Biogeochemistry. Understand the changing global biogeochemical cycles of carbon and nitrogen.
Multiple stresses. Understand the responses of ecosystems to multiple stresses.
Biodiversity. Understand the relationship between changing biological diversity and ecosystem function.
Research on global ecosystem processes motivates four broad classes of observations and experimental studies, shown below. As noted in Chapter 2, large-scale measurements in ecology tend to support all of the research imperatives above in a crosscutting fashion, with any one measurement set helping to test a variety of hypotheses. Ties of these measurement areas to the research imperatives are shown in Table 2.2. The four key measurement areas are time series observations of ecosystem state; land use and land cover change; site-based networks; and measurements of diversity, functional diversity, and ecosystem function.
Global time series of vegetation and phytoplankton state, derived from the National Oceanic and Atmospheric Administration's (NOAA) Advanced Very High Resolution Radiometer (AVHRR) and Coastal Zone Color Scanner sensors, for land and ocean, respectively, have proven their value in understanding the seasonal and spatial characteristics, interannual variability, and trends of large-scale biogeochemistry and biophysical processes.2 Space-based measurements of ecosystem state are fundamental in determining the link of terrestrial ecosystems to climate, the biogeochemistry of the land and oceans, and the impacts of climate and other disturbances. While measurements of “greenness” and ocean color are not direct ecological properties, they have proven to be highly correlated with spatiotemporal dynamics of ecosystems. Recent work3 highlights both the utility of these records and the dependence of the science on long and consistent records. Stable calibration and removal of the atmospheric signals of ozone, water vapor, and aerosols are critical to detecting ecological signals. While there is ample room for innovation in land surface remote sensing, stable calibration and correction impose stringent requirements on the sensor or sensors deployed. New instruments, while adding new capabilities, must also be “backwards compatible” to preserve time series. Atmospheric correction requires that coincident observations to quantify water, ozone, and aerosols be available for use in land surface retrieval algorithms. Spatial and temporal resolution for time series instruments are typically a compromise between sufficiently high spatial resolution to resolve ecosystem structure (0.25 to 1 km2) and swath width and data rate limitations associated with near-daily coverage. High temporal coverage is needed to ensure adequate sampling of seasonality, especially in cloudy environ-