[MARMAM] Workshop: Species Distribution Modeling with MGET, following the SMM Biennial Conference, 14 December, Dunedin, NZ
jason.roberts at duke.edu
Wed Oct 16 08:00:13 PDT 2013
We cordially invite you to attend a 1-day workshop on species distribution
modeling (SDM) with Marine Geospatial Ecology Tools (MGET), on Saturday, 14
December in Dunedin, New Zealand, following the conclusion of the Society
for Marine Mammalogy's Biennial Conference. MGET is a free, open source
geoprocessing toolbox that plugs into ArcGIS and contains over 250 tools
that are useful in a variety of research and spatial planning problems. MGET
can help you with scenarios as simple as easily acquiring oceanographic data
in GIS-compatible formats, to more complicated problems such as simulating
the dispersal of larvae by ocean currents. To learn more about MGET, please
go to http://mgel.env.duke.edu/mget.
The workshop will be a mixture of presentations and practical exercises.
Attendees will be seated at ArcGIS workstations provided by the University
of Otago and over the course of the day we will work though the steps of an
SDM example. Starting with records of sightings of a marine animal, we will
build up a geoprocessing workflow that links the sightings to oceanographic
observations, explores statistical properties of the data, fits a
multivariate statistical model relating the animal's distribution to
oceanographic covariates, evaluates the model's performance, and predicts
the animal's distribution across the seascape using remote sensing imagery.
This example will take heavy advantage the interoperation MGET establishes
between ArcGIS and the popular statistics program R, which will be used to
perform the statistical operations.
The specific model we will build depends on attendee interests, but likely
to be either models of cetacean habitat suitability and density from line
transect survey data (using distance sampling techniques), models of fish
habitat suitability and CPUE from fishery catch data, or a model of seabird
habitat use or possibly behavior from telemetry data. Regardless of which
example we select, the overall workflow will cover many of the steps needed
for all of them, and we'll discuss important difference between them.
The cost will be NZ$25-35 per person (we are still working out the exact
amount), plus lunch unless you bring your own. Space is limited. To
register, or if you have any questions, please contact Judy Rodda
(hamrodda at yahoo.com) and myself. We hope to see you there.
jason.roberts at duke.edu
Marine Geospatial Ecology Lab, Duke University
More information about the MARMAM