HR: 1330h
AN: B22A-0796    [PDF]
TI: Can we see the Forest for the Trees? Toward an Ecological/Ecosystem Basis for Remote Sensing of Carbon Exchange.
AU: * Sims, D A
EM: sims48@msn.com
AF: Ball State University, Geography Department, Muncie, IN 47306 United States
AU: Oechel, W
EM: oechel@sunstroke.sdsu.edu
AF: San Diego State University, Biology Department, San Diego, CA 92182 United States
AU: Rahman, F
EM: faiz@bsu.edu
AF: Ball State University, Geography Department, Muncie, IN 47306 United States
AU: Gamon, J A
EM: jgamon@exchange.calstatela.edu
AF: California State University, Biology Department, Los Angeles, CA 90032 United States
AB: Although a top-down approach focusing on the "forest" as a whole would seem to make sense for remote sensing estimates of global vegetation/atmosphere carbon exchange, many current models take a bottom-up, scaling approach that focuses on the "trees". This bias is understandable since most of us, including the authors of this presentation, began our careers studying eco-physiology at the leaf and single plant levels. It may also reflect the scarcity of data relating carbon fluxes to spectral reflectance at sufficiently large scales. This is unfortunate since relationships that are important at small scales are not necessarily the same ones that are most important at larger scales. Although large-scale measurements of carbon flux (from eddy covariance) and reflectance (from satellites) have been available for some time, direct comparisons have been difficult due to mismatches in their temporal and spatial scales. Development of tram systems for measurement of spectral reflectance in the footprints of eddy covariance towers (Specnet) is one approach to providing a better match between these scales. In this presentation we will compare relationships at ecosystem, plant and leaf scales. The degree of correlation between physiological and morphological (canopy structure) characteristics increased with increasing spatial scale. At the leaf and plant level photosynthetic light use efficiency (LUE) is largely independent of vegetation greenness and absorbed photosynthetically active radiation (APAR). However, our measurements at the ecosystem level suggest a strong correlation between LUE and vegetation greenness. Consequently, estimation of LUE as a parameter independent of vegetation greenness may not be as crucial as would be suggested by leaf and individual plant relationships. Although major disturbances temporarily upset the LUE/greenness relationship, we have found that vegetation rapidly re-establishes this equilibrium once the disturbance is removed. Another bias from the plant level is the assumption that greenness is not a useful parameter for estimation of carbon exchange in evergreen dominated ecosystems. However, at the ecosystem level, constant greenness is rare outside the tropics, and even in the tropics, greenness changes in subtle ways with seasonal stresses. Most evergreen coniferous systems include deciduous and annual species that contribute to seasonal variation in greenness of the ecosystem. The controls on respiration also change with spatial scale. Whereas respiration is a function of many factors at small scales, including temperature, moisture and the species of plants and microorganisms, studies at the ecosystem scale are finding respiration to be primarily driven by the input of carbohydrates from photosynthesis. The correlation between temperature and respiration rate largely disappears at larger temporal and spatial scales. At large temporal and spatial scales, acclimation of physiological processes and development of canopy structures that reflect the general availability of resources in the environment result in equilibrium relationships that hold great potential for remote sensing of ecosystem function if we can see the forest for the trees.
UR: http://vcsars.calstatela.edu/SpecNet/index.html
DE: 1640 Remote sensing
DE: 4805 Biogeochemical cycles (1615)
SC: Biogeosciences [B]
MN: 2003 Fall Meeting