The following paper was presented at the Sixth Biennial Forest Service Remote Sensing Conference, Denver, CO April 29 - May 3, 1996. Proceedings of the conference were published by the American Society of Photogrammetry and Remote Sensing (ASPRS) and can be obtained from ASPRS.


UTOOLS AND UVIEW: ANALYSIS AND VISUALIZATION SOFTWARE

 
Robert J. McGaughey
USDA Forest Service
Pacific Northwest Research Station
University of Washington, Bloedel 361
P.O. Box 352100
Seattle, WA 98195-2100
mcgoy@u.washington.edu
Alan A. Ager
USDA Forest Service
Umatilla National Forest
2517 SW Hailey Ave.
Pendleton, OR 97801
alan@oregontrail.net
 

ABSTRACT

UTOOLS is a collection of programs designed to integrate a variety of spatial data in a way that allows versatile spatial analysis and visualization. UTOOLS software combine raster, vector and attribute data into "spatial databases" where each record represents a square pixel of fixed area, and each field in the database represents a map layer, theme, or attribute. UTOOLS includes a number of common GIS functions, including procedures for calculating buffers, slope, aspect, patch size, convexity, and a view index. UTOOLS also includes UVIEW, a visualization program designed to produce two- and three-dimensional images of digital elevation models (DEM), attribute data, and vegetation patterns at watershed and landscape scales. UTOOLS programs address the daily analysis needs of resource professionals charged with managing large land areas to provide a variety of commodity and non-commodity outputs.

INTRODUCTION

Resource analysis on federal forest lands in the Pacific Northwest has become increasing complex during the last decade Analysis has shifted from procedures that emphasize single species or specific sites to more holistic approaches that concentrate on entire ecosystems and large landscapes. Recent years have seen a shortage of suitable analysis tools for use on individual ranger districts within the U.S. Forest Service. To address this shortage, the Forest Service has begun procurement and implementation of Geographic Information System (GIS) technology. However, delays in procurement and implementation have left many district offices without sufficient analysis capability to answer the questions being asked of them. Resource specialists are being asked to resolve difficult management issues but often do not have adequate analysis capabilities to support their decisions.

The assessment and study of ecosystems and their dependent species is a complex task demanding a variety of analysis tools, including geographic information and decision support systems, simulation and deterministic models, and appropriate data linkages. As part of an effort to create more appropriate spatial analysis tools to address the needs of aquatic, wildlife, and other biologists and specialists within the Forest Service, the authors have developed of a set of software programs called UTOOLS*. UTOOLS software is designed for use by field biologists and resource specialists working in district or forest offices within the U.S. Forest Service or in equivalent positions in other land management agencies. The programs have been widely applied within the Forest Service (McKinney 1995, Mellon et al 1995, Ager and Hitchcock 1992) and outside the Forest Service (Oliver and McCarter 1996) for analysis of a variety of resource issues since their first release in 1992. In this paper, we describe the capabilities of UTOOLS and discuss its application to natural resource analysis and planning.

OVERVIEW

UTOOLS is a collection of programs designed to integrate a variety of spatial data in a way that allows versatile spatial analysis and visualization (figure 1). One of the primary goals of UTOOLS is to provide analysis tools that can capture a "snapshot" of existing databases, allow the incorporation of a variety of data types, and provide analysis capabilities to address watershed-scale management issues. UTOOLS software combine raster, vector and attribute GIS data into "spatial databases" where each record represents a square pixel of fixed area, and each field represents a map layer, theme, or attribute. This raster data structure, maintained in PARADOX** format, lends itself to a wide range of analysis procedures and simplifies linkages between spatial map themes and their attributes. PARADOX queries can be used to perform many types of map operations such as intersects, joins, unions, and subject recoding. Rule bases can be applied as PARADOX queries to synthesize new layers from combinations of other layers and their attributes. The analysis capability accessed using PARADOX queries and scripts is augmented by UTOOLS programs that perform specific, raster-based, spatial analysis operations commonly used in wildlife and other resource analyses, e.g., buffering and patch identification. UTOOLS capabilites include a number of common GIS functions, including procedures for calculating buffers, slope, aspect, patch size, convexity, and a view index.

In operation, users invoke specific programs to carry out desired operations. UTOOLS programs include:
 

UCELL5 Creates PARADOX spatial databases using vector-based GIS coverages and performs common GIS functions.
UVIEW Visualization program used to compose planimetric maps and perspective images depicting data and analysis results.
FLY Command script generation utility designed to produce animation path scripts for UVIEW.
UMAP Display utility used to compose planimetric maps depicting data and analysis results.
HEIWEST Habitat effectiveness indicator calculator for Western Oregon.
HEICALC Habitat effectiveness indicator calculator for the Blue Mountains region of Eastern Oregon.
DISPLAY Landscape diversity calculator (McNeil and Flather 1992).
OVERLAY Utility to perform intersections using vector-based map coverages.
MAP2MOS Utility to convert raster-based spatial databases into vector-based map coverages.
 
UTOOLS schematic
Figure 1. Schematic of the analysis and data conversion capabilities of UTOOLS.
UTOOLS includes utilities to convert and manipulate data from a variety of sources to augment the spatial databases created by UCELL5. These utilities allow users to integrate USGS digital elevation models, raster-based GIS coverages, ERDAS GIS data coverages, and ASCII text tables containing map attributes.

UTOOLS provides data linkages to several public domain programs developed to address specific analysis problems, e.g., calculation of landscape pattern metrics (FRAGSTATS, McGarigal and Marks 1993; DISPLAY, McNeil and Flather 1992), and elk habitat assessment (HEICALC, Hitchcock and Ager 1992; HEIWEST, Ager and Hitchcock 1992).

Data in spatial databases can also be exported to ASCII raster files readable by MOSS and ARC-INFO. It is also possible to convert raster data in a PARADOX spatial database into MOSS import/export vector format. These data linkages extend the functionality of UTOOLS programs and the associated PARADOX spatial databases to encompass a broad spectrum of spatial analyses.

UTOOLS operates on IBM-compatible personal computers using the MS-DOS, Windows95, or Windows NT operating systems. Effective use of UTOOLS requires microcomputer configurations with a 66 Mhz 80486 (or faster), 4Mb of RAM, at least 300 MB of available space on the hard drive, and an SVGA display adapter. Lesser configurations are suitable for browsing the example data included with UTOOLS, but will not provide satisfactory performance for most analysis problems. UTOOLS programs can be used without PARADOX (version 3.5 and later), however, their usefulness is limited in its absence. UTOOLS programs will run from within a DOS window in Windows 3.X and Windows95. UTOOLS programs will run from within a DOS window in Windows NT but the high resolution, 256-color graphics modes will not be available in UVIEW.

BUILDING SPATIAL DATABASES FOR USE IN UTOOLS

The fundamental data used by UTOOLS programs are PARADOX "spatial databases" built by converting data layers from their native formats to the PARADOX spatial database format using the UCELL5 program. A typical database construction sequence proceeds as follows: The pixel size of a spatial database can be varied according to the needs of the project. Smaller pixel sizes result in less spatial error in the vector to raster conversion process, but result in large databases that are slow to query and map. Most UTOOLS applications for watershed-scale projects use a pixel size of 30-100 meters per side resulting in databases of 30,000-250,000 records and 10-50 Mb in size.

SPATIAL ANALYSIS WITH UTOOLS

Given the UTOOLS spatial database, operations such as overlays and unions are accomplished using simple, one-step PARADOX queries. In contrast, typical vector-based GIS systems store layers and attributes separately, and thus every layer must be overlaid and associated attribute tables must be linked to complete similar operations. While the latter is acceptable when questions are straightforward, most analyses involve repetitive querying of multiple layers and attributes. Analyses generally involve comparing and cross validating data assembled from many sources, simulating effects of management activities, or designing the spatial layout of treatment areas. Seldom are the queries and map operations needed to answer these questions perfected on the first attempt. Analysis processes generally evolve as more is learned about the quality of the resource data and the effect of the data quality on the final outcome. This evolution process is hindered when data are stored as fragments of the whole ecosystem, e.g., data cataloged according to management or mapping boundaries rather than watershed boundaries. UTOOLS programs provide the ability to integrate all of the necessary information into a single database simplifying the overall analysis process and providing a means to easily accomplish multi-resource analyses. In this way, UTOOLS provides an excellent environment for developing prototype analysis and modeling schemes for later integration into corporate level GIS environments.

LANDSCAPE VISUALIZATION CAPABILITIES IN UTOOLS

UVIEW is a display program designed to produce two- and three-dimensional images of digital elevation models (DEM), attribute data stored in PARADOX spatial databases, and vegetation patterns at watershed and landscape scales. UVIEW evolved from terrain viewing software developed for the PLANS system (McGaughey 1991), the Vantage Point visualization system (Fridley et al. 1991), and the stereoplotter visualization system (McGaughey 1992). The images produced by UVIEW provide a readily understood visualization depicting spatial analysis results and existing or desired landscape conditions.

UVIEW provides a flexible system for viewing a digital elevation model. UVIEW allows users to specify exact coordinates for the head and focus locations or interactively select a head and focus location while viewing a simple perspective representation of the DEM. UVIEW simulates camera lenses with focal lengths ranging from 15mm to 400mm and vertical exaggeration values ranging from 0.1 to 4.0. Users can also "fly" over and around a low resolution image of a DEM using a mouse controlled "virtual trackball".

UVIEW can render a DEM using a variety of methods and resolutions (figure 2). Users typically use the wire frame representations, profiles and grid, for positioning and exploration of the terrain surface. They use the solid surface representations to display attribute data from the PARADOX spatial database and simulated vegetation. Computer systems equipped with a VESA compatible graphics adapter capable of displaying 256 colors in at least 640 by 480 pixel resolution can display lighted, solid surface representations. These rendered images add realism to a scene by simulating a light source and computing various shades of color depending on the orientation of the ground surface as defined by the DEM. UVIEW identifies and eliminates hidden areas in solid surface representations by sorting and drawing DEM cells starting with the cells farthest from the head location. UVIEW displays a planimetric view of the DEM, represented as contour lines using a user-specified contour interval, along with attribute data from PARADOX databases and vector and polygon data files.

UVIEW supports interactive query development allowing the user to specify combinations of database attributes to control the color of individual cells on solid surface representations. The query system supports a variety of query operators including string pattern matching, numeric comparisons, and boolean operators. UVIEW displays query results in both perspective and planimetric views.

UVIEW provides a script language to facilitate generating sequences of images. Sequences can represent a landscape from different viewpoints or changing landscape conditions from a single viewpoint. UVIEW scripts can generate images with no user interaction providing a simple method of reproducing a standard set of images depicting alternative management scenarios. The UV360FLT and FLY utilities assist with the creation of scripts by creating animation path scripts to generate "fly-through" sequences.
 

line grid
(A) (B)
surface shaded surface
(C) (D)
Figure 2. UVIEW represents digital elevation models using (A) coarse and fine resolution profiles, (B) coarse and fine resolution grid, (C) solid surface, and (D) lighted, shaded surface.
UVIEW can model vegetation patterns to simulate existing or desired landscape conditions. The primary goal in vegetation modeling is to simulate overall landscape texture and pattern rather than specific, detailed vegetation structure. UVIEW provides two methods to model vegetation patterns: Canopy closure based vegetation modeling represents vegetation patterns over an entire landscape. This method represents differences in stand densities well but does not represent differences in stand composition and structure (figure 3). UVIEW represents all values of canopy closure using the same type and size of tree, only the density of plants varies.

Vegetation modeling based on structure definitions represents both stand density and stand composition. Stand structure definitions consist of layer descriptions with each layer in a vegetation type described by the type of plant, height, crown diameter, crown ratio, a factor describing the variability of the size parameters, and the number of plants per unit area (acres or hectares). UVIEW represents a variety of plant types ranging from grass to mature, healthy conifer and hardwood trees. Vegetation structure descriptions can consist of up to 36 layers. Practical descriptions contain two or three layers.

EXAMPLE APPLICATIONS OF UTOOLS

UTOOLS and UVIEW have been applied to a wide variety of analysis projects. Some of the applications are specific to geographic areas but most are applicable across a broad range of conditions. All applications share a common theme: resource specialists using the analysis capabilities of UTOOLS to address specific problems that were previously difficult or impossible to solve due to technical, implementation, or access limitations with existing GIS software.

Wildlife Habitat Analysis

HABSCAPES (Mellon et al 1995), developed on the Mount Hood National Forest, provides analysis capabilities for wildlife populations in a community context on large landscapes. HABSCAPES uses spatial databases created using UTOOLS and maps its results using UVIEW. The goal of the system is to predict the occurrence of all terrestrial vertebrate and aquatic amphibian species relative to landscape pattern over large geographic areas. HABSCAPES links UTOOLS spatial databases to databases containing wildlife habitat relationships and life history characteristics using custom FORTRAN programs and PARADOX scripts. Although HABSCAPES was developed for forest-level analysis on the Mount Hood National Forest, it is finding wide application for watershed analyses and adaptive management areas throughout Region 6.

Spatial Analysis of SMART Stream Survey Data

McKinney (1995) describes the application of UTOOLS to multi-scale analysis of Forest Service stream survey data stored in the stream management, analysis, reporting, and tracking (SMART) system used in Region 6. The goal of this analysis is to better understand cause and effect relationships of both in-channel and upland interactions. Prior to the adoption of UTOOLS, there was no usable methodology to analyze and view these data. McKinney (1995) developed a procedure for linking the SMART data to stream segment maps for visualization using UVIEW and methods to analyze the data using PARADOX queries and scripts. These procedures were used to assess aquatic habitat relationships within the Columbia River basin as part of the Eastside Ecosystem Management Project.
 
UVIEW image UVIEW image
(A) (B)
UVIEW image
(C)
Figure 3. UVIEW can depict vegetation patterns using canopy closure estimates for each pixel in the spatial database***. Image (A) represents a heavy harvest with aggregated retention, image (B) represents a light harvest with aggregated retention, and image (C) represents a harvest with dispersed retention.

Starkey Ungulate Research Project

The Starkey ungulate research project is a long-term study of cattle, deer and elk interactions on the Starkey Forest and Range Experiment Station near LaGrande, Oregon (Johnson et al 1991). The project uses a Loran C animal tracking system to record animal movements throughout the grazing season. A major component of this investigation is to refine existing habitat use models for cattle, deer, and elk. As part of this effort, a wide array of habitat data for the project area have been assembled in UTOOLS databases. Animal telemetry data (locations) have also been converted to spatial database format and merged with habitat data based on their coordinates. The association of habitat data with animal use in a single data structure allows for rapid exploration of the relationship between animal locations and habitat characteristics.

With the UVIEW scripting language, movies of the telemetry data are created depicting cattle, deer, and elk movements over the grazing season on a landscape image created with a UVIEW terrain model (figure 4). Vegetation can be added to these images to create a realistic image of the landscape with animal locations. This technique provides a dynamic depiction of animal behavior over time that yields valuable insights into animal-habitat relationships and inter specific interactions.
 

UVIEW image
Figure 4. Telemetry data depicting elk, deer, and cattle locations over a one month period on the Starkey Forest and Range Experimental Forest.

Landscape Management System

The Landscape Management System (LMS), developed at the University of Washington, is designed to assist in landscape level analysis and planning of forest ecosystems by automating the tasks of stand projection, graphical and tabular summarization, stand visualization, and landscape visualization (Oliver and McCarter 1995). LMS includes many separate programs that simulate stand growth and development, produce graphical or tabular displays, and store or process inventory information. The primary function of the LMS program is to connect these diverse programs into a cohesive system. LMS uses UTOOLS and UVIEW to produce landscape visualizations of forest conditions over time using a variety of alternative management strategies.

SOFTWARE DISTRIBUTION

The UTOOLS distribution consists of three files. UTOOLS1.EXE and UTOOLS2.EXE contain the UTOOLS programs and documentation and UTOOLS3.EXE contains an example data set for use with UTOOLS and the tutorial exercise described in the manual (UMANUAL.DOC in the UTOOLS1.EXE archive).

UTOOLS and UVIEW home pages is maintained on the world wide web at:

utools_uview.html

SUMMARY

UTOOLS meets a critical need for simple and efficient spatial analysis tools by removing many of the barriers commonly found in large GIS systems. UTOOLS programs address the daily analysis needs of resource professionals charged with managing large land areas to provide a variety of commodity and non-commodity outputs. Because UTOOLS is easy to learn, runs on commonly available computer hardware, and utilizes existing data; it provides an excellent learning platform for resource specialists. Using UTOOLS programs, they can explore the capabilities of GIS while conducting real-world analyses to support and enhance their resource management decisions.

ACKNOWLEDGMENTS

We are indebted to Mark Hitchcock, David Hatfield and Bill Connelly for their contributions to UTOOLS software. Ed Pugh, Jim Merzenich, Ken Tu, Shaun McKinney, Bill Connelly, Bob Clements, Dave Kendrick, Alison Reger, Jim McCarter, Jeremy Wilson, and many others provided valuable feedback on early versions of this software.

LITERATURE CITED

Ager, A.A., and M.E. Hitchcock. 1992. Microcomputer software for calculating the Western Oregon elk habitat effectiveness index. USDA For. Serv., Gen. Tech. Rep. PNW-GTR-303. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station.

Ager, A.A. and R. McGaughey. 1994. Operations manual for UTOOLS. Pendleton, OR: U.S. Department of Agriculture, Forest Service, Umatilla National Forest.

Fridley, J.L., R.J. McGaughey, and F.E. Lee. 1991. Visualizing engineering design alternatives on forest landscapes. American Society of Agricultural Engineers paper #917523. Presented at the 1991 International Winter Meeting. Chicago, IL. December 17-20.

Hitchcock, M.E. and A.A Ager. 1992. Microcomputer software for calculating elk habitat effectiveness index on Blue Mountain winter range. Gen. Tech. Rep. PNW-GTR-301. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station.

Johnson, B.K., J. Noyes, J.W. Thomas, and L. Bryant. 1991. Overview of the starkey project: current measures of elk vulnerability. In: Elk Vulnerability - A Symposium; April 10-12. Bozeman, MT. Bozeman, MT: Montana State University: 225-228.

McGaughey, R.J. 1991. Timber harvest planning goes digital: PLANS-preliminary logging analysis system. Compiler 9(3):10-17.

McGaughey, R.J. 1992. Three-dimensional terrain visualization with a real-time interface to a photographic stereomodel. In: Proceedings of the fourth Forest Service remote sensing application conference; April 6-11, 1992; Orlando, FL. Bethesda, MD: American Society of Photogrammetry and Remote Sensing. 204-211.

McGarigal, K. and B.J. Marks. 1995. FRAGSTATS: spatial pattern analysis program for quantifying landscape structure. Gen. Tech. Rep. PNW-GTR-351. Portland, OR: U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station.

McKinney, S.P. 1995. Spatial analysis of SMART stream survey data with UTOOLS. AquaTalk: Region 6 Fish Habitat relationship Technical Bulletin 10 (September). Portland, OR: U.S. Department of Agriculture, Forest Service, Region 6 Regional Office.

McNeil, B., and C. Flather. 1992. Operations manual for DISPLAY. Internal publication. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station.

Mellon, K., M. Huff, and R. Hagestedt. 1995. HABSCAPES interpreting landscape patterns: a vertebrate habitat relationships approach. In: Analysis in support of ecosystem management, Analysis workshop III. Fort Collins, CO. Apr. 10-13. USDA Forest Service. Ecosystem Management Analysis Center, Washington, DC. 135-145.

Oliver, C.D. and J.B. McCarter. 1996. Developments in decision support for landscape management. In: M. Heit, H.D. Parker, and A. Shortreid (ed), GIS Applications in Natural Resources 2. GIS World Books. 501-509.

Footnotes

* UTOOLS/UVIEW programs are in the public domain. Recipients and users may not assert any proprietary rights over the programs, documentation, or example data files distributed with UTOOLS.

** The use of trade or firm names in this publication is for reader information and does not imply endorsement by the U.S. Department of Agriculture of any product or service.

*** Data and images were produced by the students of FE 444, Winter quarter 1995, College of Forest Resources, University of Washington.


If you have comments or need additional information about this page, contact Robert J. McGaughey at the Pacific Northwest Research Station (mcgoy@u.washington.edu ).