EcoVis is software for evaluating and optimising economic and environmental management of extensive Queensland grazing properties. It is a high resolution, spatially explicit model that combines cadastral boundaries of farm management units, natural resource mapping and remote sensing, with local user knowledge of resource attributes and animal grazing behaviour. The software allows the expert user to explore strategic long-term management goals and short-term tactical decisions. EcoVis can optimise the objectives of maximising animal production returns, while minimising impact on natural resources and biodiversity, and allowing opportunities for carbon sequestration. Specific scenarios can test climate-related rules for animal movement, and farm-level stocking and de-stocking against objectives for production, soil loss and vegetation conservation.
EcoVis combines a daily time-step grazing and pasture model with an interactive, intuitive, context sensitive, and highly graphical user-interface. It has been developed to make simulation technologies more accessible to both modellers and decision-makers, and is designed as a platform for future model development. It aims to overcome limitations of existing modelling technologies through: (1) simplifying the process of configuring modelling scenarios; (2) by allowing complex spatial analyses to run on workstations and PCs (as opposed to super-computers); and (3) by providing a range of spatial and temporal visualisation tools to reduce complex outputs into easy-to-understand outcomes. Design solutions have been implemented using state-of-the-art technologies providing great flexibility for users to rapidly assess and compare the impacts of different land-uses and management practices on production, environment and ecology.
To simplify the setup and configuration of land-use management scenarios, EcoVis incorporates a range of custom-designed graphical controls, tools and reports for interfacing the input data. These include “dashboards” for composing scenarios, map-based controls for assigning data to regions, and map-editors for changing property and paddock boundaries. Model parameterisations can be progressively disclosed and edited through the interface and linked directly to geographical locations using the inbuilt mapping tools. Feedback on input data configurations is provided to users through detailed reports and “dashboards”.
EcoVis employs an innovative new graphics-card based solution to allow computationally intensive spatial models to run on PCs and workstations, as an alternative to high performance computers (HPCs). Currently, the model used in EcoVis is derived from the AussieGrass (REF HERE) grazing and pasture model, which is the culmination of 30 years of development by Queensland Government scientists. Up until now, this model required the use of HPCs limiting access to potential users and slowing its development. In contrast, EcoVis uses a new Compute Unified Device Architecture (CUDA)-based technology to run the model on discrete nVIDIA® graphics card(s), utilising their multi-core architecture to provide super-computer-like performance on PCs and workstations. This allows more people to use the technology and offers new opportunities to further evolve the model to better account for spatial associations of data (such as routing of water). Coupled with the ability to easily define new data types and rapidly assess outputs, EcoVis provides tremendous power for Queensland modellers to advance the modelling technologies.
EcoVis also incorporates a wide range of temporal and spatial analyses to simplify the interpretation of model outputs. This includes interactive analyses for visualising and comparing point-based and spatially averaged time-series data (including line-plots, scatter plots, histograms and probability distributions) as well as map-based analyses for spatial data. EcoVis allows direct comparison of spatial and temporal outputs for different scenarios through the ability to efficiently layout, stack, or overlay selected outputs, while detailed reports are available for many of the analyses. Several “operations-research” analyses have been included to aggregate, summarise and compare key outputs to aid in decision-making.