Wednesday, September 28, 2011

2011 Corn and Soybean Yields

2011 seems to have been a slightly crazy year weather-wise.  For example, Climate Progress notes that there have already been more FEMA declared disasters in 2011 than in any prior year, with three months still to go.  Between tornadoes and floods in the midwest, droughts in Texas, and hurricanes and tropical storms on the East Coast, the weather has been in the news a lot.  I was curious to know whether this showed up in a big way in crop yield data (since effects on crop yields are presumably the primary way that climate change might seriously threaten civilization).

Of the three big field crops, the 2011 yield data are available from NASS for corn and soybeans, but not yet for wheat.  The raw yield data are above.  You can see that 2011 is a bit of a jog down but doesn't look like a marked departure from trend.  My assumption is that generally the trend is dominated by improvements in agricultural technology, while the year-to-year noise is mainly due to weather.

To get a clearer idea of the context I constructed a variable which consists of the percentage change of this year relative to the average of the prior ten years.  That looks like this:

Prior to around 1940 corn yields were pretty flat and so this index fluctuates around zero with 10-20% fluctuations being pretty common.  After that, yields have been improving so the average is not zero but positive - this year is typically better than the average of the past ten - but fluctuations continue to be large.  In particular, 2011 is not the worst downturn in yield in the post 1940 era in either crop - four or five other years are worse.  However, it certainly was a worse performance than average.

So at the moment there doesn't seem to be much evidence for the proposition that the weather is getting bad enough to present a serious threat to the US food supply.  Not to say it couldn't in future.


Hal said...

I'm astounded that the data is available for 2011. They're still finishing up the soybean harvest around here.

Stuart Staniford said...

Hal - yeah I'm surprised too - around here it seems like some farmers don't harvest the last of their feed corn until November. Presumably these things are based on surveys though and at some point they think they've gotten enough responses to have confidence in their answer.

Alexander Ac said...

What is the methodology? If there are floods and crop is 100 % destroyed over certain area, is it still calculated in this "per acre"? I.e. is it calculated as total yields per total area planted?

I am also interested how this trend will continue given the over-use of soils and over-use of ground water (still ignoring CC and PO)

A Farmer said...

Those aren't the final production numbers for 2011, they have to be USDA projections. We have combined a grant total of 14 acres of corn, which is averaging about 90 bushel for this field. Our harvest will be well below average this year, and USDA won't have the final production numbers until sometime next year. As for the potential for weather to present a serious threat to food supply, I think we have some real wiggle room. There is usually enough variation in weather to allow some area of the Corn Belt to have good crops. We could stop grinding corn for ethanol. We could cut back on beef consumption, etc.

kjmclark said...

From the 9/12 crop forecast:
"Corn production is forecast at 12.5 billion bushels, down 3 percent from the August forecast but up fractionally from 2010. If realized, this will be the third largest production total on record for the United States. Based on conditions as of September 1, yields are expected to average 148.1 bushels per acre, down 4.9 bushels from the August 1 forecast and down 4.7 bushels from 2010. If realized, this will be the lowest average yield in the United States since 2005.

Soybean production is forecast at 3.09 billion bushels, up 1 percent from August but down 7 percent from last year. Based on September 1 conditions, yields are expected to average 41.8 bushels per acre, up 0.4 bushel from last month but down 1.7 bushels from last year. Compared with last month, yield forecasts are higher in the Central Great Plains and along much of the Atlantic Coast. If realized, the forecasted yield in Nebraska will be a record high. Yield forecasts are below last month across the Southern Great Plains and portions of the Southeast as hot, dry conditions persisted during August. Area for harvest in the United States is forecast at 73.8 million acres, unchanged from August but down 4 percent from 2010."

Ie. these are all just forecasts. They really don't know yet. Last Sunday's crop progress report showed only 15% of the corn crop and 5% of the soy crop had been harvested. Way too early to tell much. And as noted before, this forecast yield is just forecast production/acres harvested. Crops that drowned in the east or fried in Texas won't be counted at all.

kjmclark said...

The kicker is that the winter wheat harvest is 100% completed, and spring wheat was 96% harvested as of last Sunday. They probably don't have the data because they're about to switch over from forecasts to real data. Seems like the actual data for corn and soy aren't out until January.