HR: 15:10h
AN: B53B-07 [Abstracts]
TI: Suitability of Existing NIR and Novel SWIR Spectral Indices to Remotely Detect Water Stress in Populus
spp.
AU: * Eitel, J U
EM: eite2335@uidaho.edu
AF: Department of Forest Resources, University of Idaho, PO Box 441142, Moscow, ID 83844-1133
United States
AU: Gessler, P
EM: paulg@uidaho.edu
AF: Department of Forest Resources, University of Idaho, PO Box 441142, Moscow, ID 83844-1133
United States
AU: Smith, A
EM: alistair@uidaho.edu
AF: Department of Forest Resources, University of Idaho, PO Box 441142, Moscow, ID 83844-1133
United States
AU: Robberecht, R
EM: ecology@uidaho.edu
AF: Department of Rangeland Ecology and Management, University of Idaho, PO Box 441142, Moscow, ID
83844-1135
United States
AB:
Undetected water stress in trees within poplar plantations can result in high economic losses to regional landowners.
Although, remotely sensed spectral measures associated with variations in plant water status have been previously applied to
non-Populus species, relatively little is known about the sensitivity of such indices to identify water stress in Populus
species. To meet this need, experiments were conducted to assess whether existing near-infrared and novel short wave-infrared
spectral indices acquired at the leaf and canopy level were suitable to detect water stress in Populus species.
Relationships between several water stress indices were compared to four common measures of plant water status and analyzed
for both light-moderate and severely stressed trees. The novel Maximum Difference Water Index (MDWI) consistently exhibited
the strongest relationships with changes in plant water status at both the leaf and canopy levels. At the leaf level,
reasonable though poorer relationships were obtained between each of the Normalized Difference Water Index (NDWI), the Red
Edge Inflection Point (REIP), and the Water Index (WI) and the water status measures. At the canopy level only poor
relationships were obtained between the measures of leaf water potential (LWP) and soil water potential (SWP) with each
spectral index (i.e. r2<0.35). Strong relationships were obtained between MDWI with RWC and EWT (r2>0.55, p<0.001),
while WI produced reasonable relationships with these water status measures (r2~0.42, p<0.001). On exclusion of
severely stressed plants from the regression analysis, the only reasonable significant relationships were obtained between
MDWI and the water status measures of RWC and LWP at the canopy level (r2~0.40, p<0.001). These results indicate that
the SWIR-incorporating MDWI is more strongly correlated to changes in plant water status compared to the NIR-based NDWI and
WI indices. MDWI could be applied to multi- and hyperspectral remote sensing systems to provide large-scale inferences of
water stress in Populus species. The results of this study should be repeated on other plant species that have applied the
NIR-based indices to assess the broader applicability of the inclusion of SWIR wavelengths when assessing plant water status.
DE: 0480 Remote sensing
SC: Biogeosciences [B]
MN: Fall Meeting 2005