Evapotranspiration, seed yield and water use efficiency of drip irrigated sunflower under full and deficit irrigation conditions

Deficit irrigation occurrence while maintaining acceptable yield represents a useful trait for sunflower production wherever irrigation water is limited. A 2-year experiment (2003–2004) was conducted at Tal Amara Research Station in the Bekaa Valley of Lebanon to investigate sunflower response to deficit irrigation. In the plots, irrigation was held at early flowering (stage F1), at mid flowering (stage F3.2) and at early seed formation (stage M0) until physiological maturity. Deficit-irrigated treatments were referred to as WS1, WS2 and WS3, respectively, and were compared to a well-irrigated control (C). Reference evapotranspiration (ET_rye-grass) and crop evapotranspiration (ET_crop) were measured each in a set of two drainage lysimeters of 2 m × 2 m × 1 m size cultivated with rye grass (Lolium perenne) and sunflower (Helianthus annuus L., cv. Arena). Crop coefficients (K_c) in the different crop growth stages were derived as the ratio (ET_crop/ET_rye-grass).

Lysimeter measured crop evapotranspiration (ET_crop) totaled 765 mm in 2003 and 882 mm in 2004 for total irrigation periods of 139 and 131 days, respectively. Daily ET_crop achieved a peak value of 13.0 mm day⁻¹ at flowering time (stage F3.2; 80–90 days after sowing) when LAI was >6.0 m² m⁻². Then ET_crop declined to 6.0 mm day⁻¹ during seed maturity phase. Average K_c values varied from 0.3 at crop establishment (sowing to four-leaf stage), to 0.9 at late crop development (four-leaf stage to terminal bud), to >1.0 at flowering stage (terminal bud to inflorescence visible), then to values <1.0 at seed maturity phase (head pale to physiological maturity). Measured K_c values were close to those reported by the FAO.

Average across years, seed yield at dry basis on the well-irrigated treatment was 5.36 t ha⁻¹. Deficit irrigation at early (WS1) and mid (WS2) flowering stages reduced seed yield by 25% and 14% (P < 0.05), respectively, in comparison with the control. However, deficit irrigation at early seed formation was found to increase slightly seed yield in WS3 treatment (5.50 t ha⁻¹). We concluded that deficit irrigation at early seed formation (stage M0) increased the fraction of assimilate allocation to the head, compensating thus the lower number of seeds per m² through increased seed weight. In this experiment, while deficit irrigation did not result in any remarkable increase in harvest index (HI), water use efficiency (WUE) was found to vary significantly (P < 0.05) among treatments, where the highest (0.83 kg m⁻³) and the lowest (0.71 kg m⁻³) values were obtained from WS3 and WS1 treatments, respectively. Finally, results indicate that irrigation limitation at early flowering (stage F1) and mid flowering (stage F3.2) should be avoided while it can be acceptable at seed formation (stage M0).