Water use efficiency of field beans (Vicia faba L.)

Abstract

Two main objectives of this research were to measure the seasonal water requirements of Vicia faba L. (field bean) grown under a range of irrigation treatments in Canterbury, New Zealand and to study its water use efficiency (WUE).

After reviewing the basic physiological and physical processes which determine total dry matter production, respiration and transpiration, a model of transpiration efficiency based on shoot dry matter (TEb) model was proposed. This model might be useful to identify the important factors limiting transpiration efficiency compared to the simpler model of Tanner and Sinclair (1983). However, results from the present study were tested using the simpler model.

The seasonal water requirements of Vicia faba L. cv. Maris Bead were measured for spring sown crops during the 1981/82, 1982/83 and 1983/84 growing seasons and an autumn sown crop in 1982/83. The crops were grown on a Templeton silt-loam soil (Dystric Ustochrept) in Canterbury, New Zealand. Different amounts and timings of irrigation were applied during the vegetative, flowering and pod-filling phases of growth, with the unirrigated crops treated as controls. For the 1981/82 and 1982/83 crops during each growth phase trickle irrigation was applied weekly in amounts equal to the difference between the estimated potential evapotranspiration, adjusted for crop cover and soil evaporation, and the rainfall of the previous week. Sprinkler irrigation was applied to the irrigated spring crop of 1983/84 whenever the soil water deficit approached 50 mm. This trial was on a larger block of land than those used for the earlier crops. Crop water use was determined using the soil water balance method accounting for rainfall and irrigation.

Results obtained in the present study showed that water use of the crop under Canterbury conditions ranged from 260 to 570 mm for the spring sown crops and 540 to 790 mm for the autumn sown crop. The seasonal water use was found to vary depending on the length of the growing season from sowing to maturity, the prevailing weather conditions, and the amount of irrigation applied. The average evapotranspiration (ET) rate ranged from 3 to 5 mm d⁻¹ and attained a maximum of 9.5 mm d⁻¹ at the peak of summer during the 1982/83 growing season (autumn crop).

Potential ET estimated by the Penman's model after correction for crop cover agreed with actual ET values especially for the wetter season (1983/84). The model however underestimated evaporative demand during the dry seasons (1981/82 and 1982/83) for fully irrigated crops. The larger blocks used for 1983/84 crop compared to those used in earlier growing seasons probably also account for the observed difference between these results.

Soil evaporation (E) as a proportion of ET for the different irrigation treatments was found to range from 25 to 50% for the spring sown crops and 16 to 27% for the autumn sown crop. However, E between treatments in absolute terms did not differ much probably due to the compensatory effects of greater crop cover at the earlier stages of growth and the frequency of soil surface rewetting for the heavily irrigated crops.

The WUE of the crop was defined in terms of evapotranspiration (ETE) and transpiration (TE) based on seed and shoot dry matter yield. It was found that the WUE estimates varied according to the method of determination. Both shoot dry matter and seed yields appeared linearly related to total ET and transpiration (T) over the range of water use estimates observed. The slopes of this relationship were used to estimate ETE and TE which were found to vary from season to season. However, ETE and TE determined by dividing harvested yields by water use gave different results, especially with respect to T. The concept of WUE presupposes a simple linear relationship between yield and water use. This assumption was found not to be strictly valid for the field bean crops in the present study. Furthermore, the proposed TEb model developed from the review of the literature indicated many reasons why growth might not be simply proportional to transpiration.

For the spring crops, ETE based on seed and shoot dry matter yields averaged 6 kg ha⁻¹ mm⁻¹ and 16 kg ha⁻¹ mm⁻¹ respectively. The corresponding TE averaged 11 kg ha⁻¹ mm⁻¹ and 28 kg ha⁻¹ mm⁻¹. The ETE and TE for the autumn crop were generally lower. These figures were determined by dividing harvested yields by water use.

The present results showed clearly that daytime vapour pressure deficit (v.p.d.) affected TEb considerably. However, after correction for variations in v.p.d. (i.e. k value), transpiration efficiency still varied between and within seasons. The k value of the crop differed significantly (P< 0.001) between seasons, the spring sown and autumn sown crops showing a k value of 0.00191 and 0.0013J kPa respectively.