What is it about?
This article applies a CNGA to search for multiple configurations of spatial parameters to optimize the transition rules of land use change models.
Featured Image
Why is it important?
To the best of our knowledge, no research exists within the land use change modeling domain dealing with the task of finding multiple optimal solutions and not just one single optimum. Finding multiple “nearly” comparable quality solutions may help to reveal hidden properties in the case under study, such as spatial variations of land use change drivers. This is especially important in those cases where land use change dynamics are characterized by divergent spatial parameters across a reference area. This information is concealed by all models based on one single global optimum calculated for the entire reference area. In addition, multiple solutions constitute a step toward providing more robust solutions for future land use change predictions.
Perspectives
We believe that this work is important for many reasons. First, finding multiple optima is important to discover the relationship between specific subspaces and land use change drivers. Our results provide evidence that the effect of each driver changes magnitude in the various subspaces. Second, the existence of competing solutions should be considered in future forecasting, as each solution may not be stable over time. The case study shows that Solution #2 performs better than other solutions during the period 2000–2010, while it was Solution #1 that performed best during the period 1990–2000. This opens an avenue toward sensitivity analysis, taking into consideration various solutions for future predictions. Third, the findings are important for policy-makers, as they will help evaluate and adopt the location‐based spatial planning strategies for a given region considering several subspaces and their specificities.
Ahmed Mustafa
University of Liège
Read the Original
This page is a summary of: Benefits of a multiple‐solution approach in land change models, Transactions in GIS, October 2018, Wiley,
DOI: 10.1111/tgis.12482.
You can read the full text:
Contributors
The following have contributed to this page
