Yield Gaps, Potential Yield And Crop Production

DR. DENNIS B. EGLI

PRINCETON, KENTUCKY

Are there ‘yield gaps’ on your farm? Finding a yield gap suggests that your yields are less than they could be, so some combination of improved management practices will increase yield and reduce the yield gap. This seems like simple way to evaluate productivity. But, as usual, when dealing with a simple concept, the devil is in the details.

 I briefly discussed yield gaps in the October issue of the Corn and Soybean Newsletter. Now I want to expand that discussion to provide an in-depth treatment of the concept. 

Yield gaps represent the difference between your yield and potential yield. Potential yield was defined by a couple of Australian crop physiologists back in the 1990’s as “ the yield of a variety when grown in environments to which it is adapted; with nutrients and water not limiting and with pests, diseases, weeds, lodging and other stresses effectively controlled”. Or, to put it another way, it’s the yield when nothing – at least none of the factors we control - is limiting plant growth. 

The problem is - how do we estimate potential yield? Researchers have proposed a number of methods, but all have their weaknesses. Research plots managed to eliminate all stress, record yields, maximum farmer yields or yields from well-managed University experiments provide estimates of potential yield. Crop simulation models are a favorite tool of some scientists because they are not affected by diseases or other pests and water and nutrients can be ‘supplied’ in non-limiting amounts. Of course, their estimate is no better than the ability of the model to accurately mimic the growth of the crop.

The size of the yield gap depends, in part, on the technique used to estimate potential yield. A large yield gap, implying that there is plenty of opportunity to increase yield, is valid only if the estimate of potential yield is accurate. Responding to large yield gaps, created by unrealistically high estimates of potential yield, can result in wasting money on un-needed inputs in a futile attempt to close the yield gap.

 A number of years ago I worked with an ex-student of mine (Dr. Jerry Hatfield, recently retired as the Director of the USDA-ARS National Laboratory for Agriculture and the Environment in Ames, Iowa) to evaluate potential soybean yields and yield gaps in Kentucky. We used the highest county yields reported by the National Agricultural Statistics Service over a 40-year period (1972-2011) as an estimate of potential yield (Fig. 1). We applied this analysis to all Kentucky counties that harvested more than 10,000 acres of soybean (32 counties). Fitting a regression curve to the high yields made it possible to estimate potential yield and the yield gap for each of the 40 years. 

These estimates of potential yields represent the collective efforts of all the producers in each county in years with exceptionally favorable weather conditions. These estimates of potential yield are not as high as the classic definition because all farmers may not correctly apply the best available technology (best variety, adequately control weeds and diseases etc.). 

Potential yield (dotted line in Fig. 1) increased from 1972 through 2011 in every county, following the trend of county yields. This increase simply reflects the constant adoption of the latest high-yielding varieties and improved management practices by the producers in each county.

The average relative yield gap [((potential yield – average county yield)/potential yield) * 100] decreased as the average county yield increased (Fig. 2). The larger year-to-year variation in yield and yield gaps in the low-yield counties (Fig. 1) resulted in larger average relative yield gaps. Interestingly, there was a trend for counties with a high proportion of the soybean acres double cropped after wheat to have larger relative yield gaps than counties with little double cropping but the same yield (Fig. 2).  Apparently, stresses associated with late planting in the double-crop system contributed to a larger relative yield gap. 

The favorable weather conditions associated with the potential yield estimate did not increase the potential yield of the low-yield counties (e.g., McCrackin, Fig. 1B) to equal the high-yielding counties (e.g., Henderson, Fig. 1A). One might think that in years with the most favorable weather (probably above-average rainfall during the growing season) the yield of low-yield counties might equal high-yield counties; this did not happen, (see Fig. 1) potential yield in the low-yield counties was always less than the high-yield counties.

The year-to-year variation in the yield gap (potential yield -county yield, Fig. 1) was probably related to variation in rainfall with the largest yield gaps occurring in dry years. The larger yield gaps in the lower-yielding counties no doubt reflect the lower soil moisture storage capacity of the soils in those counties.

Yield gaps might increase over time if climate change significantly increases stress levels, but there was little evidence of this, at least not through 2011. Only 4 of 32 counties showed a significant increase in relative yield gap over time. 

The bottom line is that potential yield varied among counties and it’s reasonable to speculate that it also varied among soils within a county. The lower-yielding counties could not match the yield of the higher-yielding counties in the near-ideal weather conditions that occurred only 4 or 5 times in the 40-year period. It seems that even an all-out maximum effort to produce high yield (including irrigation) would not raise the yield of the lower-yielding counties to the levels in the highest-yielding counties.

Everyone wants to manage for high yield, but these results suggest that ‘high’ depends upon where you are. Location is important! Chasing bragging yields from a high-yield county is not the best way to maximize your bottom line if you farm in a lower-yielding county. Perhaps we should consider the advice of a South Dakota farmer – “Farm the best and leave the rest” (Quoted in ‘Prairies Vanish in the US Push for Green Energy’ by Chet Brokaw and Jack Gillum, PHYS.ORG, 2013).

Adapted from: Egli, D.B. and J.L. Hatfield. 2014. Yield gaps and yield relationships in central US soybean production systems. Agron. J. 106: 560-566.   ∆

DR. DENNIS B. EGLI: University of Kentucky