akleeman/slocum

slocum -- a tool for ocean passage planning

slocum's Issues

Forecast areas spanning date line / prime meridian

While updating rtefcst it seemed to me the 'cleanest' way to deal with the date line/prime meridian issue could be like this: The fcst Dataset stores relative lons from 0..max_lon plus a reference starting (= eastern) longitude (similar to the time dimension). All internal manipulations can work in the relative coordinate frame, with consistent sorting and no wrap around issues. A simple method or utility function converts back to absolute lons where required (analog to netDCF4.num2date).

So if I wanted absolute lons from 170E to 170W my reference lon would be +170 and fcst['longitude'].data run from 0..20.

I'm guessing though that this might have quite a knock-on effect on poseidon, emaillib and griblib? And you may already have a better/easier solution anyway...

Add probability maps.

It would be nice to allow users to choose a threshold and view, for example, the probability of wind exceeding 25 knots for a given forecast time.

Add OSCAR currents to forecast requests.

http://podaac.jpl.nasa.gov/dataset/OSCAR_L4_OC_third-deg

Allow for subscriptions.

Saildocs allows users to request that a forecast be sent regularly, this could be useful functionality.

Installation error

please add netCDF4 to the conda install documentation like so
conda install python=2.7 pip nose mock numpy scipy xray basemap netCDF4

Cheers,
-=Tucker

Improve the query syntax.

The saildocs syntax is convenient for current saildocs users, but is otherwise a bit difficult to understand. The Zygrib syntax could be an improvement.