|
|
- Info
Evaluating visible spatial diversity in the landscape
| Author: |
Palmer, James F.; Lankhorst, Janneke Roos-Klein |
| Date: |
1998 |
| Periodical: |
Landscape and Urban Planning |
| Abstract: |
It is Dutch national policy to maintain the landscape's identity. Spaciousness or landscape enclosure is the characteristic of identity investigated in this study. This study develops and validates a GIS model for evaluating and monitoring the spaciousness of Dutch landscapes. Six areas are selected to represent a range of landscape conditions. The degree of spaciousness from confined to exposed is evaluated in the field at approximately 12 views in each of the six study areas. The reliability and validity of these spaciousness ratings are demonstrated by a high correlation with ratings from 30 additional people of photographs of these 72 sites (r^2=.70). These ratings are matched with attribute data from the 25 cells corresponding to the near and middle distance zones in each of these 72 views. A raster GIS database is created for each area. The attributes recorded for each cell are the length of tree rows, and the area of woods, isolated buildings and urban blocks. Three related GIS models predicting landscape spaciousness are developed. The principle used to structure the model algorithms is that initial introduction of a small amount of landscape objects into a cell creates a rapid increase in enclosure. Translation matrices are used to combine the landscape attributes into an overall enclosure rating. The Immediate (500) and Immediate (250) models describe nearly half of the spaciousness variation seen in the field (r^2=.45 and .49, respectively). The Neighbor model is a revision of the Immediate (250) approach that incorporates the influence of the surrounding eight cells. The Neighbor ratings describe over half of the observed variation in spaciousness (r^2=.54). This study demonstrates that perceived spaciousness is closely related to the degree that the landscape is filled with objects, such as trees and buildings. It is anticipated that results will improve with availability of more accurate digital data. The next steps include: (1) applying the approach to other aspects influencing landscape identity, such as urbanization, (2) developing a system that uses this information to monitor and manage landscape quality, (3) exploring the effect of possible land use change scenarios on landscape identity, and (4) investigating the possible existence of perceptions related to differences in landscape use or region. |
|
