OPTIMISING MAIZE YIELD UNDER IRRIGATION

 

Optimising maize yield under irrigation

Optimising maize yield under irrigation requires correct integration of crop genetics, crop management and environment factors. Crop genetics relates to choosing varieties that adapt to local conditions and that have an adequate yield potential. Crop management relates to providing the crop with growing conditions that are as close as possible to optimal. This includes a variety of practices such as land preparation, sowing, soil nutrition, pest (insects, disease, and weeds) control, irrigation, etc. These practices can significantly influence crop yield, either individually or in combination, but can generally be managed, depending on farmer’s skills, knowledge and resources. The other aspect of optimising yields has to do with environmental factors, such as the weather, soil and water quality, which cannot always be totally controlled.

 

Assuming that genetics, crop management, and environmental factors are properly managed, optimizing maize yield under irrigation requires providing the crop with enough water to meet crop evapotranspiration requirements at all times. Therefore, both correct amount and timing of water applications are critical. Considerable research has shown that maize yield is linearly related to crop evapotranspiration (ET). Crop ET is the combination of water that goes through the plant (Transpiration) and water that is lost from the soil surface (Evaporation). Many researchers have reported linear relationships between maize evapotranspiration and yield. For example, Figure 1 shows the relationship adapted from two field experiments reported by Payero et al. (2008a) and Payero et al. (2008b). It shows that at that location, it took about 270 mm of evapotranspiration to start producing grain yield. This includes water needed just to produce vegetative growth and water that was lost by soil evaporation and did not contribute to grain yield.   The slope of the line indicates that after the first yield unit was produced, each mm of ET produced and additional 0.0317 Mg/ha (= 31.7 kg/ha) of grain, and it took about 663 mm of ET to produce maximum yield. These results suggest that stress at any time during the growing period will reduce ET and, therefore, will reduce yield. The magnitude of the yield reduction depends on the severity, timing, and duration of water stress.

 

Although the relationship between maize yield and ET is relatively consistent for a particular location, maize yield response to irrigation is very variable and changes with location and season. This is shown in Figure 2, from a five-year experiment conducted in Nebraska, USA (Payero et al. 2006).  In this experiment the following four irrigation treatments were compared during 1992 to 1996:

 

 

The FARMER treatment was a fully-irrigated treatment and always resulted in more irrigation. Results in Figure 2 show that during three out of the five years there was no response to irrigation at all, since those were wet years. However, even during those years the FARMER treatment still received around 500 mm of irrigation. It is also important to note what happened during the two driest years (1994 and 1995) in which there was a response to irrigation. In 1994, the deficit irrigation treatments produced around 90% of normalized yield with about 130 mm of irrigation. The FARMER treatment received 500 mm and only produced an additional 10% increase in yield. In 1995, the deficit-irrigation treatments produced 70% of the normalized yield with about 110 mm of irrigation. The FARMER treatment received 700 mm to get the additional 30% increase in yield. This illustrates that more irrigation not always result in additional yield and that sometimes full yield can be obtained with no irrigation at all.

 

 

Figure 2: Relationship between seasonal irrigation and normalized yield for surface-irrigated corn in Nebraska (from Payero et al., 2006).

 

 

How much water is needed to optimize maize yield?

ET depends mainly on weather conditions, crop cover (crop development stage and density), and available soil water. Therefore, maize ET varies daily during the growing season and with location. For example, Fig. 3 shows daily maize ET measured with an eddy covariance system in Nebraska, USA (Payero et al. 2008c). It shows that maize ET increased from emergence, peaked at about 8.5 mm/day at about 90 days after emergence and then steadily decreased. The specific shape and magnitude of the ET curve will be different for each location and for each season. Day to day variability in ET mainly reflects daily changes in weather conditions.    

 

 

Figure 3: Crop evapotranspiration for maize measured at North Platte, Nebraska, during 2001, using an eddy covariance system (Payero et al. 2008c).

 

Daily values of maize evapotranspiration (ETc, mm/d) for a given location can be estimated from weather data as:

 

ETc = ETo x Kc                                                                                              (1)

 

Where, ETo = grass reference evapotranspiration (mm/d), and Kc = crop coefficient.  ETo can be calculated from weather data using a variety of methods.  Locally-tested Kc values for different crops are currently lacking in Australia, but values have been suggested from empirical studies conducted around the world (Allen et al., 1998) that can be used to obtain a good estimation of ETc.

 

Figure 4 shows long-term average daily ETo values for different locations in Australia reported by SILO, which were calculated from weather data. Figure 5 shows the crop coefficient (Kc) curve developed from international values suggested by Allen et al. (1998). Figure 6 shows the average daily ETc for each location, calculated with equation (1) using the sowing dates shown in Table 1. Daily cumulative ETc and seasonal ETc are shown in Figs 7 and 8. Figure 5 and 6 show very similar daily and cumulative ET early in the season for all locations. However, considerable differences in ET start to show after about 70 days after sowing. These figures were generated to guide farmers as to the magnitude and timing of crop water use. However, as with any modelling approach there are uncertainties, including:

 

·         Yearly and daily deviations in weather conditions from the long-term average

·         Variations in sowing dates

·         Differences among crop varieties, especially related to season length

·         Lack of locally-determined Kc values,

·         Uncertainties in ETo since SILO uses a fixed value for wind speed of 2 m/s.     

 

Figures 5 to 8 show that maize ET requirements to produce maximum yield can vary considerably with location, with a difference in seasonal ET of 85 mm between Bookstead and St. George. Applying more water than the ET requirements of the crop will not increase yields, and can in fact reduce yields due to problems like nitrate leaching and water logging. When planning irrigations, it should be considered that irrigation is only needed to supplement ET requirements that cannot be met by water stored in the soil profile at sowing and by effective in-crop rainfall. Therefore it is critical to have some means of measuring or estimating soil water content during the season to determine when and how much irrigation is needed, if any.  

 

Location	Maize sowing date
Brookstead	15 Sep
Dalby	15 Sep
Goondiwindi	1 Sep
St. George	20 Aug
Emerald	15 Aug

 

 

Figure 4: Grass reference evapotranspiration (ETo) for different locations in Australia.

 

Figure 5: Crop coefficient (Kc) curve for maize (Allen et al., 1998).

 

Figure 6: Average daily maize evapotranspiration (ETc) calculated for different locations.

 

Figure 7: Average daily maize cumulative evapotranspiration (ETc) calculated for different locations.

 

 

 

Some local data

To illustrate the magnitude of the irrigated maize yields that can be obtained locally and how much in-crop rain and irrigation water it took to obtain those yields, we obtained data from the crop competitions conducted on the Darling Downs and the Lockey Valley. The quality of irrigation data from this dataset is quite variable and rarely measured quantities were provided. Estimates of applied irrigation water, when only number of irrigation were given, were based on the assumption that one flood irrigation was equivalent to 1 ML/ha. Data about stored soil water was seldom collected and is not included, although it should be noted that stored soil water is an important source of water to meet crop water requirements and can be very significant, especially in the heavy soils predominant on the Darling Downs.

 

Irrigated maize data from the crop competitions from 1987 to 2005 summarised in Table 2 show that maize yields averaged 11.5 Mg/ha during that period, but maximum yields ranged from 11.0 to 15.6 Mg/ha. Those yields were produced with seasonal irrigation depths ranging from 200 to 427 mm, with an average of 317 mm (100 mm = 1 ML/ha). In-crop rainfall ranged from 203-672 mm, with an average of 440 mm. Rain + irrigation raged from 503-931 mm, with an average of 767 mm.  Table 1 also shows the yield per unit irrigation (Y/Irrig) and per unit of rain + irrigation [Y/R+I)]. Although data in Table 1 do not tell us weather those yields could be obtained with less water, since that would require more in-depth analyses, they provide an indication of actual values reported by real producers.

 

		Yield (Mg/ha)		Irrig	Rain	Rain+Irrig	Y/Irrig	Y/(R+I)
Year	Max	Min	Avg	(mm)	(mm)	(mm)	(kg/ha/mm)	(kg/ha/mm)
1987	11.1	8.3	9.6	344	452	796	28.0	12.1
1988	13.9	8.5	11.0	343	588	931	32.2	11.9
1989	11.0	8.5	10.0	300	203	503	33.4	19.9
1990	12.7	9.7	11.6	340	543	883	34.2	13.2
1991	11.7	8.3	10.2	427	360	787	23.8	12.9
1992	13.4	9.1	11.3	304	591	895	37.0	12.6
1993	13.5	7.5	10.7	300	248	548	35.7	19.6
1994	11.4	8.6	9.9	358	525	883	27.7	11.3
1995	12.0	8.7	10.4	340	386	726	30.5	14.3
1996	13.1	11.1	11.9	225	672	897	52.8	13.3
1997	14.4	11.1	12.5	283	261	544	44.2	23.0
1998	13.2	8.5	10.4	313	374	686	33.2	15.1
1999	14.3	11.8	12.8	375	517	892	34.2	14.4
2000	15.3	10.3	12.8	256			50.2
2001	15.1	10.3	13.4	375			35.6
2002	14.4	10.4	12.7	327			38.9
2003	13.1	11.0	12.0	337			35.8
2004	13.8	8.8	11.6	200			57.8
2005	15.6	11.9	13.3	275			48.3
Avg	13.3	9.6	11.5	317	440	767	37.6	14.9
Min	11.0	7.5	9.6	200	203	503	23.8	11.3
Max	15.6	11.9	13.4	427	672	931	57.8	23.0

 

 

References

Allen RG, Pereira LS, Raes D, Smith M (1998) Crop evapotranspiration - guidelines for computing crop water requirements (Irrigation and Drainage Paper No. 56). Food and Agriculture Organization of the United Nations (FAO), Rome, Italy.

Payero, J.O., Klocke, N.L., Schneekloth, J.P. and Davison, D.R., 2006. Comparison of irrigation strategies for surface-irrigated corn in West Central Nebraska. Irrigation Science, 24(4): 257-265.

Payero, J.O., D.D. Tarkalson, S. Irmak, D. Davison, and J.L. Petersen. 2008a. Effect of irrigation amounts applied with subsurface drip irrigation on corn evapotranspiration, yield, water use efficiency, and dry matter production in a semiarid climate. Agricultural Water Management. In press.

Payero, J.O., D.D. Tarkalson, S. Irmak, D. Davison, and J.L. Petersen. 2008b. Effect of timing of a deficit-irrigation allocation on corn evapotranspiration, yield, water use efficiency and dry mass. Agricultural Water Management. In Review.

Payero, J.O., and S. Irmak. 2008c. Measurements of actual evapotranspiration, crop coefficient, and energy balance components of surface-irrigated corn. Irrigation Science-In Review.

 

Contact details

Dr. Jose Payero

Department of Primary Industries and Fisheries

Toowoomba, Qld

Ph: (07) 4688 1513

Email: jose.payero@dpi.qld.gov.au