Using Big Data to Inform Agricultural Decisions

The views presented in these blogs are those of the authors.

Making informed decisions at the farm or landscape scales is not easy. Critical information may be missing, consequences may not be readily identifiable, or there may be too much information to process. The agricultural sector, like all parts of our global economy, is becoming data-rich due to advances in remote and mobile measurement technologies. However, this increases the need for enhanced data management and analytical capabilities. The ability to obtain, process, and share basic data and insights gained from processed data is key to making informed farm- and policy-level decisions to improve producers’ productivity and economic returns as well as broader social and environmental outcomes.

Figure 1.  An example of precision agricultural software used in the field. Trimble FMX controller (screen on right) can independently control the rate of up to four products. Raven Envizio Pro (smaller screen on left) carries out auto steer and provides guidance. Photo by Guy Swanson.


Status of big data in agriculture: Increasingly, companies such as Monsanto and John Deere are offering services that collect detailed spatial and temporal data from farms regarding planting densities, dates, production growth, and harvesting. In return, these companies promise to evaluate the data and provide participants with information aimed at increasing farm profits by optimizing input uses and improving yields. Monsanto for example, claims that their field-specific seeding and fertilizing “prescriptions” have the potential to create billions of dollars in increased farm revenues by reducing input costs and improving yields. Monsanto’s recent purchase of The Climate Corporation, a firm specializing in site-specific weather projections, has improved Monsanto’s ability to fine tune field-based prescriptions related to weather predictions. These software developments are viewed by agribusiness companies as opportunities to help producers meet production challenges associated with greater variability and risk from a changing climate and changing economic conditions.

Next frontier for data analytics: An increase in the use of precision farming and mobile technologies along with improvements in data management software offer expanding opportunities for an integrated data infrastructure linking farm management decisions to site-specific bio-physical data and ultimately to the design of “climate-smart” policies. Field-specific data combined with recommended uses of fertilizers, seeding rates, and other inputs can be integrated with spatial landscape-scale models for fine tuning agricultural policies. For example, better quality data and models could enhance the targeting of incentive payments provided to farmers to improve water quality and conserve biodiversity.

So how might this work?: Figure 2 provides an overview of the linkages between data and decision tools that support farm decisions and landscape-scale science-based policy recommendations.

Figure 2. Linkages between data and decision tools at farm and landscape scales.

While farm-level decision making and landscape-scale analysis have different purposes, they both benefit from the same data:

  • Private data: site- and farm-specific characteristics of the land and the farm operations, and site- and farm-specific management decisions.
  • Public data: weather, climate and other physical data describing a specific location, as well as prices and other economic information.

Eric Odberg, farmer in the REACCH region using precision agriculture software in the field. Photo by Guy Swanson.

A key to achieving a smarter agricultural knowledge infrastructure is to recognize that new and better data are an asset to both private and public stakeholders, and can provide win-win situations for improving farm profits, the sustainability of our food and agricultural systems and supply chains, and the outcomes of public policies and investments. As the development of decision support tools for precision farming continues, along with the expansion of mobile technology and remote sensing capabilities, the opportunities for creative partnerships across sectors grow.

Unique opportunities created through public-private partnerships could include:

  • Reducing “respondent burdens” associated with the present system of multiple mail-based and personal interview surveys used to collect data periodically from growers and landowners (such as the National Resources Inventory, the Agricultural Census, and emerging sustainable supply chain programs). Under an integrated system, much of the baseline information can be acquired and stored once, as a part of a farm operation’s ongoing management system, rather than being collected multiple times for multiple purposes. This information could be updated in a more cost-effective way, through mobile or web-based technologies.
  • Minimizing the duplication of data collection efforts and costs, making science-based policies and precision agriculture more economically feasible.
  • Streamlining the collection of detailed data necessary for documenting organic or sustainable practices for certification, or compliance with regulatory standards.
  • Enabling private data aggregation to build more effective management tools that will help farmers understand relationships among practices and outcomes for both production and conservation.
  • Aggregating detailed data regarding management practices and associated environmental outcomes that can be used to demonstrate improvements in environmental quality at the landscape scale to regulators and the supply chain, as well as consumer packaged goods companies/brands that are increasingly interested in promoting sustainable and “green” products.
  • Facilitating and enhancing a science-based approach to agricultural policy.

Concerns to address: To make these proposed partnerships attractive to participants, key operational considerations need to be addressed. These include designing an efficient and secure data system, maintaining data confidentiality and addressing privacy concerns.

In summary, an agricultural knowledge infrastructure is an asset for supporting productivity gains and policy improvements. It is dependent upon strong partnerships among producers and public and private entities to ensure privacy and confidentiality, reliability, sustainability and usefulness for on-site management as well as science-based policies. The rapid pace of advancements in tools, technologies, and data initiatives, coupled with the increasing demand for better data, provides an ideal environment for the development of partnerships to build a viable and sustained knowledge infrastructure. As big data drives ever more demands for better policies and better management, the new tools and innovations that result will shape the sustainable management of agricultural ecosystems in a very positive way.

For more information, please see our AGree Point of View Paper, Towards a Knowledge Infrastructure for Science-Based Policy and Sustainable Management of Agricultural Landscapes, and AGree’s Working Landscapes Initiative, which includes efforts to support the integration and analysis of data collected by USDA and other federal agencies in order to quantify linkages among conservation practices and systems, soil health, yield, yield variability, crop viability during extreme weather events, and environmental indicators (e.g., water quality).

Laurie Houston, Susan Capalbo, and John Antle are researchers in applied economics at Oregon State University’s College of Agricultural Sciences.