Objectives

We find four main limitations of these existing frameworks. First, they are often based on ad hoc categories or classes, with little or minimal attention to establishing explicit measures. Second, they are based on discrete categories or classes, and so cannot resolve small differences. Also, because they are based on class (or nominal) data, they cannot be easily integrated at different spatial scales to examine landscape context. Third, the surrogate spatial variables that are used as a basis for classifying are generally too coarse grained to capture fine-grain patterns, ultimately limiting detailed characterization. Finally, and, especially, most approaches examine the in situ landscape content of a park or protected area (PPA) using levels of threat or protection. Typically, an individual PPA receives a single score, regardless of size or the types of adjacent land uses. For example, although national parks often have an intensely used, built-up portion, (e.g., Yosemite Valley in Yosemite National Park) directly adjacent with designated wilderness, typical protocols, and methods result in only a single value being applied to the entire PPA. In addition to content, we believe that the landscape context of a PPA is important as well. This aligns analytical methods with the growing concerns of land managers over these external threats. For example, two of the main threats to the National Forest System (loss of open space due to housing development and invasive species) are external (Bosworth 2004), and changes in surrounding land use are a leading threat to resources in United States national parks (GAO 1994).

We aim to address these limitations by building on the human modification framework (HMF, Theobald 2004). The HMF is based on two dimensions: the degree to which natural processes are free or controlled, and the degree to which landscape patterns are natural or artificial. To provide useful and tighter coupling of specific threats and spatial data surrogates, we refine the conceptual framework by identifying three primary types of human activities that cause human modification of natural systems and patterns. These are land uses associated with urban/built-up, recreation, and production/extraction. We detail specific metrics and common data used as surrogates that can be used to implement each of these factors to provide a stronger basis for characterizing the degree of human modification. Our overall goal in this paper is to develop a general yet simple method to characterize threats associated with human land use activities for managing landscapes and to provide metrics and more directly capture threats in a detailed, spatially explicit manner. Our objectives are to describe factors that can be used to refine the HMF, identify common surrogates of these factors that can be mapped at relatively fine grain, conduct a spatially explicit analysis of the landscape context, and illustrate these methods using a case study example for the State of Colorado.