الفهرس 
Spectral reflectance measurementsOnly 14 pages are availabe for public view
 |  | from | 155 |  |  |
| | | from | 155 | | |
Abstract
Available water is one of the most limiting factors in crop production and exposure of
plants to drought leads to noticeable decreases in leaf water potential and water content
with a concurrent increase in leaf temperature. The present study investigated the
potential of high throughput sensing methods such as spectral reflectance measurements,
laser induced-chlorophyll fluorescence measurements and near infrared temperature
measurements to detect drought stress in plants by measuring leaf water potential, plant
water content and canopy temperature. Validations were performed under controlled
conditions (growth chamber and darkroom) to assess leaf water potential, leaf water
content, canopy water content and relative leaf water content and under field conditions
to assess leaf water potential leaf, canopy water content, canopy water mass and aerial
biomass. To test the ability of spectral measurements to detect changes in leaf water
potential without changes in biomass and only small changes in leaf water content of
wheat and maize, the measurements were conducted under controlled conditions. To
study the stability of spectral reflectance measurements, fluorescence emission and near
infrared temperature measurements to detect water status in wheat under field conditions,
another series of experiments were carried out under a mobile rain-out shelter to induce
controlled water stress in plants. The results obtained showed that significant relationships (R2-values 0.74-0.92) between leaf water potential and new spectral indices ((R940/R960)/NDVI; R940/R960) were detected with or without significant changes in leaf water content of both wheat and maize under controlled conditions. The exact relationships found, however, were strongly influenced by the date of measurement or water stress induced. Thus, global spectral relationships measuring leaf water potential can probably not be established across plant development. Even so, spectrometric measurements supplemented by a reduced calibration data set from pressure chamber measurements might still prove to be a fast and accurate method for screening large numbers of diverse lines. The results of spectral measurements under controlled and field conditions showed that five spectral indices (NDVI, (R410 - R780)/(R410 + R780), (R490 - R780)/(R490 + R780), (R510 - R780)/(R510 + R780) and R600/R780) revealed to be good indicators to detect canopy water content of wheat at individual measurements and across all measurements for each cultivar in different years and with coefficients of determinations varying between 0.59* to 0.98***. As well as the NDVI and the index R600/R780 seemed to be good indicators to detect canopy water content of wheat across all measurements for all cultivars over three years of investigations (R2 ≥ 0.84***). The results of laser-induced chlorophyll fluorescence measurements under field
conditions showed that the canopy water content was closely related to chlorophyll
fluorescence at 690 nm, 730 nm and to the biomass index for each of the four
investigated cultivars with R2- values ≥ 0.83***, R2 ≥ 0.84***, and R2 ≥ 0.82***,
respectively, and averaged over all cultivars the coefficients of determinations were R2 =
0.71***, R2 = 0.74***, and R2 = 0.74***, respectively. Cultivar-specific relationships to
canopy temperature were within R2 = 0.62* to 0.97***. Fluorescence intensity at 690 nm
revealed to be a good indicator of leaf water potential for each of the cultivars, Ludwig,
Ellvis, Empire and Cubus with R2 = 0.69***, R2 = 0.66***, R2 = 0.64***, and R2 =
0.69***, respectively. This work shows the possibility to detect drought stress by laserinduced chlorophyll fluorescence measurements in wheat by measuring fluorescence intensities at 690 nm, 730 nm or the newly developed biomass index revealing to be
better indicators of drought stress than the fluorescence ratio F690/F730. In addition, canopy temperature showed a good relationship to the canopy water content (R2 ≥ 0.59*) for each cultivar at most of the measurements over three years. But the relationship was affected by the day and cultivars. Canopy temperature assessments presented good relationships to leaf water potential either in 2005 or 2007 across all measurements for each cultivar (R2 ≥ 0.58***) and all cultivars (R2 ≥ 0.65***), but weak relations were found across all measurements in 2007. Over all our results under temperate field conditions, canopy water content of plants was the most suitable physiological parameter to detect drought stress by proximal sensing methods using spectral reflectance and laser induced chlorophyll fluorescence measurements. Therefore, time-consuming destructive methods could be replaced by rapid, non-destructive methods. These technologies may open an avenue for fast, highthroughput assessments of water status in plants, which would simultaneously be useful for screening large numbers of plants (e.g., in breeding) as well as being equally important for management related actions.