Member Login


Subscribe to Our Newsletter


  • {{ error }}

Will Fewer Soybean Leaves Result in Higher Yield?

On Tuesday evening, May 2, Jan de Regt, past chairman of the MSPB, sent me a link to an article entitled “Subtract Leaves, Add Yield” that appeared in the April issue of The Furrow magazine. Naturally, the title of this article piqued my interest.

This article presents a very brief summary of the contents in an article authored by Srinivasan, Kumar, and Long that is entitled “Decreasing, not increasing, leaf area will raise crop yields under global atmospheric change” that appeared in the Global Change Biology Journal [Vol. 23 (2017)]. Below, I provide a more detailed summary of the contents of this article, and how the reported results can be translated into developing a more efficient soybean plant.

The authors determined that a detailed canopy model coupled with experimental verification under current and future elevated CO2 has not been applied to test the hypothesis that modern crops produce too many leaves. Thus, they used biochemical and biophysical approaches to model and understand whether a reduction in leaf area would change total carbon [C] gain by the crop [NPP = net primary production] through forced decreased investment in leaf construction that would cause a yield increase under current [390 ppm] and elevated [550 ppm] CO2.

They assumed, after accounting for growth respiration, that all the C diverted from the savings in leaf production will be available for pod and seed production during the reproductive phase of soybean. The yield contribution from NPP and suppressed leaf construction was summed to obtain the total seed yield and allowed a calculation of optimum leaf area index [LAI] that resulted in that maximum seed yield.


They conducted field experiments at the Univ. of Illinois using a MG III variety grown under both current and elevated CO2 on a dryland site during 2010. The site had been in a continuous soybean/corn biennial rotation using standard Corn Belt management practices. Management practices during the year of the experiment were those commonly used for a soybean crop following a fertilized corn crop.

LAI reduction was achieved by manually removing ~7 trifoliate leaves (< 2 cm long) per plant on two dates (16-21 July and 4-6 August) during the growing season. They recorded the dry weight of the removed leaf tissue.

The experiment thus was comprised of a LAI reduction treatment and a CO2 treatment [current (390 ppm) and elevated (550 ppm) CO2]. Effect of the two treatments and their interaction on LAI and seed yield was determined and used to provide results from which conclusions could be made.


The model predicted an optimal LAI of 4.2 and 4.6 to maximize seed yield at current and elevated CO2 levels, respectively, which is 40% less than the observed LAI’s of 6.8 and 7.5 for the two CO2 levels, respectively. The lower predicted LAI’s were predicted to increase seed yield by 8% and 10%, respectively.

The difference in model-predicted NPP between observed LAI and model-predicted optimum LAI was small (~3%) under both CO2 treatments; thus, the additional units of LAI above the model-predicted optimum would not significantly contribute to net C gain.

Predicted gains in yield were primarily due to decreased investment in leaf production.

As LAI is lowered below the predicted optimum value, predicted losses in NPP are detrimental to seed production.

Plants with optimal LAI are predicted to use 38% and 39% less C to make leaf tissue than plants with the observed LAI in the field under current and elevated CO2 levels, respectively.

The predicted gross primary production [GPP = total photosynthetic CO2 uptake] at midday is 13% greater with the observed peak LAI of 6.8 than at the predicted optimum LAI of 4.2 for yield. However, the predicted GPP gains are much lower in the low light levels of the remainder of the day, and these small gains are offset by the 25% increase in respiratory losses associated with the greater leaf area. This results in negligible NPP gains [difference between GPP and respiratory losses] from the 40% greater observed LAI.

The proportion of shaded leaves increases with increasing LAI; thus, at solar noon 53% of the canopy is shaded at the predicted optimum LAI, whereas 68% of the canopy is shaded at the observed peak LAI. When integrated over the entire day, an average 84% of leaves in the predicted LAI canopy are shaded, whereas an average 90% of leaves in the observed peak LAI canopy are shaded.

Light use efficiency [LUE = ratio of canopy NPP to PAR (photosynthetically active radiation) absorbed by the canopy] in the model-estimated optimum canopy is 9% and 7% greater than in the observed peak LAI canopy for the current and elevated CO2 treatments, respectively.

Reducing LAI decreases crop water demand by 11% in both the current and elevated CO2 treatments because plants with fewer leaves transpire less water without significantly changing NPP. Thus, predicted canopy WUE [water use efficiency] is 9% greater in both the current and elevated CO2 treatments.

Leaf removal in the study significantly reduced LAI throughout seedfill and resulted in an 8% seed yield increase in both CO2 treatments.


Both the model and results from the field experiment indicate that modern soybean varieties such as the one used in this study produce more leaves than necessary, and this results in reduced seed yield in both current and elevated CO2 levels.

More canopy leaves are shaded at the higher LAI’s and are thus less efficient at photosynthesis.

A canopy with fewer leaves reduces its investment in the construction of leaf tissue, and this could result in greater seed production since the reduction in leaf area is not detrimental to NPP.

The reduction in crop water demand that results from having a lower canopy LAI indicates that the also-predicted improved yield with lower LAI should increase WUE. It is also possible that a lower transpiration demand by the optimum-LAI canopy may result in fewer roots, which could result in less investment in roots by the plant. Conversely, maintaining the same root volume with a reduced crop water demand could contribute to drought tolerance.

In reality, the over-production of leaves by soybean contributes to shading of weed competitors and provides protection against leaf area loss caused by insects, wildlife, and disease pathogens that is common in the Midsouth. Thus, the reality of developing soybean plants that have a lower LAI to improve LUE, WUE, and yield must be coupled with improved, effective management strategies that prevent such leaf area loss. This is especially important since, as stated above, “As LAI is lowered below the predicted optimum value, predicted losses in NPP are detrimental to seed production”.

The development of soybean varieties with a lower LAI appears to be one avenue for realizing a next-level yield increase in soybean. This can only be achieved through breeding efforts that are directed toward this result.

The above study does not address how breeding to develop soybean varieties with smaller leaves could contribute to a lowered canopy LAI such as that achieved in the above study through leaf removal. It is possible that soybean genotypes that have the same number of nodes but smaller leaves than current genotypes can achieve the same results as those achieved/modeled through leaf removal in the above-cited study.

Composed by Larry G. Heatherly, May 2017,