|Ph.D Student||Hadar Ben Asher Liat|
|Subject||Visualization Models as Predictive, Data-Based Tools for|
Assessing Future Ecological Landscapes and
Supporting Management Decisions
|Department||Department of Architecture and Town Planning||Supervisors||ASSOCIATE PROF. Daniel Eli Orenstein|
|PROF. Yohay Carmel|
|Full Thesis text|
Human-driven landscape changes strongly influence landscape functionality and aesthetics. While landscape planners have access to biophysical data for decision-making, they often do not have the necessary information about social variables, such as aesthetic tastes, feelings, or functions of a place. Visualizing future landscapes under alternative management scenarios could be a valuable tool for aiding land management decisions. Towards these ends, empirical, quantitative ecological data on vegetation composition, pattern, and processes in a Long-Term Ecological Research (LTER) site in Israel were integrated into computerized, 3-D representations of current and future landscapes.
Our objectives were (1) to visualize landscape-shaping processes, such as wildfire, grazing, and species colonization, to assist managers, planners, and the public to envision the long-term visual significance of management alternatives; (2) to validate the similarity between the 3-D model and reality, and; (3) to study the unique contribution of the visualization tool to decision-making processes regarding natural resource management, and how such models can mediate between objective features of landscapes and the way they are perceived by different audiences.
The visual model we developed is based on 30 years of scientific knowledge and ecological data describing vegetation processes in Ramat Hanadiv, a case study of ecological conditions and processes relevant to the Mediterranean and other complex ecosystems worldwide.
Validation was performed by comparing ‘current state’ model representation with real-world photos from the perspective of the observer. The model was found to be a valid representation of reality.
The contribution of the visualization to decision making, its impact on the nature of decisions, and the confidence level of respondents from different backgrounds and organizations were examined experimentally (N-176), compared to responses when participants were provided with scientific data through conventional tools (executive summaries, graphs, and GIS maps).
The visualization significantly increased the confidence level of respondents compared to those who received only conventional tools, as it allowed respondents to see the long-term management results of their decisions and reduce the uncertainty. In contrast, our hypothesis that the visualization would influence management decisions towards greater intervention in nature was not supported. The visualization, when significant, was a moderating factor that reduced the tendency of respondents to choose an intervention management strategy.
However, the visualization did not operate as a universal language and management decisions were largely reflective of the professional background and organizational affiliation of the respondents. The visualization mainly influenced the confidence of respondents with a planning background, compared to those with a scientific background who presented high confidence level even without the visualization and were not affected.
Contrary to our expectations, the visualization did not affect the responses of the public group, who preferred executive summaries and sought further processing and mediation of the scientific information.
Looking to the future, I suggest that the ability to create future landscapes using scientific data can assist to improve decision-making processes, balancing ecological and social needs.