Tuesday, May 29, 2012

The features of our model and modelling system


The general equation of the model

Our Gross Primary Production (GPP) model design builds on the well founded and tested light use efficiency approach and incorporates recent developments in our understanding of the relationships between Carbon dioxide (CO2), Vapour Pressure Deficit, Soil Moisture and Photosynthesis. More specifically we recognise that;

→ VPD appears more predictable than previously thought.
→ SM appears to modify the VPD stomata relationship differently between plant functional types.
→ Soil dryness also reduces Vcmax and Jmax, compared to what would otherwise be predicted from stomatal conductance alone.

The general equation behind our modelling approach can be written as follows;

GPP = fAPAR . IPAR . E . f(T, CO2, VPD, SM)

Where fAPAR is the fraction of absorbed photosynthetically active radiation, IPAR is the incoming photosynthetically active radiation (400-700nm), E (epsilon) is an estimate of light use efficiency which is a function of Temperature (T), Carbon Dioxide (CO2) Vapour Pressure Deficit (VPD) and Soil Moisture (SM).


Key features of the modelling system

1.     Capable of reading in a range of datasets from the TERN and Bureau of Meteorology and re-grid them as required; a.k.a. data abstraction.
2.     Be usable stand-alone, embedded in a workflow or ported to a new model core; a.k.a. modular in design.
3.     Facilitate point-to-point calibration and validation against OzFlux tower data across a range of spatial and temporal scales.
4.     Output any metadata elements in a human readable format.
5.     Written in an open source language with cross platform interoperability.

Architecture





Technology

The computational requirements of this project are overshadowed by the need for usability and transparency- given our open access and source approach. As such, we identified the two most widely used open source platforms in our core user group: R-Project and Python. R was chosen by the development team given it is considered more widely used for benchmarking within our local user community, for example: PALS.

Data abstraction will be done using the Geospatial Data Abstraction Library (GDAL). In the R version, RGDAL bindings and the raster package will be used for low-level reading and writing of data and to facilitate the high-level model functions required for the model and modelling system functionality.



This work is licensed under a Creative Commons Attribution 3.0 Unported License.

This project is supported by the Australian National Data Service (ANDS). ANDS is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy Program and the Education Investment Fund (EIF) Super Science Initiative.

For more information visit the ANDS website ands.org.au and Research Data Australia services.ands.org.au.


Saturday, May 26, 2012

How will we know if our model works?

Simply put, models are notorious for “getting it wrong” and equally well recognised as the only way we know how to “predict the future”. That said, our GPP model isn’t predicting the future, it’s a hind-cast model, re-analysing observations and quantifying primary production based on our thorough understanding of climates past.

Spanning recent decades, our hind-cast model will produce moderate resolution monthly epoques of GPP accumulation based on remotely sensed estimates of greenness and absorbed radiation and Bureau of Meteorology and CSIRO climatology’s of the land surface. We will calibrate and validate our model against the OzFlux network of towers- the only internationally recognised way to observe carbon fluxes for biomes.

Our model provides a baseline for comparison against other approaches to modelling GPP, i.e. state-of-the-art, high-temporal resolution models capable of resolving carbon fluxes at more frequent time steps and higher resolutions- in other words, the complex approach.

Our model is uniquely simple because it can be at 5km on monthly time scales. It’s usability comes from its malleable, simple design which can be modified or repeated with ease.

Our challenge: If you can’t do it better than a simple model- then why bother?

Our simple model and modelling system will be released into the public domain so that it can be benchmarked by the world.




This work is licensed under a Creative Commons Attribution 3.0 Unported License.


This project is supported by the Australian National Data Service (ANDS). ANDS is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy Program and the Education Investment Fund (EIF) Super Science Initiative.

For more information visit the ANDS website ands.org.au and Research Data Australia services.ands.org.au.


Thursday, May 24, 2012

Who uses modelled GPP?

Consumers of GPP products and systems designed to parameterise-your-own models are traditionally ecosystem scientists, ecohydrologists and ecosystem modellers. However, these data and systems are commanding a broader audience as scientific, economic, agronomic and political communities seek to better understand and model the carbon cycle. It follows that each of these diversely different communities has a variety of end-use applications. As such, we expect our GPP product to have a variety of end uses, including, but not limited too;
  1.  Benchmarking systems for comparing ecosystem and ecohydrology models, for example, the International Land-Atmosphere Benchmarking Project (iLAMB).
  2.  Demonstration and evaluation of a range of remote sensing data streams characterising the primary drivers, absorbed photosynthetically active radiation and climatology’s.
  3.  Large-scale (i.e. continental and global) carbon balance and water resources assessment.
  4. Development of CarbonTracker Australasia and contribution to other regional carbon modelling activities.
  5. Information about primary production of natural and managed ecosystems.
  6. Demonstration as high-profile application of Terrestrial Ecosystem Research Network data streams.
Common amongst all users is the need to assess spatially explicit changes in GPP over time. Specifically, this means ability to visualise and farm data from the time series imagery and process it further. Our strategy is to focus on addressing this common need robustly rather than developing a range of potentially limited use end-user functionalities.



This work is licensed under a Creative Commons Attribution 3.0 Unported License.


This project is supported by the Australian National Data Service (ANDS). ANDS is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy Program and the Education Investment Fund (EIF) Super Science Initiative.

For more information visit the ANDS website ands.org.au and Research Data Australia services.ands.org.au.

Tuesday, May 22, 2012

Why model Gross Primary Production?


The aim of Primary Production in Space and Time is to address the national need for information on the temporal-spatial distribution of the amount of Gross Primary Production (GPP) across Australia.

GPP is the flux of carbon from atmospheric carbon dioxide into green plants through photosynthesis and we will model GPP fluxes on a monthly timescale at 5km resolution.

Quantifying GPP is critical to quantifying the dynamics of our diverse natural and semi-natural ecosystems.


This work is licensed under a Creative Commons Attribution 3.0 Unported License.


This project is supported by the Australian National Data Service (ANDS). ANDS is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy Program and the Education Investment Fund (EIF) Super Science Initiative.

For more information visit the ANDS website ands.org.au and Research Data Australia services.ands.org.au.

Sunday, May 20, 2012

The Project Team

This ANDS-funded project will fuse disconnected data sources weather, remotely sensed land-surface observations, CO2 and water flux measurements, hydrograph data and remotely sensed CO2 concentrations—to generate a time-varying field of gross primary production across the Australian continent. It will develop a software infrastructure allowing different ecosystem models to be compared with one another and with data, and a specific realization—a near real-time GPP hindcast at 0.05˚ resolution—with proven, simple light use efficiency and water balance models at its core. The project deliverables will be of strategic value to climate and carbon policy makers in Australia, and of great utility for benchmarking ecosystem models in Earth system science.


Professor Colin Prentice - Macquarie University

Project Manager
Professor Colin Prentice
Macquarie University




My work concerns the global terrestrial biosphere and its dynamic interactions with the atmosphere and climate. Much of my research centres on the modelling of ecosystem processes and scaling up from processes at the level of plants and soil micro-organisms to describe the large-scale exchanges of water, carbon dioxide and trace gases between the atmosphere and land. I have pioneered the development of global plant geography models (as the leader of the team that developed the BIOME family of models) and, since then, the development of global models to represent vegetation dynamics (the LPJ and LPX models). My present research focuses on the "next generation" of ecosystem and land surface models, incorporating new developments in ecology and ecophysiology including optimal allocation theory and plant hydraulics, and on developing tools for model evaluation and improvement using a variety of data sources ranging from eddy flux and atmospheric trace gas concentration measurements to isotopic measurements, remote sensing and palaeodata. I am also working on the quantification and analysis of sources and sinks of carbon dioxide, methane, nitrous oxide and reactive trace gases, and biogeochemical cycles and feedbacks to climate.




Professor Alfredo Huete

Scientific Advisor
Professor Alfredo Huete
University of Technology Sydney

My main research interest is the use of remote sensing to study and analyse broad-scale vegetation health and functioning. I use high frequency satellite data to observe land surface responses and interactions with climate, land use activities, and major disturbance events. I also look at vegetation dynamics and landscape phenology processes to detect shifts in seasonalities under global change conditions. My recent work on satellite-based analyses of  tropical rainforest and savanna phenology patterns was featured in a National Geographic special entitled "The Big Picture". Currently my research involves coupling eddy covariance tower flux measurements with field spectral sensors and satellite observations to better understand carbon and water cycling across Australian landscapes.  I am actively involved with several international space programs in the U.S., Japan, and Europe and lead the Sydney node facility within AusCover- TERN.

Dr Helen Cleugh - CSIRO Marine and Atmospheric Research

Scientific Advisor
Dr Helen Cleugh
CSIRO Marine and Atmospheric Research (CMAR)


Dr Helen Cleugh is leading research that explores the interactions and feedbacks between the land surface and the climate system. A scientist with CSIRO since 1994, Dr Cleugh is currently Deputy Chief in CSIRO’s Division of Marine and Atmospheric Research. She is also Deputy Director of The Centre for Australian Weather and Climate Research. The Centre is a jointly managed research partnership between the Bureau of Meteorology and CSIRO.
Dr Cleugh is also part of a dynamic and highly productive research team that maintains long-term measurements of carbon exchanges and water use in a variety of Australian ecosystems, including forests, vineyards, savannas and city suburbs. These measurements are needed to observe, understand and model the dynamics of carbon, water and energy cycles in Australian ecosystems; and explore the effects of climate variability and change – especially the vulnerability of land-based carbon sinks.





Scientific Advisor
Dr Sarah Mikaloff Fletcher
National Institute of Water & Atmospheric Research (NIWA) (NZ)






Dr Sara Mikaloff Fletcher earned her PhD at the University of Colorado, Boulder, where she used atmospheric observations and models to estimate methane emissions to the atmosphere. She employed similar techniques to determine air-sea fluxes of CO2 using ocean interior data and ocean general circulation models during her postdoctoral work at the University of California, Los Angeles. After finishing her post doctoral work, she joined the Atmospheric and Oceanic Sciences Program at Princeton University where she used atmospheric and oceanic models to study the past and present carbon cycle. In January of 2010, she and her family moved to New Zealand, so that she could take up a position at NIWA.








Dr Natalia Restrepo-Coupe - University of Technology Sydney

Dr Natalia Restrepo-Coupe
University of Technology Sydney


After receiving my PhD in 2005 I moved to the University of Arizona to work as a Postdoctoral Fellow under the supervision of Professor Scott Saleska researching the phenology and seasonality of ecosystem productivity and evapotranspiration in the Amazon Basin. At UTS, I am now working in Prof. Alfredo Huete’s C3 Ecological Modelling and Remote Sensing Group, where we are integrating remote sensing observations (from both tower-mounted optical sensors and satellites) with field eco-hydrologic and tower-based flux measurements (eddy-covariance, EC). Our goal is to study and understand seasonal and inter-annual patterns of evapotranspiration and photosynthesis in different Australian ecosystems.



Kevin Davies
University of Technology, Sydney


Kevin has a background in Information Technology and Environmental Management, having completed a B.Info.Tech (UTS) and an M.App.Sci (University of Sydney) specialising in GIS and Remote Sensing. He previously worked with the University of Sydney as GIS Coordinator for the ‘Living with Heritage’ ARC industry project supporting heritage and archaeology research in Angkor, Cambodia. He joined the Ecological Modelling, Remote Sensing, and Terrestrial Ecohydrology Research groups within the Climate Change Cluster (C3) group at UTS in early 2011, as an e-Research supported Data Manager, where he provides technical support for the management of multi-scale data generated from satellite imagery, flux-based instrument networks, tower-based phenocams, continuous optical measuring sensors, and model output data. Kevin is completing his PhD in remote sensing with the University of Sydney.



Bradley John Evans
Macquarie University




Bradley has research interests in the fusion of remote sensing and in-situ light and climate observations into simple land surface models. Bradley has researched this topic from leaf to landscape too better understand the spatial and temporal impacts of our changing climate in terms of ecosystem condition and productivity. Bradley has published a methodology for modelling the condition of individual trees using high resolution imagery and statistical modelling of in-situ assessments of crown condition. Bradley is presently completing his PhD part time and working on applying it using landscape and global scale remotely sensed imagery and similarly down-scaled methodologies. Bradley has collaborated with international groups on improving both the models behind global estimations of forest productivity (i.e. Gross Primary Productivity and Net Primary Productivity) and the uptake of these data.





Primary Production in Space and Time has officially started on April 1, 2012 with the appointment of Bradley Evans who will develop and implement the software infrastructure and Gross Primary Productivity (GPP) product, liaising as required with co-developers at CMAR and UTS. As part of the ANDS funding agreement a series of blog postings will document its development and completion.



Bradley will lead a quality control process including provision of summary statistics of goodness-of-fit to the various data sources, and identifying causes of any systematic differences from pre-existing GPP estimates including MODIS (MOD17) and BIOS2. Bradley will write documentation for the software to be posted, together with the code, on this blog.


This project is supported by the Australian National Data Service (ANDS). ANDS is supported by the Australian Government through the National Collaborative Research Infrastructure Strategy Program and the Education Investment Fund (EIF) Super Science Initiative.

For more information visit the ANDS website ands.org.au and Research Data Australia services.ands.org.au.