Duke Environmental Leadership Program

July 21-25, 2008
Levine Science Research Center, A158
Duke University
Durham, North Carolina
Please register by July 7, 2008

Program Description:

Geographic information systems (GIS) and geospatial analysis play an increasingly prominent role in conservation. The goal of this course is to introduce conservation practitioners to the capabilities of GIS-based analyses to serve as an integrating framework for conservation management, providing functionality including data acquisition, management, analysis and decision support, visualization, and documentation via metadata.

This intensive short-course will cover this spectrum of applications in the context of project-cycle workflow. The course begins by visiting common sources of geospatial data, builds working databases, attends some fundamental tasks in conservation management, and concludes with considerations for the long-term management of conservation projects in a GIS framework.

This course serves a fundamental role in the Conservation Certificate program, as it introduces several key issues that will be developed more fully in subsequent modules. The emphasis here is on geospatial methods.

Course Objectives:

The course is couched in the context of the project cycle of a conservation project, from project initiation, to analysis and decision support, to project evaluation and adaptive management. To this end, the course addresses four topical areas in turn:

1. GIS data models (including an introduction to, or refresher on, fundamental operations and best practices in ArcGIS);
2. Identifying conservation targets (including habitat classification and mapping for focal species or habitats, as well as approaches based on biophysical proxies);
3. Site characterization and prioritization (including univariate and multi-criteria decision frameworks);
4. Monitoring and assessment (change detection, assessment frameworks, and adaptive management)

Course Format:

The course is modular, consisting of lectures and hands-on tutorials in a computer lab. The modules are in 2-hour blocks over 4 ½ days. Lectures will provide background and context for the lab exercises. The lab exercises will use data from on-going conservation projects, focusing on the ModelBuilder working environment in ArcGIS.

Topical Outline:

1. The GIS Environment:  Data Models and Geospatial Analysis
GIS can be an indispensible ally in making informed decisions on how to manage lands. However, before GIS can be of any use, we need to bridge the gap between what we see and experience as a landscape and what a digital machine can interpret, i.e. between our rich analog reality and a computer's digital realm. We begin our course with a review of the two major data models used in GIS – the vector and raster data models. In doing so, we examine the different approaches to defining what exactly it is about a landscape that we are concerned about and how to represent it within a GIS. 

1.1. Data Models
Features and fields;
Vector features (points, lines, polygons; topology)
Raster fields (focal, zonal, and global operations)

1.2. Coordinate systems and projections
Project management issues

1.3. Digital landscapes
What sorts of features do conservationists find interesting?
Features (places) and phenomena (processes)

1.4. Developing conservation datasets
Standard data sources (USGS, NHD, NLCD, ...)

1.5. Geospatial data products
Hydrologic indices, greenness, potential radiation, ...
Focal and zonal statistics as integrated measures

2. Identifying Conservation Targets
Conservation targets depend on the mission of the agency, and might include rare species or communities, sites with high restoration potential, sites with high amenity value, or sites threatened with high likelihood of development or degradation. Approaches to identifying conservation targets might range from expert knowledge that pinpoints known locations, to statistical models that predict potential conservation value.

2.1. Rule-based targeting in GIS (geospatial queries and map algebra)

2.2. Habitat classification and species distribution models
CART, GLM+ROC, maxent
Classifying satellite imagery

2.3. Biophysical proxies and biodiversity surrogates
Environmental “zipcodes”

2.4. Threat-mapping and conservation feasibility
Proximity indices; development probability

3. Site characterization and prioritization
Having quantified various attributes of the digital landscape, these attributes can now provide a basis for prioritizing specific areas within the landscape for conservation or restoration. For a single criterion, this amounts to simple ranking. For multiple criteria, this must invoke a decision framework, and adjusting the relative importance of each criterion can imply different conservation priorities. Over time, management decisions also change the relative value of candidate sites, and so prioritization must be conducted in an adaptive framework.   

3.1. Conservation criteria (biodiversity, geometry, context, amenity values, ...)
Known occurrences of rarity (Heritage Element Occurrences)
Biophysical proxies and biodiversity support potential
Patch geometry (size, shape)
Connectivity analysis

3.2. Site selection algorithms
Ranking and queries
Greedy heuristics (PORTFOLIO)
MARXAN
VISTA

3.3. Weighting of criteria (multi-criteria decision frameworks)     

    
4. Monitoring, Assessment, and Adaptive Management
Conservation is a battle that is never won, and so long-term management of projects is crucial to the mission success. This implies at least two fundamental concerns in project management: the capacity to storage and manage geospatial data that accrues over time, and the ability to respond to new information appropriately, to adapt management as conditions change.

4.1. Data management and metadata
Record keeping, versioning (data updating), and metadata

4.2. Landscape change
Change detection (difference maps)
Models of landscape change (Markov processes, more complicated models)

4.3. Interpreting landscape change
An inferential framework (ordination and change vectors)

4.4. Adaptive management
Measures of success (process versus outcome monitoring)
Change points (trigger points) and adaptive response
Program evaluation

Course Facilitators:

Dean Urban, Duke University Nicholas School of the Environment and EarthSciences, Professor of Landscape Ecology.  Dr. Urban’s primary experience lies in the field of ecology with secondary expertise in biodiversity, conservation biology, endangered/threatened species, forest fires and fire ecology, forest management and silviculture, geospatial technologies, global change ecology, land use and sprawl, landscape dynamics and terrestrial ecosystems.

Jennifer Swenson, Duke University Nicholas School of the Environment and Earth Sciences, Assistant Professor of the Practice of Geospatial Analysis.  Dr. Swenson’s primary experience lies in ecology, with secondary experience in conservation biology, forest management and silviculture, geospatial technologies, land use and sprawl, terrestrial ecosystems and tropical ecology.

John Fay, Duke University Associate in Research in the Nicholas School of the Environment and Earth Sciences.

General Information:

Tuition: $1150 on or before July 7, 2008 ($1225 after July 7, 2008) 
Covers registration, instructional materials, and lunch and break foods each day on campus.

Cancellation & Refunds: All speakers and companies listed have confirmed their participation. If for unforeseen and unavoidable circumstances a speaker cannot attend, instructional time will be filled with comparable instructors. If a course is canceled due to insufficient registrations, tuition will be refunded. Duke cannot be held responsible for travel costs or arrangements made by participants.

Cancellation of registration requires 14 days written notice prior to course start date. Individuals who do not attend and fail to send written notice prior to the deadline will be charged the full tuition. You may send a substitute to attend the course in your place. Extenuating circumstances will be reviewed on a case-by-case basis.

Hotel Accommodations: Participants are responsible for their own overnight accommodations. Participants most commonly stay at the Durham Hilton Hotel (3800 Hillsborough Road, Durham, 919-383-8033).  The Hilton provides a special Nicholas School rate when reservations are made within the course room block and three (3) weeks prior to the course start date. The Duke Environmental Leadership Program provides complementary shuttle transportation between campus and the hotel at the beginning and end of each class day.

For more information visit: http://www.nicholas.duke.edu/del/continuinged/delgis.html