HR: 1340h
AN: B23A-1035 INVITED     [Abstracts]
TI: Using the Normalized Differential Wetness Index to Scale Leaf Area Index, Create Three-Dimensional Classification Maps, and Scale Seasonal Evapotranspiration Depletions in Canopies Along the Middle Rio Grande Riparian CorridorCorridor
AU: * McDonnell, D E
EM: demcdon@respec.com
AF: ReSpec, 4775 Indian School Rd NE, Albuquerque, NM 87110
AU: Cleverly, J R
EM: cleverly@sevilleta.unm.edu
AF: University of New Mexico, Department of Biology , Albuquerque, NM 87131
AU: Dahm, C N
EM: cdahm@sevilleta.unm.edu
AF: University of New Mexico, Department of Biology , Albuquerque, NM 87131
AU: Coonrod, J A
EM: coonrod@unm.edu
AF: University of New Mexico, Department of Civil Engineering, Albuquerque, NM 87131
AB: This research creates temporally and spatially explicit data layers of vegetation, leaf area index (LAI), three dimensional (3D) vegetation classification maps, and seasonal evapotranspiration (ET) depletions along the middle Rio Grande riparian corridor. The first part of this work produces two dimensional (2D) classification maps of native and non-native canopy vegetation using temporal patterns and the decision tree classifier in ENVI 4.0 (Research Systems Inc. Boulder, Colorado). The second part of this work correlates the normalized differential wetness index (NDWI) with field measurements of plant area index (PAI), stem area index (SAI), and leaf area index (LAI) using the LAI-2000 Plant Canopy Analyzer (PCA) (LICOR Inc., Lincoln, Nebraska). SAI is measured in winter to capture only branches and stems. PAI is measured during the growing season. Field measurements taken within 10 days of image capture dates provide adequate correlations though the closer the dates the better the correlation. LAI represents the surface area of active green leafy vegetation. NDWI correlates with both PAI and estimated LAI in both Tamarisk chinensis and Populus deltoides ssp. Wislizeni sites better than the more traditional normalized differential vegetation index (NDVI). This study also suggests that winter PCA measurements approximate SAI which should be subtracted from PAI in woody vegetation like T. chinensis and Salix exigua stands. The results show that correcting for leaf geometry by multiplying T. chinensis areas with cylindrical cladophylls by pi and the remaining flat leaf vegetation by two yields the best relationship between NDWI and total LAI. The 2Dclassification maps can be placed on top of relief maps of LAI to produce 3D classification maps. The final part of this research scales ET from four 3D eddy covariance towers located in two T. chinensis and two P. deltoides study sites. ET is regressed with LAI, percent daylight (PD), and average hourly incoming net Radiation per day (Rn). The best relationship results when stomatal side surface area is considered. Cladophylls have stomata on all sides while flat leaves have stomata on the abaxial side only. The correction multiplies LAI by .5 in flat leaf vegetation and 1 in T. chinensis vegetation. That means that transpiration rates are two times greater in T. chinensis sites compared with P. deltoides for the same leaf area. The resulting equations are put into a model to quantify ET for each available image and then for the entire growing season.
DE: 0439 Ecosystems, structure and dynamics (4815)
DE: 0480 Remote sensing
DE: 0483 Riparian systems (0744, 1856)
DE: 1813 Eco-hydrology
DE: 1818 Evapotranspiration
SC: Biogeosciences [B]
MN: Fall Meeting 2005