In order to provide timely and reliable assessments of the global crop market, it is necessary to move from limited, in situ crop assessments to operational crop monitoring on regional and global scales with multi-temporal remotely sensed (RS) data. Due to timeliness requirements and large volumes of imagery needed for operational agricultural monitoring, it is important that data processing and management, crop models, decision support systems, and output products are automated to the fullest extent possible. Lack of automation at the analytical step is one of the main issues precluding the introduction of RS time-series data into operational, near-real time settings. Development of an advanced, innovative system in support of timely assessments of emerging market opportunities for U.S. commodity crops is proposed. Remotely Sensed time-series data will be used to monitor crop conditions, compare given year conditions with those of the previous years, and to forecast relative crop yield on a bi-monthly basis. The overall results of the project will enhance regional and global agricultural monitoring and improve the timeliness and accuracy of current and projected crop commodity information relative to temporal marketing opportunities. The algorithms will improve FAS operational processing of time-series data and will contribute to FAS marketing and trade assessments. OBJECTIVES: The objectives of this project are as follows: Create capabilities to extract summary crop statistics from large spatial time-series datasets for near-real time generation of analytical data products, Enhance regional and global crop commodity monitoring, forecasting, marketing, and trade decision-making, Improve the accuracy, frequency, and spatial resolution of statistical crop assessments and forecasts and crop intelligence delivery for decision making, Provide timely assessments of emerging market opportunities for U.S. commodity crops, Boost ability of US agro-businesses to rapidly respond to and capitalize on new market opportunities both nationally and globally, Make the resulting analytical crop data products ready for integration into the operational framework of USDA Foreign Agricultural Service (FAS), NASA-FAS-UMD Global Agriculture Monitoring (GLAM) Project, U.S. Agribusiness, and commodity exchanges, and Contribute to the USDA strategic goal of enhancing economic opportunities for agricultural producers by providing accurate, objective, reliable, and timely assessments crop conditions and predictions of crop yield and market opportunities for U.S. agriculture. APPROACH: GDA will prototype a set of automated, decision support algorithms for near-real time, analytical assessment of large volumes of satellite derived time-series data on crop conditions. For development, the project collaborates with the Production Estimates and Crop Assessment Division (PECAD) of the USDA Foreign Agricultural Service (FAS). Phase I of the project will utilize current near-real time and historical publicly available NASA MODIS data from the Global Agriculture Monitoring Project (GLAM), a joint project between NASA, USDA FAS, and the University of Maryland. These data will be used to monitor crop conditions, compare given year conditions with those of the previous years, and to forecast relative crop yield on a bi-monthly basis. The proposed approach will be validated on the U.S.A. pilot sites against USDA NASS Agricultural Statistics and on international pilot sites against agricultural statistics from FAS
