Integration of near-surface remote sensing and eddy covariance measurements: new insights on managed ecosystem structure and functioning

Ground-based, visible light imagery has been used for different purposes in agricultural and ecological research. A series of recent studies explored the utilization of networked digital cameras to continuously monitor vegetation by taking oblique canopy images at fixed view angles and time intervals. In our contribution we combine high temporal resolution digital camera imagery, eddy-covariance, and meteorological measurements with weekly field-based hyperspectral and LAI measurements to gain new insights on temporal changes in canopy structure and functioning of two managed ecosystems in California’s Sacramento-San Joaquin River Delta: a pasture infested by the invasive perennial pepperweed (Lepidium latifolium) and a rice plantation (Oryza sativa). Specific questions we address are: a) how does year-round grazing affect pepperweed canopy development, b) is it possible to identify phenological key events of managed ecosystems (pepperweed: flowering; rice: heading) from the limited spectral information of digital camera imagery, c) is a simple greenness index derived from digital camera imagery sufficient to track leaf area index and canopy development of managed ecosystems, and d) what are the scales of temporal correlation between digital camera signals and carbon and water fluxes of managed ecosystems? Preliminary results for the pasture-pepperweed ecosystem show that year-round grazing inhibits the accumulation of dead stalks causing earlier green-up and that digital camera imagery is well suited to capture the onset of flowering and the associated decrease in photosynthetic CO2 uptake. Results from our analyses are of great relevance from both a global environmental change and land management perspective.