Environment and Planning B: Urban Analytics and City Science

In this paper, part 2 of a three-part series, we present the formal specification and calibration results of the land-use change component of the second-generation California urban futures model. The land-use change component consists of a series of nonordinal multinomial logit models of site-specific land-use changes. These models incorporate spatial measures (for example, mix of adjacent land uses, and proximity to similar activities) as well as local policy and accessibility measures. Various model forms are developed and explained, as are multiple approaches to evaluating the equation for goodness-of-fit.

Architectural mass is widely discussed and regulated but there have been few experimental studies on just which design features influence impressions of architectural mass. I investigated the effect of four design features (visual area, partitioning, fenestration, and facade articulation) on comparative choices made on the basis of the criterion of mass. Results suggest a large effect for visual area, a moderate effect for fenestration, and very small effects for facade articulation and partitioning.

In this study Ming-style chairs are used as examples to investigate the critical factors constituting the style of products. In an experiment, subjects were asked to compare varied pairs of Ming-style chairs. These comparisons were then analyzed by using the methods of multidimensional scaling and multiple regression analysis. The results show that the perception of Ming-style chairs between designer subjects and nondesigner subjects are similar and that the three main dimensions constituting the subjects' perceptual space are ‘elementary feature’, ‘emotional image’, and ‘use feeling’. There exist some common critical elementary features among Ming-style chairs. The degree of style similarity (being perceived as the same style) between a pair of chairs will increase with an increase in the number of critical features shared. The influential powers to this perception of similarity are different among these features.

Existing parameters that measure the connectivity of elements in landscape networks are dependent upon the size of the network, defined as the number of locations in it. These parameters are unsuitable for comparing the connectivity in networks that have different numbers of locations. The objective of this paper is to extend the existing parameters, so that the degree of connectivity of elements in a variety of networks can be compared. These parameters can be used to test relationships between the connectivity of locations and their function in the region and to compare these relationships across several locations in different regions as well as in one location in time. In simulations, we show that the size of the graph affects the values of the parameters derived. However, in contrast to the existing parameters, the degree of connectivity of elements in differently sized networks can be usefully compared. The parameter based on shortest-weighted paths can also differentiate between differently spaced networks.

It has been argued that urban sprawl leads to inefficient land-use patterns with leapfrogging and low-density development in the urban fringe. Communities can implement a number of growth-management programs to encourage more efficient and environmentally sound development patterns. However, a long time period is needed to monitor the progress of growth-management techniques and to determine if they have achieved the intended effects. Unfortunately, urban development is mostly an irreversible process that leaves no room for experiment. As a result, we use GIS to simulate the impact of various growth-management policies on future land-development patterns in Portage County, Ohio. A comprehensive database and population growth forecasts are compiled to support the GIS analyses that generate a set of alternative residential growth scenarios. As a spatial decision support system, the GIS is found to be an efficient tool to simulate the form and process of development for detailed assessment and planning. As an educational tool, development simulation with GIS allows the generation of alternative scenarios to help citizens and local officials make sound planning decisions today, define future policies, and plan for new growth.

Under a typical riparian landscape structure where a vegetated buffer is sandwiched between a greenway trail and the river bank or lake shore, a trail user's perceptions about riparian scenic quality are influenced by, among other factors, buffer-induced setback effects. As a trail is set back to leave enough space for the vegetated linear area, both the subject and the composition of a trail user's viewshed can change significantly. This may in turn affect his or her perceptions of the lakescape or riverscape. For greenway trail planners, the buffer-induced setback effects become most relevant when they have the option of allocating trails behind either a legislature-regulated buffer with constant width or a scientifically recommended buffer of variable width. In North Carolina's Piedmont region, the discrepancy in width between constant and variable buffers may range from 10 to 200 m, large enough to cause significant visual variances. In this paper a method to investigate these setback effects is presented. Based on Shafer's landscape-photography approach, a technique called BISEA (buffer-induced setback effect assessment) is designed to inquire and represent two pieces of information about the setback effects. These are: (1) the subject and composition changes in prospective trail users' viewsheds caused by buffer-width variations; and (2) the impacts of these changes on riparian scenic quality as perceived by trail users. Implementation of the method is facilitated by a computer-based support system that comprise tools for landscape photography (a digital camera), survey [a global positioning system (GPS) receiver], image processing (a digital image-processing program), database management and visualization (a GIS), and scenic quality assessment (a spreadsheet program).

Wayfinding is a basic activity that people do throughout their lives as they navigate from one place to another. Many theories of spatial cognition have been developed to account for this behavior. However, most of the computational models focus on knowledge representation (for example, cognitive maps) and do not consider the process of how people structure wayfinding tasks and space. In this paper we present a computational method to compare the complexity of wayfinding tasks in built environments. As a measure for such complexity, we use a simple wayfinding model that consists of two critical elements: choices and clues. We show that elements of people's perception and cognition can be used to determine the elements of the wayfinding model and, therefore, to compare the complexity of wayfinding tasks in built environments. A case study of wayfinding in airports demonstrates the applicability of the method. The integration of this method into the computational design process of built environments will help to identify architectural problems with regard to wayfinding prior to construction.