Vol. II. Research in the Application of Spectral Data to Crop Identification and Assessment

Four investigations of the application of spectral data to crop identification and assessment are discussed in this volume. Part A discusses the development of spectromet crop development stage models. Several state-of-the-art development stage models for corn and soybeans were reviewed. One photothermal model and four thermal models were selected and evaluated. Part B describes an investigation of spectral data as a source of information for crop yield models. Intercepted solar radiation and soil productivity are identified as factors related to yield which can be estimated from spectral data. Several techniques for machine classification of remotely sensed data for crop inventory are evaluated in Part C. Early season estimation, training procedures, the relationship of scene characteristics to classification performance, and full-frame classification methods have been studied. In part D, a task to determine the optimal level for combining area and yield estimates of corn and soybeans is discussed. The optimal level is assessed utilizing current technology: digital analysis of Landsat MSS data on sample segments to provide area estimates and regression models to provide yield estimates