Issue link: https://read.dmtmag.com/i/75042
analysis of data representing other habitat character- istics, such as vegetation indices. Filling the Holes An extremely common analytical method in wildlife research is to spatially map the probable area of a species' activity. Excellent third-party extensions (e.g., Home Range Tools and Geospatial Modelling Environment) also exist for creating home ranges as minimum convex polygons (MCPs) and kernel-density esti- mations (KDEs). However, ACCRU research has required batch processing with random iterations (MCPs) or alternative bandwidth-selection techniques (KDEs) that aren't available in the alternative software. Many opportunities exist with GIS to gener- ate future or alternate landscapes. Kerri Lappin's master's project at the University of Alberta, for example, involved random placement of theoretical wetlands in a coal-reclamation area. A script applied to the polygon shape attribute within ModelBuilder did the trick. ACCRU Tools prototypes also assisted Scott Nielsen's extensive work with the Great Sand Hills Environmental Study, Saskatchewan, where part of the study analyzed how the biodiversity may be affected by the introduction of more oil/gas wells. The methods developed for the new associated roads simulation have since been automated. Data files for input into the FragStats landscape-metrics package, streamlining satellite-image classification, pre- and post-habitat modeling via regression-based statistical methods, extracting zonal statistics for multiple rasters, processing large numbers of climate grids, calculating vegetation indices, and batch-processing data conversions are among the additional solutions incorporated in to the collection of ACCRU Tools. Cooperation ACCRU Tools evolved from the various computational efficiencies developed to solve the conservation- research challenges in the Ecology RIG. The toolbox was built by expanding Esri ArcGIS Desktop software, using the ModelBuilder framework and Python pro- gramming language—both transparent and open, per- mitting anyone to examine the custom tools to learn from them and adapt them to their own needs (see Figure 5). But that can't be said about the compiled exten- sions that aren't modifiable by end users. For example, commands added in as a toolbar can't be used in the geoprocessing framework. And if the functionality needs to be tweaked to better fit the data or randomly iterated for simulations and hypothesis testing, then researchers must "babysit the GUI," patiently wait- ing for tool completion before continuing on with the results in the next analytical steps. The headaches of Figure 5. ModelBuilder and Python effectively linked, leveraged and expanded upon the built-in geoprocessing tools for a variety of research solutions. ACCRU Tools are built on these open methodologies and organized by numerical category to allow for efficient indexing and searching in ArcToolbox. installation are another problem associated with many of the compiled extensions. Think of how fast advancements could be made in research using GIS if everyone collaborated and shared their source of analytical methods. It's still mostly a one-way street for authors: researchers ask questions, GIS solutions are developed that may or may not end up as a custom tool but get the required analyses done, and researchers continue merrily along with their work on their own, sometimes even expand- ing on the methods they learn. Author's Note: If you discover something that will help address conservation-ecology questions, please join the growing coop- erative spirit by developing open and transparent tools (similar to the Model and Script Gallery at resources.arcgis.com and the humble but ever-evolving ACCRU Tools). To learn more and download the freely available ACCRU Tools, visit the Web at www.biology.ualberta.ca/facilities/gis/?Page=3063. Charlene Nielsen is a GIS analyst, Biological Sciences, University of Alberta; e-mail: charlene.nielsen@ualberta.ca. J UL Y 2O12 / WWW . GEOPLA CE . COM 29