Biodiversity and Ecosystem Informatics: VLDB Scientific Research or - - PowerPoint PPT Presentation

Biodiversity and Ecosystem Informatics: VLDB Scientific Research or an Applications Area for Computer Database Technology?

Kathleen M. Bergen, Ph.D. School of Natural Resources and Environment The University of Michigan

Biodiversity and Ecosystem Informatics: Three Questions

• 1. Is this a Very Large Database problem?
• 2. Are there research challenges for VLDB researchers or

is this another applications situation?

• 3. Is BDEI a valid emerging field of science and

applications research?

• 1. Is This A Very Large Database Problem?

– Biological Species Diversity

• N of named species is approximately 1.5 million to 1.8 million.
• The full N of species may be up to 10 million
• Large numbers of species specimens and locational observations are

distributed throughout the world in numerous museums, herbaria,

• rganizations, and agencies

– Ecosystems

• In-situ and traditional ecosystem measurements by agencies and

scientists

• NASA Earth Science satellites beam home image data at the rate of 1.2

terabytes per day.

• Currently at 500 terabytes, NASA digital image archives will grow to

greater than 10 petabytes by 2010

• Scientists and organizations create derived datasets

• 2. Are there Research Questions for VLDB in BDEI?
• Research Problem: Biologists and managers want to

map the location of known species occurences, and, given known occurances, model, predict and map where additional members of that species may live, based on

• ther ecosystem variables AND then biologists want to

know what will be the effect of change in these ecosystem or environmental variables on the persistence of species and their ranges.

University of Michigan BDEI Project Methods:

Bird Species Habitat

Assemble Spatial Database of Ecosystem Variables

• climate
• vegetation type
• hydrology
• topography

Envelope Atomic GARP Logistic

Derive New Multi-dimensional Vegetation Structural Characteristics from Remote Sensing Data and Test: Improvement

• height
• basal area
• biomass/carbon
• overstory/

understory Assemble Species Occurrence point samples from field or museum collections with geographic coordinates Develop and Apply Algorithm e.g. GARP Method (Genetic Algorithm for Rule-Set Production), ISOCLUS or other Species Habitat Range and Suitability Models and Spatial Datasets

Globe Biome Ecosystem Organism Cell Gene

Levels of Organisation in Biology

• 3. Is Biodiversity and Ecosystem Informatics a Field of

Scientific Research or an Application of Computer Database Technology?

Biodiversity and Ecosystem Informatics: Three Questions

• 1. Is this a Very Large Database problem?
• 2. Are there research challenges for VLDB researchers or

is this another applications situation?

• 3. Is BDEI a valid emerging field of science and

applications research?