Environmental Modeling

Investigating the complexities of our environment requires innovative tools. Innovation has been at the forefront of our success and dominates the development goals of Clark Labs.

Following this commitment to innovation, IDRISI includes a wide range of cutting-edge modeling tools. The integrated Earth Trends Modeler (ETM) application provides new tools for the observation, exploration and analysis of time series data to uncover trends in the environment, particularly as they relate to global climate change.

The integrated Land Change Modeler (LCM) application guides the practitioner through the complex and challenging steps of modeling land use change scenarios. Using state of the art tools and algorithms, past land cover change can be utilized to inform projections of future land cover change.  This functionality is critical for initiatives such as REDD (Reducing Emissions from Deforestation and Forest Degradation). LCM also includes tools to project how these change scenarios will impact biodiversity.

A variety of other modeling utilities are also available such as Macro Modeler, which employs a graphical modeling environment for executing multi-step functions and Windows COM-compliancy for developers to employ batch processing using Python or any COM-compliant programming interface.

Application areas include:

  • Environmental management
  • Land cover change analysis and prediction
  • Land planning
  • Habitat impact analysis
  • Climate trend analysis

Analytical Examples:

 
EEOT-Earth-Trends-Modeler-S

Earth Trends Modeler in IDRISI is used to uncover trends in the environment, particularly as they relate to global climate change. This graphic is the result of Extended Empirical Orthogonal Teleconnection analysis of anomalies in sea surface temperature and anomalies in lower tropospheric temperature over the 1982-2010 period.

Back to Top

 
Land-Change-Modeler-SimWeight-S

Land Change Modeler in IDRISI models land use change scenarios. The SimWeight empirical transition potential modeling procedure in LCM based on a modified K-nearest neighbor machine learning algorithm, shown in this graphic, can be used to model these scenarios along with Multi-Layer Perceptron neural network and logistic regression options.

Back to Top

Clark University