Marine GIS Data Model Example data problems for marine analysis

1
Marine GIS Data ModelExample data problems for
marine analysis

• P.N. Halpin
• E.A. Treml
• Nicholas School of the Environment
• And Earth Sciences
• Duke University

2
Marine GIS Data Model
Georelational model
Geodatabase model
Halpin Treml 2002
3
Marine GIS Data Model sampling

• Marine data is collected through a wide range of
• Methods
• Instruments
• Time scales
• Periodicity
• Precision
• Accuracy

Halpin Treml 2002
4
Marine GIS Data Model

• Marine GIS Applications are especially complex
  due to
• Wide array of sampling media
• Data types
• Data dependencies (e.g. time, effort)

Halpin Treml 2002
5
Marine GIS Data Model
Recently a team got together to begin the
development of a standard marine data model
The central issue How do we do a better job of
analyzing a 4 dimensional marine world with 2.5D
GIS software
Halpin Treml 2002
6
Marine GIS Data Model Process
Halpin Treml 2002
7
Marine GIS Data Model
What is a GIS data model? An ArcGIS data model
provides a basic template for implementing GIS
projects (i.e., inputting, formatting,
geoprocessing, and sharing data, creating maps,
performing analyses, etc.) For developers, it
provides a basic framework for writing program
code and maintaining applications
Halpin Treml 2002
8
Marine GIS Data Model
How does a GIS data model work? An ArcGIS data
model is written in a graphical model language
(Visiotm) that provides both a diagram of the
data structure as well as UML (Universal Modeling
Language) code. UML code can be directly ported
to the GIS to create an empty database structure
defining all of the necessary relationships and
dependencies.
Halpin Treml 2002
9
Marine GIS Data Model
Halpin Treml 2002
10
Halpin Treml 2002
11
Marine GIS Data Model
Georelational model
Geodatabase model
Halpin Treml 2002
12
Marine GIS Data Model
An extension of GIS features to better fit marine
applications
This is a conceptual diagram How implement this
is open for discussion
Halpin Treml 2002
13
Marine GIS Data Model Example
Goal To begin with example data that fills as
many different data types as possible.
Halpin Treml 2002
14
Marine GIS Data Model Example
Gulf of Maine Selected due to rich history of
data collection across a wide range of marine
science and management disciplines.

Halpin Treml 2002
15
Gulf of Maine Sounding and Elevation Data
National Ocean Service (NOS) Hydrographic Data
Base
Shiptrack bathymetry data (GLOBEC)
Grid bathymetry data (GLOBEC)
http//pubs.usgs.gov/openfile/of98-801/bathy/data.
htm
Halpin Treml 2002
16
Gulf of Maine Bathymetry GRIDS Contours
15 arc-sec bathymetry rasters 100m contours
vector
http//pubs.usgs.gov/openfile/of98-801/bathy/data.
htm
Halpin Treml 2002
17
Gulf of Maine
Dynamic hydrographic circulation models
Streamfunction
Finite Element Model Model TIN nodes
Dartmouth College, Hanover, New Hampshire, U.S.A.
http//www-nml.dartmouth.edu/circmods/gom.html
Halpin Treml 2002
18
Gulf of Maine
Dynamic hydrographic circulation models
Drifter tracking simulation
Dartmouth College, Hanover, New Hampshire, U.S.A.
http//www-nml.dartmouth.edu/circmods/gom.html
Halpin Treml 2002
19
Gulf of Maine
SST Gradients Fronts
Synoptic data collection of SST (SeaWiFs-AVHRR)
processed to develop gradient and distance to
front raster data sets.
SST Fronts
Distance from SST fronts (surface)
Halpin Treml 2002
20
Gulf of Maine
Fisheries survey data Ground fish stocks
Halpin Treml 2002
21
Marine GIS Data Model Example
Goal To begin with example data that fills as
many different data types as possible.
Halpin Treml 2002
22
Gulf of Maine
Fixed X,Y,Z Data NMS/NOAA Baseline Point
Data Offshore boundaries
Data source USGS-NOAA
Halpin Treml 2002
23
Gulf of Maine
Fixed X,Y,Z Data NOS Bathymetric Point Data
Data source USGS-NOS
Halpin Treml 2002
24
Gulf of Maine
Cruise Survey Observation Data Harbor Porpoise
(Fall 1999).
Data source Debi Palka, NMFS
Halpin Treml 2002
25
Gulf of Maine
Cruise Survey Observation Data 1) SPP all
species are harbor porpoises 2) strata. This
could be high density (high), intermediate
density (interm), inshore, offshore, and
extra.   3) team. This can be either upper
or lower. Some of the groups of porpoises were
seen by only the upper team, others by only the
lower team, and some were seen by both teams.
All sightings are included no matter what its
duplicate status.   4) year, month, day.
Data source Debi Palka, NMFS
Halpin Treml 2002
26
Gulf of Maine
Cruise Survey Observation Data 5) transect.
This is a straight piece of effort, which could
be made of several legs.   6) leg. This is part
of a transect where everything is the same, ie.
observer placement, weather.   7) obsnum. This
is a unique sighting number for an animal group
seen by a team on a particular day.   8)
sighttype. This can be either O for original
or F for follow-on sighting. That is O is
the first time that group was seen and F is the
same group but a later surfacing. So for a group
that had two recorded surfacing, the obsnum is
the same but the first has sighttypeO and the
second has sighttypeF.
Data source Debi Palka, NMFS
Halpin Treml 2002
27
Gulf of Maine
Cruise Survey Observation Data 10) sizeb. This
is the best estimate for the size of the
group.   11) raddist. This is the estimate of
the distance (in meter) between the ship and the
group.   12) angle. This is the estimate of the
angle between the track line (0 degrees) and line
of sight to the group. 90 degrees is 90 degrees
to the starboard, 270 degrees is 90 degrees to
the port.   13) swmdir. Estimated direction the
group was swimming at the time of the recording.
Use same compass angles as in 12) angle.
Data source Debi Palka, NMFS
Halpin Treml 2002
28
Gulf of Maine
Cruise Survey Observation Data 14) sightlat and
sightlong. Position of the ship at the time of
the sighting. NOT the exact position of the
group. Format is degree.fraction of a degree.
For example, 43 degrees 30 minutes is
40.50.   15) beaufort. Beaufort sea state. Not
necessarily the wind speed. Format is the normal
beaufort sea state (usually an integer) and an
estimation of the percentage between the integer
beaufort sea states. So beaufort 2.9 is almost a
beaufort 3 but not quite.   16) shipspeed.
Ships speed through the water.   17) smg.
Speed made good over the ground.
Data source Debi Palka, NMFS
Halpin Treml 2002
29
Gulf of Maine
Recording Buoys Hydrodynamic Data
Data source
Halpin Treml 2002
30
Gulf of Maine
Recording Buoys Selected Buoy data attributes
(1978 to present) Wind Direction (WDIR)
measured ever 10 minutes 24hours a day Wind Speed
(WSPD) measured ever 10 minutes 24hours a
day Wind Gust (GST) measured every hour 24hours
a day Wave Height (WVHT) measured every hour
24hours a day Dominant Wave Period (DPD)
measured every hour 24hours a day Atmospheric
Pressure (PRES) measured every hour 24hours a
day Pressure Tendency (PTDY) measured every hour
24hours a day Water Temperature (WTMP) measured
every hour 24hours a day
Note for standard meteorological data, there are
approximately 200k measurements since 1978,
approximately 2400 for the duration of this
marine mammal study.
Data source http//www.ndbc.noaa.gov/station_hist
ory.phtml?station44005
Halpin Treml 2002
31
Gulf of Maine
Harbor Porpoise Telemetry Data 5 animals
tracked as point observations over time (Fall
1999).
Data source Tara Cox, Duke University
Halpin Treml 2002
32
Gulf of Maine
Shoreline Shoreline Vector Data NOAA Medium
Resolution Shoreline .
Data source NOAA
Halpin Treml 2002
33
Gulf of Maine
Bathymetry Research cruise route
Data source Anonymous
Halpin Treml 2002
34
Gulf of Maine
Prey (Silver Hake lt 40cm) Trawl Data Silver
Hake lt 40cm (Fall 1999).
Data source Tara Cox, Duke University
Halpin Treml 2002
35
Gulf of Maine
Prey (Silver Hake lt 40cm) Trawl Data Silver
Hake lt 40cm (Fall 1999).
Example Time-duration vector
Data source Tara Cox, Duke University
Halpin Treml 2002
36
Gulf of Maine
Cruise Survey Effort Data Time duration effort
(Fall 1999).
Data source Debi Palka, NEFMS
Data source Debi Palka, NMFS
Halpin Treml 2002
37
Gulf of Maine
Cruise Survey Time duration effort data (Fall
1999). Same as in sighting data for variables
strata, year, month, day, transect, leg, and
beaufort.   1) begtime, endtime. Time the leg
began and ended. Format same as sighttime
above.   2) beglat, beglong. Corresponding
latitude and longitude for the time the leg
began. Format same as sightlat above.   3)
endlat,endlong. Corresponding latitude and
longitude for the time the leg ended.
Data source Debi Palka, NMFS
Halpin Treml 2002
38
Gulf of Maine
SBNMS Boundary Boundary / Area Data Stelwagen
Bank (SBNMS)
Data source NOAA
Halpin Treml 2002
39
Gulf of Maine
Time-Area Closures Time-area Data
Data source
Halpin Treml 2002
40
Gulf of Maine
Bathymetry Depth Data 15 minute bathymetry
raster grid .
Data source
Halpin Treml 2002
41
Gulf of Maine
Distance from Fronts SST / Front / Distance Data
Raster Data Model?
?
Data source Rob Schick NEA http//marinegis.org
Halpin Treml 2002
42
Marine GIS Data Model
Example Data and the current data model The
existing data model defines a generic,
conceptually designed framework. This current
structure requires extensive manipulation /
customization to allow for the incorporation of
real data.
Halpin Treml 2002
43
Marine GIS Data Model
Example Data Questions What types of data are
missing? Should be included? What relationships
should be included? What should be optional vs.
required?
Halpin Treml 2002
44
Marine GIS Data Model
Questions for Discussion Should temporal
components be treated as optional attributes
within a geodatabase or as required variables of
new features classes? Should there be a more
generic dynamic / temporal dependent feature
class extensions (useful to marine, atmospheric,
hydrologic disciplines?). Should there be a
single marine data model or more field specific
models benthic mapping, navigation,
pelagic-animal tracking, oceanographic, bottom
mapping?
Halpin Treml 2002
45

• Questions?