Historical gains in soybean (Glycine max Merr.) seed yield are driven by linear increases in light interception, energy conversion, and partitioning efficiencies

Abstract

Soybean (Glycine max Merr.) is the world's most widely grown leguminous crop and an important source of protein and oil for food and feed. Soybean yields have increased substantially throughout the past century, with yield gains widely attributed to genetic advances and improved cultivars as well as advances in farming technology and practice. Yet, the physiological mechanisms underlying the historical improvements in soybean yield have not been studied rigorously. In this 2-year experiment, 24 soybean cultivars released between 1923 and 2007 were grown in field trials. Physiological improvements in the efficiencies by which soybean canopies intercepted light (εi), converted light energy into biomass (εc), and partitioned biomass into seed (εp) were examined. Seed yield increased by 26.5kg ha(-1) year(-1), and the increase in seed yield was driven by improvements in all three efficiencies. Although the time to canopy closure did not change in historical soybean cultivars, extended growing seasons and decreased lodging in more modern lines drove improvements in εi. Greater biomass production per unit of absorbed light resulted in improvements in εc. Over 84 years of breeding, soybean seed biomass increased at a rate greater than total aboveground biomass, resulting in an increase in εp. A better understanding of the physiological basis for yield gains will help to identify targets for soybean improvement in the future.

Keywords: Energy conversion efficiency; harvest index; light interception efficiency; partitioning efficiency; radiation use efficiency; yield potential..

Fig. 1.

Meteorological data for the 2012 and 2013 experimental growing seasons (planting date to 30 September): daily total solar radiation (A and B), daily maximum, mean, and minimum temperatures (C and D), and rainfall and irrigation events and accumulated precipitation across the growing season (E and F).

Fig. 2.

Seed yield, ε_i, ε_c, and ε_p with soybean cultivar year of release (YOR) for the 2012 and 2013 growing seasons: seed yield (A and B), seasonal interception efficiency (ε_i, C and D), conversion efficiency (ε_c, E and F), and partitioning efficiency expressed in energy content (ε_p, G and H). Lines represent significant least-squares regression. m, slope; r, Pearson correlation coefficient; *, **, and *** denote significance at P<0.05, P<0.01, and P<0.001, respectively.

Fig. 3.

Interception efficiency (ε_i) across the growing season in 2013 for each of the 24 soybean cultivars grouped by YOR. DOY, day of year.

Fig. 4.

Accumulated aboveground biomass versus cumulative PAR_i in 2012 (A) and 2013 (B). Lines represent least-squared regression between dry biomass versus cumulative PAR_i. The slope of each line (m) is ε_c. Each point represents the biomass and cumulative PAR_i for the five oldest cultivars and the five most recently released cultivars.

Fig. 5.

Determinants of partitioning efficiency (ε_p) versus YOR at growth stage R8 plotted against cultivar YOR in 2012 and 2013: seed biomass (A and B) and total biomass (C and D). Lines represent significant least-squares regression (*** P<0.001).

Fig. 6.

2012 correlation matrix of yield and Monteith efficiencies. ε_p is expressed in terms of biomass (g seed/g total aboveground biomass). Scatterplots and correlation coefficients are plotted in a matrix where lines represent significant least-squares regression. Bold indicates significant results.

Fig. 7.

2013 correlation matrix of yield and Monteith efficiencies. ε_p is expressed in terms of biomass (g seed/g total aboveground biomass). Scatterplots and correlation coefficients are plotted in a matrix where lines represent significant least-squares regression. Bold indicates significant results.