Integration of plant-based canopy sensors for site-specific nitrogen management.

• Ano de defesa: 2011
• Autor(a) principal: SHIRATSUCHI, L. S.
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Palavras-chave em Português: Sensoreamento; Fertilization; Agricultura de Precisão; Nitrogênio; nitrogen; precision agriculture
• Link de acesso: http://www.alice.cnptia.embrapa.br/alice/handle/doc/928933

The soil's nitrogen (N) supply can vary drastically in the field, spatially as well as temporally making any soil prediction difficult even with very detailed mapping. Consequently, a plant-based approach wherein the measured canopy can indicate the N needs in a reactive and spatially-variable way can be a better approach than mapping, because integrate the soil N supply and translate the crop need on-the-go. The first experiment evaluated the performance of various spectral indices for sensing N status of corn, where spectral variability might be confounded by water-induced variations in crop reflectance. We found that water and previous crops effects on vegetation indices (VI) must be considered, and also that some VIs are less susceptible to water with good ability for N differentiation. In the second experiment, the objective was to develop an approach that relies an local soil conditions as well as on active canopy sensor measurements for real-time adjustment of N application rate. We found that local variations in plant N availability must be considered to determine the optimal N rate on-the-go, and that the localized reference incorporated the spatial variability of the N-rich plot. Next, we determined the correlation between active canopy sensors assessments of N availability and ultrasonic sensor measurements of canopy height at several growth stages for corn. We found strong correlations between both sensors and that they had similar abilities to distinguish N-mediated differences in canopy development. The integrated use of both sensors improved the N estimation compared to the isolated use of either sensor. Based on these strong correlations, we developed an N recommendation algorithm based on ultrasonic plant height measurements to be used for on-the-go variable rate N application. Lastly, we evaluated the crop water status using infrared thermometry integrated with optical and ultrasonic sensors, we concluded that the integration of sensors was beneficial to detect water-stressed zones in the field, affecting yield and possibly promising to delineate zones for N and Water management.