Researchers applied machine learning to satellite-derived datasets to analyze tillage practices and crop yields in the US Corn Belt between 2005 and 2017. The purpose of this study was the understand what benefits are derived from conservation tillage, the agricultural practice of minimum soil disruption to enhance crop production.
Tilling is done by farmers before a crop is planted. The intent of tilling is to break up compacted soil, remove weeds, and mix nutrients.
The practice can boost short-term crop production but soil is degraded over time.
A report from the Food and Agriculture Organization of the United Nations published in 2015 found that a third of the world’s food producing land in the past 40 years has been lost to degraded soil.
Conservation tilling, in contrast, boosts soil nutrient levels and prevents erosion, by leaving the previous year’s remnant vegetation in place.
Conservation tillage is the agricultural practice of leaving at least 30% of crops remnants on the soil to reduce soil erosion, increase water retention and drainage. On the other hand, for areas experiencing very early wet seasons, conventional tillage can help to dry water-logged soils which can delay the timing of planting crops.
Despite the observed benefits for many types of agricultural practices, many farmers are reluctant to incorporate conservation tillage in their agricultural practices, fearing a loss of crop yields.
Since most studies on conservation tillage and yield production had only been done at the level of local experiments, researchers from Stanford wanted to analyze how tilling practices affected year-over-year crop yields on a larger scale.
Their study area involved the US Corn Belt, an area spanning 1 million square kilometers across 12 states in the midwestern United States.
This study focused on nine of those states: South Dakota, Minnesota, Iowa, Missouri, Wisconsin, Illinois, Indiana, Michigan, and Ohio.
The dominant crops planted in this area are maize (corn) and soy. The researchers looked at nine states for corn production and three states for soy.
Researchers used two satellite-derived datasets to apply a machine-learning causal inference approach:
The first was a gridded dataset of annual tillage practices for the north central US from 2005 to 2016 by Azzari et al , 2019.
The second dataset used was the Scalable Crop Yield Mapper (SCYM) which the researchers used to produce yield maps for maize and soybean.
The computer model was then trained to compare yields based on tillage practice.
What the researchers found is that conservation tillage resulted in a modest increase in crop yields (an average of 3.3% and 0.74% yield increase for maize and soybeans, respectively). The crop yield varied with some areas experiencing up to an 8.1 percent increase for corn and 5.8 percent for soybeans. Other fields produced a negative effect 1.3 percent for corn and 4.7 for soybeans.
Factors such as local climate and water soil levels played an influential role in crop yields.
“Figuring out when and where reduced tillage works best could help maximize the benefits of the technology and guide farmers into the future,” said study senior author David Lobell, a professor of Earth system science in the School of Earth, Energy & Environmental Sciences and the Gloria and Richard Kushel Director of the Center on Food Security and the Environment.
In addition, the full benefits of conservation tillage can take over a decade to realize.
The researchers note that full benefits for corns farmers takes eleven years.
Full benefits for soybean production takes over two decades.
In the meantime, farmers can realize immediate benefits of conservation tillage in the form of reduced need for labor, fuel, and farming equipment.
Conservation Tillage Study and References
Deines, J. M., Wang, S., & Lobell, D. B. (2019). Satellites reveal a small positive yield effect from conservation tillage across the US Corn Belt. Environmental Research Letters, 14(12), 124038. https://doi.org/10.1088/1748-9326/ab503b
Reduced soil tilling helps both soils and yields, Stanford researchers find. Stanford, December 6, 2019
Azzari, G., Grassini, P., Edreira, J. I. R., Conley, S., Mourtzinis, S., & Lobell, D. B. (2019). Satellite mapping of tillage practices in the North Central US region from 2005 to 2016. Remote sensing of environment, 221, 417-429. https://doi.org/10.1016/j.rse.2018.11.010