WIS hindcast data for winds/waves (data source)

[birdage]

WIS provides hindcast data for a large number of nodes (see image), i dont think this information is available in the catalog, it is however available on their website in txt format.

Extract from an available text file

Atlantic
61001_v01               MEAN MONTHLY                            MAXIMUM CONDITIONS AT HMAX
  YRMT   No.OBS       HMEAN      TPMEAN        HMAX         TP     WAVE DIR     WND SPD    WIND DIR
 198101    240         1.45        8.29        2.57       10.79        5.00        7.90       58.00

[birdage]

@Bobfrat any thoughts?

[rsignell-usgs]

Text files. Yeesh. Crying out for ingest to 52 north & SOS, no?

[kwilcox]

Do the local RAs have this data available in their SOS servers?

[birdage]

@kwilcox not sure, me and bob were chatting about data sources and this was one that came up that i dont think its available through the catalog

@rsignell-usgs would be good if we could get at them a different way, there is close to 2500 nodes and we wouldnt want to have to pull down every text file.

[emiliom]

I'm intrigued. I wasn't aware of these hindcasts, and don't recognize what "WIS" is. Can someone post a link or two to the data files and a description of WIS? Thanks.

[Bobfrat]

http://wis.usace.army.mil/

US Army Corps of Engineers Wave Information Studies for hourly, long-term (20+ years) wave climatologies along all US coastlines.

[emiliom]

Thanks. It looks like a useful resource. I'll bring it up in NANOOS in our weekly call today.

[birdage]

@emiliom @Bobfrat i think im going to parse the resources and add them to the waves notebook.

[kwilcox]

Long term, we should think about adding connectors to this data in pyoos and sensor-web-harvester

[dpsnowden]

Before you parse these send a note to Julie Thomas at SCCOOS. She's also the CDIP manager. I think she did something with this data.

[birdage]

@kwilcox that would be a good plan

[birdage]

@dpsnowden @emiliom i had some time last night so wrote a parser for the wis data, it captures all the station information and monthly mean/max points and stores it in a DB that can be queried. im going to drop Julie Thomas an email today when i have a sec.

http://nbviewer.ipython.org/github/birdage/system-test/blob/76315348f1678719b879aebea020fa4351a8c3e4/theme2_extreme_events/ioos_extreme_events_waves.ipynb

[birdage]

@dpsnowden do you have contact info for Julie Thomas

At the minute rather than the DB approach which is not very sustainable long term, we are looking at adding a collector to pyoos to handle wis requests, this has been created as an issue in pyoos

[robragsdale]

@birdage Julie Thomas' email is jot@cdip.ucsd.edu

[birdage]

@robragsdale thanks!

[dpsnowden]

Please review the documenting issues section. I think it is more important to log this issue and move on than to continue to find clever work arounds. The project goals focus on identifying things like this and calling them out, rather than spending too much time making up for the deficiencies. A good outcome from this notebook would be a well documented issue saying something like:

• USACE should consider publishing WIS data via a web service and then creating and registering a standards compliant metadata record in an agency metadata repository.

Then, we wrap up the notebook and move on to another topic. I don't think that having the plots on the notebook adds enough value to warrant you spending time on it.

[dpsnowden]

I labeled this issue as wontfix and assigned it to the "Post SIT Enhancements" milestone to indicate it's not something we can do during this project but still needs to be done.

[kknee]

Agree with the WIS strategy. Will document the notebook appropriately and will create a separate issue here without all this extraneous discussion.

However, let's not give up on waves entirely. At a minimum we can use CDIP stations for expanding the model-obs comparison note books. Also looks like a handful of west coast stations have enough data to continue the extremes notebook.

[birdage]

@dpsnowden @kknee regarding the WIS data, i actually wrote a pyoos wis extractor one evening last week, and it will be merged in to the main pyoos code when i do some tidy up.

link: https://github.com/ioos/pyoos/pull/33