246 
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B7. Beijing 2008 
o o o o o o o 
r--com-<frcocMT-o 
CM 
(%) aoue*oa|jsy 
Figure 3. Reflectance of winter barley at site 06A West of 
pipeline along stress transect on 7 th of May 2006. Each line 
represents reflectance at 10 m intervals along the transect. The 
noisy regions in the spectra are the result of atmospheric water 
vapour. 
The 2006 first derivative results for winter barley at test site 
06A reveal a double peak in the red-edge as reported by Zarco- 
Tejada et al. (2003) and Smith et al. (2004). First derivatives of 
the red-edge region change in magnitude, position and shape 
between the peaks at 700 (A) and 723 nm (B). Peak B becomes 
more pronounced relative to peak A, which maintains the same 
shape and position with proximity to the pipeline (Figure 4). 
The West transect limb exhibits notable differences between 
peaks A and B of up to 0.34 from 0 to 15 m and 0.24 from 20 to 
70 m respectively (Figures 4a and b). An exception to this trend 
is 60 m West, exhibiting a lower red-edge peak, differing by up 
to 0.34 at peak A and 0.53 at peak B (Figure 5a). The same 
differences between peaks A and B were also evident along the 
East transect limb, although less marked. The contrast in the 
East and West transect limbs could be explained by the location 
of the vehicle running track to the West of the pipeline (Figure 
2). The heavy machinery used to install the pipeline could have 
led to sub-soil compaction, leading to long-term soil infertility 
and impeded root penetration which can cause vegetation stress 
(Rowell and Florence, 1993; Llewellyn and Curran, 2005). 
These observed differences in first derivatives for winter barley 
are of the same order of magnitude as those reported by Zarco- 
Tedjada et al. (2003), Llewellyn and Curran (2005), Smith et al. 
(2004), and Smith et al. (2005) for a range of vegetation stress 
inducers (elevated levels of soil gas, herbicide, shade, water, 
temperature and humidity). 
(a) 
Wavelength (nm ) 
(b) 
Wavelength (nm) 
Figure 4. VIS-NIR first derivative of reflectance of winter 
barley at site 06A along stress transect, on 7th of May 2006: (a) 
West of pipeline 0-15 m; and (b) West of pipeline 20-70 m. 
Each line represents reflectance at 1 and 5 m intervals for (a) 
and 10 m intervals for (b) along the transect. 
More marked differences in magnitude between peaks A and B 
along the stress transect (Figure 4a and b) were observed 
relative to the control transect. The relative consistency 
between peaks A and B along the control transect, coupled with 
their associated proportions of soil background depicted in the 
nadir photographs at 20 and 40 m West (Figure 5b and c), 
strongly infer first derivative analysis is detecting vegetation 
stress and not soil background at test site 06A. An exception to 
this is at 0 m in close proximity to the pipeline where the 
proportion of soil background is very marked (Figures 4a and 
5a). Moreover, the influence of poor atmospheric conditions at 
10 and 20 m West have little influence on differences between 
first derivative peaks A and B (Figure 4a and b).