| Título: | Physiological and Spectral Response to Water Stress Induced By Regulated Deficit Irrigation On Pear Trees |
| Autores: |
Raymond Ralph Struthers; KULeuven Jamshid Farifteh; KULeuven Rony Swennen; KULeuven Pol Coppin; KULeuven |
| Fecha: | 2013-01-28 |
| Publicador: | Applied Remote Sensing |
| Fuente: |
 |
| Tipo: |
Modeling |
| Tema: |
Geomatics, Physiology Hyperspectral remote sensing, flower buds, shoot growth, soil moisture, vegetation indices Hyperspectral Remote Sensing |
| Descripción: | This paper illustrates a pilot study designed to examine spectral response of ‘Conference’ pear trees (Pyrus communis L.) to water stress induced by regulated deficit irrigation. The main focus is to investigate the possibility of detecting early stage water stress before permanent and irreversible injury occurs to the trees. In addition, the variation in tree characteristics such as shoot lengths and flower buds in relation to induced water stress has been studied to define their effects on the spectral measurements. Potted pear trees were grown in the open air, receiving three different treatments: 1) Control treatment that is well watered throughout the period of the experiment, 2) mild stressed where average soil moisture is 9% lower than control treatment and 3) severe stress where the average soil moisture is 47% lower than control. Soil moisture contents, spectral reflectance and phenological data were collected on seven dates beginning June 22 through September 15 in 2010. Three different spectral vegetation indices were applied to three different wavelengths: visible (VIS), very near-infrared (VNIR) and short wave infrared (SWIR). Physiological measurements showed that severely stressed treatment had 17% less shoot growth in 2010 than the control treatment but 51% more flower buds in 2011. Vegetation indices suggest that the SWIR domain detected water stress about 19 days earlier than the VIS and NIR domains. However the data derived from the VIS and VNIR domains proved valuable in determining secondary effects of water stress using the plant senescence reflective index (PRSI). The results support the discrimination of non-stressed and severely stressed trees by using red edge position (REP). The REP results indicate that the non-stressed trees (Control treatments) shift towards longer wavelengths compared to the severely stressed treatment. The results support that the utilization of vegetative indices for application to water stress can be tailored to obtain information about the water stress states of the trees at a particular stage of the growing season. It also illustrates the fact that the canopy architecture, pigmentation and water content are dynamic, increasing the complexity of using hyperspectral data and the ability to distinguish between water stress and growth and development. |
| Idioma: | Inglés |