Comparing NDVI from Plot to Satellite Scales
Wylie et al (2000) developed relationships between plot bases NDVI with leaf area index (LAI), biomass and fraction of absorbed photosynthetically active radiation (fPAR) over central and northern Great-Plains grasslands in 1995. The authors estimated these parameters for ground pixels under three grazing and burning regimes, subsequently interpolated them spatially and regressed them against Landsat Thematic Mapper (TM) NDVI. Good agreement was found between scaled-up grid based parameters and satellite NDVI with a r-square values .92-.94 (Wylie et al, 2000).
Fig. 1: The 1995 ground radiometer to biophysical parameter [biomass (g/m2), fPAR, and LAI] regression 95% confidence intervals for observations compared with 1996 data (UU = unburned ungrazed treatment) (Wylie et al, 2000).
Fig. 2: Herbaceous grid-based aboveground biomass regressed on Landsat TM NDVI (Wyllie et al, 2000).
Estimation of Evaporation
NDVI has been shown to be a good predictor of evaporation (ET) over grassland, though spectral differences between growing stage and mature vegetation should be taken into account. Kondoh and Higuchi (2001) found that Landsat TM derived NDVI had NIR increase during the growing season with concomitant red band decrease. The red band decrease is indicative of increase in absorption due to photosynthesis in this period, though at the mature stage ET may be best correlated with NIR, which is a proxy for insolation (Kondoh and Higuchi, 2001). This may be a good approach for grasslands since the surface roughness is low compared to larger vegetation which can lead to a surface boundary layer with relatively low vapor pressure deficit (VPD), increasing the influence of insolation on ET.
Fig.3: Daily evapotranspiration and TM-Based NDVI. The data from 1984 to 1988 are plotted on the figure (Kondoh and Higuchi, 2001).
Fig. 4: Relationship between the daily evapotranspiration and TM-derived NDVI. The soil line is the regression line for the growing season (DOY 100-200), and the dashed line is the regression line for the mature season (DOY 240-330) (Kondoh and Higuchi, 2001).
Fig. 5: Seasonal variation of red and near-infrared bands of TM as expressed by digital number (Kondoh and Higuchi, 2001).
Drought Indicator
Time series analysis of NDVI allows establishment of a baseline for normal vegetation productivity for a region. Interannual increase or decrease relative to this baseline can serve as an indicator for drought intensity.
Fig. 6: The difference between the average NDVI for a particular month of a given year (such as August 1993, above) and the average NDVI for the same month over the last 20 years is called NDVI anomaly. The above image shows the NDVI anomaly in the U.S. for August 1993. In that year, heavy rain in the Northern Great Plains (North and South Dakota, Alberta, and Saskatchewan) led to flooding in the Missouri River. The resulting exceptionally lush vegetation appears as a positive anomaly (green). Concurrently, in the Eastern U.S., rainfall was very low, and the region exhibited a strong negative anomaly (dark red) (Nasa Earth Observatory Web Reference [2]).
Evaluation of Rangland Forage Quantity and Quality
Livestock managers have a vested interest in knowing the amount and quality of forage that is available on rangelands. A recent investigation predicted live, dead standing and total biomass as well as nitrogen concentration and standing nitrogen, which showed good correlations for all except nitrogen concentration (Thoma et al, 2002). The authors suggest that NDVI is a good estimator of total forage but not of nitrogen concentration, or forage quality.
