Data – The Farmer’s New Edge
Farming is a centuries-old industry that continues to play a pivotal and understated role in the landscape of today’s society. Modern agriculture companies are turning to big data to innovate the farming process and make it sustainable for the future.
The United Nations estimates that the world population will reach 9.7 billion in 2050, an increase of over 2 billion from the current population of 7.3 billion [1]. Farming practices are swiftly developing to feed this growing population. However, there are growing concerns over some of these practices such as wide-spread herbicides usage [2]. Companies are introducing innovative technologies in the production of food commodities while at the same time improving sustainable farming practices.
Deere & Company (known to most by its brand name, John Deere) is looking to be a first mover in applying machine learning to agricultural practices. In September 2017, Deere & Company purchased Blue River Technology, a California-based company that applies machine learning to optimize crop spraying technology and equipment [3]. A growing concern in the agriculture industry is the continued widespread use of herbicides. Herbicides are used broadly and widely on farmland as farmers are not able to specifically target weeds and must apply herbicides to the entire crop. This promotes and requires widespread herbicide use which increases the cost of crop inputs for farmers, creates herbicide resistance in crops, and contributes to the growing health concerns involving toxicity of herbicides [4]. Through Blue River’s See & Spray platform, machine learning is used to identify and spray weeds, not crops. See & Spray uses visual cues to distinguish between weeds and crops, and, if it identifies a weed, applies an appropriate amount of herbicide specifically to the weed, avoiding surrounding crops. See & Spray also gathers data as it visually assesses fields to continually improve the overall dataset available for assessment. Blue River claims that See & Spray can reduce herbicide use by up to 90% [5],[6]. Deere & Company intends to integrate this technology into its farming equipment in the future, helping drive a new wave of “smart” farming equipment that will enable farmers to produce crops more efficiently [7].
See & Spray is currently not widely available in the market. In the short-term, Blue River will focus on refining the product by partnering with local farmers in field tests. Blue River recently began introducing and demonstrating the technology to the broader agriculture community. It will also work on refining the prototype to make it larger (able to spray more rows at once) and increasing the speed of application [8]. Over the next two years, the company will also need to improve the product by continuing to refine the algorithms used to distinguish between weeds and crops. This will be done through continued collection of data during field tests [9]. Over the next decade, Blue River will have to focus on expanding the technology to be used on a broader range of crops. Currently, See & Spray is being tested on the cotton belt, however, farmers will demand the technology be available for a wide variety of crops, especially corn and soybeans [10]. As a result of this expansion, Blue River will have to continue to manage an ever-increasing database of weed and crop information.
Deere & Company can also consider other actions to improve and reduce the use of herbicides. It has historically been known for manufacturing farm equipment but has more recently expanded into the broader agricultural solutions realm [11]. The company should consider partnering with agriculture chemical manufacturers in the development of herbicides that can be used specifically with the See & Spray technology. The goal would be that new formulations could be less toxic as they would focus only on the genetics of the weed and not consider the genetics of the entire crop. Blue River can also consider incorporating satellite and drone imagery to build its databases. One of the challenges with See & Spray is that it must go into the field to collect data. However, if satellite images are used, Blue River can continually collect data over extended periods of time, which will improve the accuracy and availability of data used to develop the algorithms that distinguish between weeds and crops.
While this technology is clearly a positive move, I question whether there will be rapid adoption once it is released commercially. Many farms are struggling financially, and new equipment is expensive. I question how Blue River will ensure the technology can be easily and affordably introduced on a large scale to the farming industry. I also wonder if this technology will be negatively received by herbicide manufacturers, as it clearly results in less herbicide volume used which in turn results in lower profits for these manufacturers. Herbicide manufacturers spend billions of dollars developing herbicides and no doubt want a reasonable return on their investment [12]. How can Deere & Company work with herbicide manufactures to ensure that these companies will be accepting of the technology? (789 words)
ENDNOTES
[1] United Nations, “World population projected to reach 9.7 billion by 2050,” http://www.un.org/en/development/desa/news/population/2015-report.html, accessed November 2018.
[2] Charles M. Benbrook, “Trends in glyphosate herbicide use in the United States and globally,” Environmental Sciences Europe, February 2, 2016, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5044953/, accessed November 2018.
[3] “Deere to Advance Machine Learning Capabilities in Acquisition of Blue River Technology,” press release, September 6, 2017, on John Deere website, https://www.deere.com/en/our-company/news-and-announcements/news-releases/2017/corporate/2017sep06-blue-river-technology/, accessed November 8, 2018.
[4] Charles M. Benbrook, “Trends in glyphosate herbicide use in the United States and globally.”
[5] Source: Chostner, B., “See & Spray: The Next Generation of Weed Control,” Resource, July/August 2017, ABI/INFORM via ProQuest, accessed November 2018.
[6] Blue River Technology, “See & Spray,” http://smartmachines.bluerivertechnology.com/, accessed November 2018.
[7] Source: “Investing in Ag Technology,” AgriMarketing, October 2018, ABI/INFORM via ProQuest, accessed November 2018.
[8] Margy Eckelcamp, “Blue River Expands See & Spray Testing Before Commercial Launch,” March 5, 2018, Farm Journals’ Ag Pro, https://www.agprofessional.com/article/blue-river-expands-see-spray-testing-commercial-launch, accessed November 2018.
[9] Matthew J. Grassi, “How Blue River’s See & Spray Technology Could Change Agriculture Forever,” August 5, 2017, Precision Ag, https://www.precisionag.com/service-providers/how-blue-rivers-see-spray-technology-could-change-agriculture-forever/, accessed November 2018.
[10] Margy Eckelcamp, “Blue River Expands See & Spray Testing Before Commercial Launch”.
[11] “Investing in Ag Technology”.
[12] Carey Gillam, “Monsanto to invest more than $1 bln in dicamba herbicide production,” June 24, 2015, Reuters, https://www.reuters.com/article/monsanto-dicamba-idUSL1N0ZA1XN20150624, accessed November 2018.
Good questions re: how pesticide manufacturers will react to this new innovation. I actually see them as potential acquirers of this software, where they can pitch it as a bundled sale with pesticide to spray just the correct amount. Re: farms struggling financially, one could argue that if this product helps them save on excessive and wasteful use of pesticides, then it may be worth the upfront cost to purchase. It likely all depends on the price point of the product and what the value of the marginal benefit actually is.