Printer Friendly

A conceptual model for assessment of climate extremes that affect corn yields.

ABSTRACT. The USA Corn Belt was examined to assess the impacts of observed climate change on corn production for the period 1960-2012. Given a modified definition of a Corn Belt State, 13 states were included in the study. Temperature and precipitation trends during the growing season (April-September) showed the following: a) slight warming of 0.73[degrees]F, b) increase in growing season of 9 days, and c) mean precipitation increase of 5.51 cm (2.17 in), which along with technological advancements, support the observed increase of 1.7 bushels per acre per year for the period. A conceptual model assessed the impacts of extreme weather and climate on corn yields in bushels per acre. This model is represented by an Upper Bound (based on technological advancements), and a Lower Bound that is defined as the difference between the mean production and the Upper Bound. All values that fall below the Lower Bound were defined as extremely poor yields that can be attributed to extreme weather and climate events. The model was applied to the entire Corn Belt Region for the period 1960-2012. The years 1983, 1988, 1993, and 2012 were identified as extreme events (which are well-recognized in the agroeconomic community). The benchmark model framework can be extended through the 21st century to monitor the number of extreme events and the magnitude of their departure from the Lower Bound, and it is presented as an instrument for assessing climate change impacts on corn yields.

Keywords: Corn Belt, extreme weather/climate, yields


The "Corn Belt" of the United States of America (Fig. 1) is a region of the world known for food production (see Kucharik & Ramankutty 2005). The soil constituents, combined with a favorable climate, have made this area a highly productive food source region for humankind for over five decades. Increasing corn yields from this area are well-recognized, which can be largely attributed to technological advancements. Genetic improvements and enhanced mechanization in the 1930s started the climb in corn yields that achieved record maxima in 1960 and again in 1982 (Thompson 1986). An important question, that continues to gain attention, focuses on the potential role of climate change on corn yields (including current climate trends as well as the prediction of continued global warming through the 21st century). Kaufmann & Snell (1997) estimate that 19% of the variability in corn yield is due to climate variables, and approximately 74% is explained by technology and related factors (such as fertilizers, pesticides, seed varieties, planting methods, labor, and capital). Lobell & Asner (2003) used county level USDA yield information from 1982 to 1998 to study the impacts of climate change on the overall trends in crop yields. They concluded that previous estimates of increased corn yields attributed to technological advances were overestimated by approximately 20% due to climate-driven increases in yield.

It is further noted that the complexity in explaining the various causes of rising corn yields is somewhat difficult. Sacks & Kucharik (2011) have provided a nice overview of this complexity and the contributions of many factors, such as earlier planting dates due to an increase in the number of Growing Degree Days (GDD) necessary for corn maturation. Additional intricacies include the factors that affect leaf and plant development, as well as the grainfill period. Realizing these and other complexities, this analysis presents a somewhat simple approach to separate direct climate effects (i.e., trends in temperature, length of the growing season, and precipitation in the Corn Belt) from all other factors. Specifically, a conceptual model is developed and implemented with observational data to estimate the roles played by technology advancements, mean weather and climate conditions, and extreme weather events, on a climate time scale.


The Corn Belt Region for this study includes any state that exceeded an average annual yield of 150 million bushels for the decade 2000-2009. This resulted in the inclusion of 13 states, including North Dakota, which has not frequently been included in past studies (e.g., Kucharik 2006). Figure 1 shows the vegetative cover for the Corn Belt (identified by the dashed line), as well as the location of 38 stations selected for compiling the climatic data set. These stations have a mean minimum spacing of 145 km, with an average departure from this mean of 20 km. Efforts were made to avoid the clustering of data points, as well as to maintain equal spacing. Annual corn yield data were taken from the USDA's National Agricultural Statistics Service to assess trends through time (NASS 2014). Figure 2 is presented to show the total annual production in the 13-state Corn Belt Region for the period 1960-2012. Recently released 2013 corn data reveal a new record productivity, which is depicted in Fig. 2 as an open circle. Annual productivity has increased from around 3 X [10.sup.9] bushels in the 1960s to a current production of around 11 X [10.sup.9] bushels. Realizing the intricacies of farming practices, as well as acres planted for different crops and other considerations, it was decided to examine trend lines and weather/climate effects based on bushels per acre of corn production. Accordingly, Fig. 3 is presented along with a least squares linear fit that shows an r value of 0.934, with an average annual increase of 1.7 bushels/acre. Again, the recent 2013 yield data are included as an open circle. Noteworthy is a record departure below the trend line of 37 bushels per acre for the 2012 growing season, eclipsing previous record departure years of 1988 (26 bushels/acre), 1993 (23 bushels/acre), and 1983 (23 bushels/acre). However, the drought year of 1988 remains the greatest percent departure from the trend.

Spatial and Temporal Variability.--To assess trends and variability in temperature and precipitation throughout the Corn Belt, data from the 38 stations for the six months (April-September) most essential to corn production were collected from the National Climatic Data Center (NCDC) and analyzed. Further, changes in the length of the growing season, defined as the number of days between the last spring freeze (32[degrees]F) and first fall freeze, were computed. These analyses allowed for the opportunity to address the potential role of climate on corn production.

Temperature and length of growing season: The trend in mean air temperature, considering all Corn Belt stations for the period April-September since 1960 (Fig. 4) revealed considerable variability, ranging from 62.4[degrees]F in 1992 to 67.4[degrees]F in 2012 (with a 0.73'F increase in the mean). In addition to this variability, and even more interesting, is the variability in mean temperature among the 38 selected stations in the Corn Belt (which ranges from +3.7[degrees]F in northern Minnesota to -1.7[degrees]F in eastern Iowa). The complexity of such spatial variability has also been noted in the study by Kucharik & Serbin (2008), and further treatment of this topic is suggested. The change in the length of the Corn Belt growing season, as defined above, is approximately 9 days, with over 8 of these days gained as a result of the last spring freeze occurring earlier. This increase is comparable to the trend toward earlier start dates for planting of 10 to 12 days (Kucharik 2006; Sacks & Kucharik 2011). Thus, it is reasonable to conclude that some fraction of the increased growing season length can be attributed to a longer freeze-free growing season. When growing season lengths for the 38 stations throughout the Corn Belt Region are examined, large spatial variability is present as expected, ranging from 198 days in southern Missouri to 122 days in northern Minnesota.

Precipitation: Generally speaking, cooler and wetter summers in the Cora Belt favor increased yields, while hot and dry summers are detrimental to production, as seen in the 2012 disastrous crop season (Thompson 1986; Neild & Newman 1990). Although the Corn Belt shows a rather modest growing season mean temperature increase of 0.73[degrees]F from 1960 to 2012 (for the fitted line in Fig. 4), a greater percent increase is noted when precipitation trends are considered. Figure 5 shows an increase in mean precipitation for April-September of 5.51 cm (2.17 in) for the period, but again large variability and weak correlation in the trend line are evident.


In view of the above discussion on temperature and precipitation variability, it might seem impossible to document the effects of extreme weather and climate on corn yields. However, a unique approach was taken to diagnostically assess the complexities of technology and weather extremes on the climate time scale that affect corn yields. The conceptual model is presented in Fig. 6, which defines an Upper Bound for production that is based on the cumulative effects of technology (which is illustrated in the next section with the 19602012 data set). Next, the annual mean for the yield record was plotted, and the difference between the mean line and the Upper Bound defined the equivalent range of corn yield for the Lower Bound. Both the upper and lower bounds are equidistant from and parallel to the mean trend line, and the 53-year data set established these fixed boundaries. Data points in the record that fall below the Lower Bound were defined as low yields attributable to extreme weather and climate events. Increasing severity of extreme weather and climate events results in greater yield departures from the trend line, as noted by the lower dashed line tracing the most extreme low yield values, which fans out through time (Fig. 6). The frequency of seasonal yields that fall below the Lower Bound may also increase with more extreme weather and climate events, a hypothesis worth monitoring in future years. It is again emphasized that the model, and its application, is an instrument for diagnostic assessment of climate change impacts.

Application of the Model.--To show the usefulness of the conceptual model, it has been applied to the entire Corn Belt Region (Fig. 7). Four years of poor yield (1983, 1988, 1993, and 2012) fall below the Lower Bound and are thus classified as occurring under extreme weather and climate events that negatively impacted yields. Not surprising, and as well-known, the two worst years for corn productivity (2012 and 1988) were both characterized by extremely hot and dry conditions during the spring and summer months (Namias 1991; NCDC 2015). The other two years of poor corn productivity displayed other noted scenarios, namely extremely cool and wet conditions throughout the 1993 growing season (Kunkel et al. 1994) and cool, wet conditions early in the growing season followed by hot and dry conditions late in the 1983 growing season.

The Upper Bound, as defined in this study, illustrates the limitations of extraordinary technological advancements. Five years covering a large span of time (1972, 1979, 1994, 2004, and 2009) had yields near or on the Upper Bound, showing that increases in technology have not been so revolutionary as to reveal a consistent increase in the number of years abutting the Upper Bound, at least since 1960. The four years below the Lower Bound are successfully captured by the methodology of the conceptual model. Finally, the domain of extreme events, highlighted by the crippling 2012 season (Fig. 7), fans out below the Lower Bound, another feature illustrated in the model. As such, the diagnostic methodology in this model is prepared to document the cumulative effects of 21st century extreme weather and climate on corn yields in the USA Corn Belt Region.


Data for total corn yields, as well as yields of bushels per acre, for the USA Com Belt region were assessed for the period 1960-2012, along with relevant weather and climate data records. A definition for a "Corn Belt State" was introduced, which qualifies 13 states for analysis. Weather records from 38 selected stations across the Corn Belt were used, which have a mean station spacing of 145 km, with an average departure from the mean of 20 km. The observational data showed that the essential period during which corn is grown in the Corn Belt (April-September) has become wetter and slightly warmer through time, although considerable spatial and temporal variability was noted and can be expected (as seen in the cause of the disastrous 2012 crop season). More research into spatial and temporal variability is warranted, especially with the anticipated continuation of global warming and the associated extreme weather events. The precipitation increase for April-September found in this study is also consistent with the expected increase in conditions that support summertime convective precipitation (namely humidity and CAPE (convective available potential energy), as seen in the study by Trapp et al. 2007).

Based upon the conceptual model, 53 years of data were assessed for the occurrence and effect of extreme weather and climate events on corn yields. This model worked as envisioned, even in the presence of continued increase in annual yields (1.7 bushels per acre per year) contributed in part by technological advancements. Application of the model identified four events that represent extreme weather detrimental to corn yields. These are (from the most extreme to the least extreme in terms of departure from the trend line) 2012, 1988, 1993, and 1983 for the entire Corn Belt Region.

Further, it is important to realize that the model framework presented in this study can be 2012, 1988, 1993, and 1983 for the entire Corn Belt Region, extended through the 21st century (assessing any global warming effect) to monitor and quantify the number of extreme events. Specifically, if there are more extreme weather and climate events (e.g., heat, drought, excessive rainfall), the number of events that fall below the Lower Bound for corn yields may increase, and there also exists the opportunity for more record departures from the mean trend line (as seen in 2012). Thus, this model provides a diagnostic assessment of corn yields, categorizing them based on both technological advancements and climate change.


Kaufmann, R.K. & S. Snell. 1997. A biophysical model of corn yield: integrating climatic and social determinants. American Journal of Agricultural Economics 79:178-190.

Kucharik, C.J. 2006. A multidecadal trend of earlier corn planting in the Central USA. Agronomy Journal 98:1544-1550.

Kucharik, C.J. & N. Ramankutty. 2005. Trends and variability in U.S. corn yields over the twentieth century. Earth Interactions 9:1-29.

Kucharik, C.J. & S.P. Serbin. 2008. Impacts of recent climate change on Wisconsin corn and soybean yield trends. Environmental Research Letters 3:1-10.

Kunkel, K.E., S.A. Changnon & J.R. Angel. 1994. Climatic aspects of the 1993 Upper Mississippi River Basin flood. Bulletin of the American Meteorological Society 75:811-822.

Lobell, D.B. & G.P. Asner. 2003. Climate and management contributions to recent trends in U.S. agricultural yields. Science 299: 1032.

Namias, J. 1991, Spring and summer 1988 drought over the contiguous United States-causes and prediction. Journal of Climate 4:54-65.

NASS (USDA National Agricultural Statistics Service). 2014. Quick Stats Tools. At: http://nass. (Accessed 1 September 2013)

NCDC (National Climatic Data Center, NOAS). 2014. Land-Based Station Data. At: http://www. (Accessed 12 March 2013)

Neild, R.E. & J.E. Newman. 1990. Growing season characteristics and requirements in the Corn Belt. National Corn Handbook (NHC-40). At: https:// (Accessed 15 April 2014)

Sacks, W.J. & C.J. Kucharik. 2011. Crop management and phenology trends in the U.S. Corn Belt: impacts on yields, evapotranspiration and energy balance. Agriculture and Forest Meteorology 151: 882-894.

Thompson, L.M. 1986. Climate change, weather variability, and corn production. Agronomy Journal 78:649-653.

Trapp, R.J., N.S. Diffenbaugh, H.E. Brooks, M.E. Baldwin, E.D. Robinson & J.S. Pal. 2007. Changes in severe thunderstorm environment frequency during the 21st century caused by anthropogenically enhanced global radiative forcing. Proceedings of the National Academy of Science 104:19, 719-19, 723.

Manuscript received 14 August 2014, revised 22 May 2015.

Ernest M. Agee (1) and Samuel Childs: Department of Earth, Atmospheric & Planetary Sciences, Purdue University, West Lafayette, IN 47906 USA

(1) Corresponding author: Ernest M. Agee, 765-494-3282 (phone),
COPYRIGHT 2014 Indiana Academy of Science
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Agee, Ernest M.; Childs, Samuel
Publication:Proceedings of the Indiana Academy of Science
Date:Aug 20, 2014
Previous Article:Relationships between multi-scale environmental and land-use factors and summer demographics of the Northern Clearwater Crayfish, Orconectes...
Next Article:A comparison of the efficiency of mobile and stationary acoustic bat surveys.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters