2022 Seed Grant Recipients

Investigators:  Dr. Christie Bahlai (College of Arts and Sciences), Dr. Kayla I. Perry, and Lama Y. Tawk (College of Arts and Sciences)

Flowers with a building in the background

Differences in the conceptualization of ideas, including language and methodological approaches, can mask commonalities and introduce variability among studies. These challenges limit large-scale synthesis of research, which is critical for identifying knowledge gaps and informing management or policy. Moreover, the sheer vastness of available information makes complete evaluation of a body of sources out-of-reach for researchers to complete using traditional methods.  To address these challenges, computational informatics approaches that facilitate rapid synthesis of content have been adopted for analyzing the scientific literature. Text mining through machine learning techniques can perform automated analysis of large quantities of published content by identifying and extracting key information from text. Text mining provides a toolkit to help us understand how complex systems, where many possible drivers interact with many possible outcomes. For instance, within human-natural systems, the sustainability of residential landscapes is shaped by drivers at household, neighborhood, municipal, and regional scales. These drivers influence landscape sustainability through the provision of ecosystem services. Regulatory documents developed by municipalities strongly influence residential landscape practices, and thus can broadly facilitate or constrain ecosystem services, but they differ across geographies, landscapes and cultures. However, these differences are locked in vast quantities of text to archive laws, regulations, and guidelines for many human activities.

This team is text mining data within landscape ordinances as a way to understand patterns in how these documents are used and the relative importance of regional attributes (such as climate and topography) and cultural factors (such as policy) in the formation of ecological ideas about landscape management.  This project investigates the link between municipal policy and ecosystem response (and, indirectly ecosystem service provision, where data supports exist) via landscape design guidelines provided by municipal ordinances.  By creating a database of the metadata, the team will characterize geographic trends in how landscape ordinances are implemented.  Specifically, the team will:

  1. identify geographically-clustered trends in phrasing and structure from the landscape ordinances,
  2. establish how landscape ordinances both shape and are shaped by local ecological processes, and
  3. show how landscape ordinances that incorporate more ecological concepts are associated with disturbance-prone landscapes.