Close Window

CD-09 Abstract

A Modeling Synthesis of the Impacts of Tropical Forest Conversion on Carbon Fluxes and Storage, and on Nutrient Dynamics in Amazonia

Edward B. Rastetter — Marine Biological Laboratory (US-PI)
Yosio Edemir Shimabukuro — INPE - Instituto Nacional de Pesquisas Espaciais (SA-PI)


Our objectives are to quantify and improve our understanding of the changes in C

storage, C exchange, and nutrient dynamics in response to human-induced environmental

change in Amaz˘nia. We will address these goals using models of canopy gas-exchange and

ecosystem biogeochemistry. These models will allow us to use existing information and

information collected during the intensive field studies to generate regional

extrapolations, thus creating linkages in LBA between the specific point studies and the

remotely sensed data.

We will use a hierarchy of models to scale in space from point to region and in time

from hours to decades.

  1. At the finest scale we will employ a process-based multi-layer canopy,

    soil-plant-atmosphere model (SPA) that predicts hourly gross photosynthesis and

    transpiration. This model is parameterized with fine-scale measurements of canopy

    structure and meteorology, and can be tested against independently collected tower eddy

    covariance flux measurements. We will also develop a transect version of the model to

    generate predictions of gas exchange along an aircraft flight path. The transect runs will

    necessarily be parameterized with less detailed data, and will allow us to investigate the

    effects of any simplifying assumptions on model efficacy.

  2. At the next coarser scale, we will use a simpler, aggregated model of daily whole-canopy

    carbon ę fixation and transpiration. This model is derived wholly from the process-based

    model, capturing its aggregated behavior, while operating with daily input and at

    significantly higher computing speeds. This aggregated model serves two purposes. Firstly,

    it can be employed within a gridded GIS database, so that we can derive regional estimates

    of C and energy fluxes under specified land-use and climate change scenarios. Secondly, it

    can be inserted into our ecosystem biogeochemistry model as a validated

    photosynthesis/transpiration component.

  3. The third model, of ecosystem biogeochemistry (MBL-GEM), will be used to examine the

    coupling between nutrient dynamics and longer-term C storage and exchange. Our primary

    goal will be to test whether we can predict the long-term responses of biomass and soil

    organic matter to land-use change using chronosequence data for validation. This model

    will also be coupled with a GIS-database model to generate regional extrapolations. In

    this instance our predictions of regionally varying biomass and leaf area can be tested

    against remotely sensed data as a further test.

The proposed work will represent a significant step toward an understanding of

Amaz˘nia for three important reasons. Firstly, by identifying key functional differences

in canopy gas exchange characteristics our soil-plant-atmosphere model will help to

develop effective inter-site comparisons. Secondly, our models represent a logistically

effective means for generating linkages among components of LBA, particularly in terms of

linking the point measurements of towers with remotely-sensed regional data-sets. Our

aggregation protocols will help to generate coarser-scaled predictions, with more modest

data requirements, better suited for extrapolation. Thirdly, we will integrate the

understanding derived from tower-based flux measurements into our ecosystem biogeochemical

models. Understanding the linkages between C and nutrient cycles is crucial for developing

long-term predictions of the response of biomass and soil organic matter to land-use and

climate change.

We hope to work closely with those teams making eddy flux measurements, both from

towers and aircraft. We are also interested in collaborating with groups developing

regional data sets of land-cover, vegetation type and climate variables. We are searching

for a Brazilian or South American collaborator to work with us on model parameterization,

validation, analysis, and extrapolation.

Close Window