Abstract

Media summary

Key Words

Sprinkler-line source, Yield, Harvest Index (HI), Irrigation Water Use Efficiency (IWUE)

Introduction

Water scarcity is an increasingly important issue in many parts of the world. Efficient use of water in agriculture is needed for the conservation of existing water supplies. Increase in water use efficiency (WUE) can be achieved by deficit irrigation practices, contributing to water conservation.

Maize is a major summer crop in the irrigated areas of the Mediterranean region. It has high water requirements and it is highly productive if water and nutrients are not limiting. However, it is very sensitive to water stress (Rhoads and Bennet 1990). In contrast, sorghum, which is grown as a summer crop in similar agronomic conditions to maize, is a drought tolerant crop (Krieg and Lascano 1990). Its drought resistance is attributed to its prolific root system and its phenology and osmotic adjustment to water stress (Girma and Krieg 1992; Singh and Singh 1995).

A field experiment, with a range of irrigation treatments, was performed to compare the response of maize and sorghum to water deficit and to study sorghum as an alternative crop to maize under water limiting conditions.

Methods

A field experiment was conducted in 1995 in Zaragoza, Spain (latitude 41° 43’ N, longitude 0° 49’ W) on a loam soil classified as Typic Xerofluvent. The average volumetric soil water content at drainage upper limit and lower limit in the 0-1 m profile was 27.8% and 8.9%, respectively. Average temperature during the growing season (May-September) ranged from 19.6 to 29.6 °C. Rainfall from May to September was 32 mm. Reference evapotranspiration (ET) from sowing to maturity was 695 mm. Crops were sown in May, at 0.75 m row spacing and achieved final plant densities of 5.2 pl/m² and 21.7 pl/m² for maize and sorghum, respectively. Calculation of fertilizer amounts were based on soil fertility and expected yields so that nutrients were not limiting. Weeds, pests and diseases were controlled.

The sprinkler line-source technique (Hanks et al. 1976), which provides a continuous variable water gradient perpendicular to the sprinkler line, was used to establish six irrigation treatments (T-1 to T-6). The applied irrigation water decreased linearly from the sprinkler line (T-1) to the edge of the plot (T-6). Irrigation schedule was calculated so that T-1 received the full water requirements and that therefore T-2 to T-6 received decreasing amounts of water.

Water applied and crop water uptake were monitored during the season. Phenological stages in maize and sorghum were recorded. Fraction of light intercepted (fIPAR), leaf area index (LAI) and total above-ground biomass were measured at different times during the growing season. At maturity, total biomass, yield and harvest index (HI) were obtained.

Results

Soil water depletion (difference between soil water content at the beginning and at the end of the growing season) in T-1 was relatively small in both crops (Fig. 1). In T-3 and T-5 soil water uptake in sorghum was greater than in maize. In maize most of the water uptake was from 0-50 cm profile, while sorghum extracted water to 100 cm deep and to lower values of lower limit than maize (Fig. 1). Similarly, in an experiment with limited water supply, Singh and Singh (1995) found that sorghum extracted more water from deep soil layers and maize from top soil layers.

Fig. 1. Soil volumetric water content (θ_v) at the beginning (dashed line) and the end (solid line) of the growing season for treatments T-1 (circles), T-3 (squares) and T-5 (triangles) for maize (MZ95) and sorghum (SG95).

Maize had longer growth duration than sorghum. Flowering and maturity were delayed in both crops under water deficit conditions. Reduction in the fraction of intercepted light as irrigation level decreased was similar in both crops (data not shown). Reduction in LAI in response to deficit irrigation was more important in maize than in sorghum (data not shown). Light extinction coefficient (K) values (estimated from fIPAR and LAI measurements) showed differences in leaf architecture between crops, both under full irrigation and under water deficit conditions. Sorghum had canopy characteristics, such as low K and more erect leaves, which conferred better adaptation to water stress (Farré 1998).

Final above-ground biomass, yield and HI were higher in maize than in sorghum in T-1 and T-2, treatments with nil or minimal water deficit. However, water deficits reduced biomass and yields more in maize than in sorghum, resulting in higher yields and HI in sorghum in T-3 to T-6, treatments with moderate or severe water deficit (Table 1). The effects of deficit irrigation on yield were greater than on total biomass, as reflected in the lower HI under deficit irrigation (Table 1). At high irrigation levels, maize HI was higher than that of sorghum, but decreased markedly with increasing water deficit. In contrast, in sorghum there was a range of moderate deficit irrigation treatments over which HI was unaffected, and HI was only reduced in the severe deficit irrigation treatments (Fig. 2). Sorghum was therefore, more efficient in converting biomass into grain yield than maize under irrigation deficit conditions. Similarly, Garrity et al. (1983), working with sorghum reported no change in HI for a range of water deficits.

Traditionally the relationship between yield and ET has been studied in WUE analysis. However, not all irrigation water is used in the ET process (i.e. runoff, deep percolation) and a fraction of the ET comes from sources other than irrigation (i.e. stored soil water, rainfall). At high irrigation levels, no linear relationships have been found between yield and irrigation (Y-I) (Mantovanni et al. 1995). In this study however, within the range of water applied, a linear model best fitted the Y-I relationship. Maize outyielded sorghum at the highest level of water applied, but the slope of the relationship was greater in maize, giving higher yields for

Table 1. Final above-ground biomass (Biomass), grain yield (Yield) and harvest index (HI) for the different irrigation treatments in maize and sorghum in 1995. Data are expressed as oven-dried biomass.

Treat	Biomass( g m-2)		Yield ( g m-2)		HI
	Maize	Sorghum	Maize	Sorghum	Maize	Sorghum
T-1	2140.1	1837.7	1082.2	853.8	0.51	0.49
T-2	1740.6	1638.3	878.8	741.8	0.50	0.47
T-3	1099.8	1299.8	480.1	629.5	0.43	0.46
T-4	700.3	1072.7	195.1	488.5	0.28	0.46
T-5	485.1	728.4	55.8	265.1	0.12	0.37
T-6	356.6	522.3	9.5	64.3	0.03	0.13

Fig. 2. Harvest index vs seasonal irrigation for maize (○) and sorghum (●).

sorghum over a wide range of water irrigation applied (Fig. 3). In this study, the cross-over point below which sorghum outyielded maize was 460 mm of net water application. Muchow (1989) repored similar findings working with maize and sorghum. The superior yield of maize under favourable water supply is due to its longer growth duration and higher radiation use efficiency than sorghum (Muchow 1989).

Apart from applied water, soil characteristics affect crop water use, growth and yield (Tolk et al. 1997). Therefore, soil type should be taken into account when studying deficit irrigation strategies. Stockle and James (1989) found that large soil water holding capacity, high soil water content at sowing and deep root systems were important factors for successful implementation of deficit irrigation.

Fig. 3. Relationship between grain yield (Y) and seasonal irrigation (I) applied for maize (△) and sorghum (▲). Regression equations for maize, Y=-358.2 + 2.5 I R²= 0.91 and sorghum, Y= -40.2 + 1.8 I R²= 0.95.

The irrigation water use efficiency (IWUE), expressed as the yield-irrigation water (Y-I) relationship, is an economically important concept under water limiting conditions. Maize and sorghum differed in the patterns of IWUE response to deficit irrigation. IWUE in maize linearly decreased with decreasing amount of water applied in irrigation (R² = 0.93). In contrast, IWUE in sorghum remained stable for decreasing amounts of water applied for a wide range of deficit irrigation treatments (T-1 to T-5), and only significantly decreased at T-6 (Fig. 4). For similar yield and irrigation level, the differences in IWUE between crops can be explained by the greater ability of sorghum to extract water to lower soil water contents and to greater depths in the profile.

Fig. 4. Irrigation water use efficiency (IWUE), expressed as the ratio of grain yield to seasonal irrigation applied, for the different irrigation treatments in maize (full bars) and sorghum (empty bars).

Conclusions

Maize and sorghum differed in their responses to deficit irrigation. Maize outyielded sorghum for high levels of water applied. However, irrigation deficit reduced biomass, yield and HI more in maize than in sorghum, giving higher yields for sorghum under moderate or severe water deficit treatments. The sorghum performance under deficit irrigation conditions was associated with a great ability to extract water from deep soil layers, shorter growth duration and leaf characteristics more favourable to drought. Sorghum was more efficient in the use of irrigation water. The choice of maize over sorghum will not only take account of relative yields but also prices and costs. Assuming equal net returns per unit of yield produced for both crops, the results of this study suggest that maize is preferable to sorghum if water supply is not limiting, but sorghum could be an alternative to maize under irrigation deficit conditions.

Acknowledgments

We thank Miguel Izquierdo and Jesús Gaudó for their help in the field work. The research was supported by Comisión Interministerial de Ciencia y Tecnología and Instituto Nacional de Investigación Agroalimentaria.

References

Farré I (1998). Maize (Zea mays L.) and sorghum (Sorghum bicolour L. Moench) response to deficit irrigation. Agronomy and modelling. PhD University of Lleida (Spain), 150p.

Garrity DP, Watts DG, Sullivan CY and Giley JR (1983). Moisture deficits and grain sorghum performance: Drought stress conditioning. Agronomy Journal 75, 997-1004.

Girma FS, and Krieg DR (1992). Osmotic Adjustment in Sorghum. I. Mechanisms of diurnal osmotic potential changes. Plant Physiology 99, 577-582.

Hanks RJ, Keller J, Rasmussen VP and Wilson GD (1976). Line Source Sprinkler for Continuous Variable Irrigation-Crop Production Studies. Soil Science Society America Journal 40, 426-429.

Krieg DR, and Lascano RJ (1990). Sorghum. In: Irrigation of Agricultural Crops. BA Stewart and DR Nielsen (Eds.). American Society of Agronomy, Madison, Wisconsin, USA, 719-740.

Mantovani EC, Villalobos FJ, Orgaz F and Fereres E (1995). Modelling the effects of sprinkler irrigation uniformity on crop yield. Agricultural Water Management 27, 243-257.

Muchow RC (1989). Comparative productivity of maize, Sorghum and Pearl millet in a semi-arid tropical environment. II. Effect of water deficits. Field Crops Research 20, 207-219.

Rhoads FM and Bennet JM (1990). Corn. In: Irrigation of Agricultural Crops. BA Stewart and DR Nielsen (Eds.). American Society of Agronomy, Madison, USA, 569-597.

Singh BR and Singh DP (1995). Agronomic and Physiological Responses of Sorghum, Maize and Pearl Millet to Irrigation. Field Crops Research 42, 57-67.

Stockle CO and James LG (1989). Analysis of deficit irrigation strategies for corn using crop growth simulation. Irrigation Science 10, 85-98.

Tolk JA, Howell TA, Steiner JL and Evett SR (1997). Grain sorghum growth, water use and yield in contrasting soils. Agricultural Water Management 35, 29-42.