Ecosystem modelling, scaling, benchmarking and data assimilation for the Australian continent
Abstract
Modelling terrestrial biosphere processes in the world's driest inhabited continent presents some unique challenges, but also excellent opportunities to capitalize on major data infrastructure projects led by the Terrestrial Ecosystem Research Network (TERN). TERN, Australia's counterpart to NEON, has charged its ecosystem Modelling And Scaling InfrasTructure facility (e-MAST) with enhancing the national capacity for model benchmarking, data assimilation and data-model integration. e-MAST models the Australian terrestrial biosphere in space and time from various disciplinary perspectives. The foundation for all e-MAST modelling is ANUCLIMATE, a set of well-founded, high-resolution estimates of key climate variables (minimum and maximum temperature, precipitation, vapour pressure) from the 1970s up to near the present. Using all reliable meteorological measurements, we have enhanced the standard (ANUSPLIN) approach to allow interpolation of all variables at daily time resolution to a common 0.01 degree grid. Satellite data assimilation is used to provide continent-wide fields of derived variables including stomatal conductance, evapotranspiration and soil moisture. Plant productivity has been modelled by fusion of eddy-covariance CO2 flux measurements with satellite reflectance data, exploiting the well-tested light-use-efficiency modelling approach. Working towards the next generation of robust, process-based ecosystem models, we are synthesizing observations of plant biophysical and physiological traits; developing gridded surfaces of these traits; and working with TERN's MultiScale Plot Network to improve national coverage of trait measurements. Evaluation and benchmarking of models is based on the Protocol for the Analysis of Land Surface Models (PALS), which is being extended from its original core of flux data-model comparison ';experiments' to encompass more data types, including remote atmospheric CO2¬ concentrations and streamflow measurements, which (when coupled through atmospheric and river transport models, respectively) can provide powerful constraints on the large-scale behaviour of ecosystem models. TERN is built on data sharing and we promote open-source models, data and publishing as a matter of principle. e-MAST invites you to join us in tackling the challenges facing Australia's unique landscape and the flora and fauna it supports.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2013
- Bibcode:
- 2013AGUFM.B14D..01E
- Keywords:
-
- 0439 BIOGEOSCIENCES Ecosystems;
- structure and dynamics;
- 1615 GLOBAL CHANGE Biogeochemical cycles;
- processes;
- and modeling;
- 1813 HYDROLOGY Eco-hydrology;
- 1640 GLOBAL CHANGE Remote sensing