Font Size:
Geomancer
Last modified: 2011-09-10
Abstract
The value of digitized biodiversity information is greatly increased if it is georeferenced. Georeferenced information can be viewed on a map and to be used to integrate with other spatial data sets such as environmental variables and species range maps. Georeferenced primary species occurrence records can be used to answer a wide range of questions. For records that don’t have explicit coordinates provided by the collectors, we are left with the need to provide the spatial data retrospectively based on the original textual descriptions. Taken one at a time, this process can take as long as digitizing all of the rest of the information about a specimen or observation. Tools to make the georeferencing process more efficient are therefore increasingly important as more primary data begin to be digitized.
The BioGeomancer project was funded in 2005 by the Gordon and Betty Moore Foundation to improve existing tools and create new resources that serve to partially automate the process of georeferencing. The resulting BioGeomancer web application was deployed to a local server with a spatial database (PostgreSQL with PostGIS) and it supports single and bulk georeferencing using shared projects as a collaboration point between users. Like many server-based applications however, it suffers from performance and scaling limitations when many users access the system.
Geomancer, a project funded by the Canada California Strategic Innovation Partnership with Canadensys and UC Berkeley, brings BioGeomancer into the cloud. It leverages algorithms developed in BioGeomancer while overcoming its performance and scaling challenges by running it on Google App Engine and leveraging Google Application Programming Interfaces for Prediction, Translation, and Geocoding. Geomancer is an open source project hosted on GitHub. In this talk we will describe the conceptual and technical framework for building Geomancer, and we will discuss the key challenges of building an application that uses both local and cloud-based computing resources to keep costs low. We will also provide a live georeferencing demo that uses Google Fusion Tables to visualize the results.
https://github.com/GeomancerProject/Software
The BioGeomancer project was funded in 2005 by the Gordon and Betty Moore Foundation to improve existing tools and create new resources that serve to partially automate the process of georeferencing. The resulting BioGeomancer web application was deployed to a local server with a spatial database (PostgreSQL with PostGIS) and it supports single and bulk georeferencing using shared projects as a collaboration point between users. Like many server-based applications however, it suffers from performance and scaling limitations when many users access the system.
Geomancer, a project funded by the Canada California Strategic Innovation Partnership with Canadensys and UC Berkeley, brings BioGeomancer into the cloud. It leverages algorithms developed in BioGeomancer while overcoming its performance and scaling challenges by running it on Google App Engine and leveraging Google Application Programming Interfaces for Prediction, Translation, and Geocoding. Geomancer is an open source project hosted on GitHub. In this talk we will describe the conceptual and technical framework for building Geomancer, and we will discuss the key challenges of building an application that uses both local and cloud-based computing resources to keep costs low. We will also provide a live georeferencing demo that uses Google Fusion Tables to visualize the results.
https://github.com/GeomancerProject/Software