Near-Realtime Estimation and Forecasting of National and Global Specialty Crop Supplies

Global production of perishable food crops is increasingly vulnerable to supply chain interruptions given climatic changes in key growing areas, with associated pressures such as reduced water stores, increasing incidents of new diseases and pests, and disruptive events such as wildfires and floods. Detailed knowledge of crop supply at any given time and an ability to forecast the anticipated supply into the near future are key to increasing the resiliency, efficiency and sustainability of perishable food crop production nationally and globally. This CCRPP project will refine and commercialize capabilities developed during Phase II to determine current day supply of important vegetable and berry crops and to forecast their future supply. We use novel techniques to classify the crops, which are grown in many relatively small fields in close proximity and with relatively short and staggered growth periods - attributes which historically have prevented their reliable detection and tracking with satellite imagery. We apply an innovative approach to generating larger sets of labeled data given the typical dearth of labels from on-the-ground information, which promises to make our capability highly scalable and efficient. Applying machine learning and computer vision techniques to time series of satellite imagery, the system forecasts harvest events, on a field by field basis, several weeks in advance and with an error of about one week. The outcome will be an operational capability to deliver up-to-date estimates of crop supply and risk monitoring and forecasts of the associated supply chain for these crops into the near future, for commercial stakeholders of the fresh produce industry and with applications to global food security. Knowledge of current supply and forecasted future supply of perishable food crops are key to increasing the resiliency, efficiency and sustainability of their production. This CCRPP project proposes to refine and advance the commercialization of the capabilities developed during our Phase II-Extended to estimate current supply of several important, high value perishable crops and to forecast their future supply. The spatial and temporal growing patterns of such crops historically have prevented their reliable detection and tracking with satellite imagery. We apply novel classification techniques and an approach to generating large sets of labeled data that makes the capability highly efficient and scalable.  We expect to operationally forecast harvest events weeks in advance and with an uncertainty of about a week. Applying historical yields per area for each crop, these performance metrics will provide accurate yield forecasts to stakeholders in the commercial produce industry and with potential applications to global food security. The principal objective of this project is to mature the capabilities developed during the associated Phase II-Extended project to commercialize both the detection of crop types and detection and forecasting of harvest dates, for crop fields that are typically rotated between different crops in close succession. Following are the tasks to be completed to accomplish this objective: Define default AOIs and a capability to update, expand and add new AOIs. Design and test a scalable deployment infrastructure. Design, test and deploy backend storage for AOIs and results. Develop a user interface to the web applications. Conduct customer outreach, training, and marketing. Add new crop types and geographies based on customer demand. At the end of the contract, a demonstration of the web services will be available.

Potential applications include the NASA Harvest consortium for global ag and food security and NASA Acres for US ag. Labeled imagery data will be relevant for training other NASA land class models through repositories such as MLHub. Frequently updating crop acreages will address the data gap in the Cropland Data Layer (CDL) and would be of direct benefit to critical federal and state programs, including the national GHG MMRV strategy, the U.S. Bureau of Reclamation, the National Land Cover Database (NLCD), and state water boards. Commercial customers are crop producers, food service and grocery retailers, and financial service providers of the high-value, specialty crop agriculture industry. The short growth cycles, and relative vulnerabilities of these highly perishable crops to weather, pests, weeds and diseases have been a barrier to accurate near-term forecasting to date.

This is a mature technology (TRL 7+) — the realistic path in is usually NASA's Technology Transfer Program: licensing an existing NASA patent, or a Space Act Agreement to use NASA facilities/expertise directly. NASA also runs a startup licensing program with no upfront fee for companies formed to commercialize a specific NASA technology.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.