Comparing Environmental and Ecological Cross-Scale Structure Through Data Modelling of State Space

Authors: J. Stallins*, University of Kentucky
Topics: Geographic Theory, Environmental Science, Biogeography
Keywords: Scale, data, topology, resilience
Session Type: Poster
Day: 4/5/2019
Start / End Time: 3:05 PM / 4:45 PM
Room: Lincoln 2, Marriott, Exhibition Level
Presentation File: Download


One way to account for ecological scale in geographical analyses is to hierarchically partition data collection and/or analyses into discrete, a priori units of known extent and resolution. In this way, scalar extent and resolution become decomposed into the framing of questions and deployment of the analytical framework. A second way to account for ecological scale is to use spatial autocorrelation functions to identify the extents at which correlative relationships between variables are maximized. However, both approaches prioritize, not unproblematically, singular discontinuous scalar extents and resolutions. While it is important to identify the key scales at which a specific property of a habitat or landscape is expressed, they overlook the explicit cross-scaling of structure and processes that operates in ecosystems. Consequently, emergent discontinuities may be replaced or obscured by those derived through human-imposed measurement levels or by spatial autocorrelation distances that collapse cross-scale processes into a single metric. We present a data modelling approach whereby multiple scalar extents and resolutions can be embedded in the data structure for a particular phenomenon. By creating a multidimensional state space from different measurement levels and aggregations of data informed by particular conceptual paradigms, the topology of the data can be used to infer the presence of less artificial discontinuities across scalar extents. This data modelling approach facilitates simultaneous use of different kinds of variables at varying extents and resolutions, along with the conceptual paradigms affiliated with them. Topography from barrier island dunes is used to illustrate this data modelling method

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login