ESTIMATING POTENTIAL AND ACTUAL CROP GROWTH AND YIELD BASED ON COMBINATION OF REMOTE SENSING AND GROWTH MODELING - A CASE STUDY FOR IRRIGATED RICE

Authors: INOUE, YOSHIO - INST AGR ENV SCI, JAPAN; Moran, Mary; HORIE, TAKESHI - KYOTO UNIV, JAPAN

Submitted to: Journal of Plant Production Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/15/1998
Publication Date: N/A
Citation: N/A

Interpretive Summary: Crop managers use many tools for diagnosing and managing plants and soils to improve profits and improve environmental conditions. One tool that is gaining acceptance is the crop simulation model, which can help the farmer predict crop growth and yield. However, such models require a great deal of detailed information about crops, soils, and management practices, and it is tedious and sometimes impossible to collect all such model parameters. A new approach was tested to use on-site radiometers to measure field crop conditions and incorporate this easy-to-obtain information into the crop simulation model. With this approach, radiometric measurements (such as crop temperature) were used to optimize the model inputs and allow accurate predictions of future crop growth and potential yield. Results showed that this was a viable tool for improved management of rice crops. This approach could be put into operation utilizing radiometric measurements from sensors aboard aircrafts to cover all the farmer's fields in a set of images. Such information from aircraft- and satellite-based sensors is currently available at a reasonable cost. The combination of crop simulation models and remote sensing can provide farmers with an operational means of acquiring information for optional crop management.

Technical Abstract: Spectral reflectance of differentially-managed rice canopies was measured over an entire growing season and analyzed with special attention to linking remotely sensed information with a simple growth model. The fraction of absorbed photosynthetically active radiation(fAPAR), which is often used as a key variable in simple process models, was well correlated with spectral vegetation indices (VI). VIs, such as NDVI and SAVI, were derived from the ratio of reflectance at two wavelengths (R660nm and R830nm) and a new VI, termed the normalized difference of R1100nm and R660nm divided by their sum. These indices became less sensitive to fAPAR when fAPAR was larger than 0.4. The use of R1100nm and R1650nm with R660nm and R830nm in multiple regression significantly improved the prediction accuracy of fAPAR. A close linear relation was found between a spectral ratio R830nm/R550nm and leaf nitrogen content during the ripening period although it was not the case before heading. Multiple regression analysis showed that a combination of four spectral bands--R550nm, R830nm, R1650nm, and R2200nm--was useful for estimating the total amount of leaf nitrogen. Such remotely-sensed nitrogen variables would be potential model parameters in a simple model. A real-time calibration module based on a simplex algorithm was developed and proved effective in linking the remotely-sensed fAPAR with a simple model. This approach was also useful for inferring the physiological parameters such as radiation use efficiency for each rice canopy without destructive sampling. The re-parameterization and/or re-initialization with remotely-sensed information was demonstrated to be a practical and effective approach, especially for operational purposes.