This is a guide to the bathymetry service used by ComMIT to
download sample grids for training purposes.
The bathymetry service responds to HTTP GET requests in the following format:
http://serverpath/ComMIT/bathymetry/{verb}.{format}?params
Fields are as follows:
- verb: This is the basic command. Current verbs are
"search" and "create". See below.
- format This is the output format. For search
command, "csv" is supported. For create command, "most" is
supported.
- params: These are the URL parameters. They have slightly
different meanings depending on the command.
- xmin, ymin, xmax, ymax. The extent of interest,
in decimal degrees, 0-360.
- cellsize
This is the desired cell size, in decimal
degrees. Appending a "s" specifies the cell
size in arc-seconds, and a "m" specifies it in
arc-minutes; these are recommended to prevent
rounding errors. For example, "3s" and "0.000833333"
are roughly equivalent, but with the former,
nodes are guaranteed to be exactly 3 arc-seconds
apart.
- xcellsize, ycellsize
Same as "cellsize" but with different grid spacing in
x and y. For the search feature, the minimum of these
is used for searching; they are only used separately
when generating the grid.
- nx, ny
If specified along with xmin, ymin, and a cellsize,
the extent is calculated to produce a grid
with the exact node count in x and y specified.
xmax and ymax will be ignored in this case.
search command
Returns a list of source grids available, optionally filtered by
extent and cell size.
If the extent is specified with the xmin, ymin,
xmax, ymax parameters, only grids that overlap (partly or completely)
the specified envelope are returned.
If, in addition,
the cell size is specified, a minimum set of grids required
to cover the request extent at the desired resolution is
returned. This is the set that will be used for
gridding. If a valid set
of extent parameters (xmin/ymin/xmax/ymax) are not
specified, these are ignored. If cellsize is not specified,
all overlapping grids are returned. If xcellsize and ycellsize
are used, the minimum of these is used when searching;
they are only used separately when actually generating
the grid.
Response format (CSV):
filename, dirname, xcellsize, ycellsize, xmin, xmax, ymin, ymax
create command
Requests creation of the specified grid. The current file
format supported is "most". Both extent and cell size
parameters must be set.
Responses:
The grid can take some time to create, so a response is
sent immediately with a status update if the grid is still
being processed. The HTTP status code indicates the status:
- 200 (HttpServletResponse.SC_OK): The grid is complete. The response will contain
the requested grid content in its entirety.
- 202 (HttpServletResponse.SC_ACCEPTED): The grid is in
progress. The content will look like:
{percentdone} {message}
The client should wait a few seconds then repeat the exact
same request for an update. percentdone and message can be
used for a progress bar.
- 500 (HttpServletResponse.SC_INTERNAL_SERVER_ERROR) An error
occurred while generating. This is reported once; repeating
the request will start a new grid generation process.
grid command
Downloads an entire source grid: grid/dirname/filename
About the dataset
The ComMIT grid tool provides combined bathymetric/topographic
gridded digital elevation models for tsunami modeling. While the dataset is useful for training and
demonstration purposes, it was automatically generated primarily from
low-resolution bathymetric data sources, and has received limited
manual review. It should not be used for hazard mapping or similar
purposes.
The high-resolution (3-arc-second) dataset is primarily derived
from two sources:
- Bathymetry is from the GEBCO_2014
30 arc-second grid version 20141103, produced by the General Bathymetry Chart of the Oceans (GEBCO) operating
under the Intergovernmental Oceanographic Commission (IOC) (of UNESCO) and the International Hydrographic Organization (IHO).
The GEBCO dataset is compiled from many different global and regional data sets, evaluated and edited, and shifted to
reference mean sea level vertically, and WGS84 horizontally. This data has been INTERPOLATED FROM 30 ARC-SECONDS TO 3 ARC-SECONDS
to match the topography data set. This makes it inappropriate for any use except training.
- Topography is from the CGIAR
SRTM 90m version 4.1 digital elevation model produced by the CGIAR Consortium
for Spatial Information. This 3 arc-second (~90 meters at the equator) dataset
is a processed version of the Shuttle Radar Topography Mission (SRTM) data created by NASA.
The ComMIT bathymetry dataset was created by first removing land values in the GEBCO data, then interpolated from
30 arc-seconds onto the 3 arc-second SRTM grid with a simple bilinear interpolation. Land values are then simply replaced with the SRTM data.
This simple process preserves the coastline well, and produces a dataset is relatively inaccurate in bathymetry, but tends to be
stable for most tsunami sources in the NCTR propagation database, and is therefore useful for
training purposes. NCTR recommends it be used only for this purpose.