An Evaluation of Benbrook’s Pesticide Use Study and Evolving Super Weeds

The study, “Impacts of genetically engineered crops on pesticide use in the U.S. — the first sixteen years” by Charles M Benbrook of Washington State University was published September 28th in peer reviewed “Environmental Sciences Europe 2012″. Reuters covered the report in this article: “Pesticide use ramping up as GMO crop technology backfires”. “Yes on 37 For Your Right to Know” advocates followed by publishing a press release titled “Damning New Study: GMOs Cause Massive Overuse of Pesticides; Data Sheds Light on Why Pesticide Companies Lead Opposition to Prop 37″.

The purpose of this post is to sort through the study’s strengths and weaknesses.

For background, Roundup ready soybeans were introduced in 1996 as the first HR (herbicide-resistant) crop. In 2011, an estimated 94% of the soybean area planted, 72% of corn, and 96% of cotton were planted to HR varieties, while about 65% of corn and 75% of cotton in the U.S. were planted to Bt varieties. See charts (below).

USDA Graph Showing Adoption of Genetically Engineered Seeds from 1996 through 2011 for Soybeans, Cotton, and Corn

Biotech Share of U.S. Corn Acres in 2011

Benbrook’s study looks credible in the reporting of pesticide use, but he never took into account crop yields in his results. It reports that between 1996 and 2011, herbicide use has increased while insecticide use decreased for a net gain in pesticide use of 404 million pounds, or about seven percent. Pesticide use is reported per acre, but the amount used per bushel of yield is ignored by the study.

The conclusion of the study is this:

Contrary to often-repeated claims that today’s genetically-engineered crops have, and are reducing pesticide use, the spread of glyphosate-resistant weeds in herbicide-resistant weed management systems has brought about substantial increases in the number and volume of herbicides applied. If new genetically engineered forms of corn and soybeans tolerant of 2,4-D are approved, the volume of 2,4-D sprayed could drive herbicide usage upward by another approximate 50%. The magnitude of increases in herbicide use on herbicide-resistant hectares has dwarfed the reduction in insecticide use on Bt crops over the past 16 years, and will continue to do so for the foreseeable future.

I’d like to first comment on his results and conclusion (above).

1. Within the report, Benbrook appropriately cites the application of herbicides per acre. But telling us how many pounds gain in use of pesticides since the introduction of HR crops were introduced in 1996 falls short by not taking into consideration crop yields.

2. In Benbrook’s conclusion, he shows that GE crops have led to a larger use of pesticides when we were promised they would use less, especially since pesticide resistance has evolved. The claim that I am more familiar with is that glyphosate is a less toxic chemical than those herbicides used previously. After-all, HR seed technology, by definition, requires application of the herbicide glyphosate. Additionally, when I did average yield calculations (below) for corn and soybeans, results suggested that the time period of Benbrook’s 16-year study did use fewer pesticides when yield was taken into account.

3. The main point of Benbrook’s conclusion is that glyphosate-resistant weeds will require an increase in pesticide use. An obvious exception to this would be the increased hand-removal of weeds from fields which is already happening. Today’s growing weed and pest resistance to pesticides is causing a return to the use of the original and more toxic chemicals such as 2,4-D and the promotion of GE seeds with more stacked traits and higher concentrations of Bt. This is of great concern to everyone and he does a good job covering this emerging story. Weed-resistance has huge implications and it has been in the news repeatedly over the past several years, so in that sense this study presents nothing new. There are many unknowns surrounding future crop production and crop production methods as a result of pesticide resistance and his overview of today’s situation towards the end of the study is worth reading.

In his study, this paragraph tells us that pesticide use is increasing with each passing year of GE crop use:

Taking into account applications of all pesticides targeted by the traits embedded in the three major GE crops, pesticide use in the U.S. was reduced in each of the first six years of commercial use (1996–2001). But in 2002, herbicide use on HR soybeans increased 8.6 million kgs (19 million pounds), driven by a 0.2 kgs/ha (0.18 pounds/acre), increase in the glyphosate rate per crop year, a 21% increase. Overall in 2002, GE traits increased pesticide use by 6.9 million kgs (15.2 million pounds), or by about 5%. Incrementally greater annual increases in the kilograms/pounds of herbicides applied to HR hectares have continued nearly every year since, leading to progressively larger annual increases in overall pesticide use on GE hectares/acres compared to non-GE hectares/acres. The increase just in 2011 was 35.3 million kgs (77.9 million pounds), a quantity exceeding by a wide margin the cumulative, total 14 million kg (31 million pound) reduction from 1996 through 2002.

The above paragraph suggests that with increased HR crop acres, there is a corresponding increase in the use of glyphosate. That would certainly be expected, would it not? Most resistant weeds weren’t documented until 3-4 years after 2002, so I find his mixing of these figures and timeline confusing. It would have been valuable if the study had included a timeline of the rate of adoption for each GE crop for a baseline of comparison to pesticide use. (See weed reference list link)

Taking into Account Crop Yields
Using this USDA chart as my reference, I averaged the corn yield per bushel for the 16 years leading up to 1996, and compared that yield to the yield during the 16 years of Benbrook’s study, 1996 through 2011. From 1980 through 1996, I came up with an average yield of 110 bushels of corn per acre. From 1997 through 2011, the average yield was 143 bushels of corn per acre, a 30 percent increase. A few variables worth mentioning that are not taken into account are weather, an increase in irrigation, advancements in machinery, and the rate of adoption of HR varieties of corn by year. Yet, one can see that the yield increase is far greater than the seven percent of Benbrook’s reported increase in pesticide use during this study’s sixteen-year time period.

Doing the same exercise for soybeans and using a yield chart from Iowa State, from 1980 through 1996 the average U.S. soybean yield was 30 bushels per acre. From 1997 through 2011 the average yield was 40 bushels per acre. This is a 33 percent increase during the time period of Benbrook’s study, again, clearly well-above the seven percent increase in pesticide application that he cites in the results of his study.

Resistant Weeds
This chart from lists the 24 glyphosate-resistant weeds, when they were first seen, and which states they inhabit. Perhaps two of the worst problems right now are the glyphosate-resistant Palmer amaranth in cotton and Bt resistant western corn rootworm. Last year, 500,000 cotton acres in Georgia were weeded by hand for pigweed (Palmer amaranth). On a personal note, my own family had to hoe marestail out of soybean fields this year on our farm in Northeast Nebraska. Glyphosate resistant weeds are becoming ubiquitous across U.S. corn, soybean, and cotton cropland — spanning millions of acres.

Deere Boom Sprayers can span 100 feet or more

Overuse and unregulated management of crops using glyphosate and Bt genetically modified technology has greatly contributed to these emerging resistant weeds and pests. Mandated ethanol use has driven up corn prices and resulted in unrotated corn on corn acres, where rootworm resistance is seen. An aging farmer demographic along with depopulation of the rural areas are conditions which have embraced the use of genetically modified technology which has saved labor while also enabling operators to work off-farm for income. Custom boom sprayers spanning 100-feet or more have done the work of weed-removal. Aerial spraying of pesticides continues to be widely used. The no-till which is practiced by using Roundup saves fuel costs and labor.

When one puts all of these facts together it is worth wondering whether we are nearing the end of the easy years of crop production in the U.S., the end of industrial agriculture as we know it. Will there be a decrease in crop yields? Could this kill the biofuels programs? Will farms need to become smaller again so that farmers can manage weeds? Will teenagers (or migrant laborers) be employed by roguing corn and bean fields? So far, Monsanto has not been very informative about its outlook concerning this issue. In Ag decision-making, in the end, it always comes down to economic viability and policy. At least for the short term, it looks as if more expensive and more hazardous chemical inputs will continue to increase in use for corn, soybeans, and cotton because of pesticide resistance which is growing at an alarming rate.

11 thoughts on “An Evaluation of Benbrook’s Pesticide Use Study and Evolving Super Weeds

  1. K.M. Post author

    Thanks, Steve.

    I received the following in an email from a reader:

    I have read with great interest (this post) and as a person who has been involved with agriculture for more years than I care to remember I have never been impressed by what the Monsanto’s of the world claim.

    There have been increase in yields yes, but due to breeding and not thru gene manipulation.

    What is pertinent is the problems that are now rearing their ugly heads and as I am involved with countries like India, SE Asia, and Africa and have seen what GM has meant to poor farmers with some 200k suicides in India, what is now appearing could be the tip of the iceberg – we shall see.

    I am one of your readers who appreciate what you are doing in collating info about agriculture. It is about time the industry was honest – we are in great need of increase in crop yields and there are ways and means to do this with less fertilizer and water. Everything that GM suppliers do not want to see – as a last swipe – why does the Gates Foundation see it’s fit to invest in Monsanto??

    1. v

      the indian suicides have been linked to GM cotton, but is it valid? 2 points.

      First is that the suicides started before GM cotton was introduced. The suicides do happen soon after subsidies were removed/reduced.

      Second. If suicide was due to the GM cotton inducing economic hardship then why are african cotton farmers of the same time period who dont use GM cotton also suffering economic hardship? An explanation that encompasses both india and africa is the artificially low prices due to lack of subsidies to indian and african farmers while others like chinese and american farmers have subsidies

  2. ramy

    I wonder whether looking at pesticide application per yield gives us useful data or not. If we are concerned with pesticide runoff or impact on the environment, acreage (+toxicity of the chemical) is the critical variable. If we are interested in the amount of pesticide we are ingesting, yield is the right variable. I don’t think that is what he is asking/answering.

    1. K.M. Post author

      I think your points are good. But, think about it in reverse. If yields had declined by 30% and we had increased pesticide use, would that be worth including?

      1. Brett

        I don’t think we’d be talking about pesticides per bushel even if yield declined. It simply is not relevant to what we care about when we measure the pesticides–how much goes on the land, and how much leaves the land. Conservation is not measured by the pound, it is measured as it is applied to an acre.

        If we want to start measuring conservation by the pound, then we need to broaden the examination. Much of the yield increase in the Midwest is due to drainage. What are we going to use for an offset on that? Drainage carries many more pesticides directly to waterways. Is the measure pesticides per bushel applied to the field, or pounds of pesticides in the water supply per bushel? Most people do not care how many pounds go on the field–they care about how many pounds run off the field, and how many pounds are retained in their food. If you are going to change the approach to measurement, you have to also change the factors you measure–you can’t use measures designed for an acreage-based approach in a bushel-based approach.

        And I really don’t think big ag wants to go as far down this road as they think they do. If we want to measure conservation by the pound, let’s look at another stat: At 5.2T (Iowa’s “acceptable” erosion rate) and around 180 bpa yield, for every pound of corn we harvest, more than a pound of soil erodes. If you look at the whole production cycle of corn on a “pound per pound” analysis, it really highlights what a terribly resource-intensive crop it is.

  3. dc

    In looking at your yield for corn especially, I’m wondering if you could at least tease yield increased due to increased fertilizer use? As the corn price has risen, we find it much more profitable to apply higher amounts.

  4. SB

    I read with great interest your article. You mentionned the importance of the issues of economic viability and policy. Let me indicate a complementary paper among others that addresses these issues of pesticide use and evolving resistant weeds with GM herbicide-tolerant crops, at least in part.
    This paper deals with GM crops, herbicide use, glyphosate-tolerant crops, and weed resistance to glyphosate in the USA, but not with all GM crops, mainly with herbicide-tolerant soybean.
    Let me give the references of this paper that may be of interest to some readers.
    Bonny S. 2011. Herbicide-tolerant Transgenic Soybean over 15 Years of Cultivation: Pesticide Use, Weed Resistance, and Some Economic Issues. The Case of the USA.
    Sustainability, 3(9), 2011, pp. 1302-1322. DOI:10.3390/su3091302

    Freely available at:

    S BONNY, INRA (National Institute for Agricultural Research, France)

    1. K.M. Post author

      Thanks for your note. I will be doing a follow-up to this because there is some additional information that needs to be presented.

  5. Jeff Schulte

    While I see how analyzing pesticide use against yield is informative economically. That is: it is profitable to use GMO crops even with increased inputs. I don’t see how this is relevant environmentally. While there is no doubt a trade off with yield and inputs (including irrigation, fertilizer, and pesticide), it remains to be seen whether this is the most EFFICIENT use of resources. Over fertilization is common because it is cheap insurance against field loss of fertilizer. And even though yields continue to go upwards with increased fertilizer usage, they do level off, and the most efficient usage for fertilizer is significantly lower than is typically applied. Following the same line of reasoning,: many pesticides use oil as a chemical base and additional energy to synthesize. Same for other inputs including pumping water for irrigation and fertilizer production. The relevant questions environmentally is: what does the yield curve look like?
    At what point is pesticide most profitable? At what point is the yield per input maximized?


Leave a Reply

Your email address will not be published.

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>