Spatial Statistics: What’s so HOT about Spatial Pattern Analysis?

gpbheader[ This blog post was written by Lauren Scott, Geoprocessing/Spatial Statistics Product Engineer in the Software Products Group at ESRI in Redlands, and originally posted on the ESRI Geoprocessing Blog. ]

Hot Spot Analysis is just one of the pattern analysis tools in the Spatial Statistics Toolbox.  You can use these tools to explore spatial patterns in order to answer questions like:

  • Where are crime rates unexpectedly high?
  • Are there regions in the country where people live longer
  • Where do we find anomalous spending patterns?
  • Are there sharp boundaries between affluence and poverty?
  • Is the disease remaining geographically fixed or is it spreading?
  • Which features are most concentrated?
  • Does the spatial pattern of the virus mirror the spatial pattern of the population at risk?
  • Which site is most accessible?
  • Where is the population center?
  • Which species has the broadest territory?

To learn more about spatial pattern analysis, check out some of these resources:

University of Vermont Spatial Analysis Laboratory Named Center of Excellence

uvtowerThe University of Vermont’s Spatial Analysis Laboratory, part of the Rubenstein School of Environment and Natural Resources, has received two prestigious honors in recent months.

The Definiens corporation, founded by Nobel Prize laureate Gerd Binnig, recently designated the lab one of eight international Centers of Excellence, based in part on the Urban Tree Canopy (UTC) assessment work the lab carried out in collaboration with the U.S. Forest Service. The other seven Centers of Excellence organizations are among the most well-respected and well-funded remote sensing labs in the world.

In addition, ESRI, a leading developer of GIS software, last spring named the lab one of the first ESRI Development Centers.

Both honors have benefits for the UVM community. “The Definiens designation allows us to receive software with a commercial value of $80,0000 and gives us priority access to technical support and the ability to participate the beta software releases,” said Jarlath O’Neil-Dunne, geospatial analyst at the lab. “Thanks to our ESRI Development Center status, all UVM students, staff, and faculty can now install full versions of ArcGIS on their personal computers,” he said. “A convenience in normal times, having access to GIS software at home will be a necessity if UVM is impacted by the H1N1 virus.”

“These are both truly impressive honors,” said Mary Watzin, dean of the Rubenstein School of Environment and Natural Resources. “They are a testament to the international stature of the Spatial Analysis Laboratory, placing it among a handful of the most well-respected GIS and remote sensing labs in the world. The designations will also bring tangible benefits to UVM students, faculty, and staff.”

[ Source: University of Vermont Communications ]

Bayesian Modeling in Ecology

uwFriday Forum: Graphical models and Bayesian model in ecology: a meta-analysis application
Friday, September 18, 2009, 12:00 p.m. – 1:00 PM
University of Wyoming, Ag C Building Room 316
Presenter: Kiona Ogle, Assistant Professor of Botany

This talk will introduce basic concepts of Bayesian statistical modeling. The concepts and methodologies will be illustrated in a meta-analysis application that synthesizes literature information on specific leaf area (an important plant “functional trait”) of 305 tree species occurring in the US. Graphical models will be overview as a tool for understanding relationships between different data sources, parameters, and latent processes.

GIS Supports DOE Awards for Studying Ocean Thermal and Current Resource Potential

doeU.S. Department of Energy Secretary Steven Chu today announced that 22 advanced water power projects will receive up to $14.6 million in funding to advance the commercial viability, market acceptance, and environmental performance for new marine and hydrokinetic technologies as well as conventional hydropower plants. The projects selected today will further the nation’s supply of domestic clean hydroelectricity through technological innovation to capitalize on new sources of energy, and will advance markets and research to maximize the nation’s largest renewable energy source.

Among the projects awarded funding from the DOE, two involving ocean thermal and current resource potential have GIS components:

Georgia Tech Research Corporation (Atlanta, GA) will perform an ocean current resource potential database, which will then be used to develop a web-based interface and GIS (Geographic Information System) tools for understanding the locations and practical amount of energy that can be extracted from ocean currents. DOE share: up to $500,000; Duration: up to one year

Lockheed Martin Corporation (Manassas, VA) will develop a GIS-based dataset and software tool to assess the maximum practicably extractable energy from the global and domestic U.S. ocean thermal resource and identify regions viable for OTEC and Cold Seawater Based Air Conditioning. DOE share: up to $500,000; Duration: up to one year

Update: Lidar Solutions in ArcGIS

lidar[Update: since first posting this back on May 1st, Clayton has added three new posts in this series.]

Clayton Crawford, Product Engineer in ESRI’s Software Products Group’s 3D Team.   He has been writing a series of posts on the Geoprocessing blog called “Lidar solutions in ArcGIS”.  These posts cover Lidar processing tasks and workflows, and will show how to manage these vast point collections and outline approaches for mining information from them.

Here is a list of topics Clayton plans to cover, with links to the seven posts already completed:

Interview: Carl Steinitz on GIS and Design

cslargeESRI writer Jim Baumann recently interviewed Carl Steinitz on the integration of GIS and design, and we share a portion of that interview here.  Steinitz, Alexander and Victoria Wiley Research Professor of Landscape Architecture and Planning, has been teaching at the Harvard Graduate School of Design since 1966.   His interests are reflected in his teaching and research on landscape change, methods of landscape analysis, visual quality, and landscape planning and design.  In 1984, he received the Outstanding Educator Award of the Council of Educators in Landscape Architecture; he also received the 1996 Distinguished Practitioner Award from the International Association for Landscape Ecology (U.S.A.).  In 1997, he was chosen by the student body to receive the annual Graduate School of Design Teaching Award.

Baumann: You have stated, “At large scale, you are dealing with strategy, at middle scale you are dealing with tactics and as small scale, you are dealing with details.”  Can you elaborate?

Steinitz: It is a generalization. When planning at regional scale, with changes such as new infrastructure, urbanization and conservation, or when looking at a regional scale plan, nobody cares if a village has two or three story buildings. When planning a village or looking at a village plan, nobody cares if the garden of a home has an apple tree or a pear tree. But when we were designing our garden, we preferred a cherry tree to an apple tree.  Focus is typically a function of the lens of scale.

Baumann: How can GIS best be used in relationship to design when considering these telescoping scales?

Steinitz: We need to be able to work at several scales of resolution in space and classification.

Baumann: At what point should GIS be introduced into the design process?

Steinitz: GIS does not have an automatic role. If it is to play a role, it must be considered as part of the process of ‘designing’ the methodology of any study.

Baumann: How can GIS be used more effectively as a tool in architectural and landscape design?

Steinitz: This depends on many things: e.g. computer technologies, intuitive, easy-to-use interfaces, relevant software, appropriate data scale(s), needs for analysis, available or adapted models, trained people, and even fear and mistrust.  The larger and riskier the design project, the more GIS is likely to be used.

Baumann: How would GIS tools that simulate dynamic processes be best used in the design process?

Steinitz: They can be useful in impact assessment when comparing alternatives, but would “best” be used in making designs—change models—iteratively, in immediate feedback interaction with impact evaluations.

Baumann: Would the development of a GIS data model for GeoDesign be feasible?  If so, what would be included in it?

Steinitz: Not “a” data model, but rather capabilities for very flexible data models to meet particular data needs in time, space, and classification to be adapted to meet the needs of any particular design project. This will only be useful if the methodological framework for design can USE the data flexibilities in its several stages.

Baumann: Does a digital environment/alternative reality like Second Life play a role in landscape architecture?

Steinitz: I am not a fan of Second Life and consider it a sad retreat from real life. However, it has a potentially useful software base and is of interest to researchers at the Centre for Advanced Spatial Analysis at University College London where I am Visiting Professor, and at other research groups.  I expect it to be used more frequently to simulate projected designed environments.   It can be adapted to be the medium of design.

Baumann: Michael Goodchild recently posed the question, “If spatial dependence in the form of Tobler’s First Law—nearby things are more related than distant things—is a general and fundamental principle of geography, what is its meaning in design?”

Steinitz: I have no idea what it means in design. But I do know that more and more designers are NOT following this law, and that technologies are increasingly enabling real time multi-user collaborations in Web-space. I have done this in teaching and research for years, and not with “local” collaborators.

Baumann: What does the future hold for the use of technology in design?

Steinitz: We will increasingly see experimentation in participatory design methods that are technology-driven, and that directly link to data acquisition at the beginning and to constructing changes at the end.  Will the results be more “successful”?  Who knows…but one can try.