Frolking, S, Milliman, T, McDonald, K, Kimball, J, Zhao, MS, Fahnestock, M (2006). Evaluation of the SeaWinds scatterometer for regional monitoring of vegetation phenology. JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 111(D17), D17302.
[ 1] Phenology, or the seasonality of recurring biological events such as vegetation canopy development and senescence, is a primary constraint on global carbon, water and energy cycles. We analyzed multiseason Ku-band radar backscatter measurements from the SeaWinds-on-QuikSCAT scatterometer to determine canopy phenology and growing season vegetation dynamics from 2000 to 2002 at 27 sites representing major global land cover classes and regionally across most of North America. We compared these results with similar information derived from the MODIS leaf area index (LAI) data product (MOD-15A2). In site-level linear regression analysis, the correspondence between radar backscatter and LAI was significant ( p < 0.05) at most but not all sites and was generally higher (R-2 > 0.5) for sites with relatively low LAI or where the seasonal range in LAI was large (e.g., > 3 m(2) m(-2)). The SeaWinds instrument also detected generally earlier onset of vegetation canopy growth in spring than the optical/near-infrared (NIR) based LAI measurements from MODIS, though the timing of canopy senescence and the end of the growing season were more similar. Over North America, the correlation between the two time series was stratified largely by land cover class, with higher correlations (R similar to 0.7 - 0.9) for most cropland, deciduous broadleaf forest, crop/natural vegetation mosaic land cover, and some grassland. Lower correlations were observed for open shrubland and evergreen needleleaf forest. Overall, the results indicate that SeaWinds backscatter is sensitive to growing season canopy dynamics across a range of broadleaf vegetation types and provides a quantitative view that is independent of optical/NIR remote sensing instruments.