Farmers and society need new ways to grow more food more sustainably. Digital agriculture (DA) applies new technologies such as data science, advanced sensors in the field and from space, digital communication channels and automation on the field, to give farmers real-time access to better insights to make optimal decisions, increase yield, reduce waste and drive up wealth in rural areas, especially in emerging markets.

Vineyard mapping technology created by Cornell's Efficient Vineyard program
Vineyard mapping technology created by Cornell’s Efficient Vineyard program

What is digital agriculture?

Digital agriculture (DA) is a holistic, systems-level approach to agriculture that integrates the rapidly expanding power of automation, remote sensing, communication and information technologies to enhance the sustainability, productivity and profitability of farming. DA affects all components of the food system and provides new methods and tools for data collection, analysis, distribution, policy analysis and decision-support. DA’s impact is broadly felt in instances such as refining crop and animal genetics, increasing productivity and profitability on farms both large and small, reducing environmental impacts, streamlining transportation logistics, tracking produce from farm to table and helping farmers remain productive in a changing climate.

Why is digital agriculture important?

Sustainably feeding people, improving livelihoods and protecting the environment are among the most significant challenges of our time. Making meaningful contributions to address these
challenges requires systems-level thinking and an interdisciplinary, holistic approach. Digital agriculture provides the conceptual and operational foundations to drive innovation and tackle challenges at scale – farm, region, nation, globe — and to integrate solutions and evaluate impact across the technical, social, economic, policy and environmental dimensions of agriculture.

 Hydroponic tomatoes growing in a Cornell greenhouse
Hydroponic tomatoes growing in a Cornell greenhouse. Photo by R.J. Anderson / CCE.

What will the future of agriculture look like with digital agriculture?

Digital agriculture (DA) will play a critical role in improving management of existing agri-food systems and enable design of entirely new systems. DA reflects a shift from fragmented, generalized management of farm resources to individualized, real-time, integrated and data-driven management. The realization of mature DA approaches will create unprecedented opportunities for achieving efficiencies and targeted impact via optimized and individualized handling of biological (crop, livestock, pest and pathogen), environmental, social and economic components of agriculture and will facilitate outcome-based policy analysis and design.

What is Cornell’s vision of radical transformation through digital agriculture?

We need new technologies, analytical processes, and models of governance to advance higher productivity, efficiency and sustainability of our agri-food systems. Similarly, the challenges of improved nutrition, food quality and food safety require new thinking, new tools and new socioecological relations. Many of the improvements that are needed can be facilitated by digital innovations in agriculture that support the optimization of farming practices to meet production, distribution and sustainability goals.

DA activities at Cornell are organized into three interconnected research cores — systems analytics, digital innovations, and discovery and design — that aim to address the grand challenges facing global agri-food systems by 2030 and beyond.

1. What is systems analytics? Systems analytics involves the use of advanced computing approaches that leverage artificial intelligence, machine learning, and systems engineering to organize and examine large data sets and draw conclusions about the information they contain. In digital agriculture, systems analytics accelerates scientific discovery and serves as a critical bridge for connecting all hardware and software pieces within DA. The operational vision of the future farm is rapidly shifting toward human/machine learning interfaces to optimize innovation and management at scale in both digital and physical spaces. As sensor-based data collection increases in frequency and volume, machine learning and artificial intelligence help integrate these data into crop/animal production models that support on-farm decision-making. Models can be improved based on sharing of data across farms to improve accuracy and impact, where encryption and blockchain technology are used to avoid compromising data confidentiality and integrity. Models can be further improved by incorporating policy and socioeconomic factors. The maturation of systems analytics will transform the way data drives agricultural decisions, allowing for increased individuation and temporal resolution and enhancing the potential for long-term agricultural planning and stakeholder involvement to manage resources more sustainably, improve distribution networks and pursue socioeconomic prosperity.

A sensor installed in the trunk of a grapevine.
A sensor installed in the trunk of a grapevine. After the installation of the sensor, the researchers cover and insulate the trunk. Photo by Alan Lasko/Provided.

2. What do we mean by digital innovations? Digital innovations of interest include (i) next generation communications and computing technologies to provide affordable, high-performance connectivity to farms, transportation infrastructure and markets, (ii) scale-bridging sensing technologies to enable automated collection and transmission of data to monitor the status of biological (crop, livestock, pest and pathogen), environmental (soil, water, air and barn) and sociological systems affecting the production, processing and distribution of food, and (iii) smart, autonomous systems that benefit, in real time, from the coordinating power of analytics to provide optimized and individualized handling of all elements within agri-food systems. An important concept to guide the development of digital innovations will be to minimize cost and promote accessibility and adoption in different farm settings by optimizing the compatibility of sensors, artificial intelligence and machine learning, robotics and communications protocols, and providing common infrastructure (i.e., edge computing) that allows efficient transfer of data streams for deployment of customized management and decision-support tools.

3. What do we mean by discovery and design? Digital agriculture at Cornell will pursue discovery and design across the interconnected scales of agri-food systems and will translate discoveries into field applications. The development of plant-, animal- and microbe-specific monitoring hardware – from nano- to macro-scales – will generate new data streams for the elucidation of physiological, ecological and socioeconomic phenomena on the farm and beyond. This pipeline of data will generate new insights and models that will feed back into our systems analytics and digital innovations cores to form a foundation for the efficient, inclusive and sustainable management of the agricultural systems of tomorrow. In particular, DA will provide paths to a deeper understanding of the relationship, for example, between genotype and phenotype within a molecular breeding program; cell biology to disease management practices in an orchard or dairy operation; or production to distribution and consumption patterns in a socioeconomic context. Emerging understanding of the biology that underlies yield, nutritional quality or environmental resilience will also help define new targets for digital innovations for measurement and intervention. This pipeline of new technologies, strains and integrated analytical tools is essential for developing the efficient, sustainable farms of tomorrow.

What are some anticipated projects?

1.   New Hardware and Software to Sense and Act

  • Design nano-biosensors to “measure the immeasurable.”
  • Engineer smart, autonomous machines that can evaluate and treat individuals with minimal invasion.

2.   Systems Engineering for Big Data and Farm Operations

  • Develop methodologies to manage and deploy disparate, complex data structures and streams (climate, local weather, soil, plant/animal, machine/human) for revenue and sustainability.
  • Optimize agricultural operations from research and development, through field practices, to distribution and markets.

3.   New Connectivity Solutions

  • Expand the Cloud to the edge of the farm, the Edge Cloud, requiring seamless and continuous computation and communication despite severely limited and/or intermittent energy supply, and despite intermittent and/or sparse cellular network connectivity.
  • Reconcile the modes and idiosyncrasies of a highly heterogeneous, highly granular collection of analog and digital, as well as hardware and software components.

4.   Artificial Intelligence (AI) and Computing

  • Utilize available data, network and computing resources to reason autonomously and support robots and humans to optimally cooperate safely and securely in real-time.
  • Compose algorithms that bring scientific solutions to agriculture and food system management.

5.   Biology, Physiology and Bioinformatics

  • Develop and deploy micro, nano, and molecular technologies for the transmission of biological information from organisms to computing systems in real-time.
  • Exploit new data streams to advance genetics, phenology, and breeding for health, stress mitigation and local adaptation.
  • Integrate large scale data with genetic and biophysical information to inform climate-ready varieties and adaptation techniques.

6.   Policy Analytics, Economics, and Society

  • Support development of real-time, high-resolution yield forecasting, disease and practice modeling at-scale utilizing AI, spatial statistics, and machine learning approaches.
  • Develop policy and risk analytics to improve sustainability and inform decision-making both at the farm and policy level.
  • Examine how off-farm service providers impact development and adoption of digital agriculture and how adoption may disrupt existing social and economic relationships.
  • Analyze how data ownership, data access and privacy concerns mediate development of digital agriculture.