 Workshop, Jois, AT, 6th–8th October, 2008
,
Adagio
,
Cecilia
.
Tariq
J. A.
Usman
K.
2009
Regulated deficit irrigation scheduling on maize crop
.
Sarhad Journal of Agriculture
25
,
441
450
.