Current 3D software is highly adapted for modeling artificial objects, such as structures and buildings. However, a significant part of landscape design is about the natural environment, which creates a challenge for the existing 3D modeling software. This thesis has studied the complexity of the natural environment with virtual landscapes, through testing the modeling of a traditional landscape architectural project, a park. The aim was to research, how are the modern modeling tools capable to depict the intricacy of landscapes. Previous research of landscape modeling and virtual landscapes was surveyed, and the scarceness of earlier studies support the need for this thesis.
As the digital world mimics the physical world, the elements of a real landscape were discovered first. Multiple sources were surveyed, and a categorizing of five different elements of a landscape was established: landform, structures, water, vegetation and atmosphere. These five components functioned as the base of this research.
The modeling of a landscape started by researching earlier studies about landscape modeling, to explore the current state of 3D tools. Thereafter, the surveyed theory was connected to the practice, and the tools were tested by replicating the five elements of a landscape with a 3D software. The purpose was to recognize the most problematic components to transform into digital form, and to find an explanation for the problems. The common nominator for the complications were found to be caused by the changing appearances of the elements though time, in other words the more dynamic the element, the more difficult it was to digitally duplicate.
After the problem was recognized, the research proceeded to study how could these dynamics be presented with other virtual methods. The complexity of the matter required an untraditional tool, a virtual reality software with its four-dimensional approach to be used. By testing the replication of the five elements again, this time with VR, the most problematic components were once again categorized. VR was able to depict the previously recognized, problematic elements without major complications, but the scale of difficulty was still affected by the dynamism factor.
A comparison of the results examined the differences and similarities between the methods, and two scales were concluded. A scale of modeling difficulties showed that although some of the elements are easier to model in 3D, the hardest elements, vegetation and atmosphere, are both created much easier with VR. The scale of how well the dynamism and realism were depicted with the modeling methods was unambiguous, the VR was superior in presenting a realistic landscape.
The objective of this research was to discover a technique to present the natural environment equally realistically than the static, artificial world. Although VR is good at displaying life-like scenarios to the viewer, these types of software are not intended to be used as an exact designing tool currently. These limitations create a further need for 3D software. Yet, 3D is not able to display the dynamism of nature, which may impact the way the design is presented. All in all, VR software have a lot of potential to develop towards useful designing tools for landscape architects. Perhaps all the currently required programs can be found in one package in the near future.
https://aaltodoc.aalto.fi/handle/123456789/101838
