Soybean yield and biomass responses to cumulative transpiration: Questioning widely held beliefs

Yield under drought for several crops has been established as a linear function of the cumulative water transpired during the growing season. For well-watered crops, however, there are no published data on how the duration of the cropping cycle and plant population affect the relationship between yield and transpiration. We evaluated the relationship between yield and estimated cumulative transpiration (T) or evapotranspiration (ET) for well-watered soybean (Glycine max [L.] Merr.) over a wide range of maturity groups (MG, 00–VI) and population densities (10–100 plants m⁻²) for 3 years. Daily T was estimated by determining the potential ET for a given day and multiplying this by the fraction of radiation intercepted by the crop, and a crop coefficient. Soil evaporation estimates were also made using an energy-balance approach after first subtracting the amount of radiation intercepted by the canopy. Daily values of T and ET were summed from emergence to R6. For all MG, cumulative T increased linearly with increasing population density (1.30 mm plant⁻¹ m⁻²), but predicted T at low populations (y intercept) more than quadrupled with increasing maturity, from 121 mm (MG 00) to 584 mm (MG VI). In contrast to the linear increase of yield to cumulative T for crops under drought stress, yield response to cumulative T for fully irrigated soybean differing in maturity was described well by an exponential model, predicting that 90% of the asymptotic yield would be obtained at 444 mm of T. Accounting for differences in harvest index and vapor-pressure deficit during the season among cultivars of differing maturity did not resolve the non-linear response of yield or biomass to cumulative T. These data indicate that for water-replete conditions, decreased T associated with short-season soybean need not result in decreased yield relative to full-season cultivars.