Speaker:
Isaac Kwadwo Nti and Philip Sallis

Date(s):
03 Sep 2012 to 03 Sep 2012

Topic:

Land-Use/Cover Change (LUCC) modelling helps to depict complexities and interdependencies of components inherent in geospatial systems where potential terrestrial modifications may be made. Also Dynamic LUCC models are purposely modelled to simulate the changes of natural and or artificial environment. They can provide valuable insights into possible land-use configurations for the future.  A LUCC model framework for Auckland Region is required for policy makers to know the empirical trends of landscape change to make future decisions for the region that are coupled with socio economic data.

In this paper, a demonstration of the working process and results of LUCC modelling of Auckland Region is presented. Using a Land Cellular Automata (LAC) Modelling approach coupled with remote sensing technologies and GIS for data preparation and analysis, Markov Transition Matrices and Weights of Evidences (WoE) of coefficients were used to calibrate the model. Multiple Window Constant Decay Function of map comparison method was used to validate the model.  Empirical Land Use Maps for 1990 and 2006 were used to calibrate the LUCC whilst maps of 2008 were used for validation. Using discrete time steps simulations of land-use maps into the future are generated in 2D maps and or animated formats.

The raster land-use map of Auckland Region is categorised into seven(7) classes – Natural Forest, Planted Forest, Grassland, Cropland, Wetlands, Settlements and Other Lands. Each land-use cell of the raster map has the probability of transiting into another state of land-use at each discrete time step based on the neighbourhood cells and transition rules generated during calibration. In case of continuous time, transitions occur one after another, where the current transition is used as the seed time and map for subsequent change.

This framework can be used for environmental impact assessment and socio economic scenario building to determine future trends.

 

Download the Presentation [PowerPoint 2007 Presentation - 4.2 MB]