search

glos / myglos

Repository for MyGLOS (GLOS Data Portal)
GNU General Public License v3.0
2 stars 2 forks source link

Ice Cover #183

Open kkoch opened 6 years ago

kkoch commented 6 years ago

Interest / hot topic is ice cover for the GL. We currently have the GLCFS ice thickness model.

See:

  1. Whether we could tweak the existing GLCFS display on the portal. Ed brought up that he would prefer gray-black color scheme instead of the current blue. Also that the legend is in metric and is not currently affected if the user changes their preference to English units. RPS thinks this would be fairly easy to accomplish but will provide an LOE.

  2. We also looked at GLERL’s ice cover nowcasts https://www.glerl.noaa.gov/data/ice and determined that there are GRIB files behind their maps. Kelly will look at that and see how easy it would be to pull that data into the portal. Another idea she had was the possibility of creating a specialty map that users could get to at portal.glos.us/ice and the additional possibility of having a linked theme from the map itself. She think the LOE would be minimal but will also cost that out for Tad before jumping into.

  3. Erddap info from glerl > https://coastwatch.glerl.noaa.gov/erddap/tabledap/glerlIce.csv?time%2CSuperior%2CMichigan%2CHuron%2CErie%2COntario%2CSt_Clair%2CGL_Total&time%3E=2001-07-01&time%3C=2018-07-01

@kknee will investigate and provide LOE

kknee commented 6 years ago

  1. Adjusting the color scheme only take a few minutes, please let me know what you would like. The EDS legends should be responsive to units, I'll have the EDS team go ahead and fix that.

  2. Need to clarify this one. The GRIB files are from the U.S. National Ice Center (http://www.natice.noaa.gov/products/grib.html). Is that the data that is behind the GLERL maps? I thought we didn't have access to the GLERL data and were waiting for it to be added to ERDDAP? Regarding the GRIB files, they look pretty standard - next step is to determine what variables are available, which one to use, and how to visualize.

  3. I didn't see anything on the Coastwatch ERDDAP server that would allow us to map ice cover. Did we want to do something else with that CSV?

kknee commented 6 years ago

Regarding the GRIB files, they have a single variable, called “Ice_cover_1land_0sea”, description reads “Ice cover 1 land 0sea @ surface”. The units attribute is blank.

kkoch commented 6 years ago

Decided at DMAC meeting for RPS to go ahead with testing on the GRIB's. @tslawecki to verify with Ed on color scheme for modifying the EDS legends. @tslawecki also working on a tracker similar to the maumee nutrient tracker for ice.

tslawecki commented 6 years ago

@kknee - what would it take to show GRIBs in portal - is that still an EDS possibility?

tslawecki commented 6 years ago

Posted working development version of a viewer for ice cover data at http://34.213.217.93/GL-ice.html. I am looking for feedback before refining the look, especially on what the introduction and information text should say.

There are several planned tweaks - may turn this into a separate Github issue for tracking ... or better yet its own repo?:

tslawecki commented 6 years ago

@kknee Question, some guidance on showing spatial extent of ice in portal:

kknee commented 6 years ago

@tslawecki thanks for the guidance.

on GLCFS, confirmed that we collect concentration in addition to thickness. Will make the switch.

kknee commented 6 years ago

Here's a screen shot of the NIC data - still working out units/legend:

image

kknee commented 6 years ago

Also, colors have been updated to greyscale for ice thickness:

image

kknee commented 6 years ago

@tslawecki @kkoch do we want to use the same legend for the NIC data as they do, see: http://www.natice.noaa.gov/pub/special/great_lakes/2018/charts/composite_east/el180206color.jpg

kknee commented 6 years ago

Also, I called both phone numbers listed on the image (http://www.natice.noaa.gov/pub/special/great_lakes/2018/charts/composite_east/el180206color.jpg)to inquire about the units for 'tenths'. The US number is out of service and the Canadians had no idea what I was talking about...

The way the data is described, seems like it's a concentration, or percent, but I'm not 100% sure.

kkoch commented 6 years ago

I found this > http://www.natice.noaa.gov/products/egg_code.html

Does that help?

ColinNBrooks commented 6 years ago

Does this legend help too: https://coastwatch.glerl.noaa.gov/glsea/cur/glsea_cur.png

kknee commented 6 years ago

ok, according to the 'egg code' seems like tenths is standard for sea ice and what we're looking at is concentration reported in tenths. Is that the best way to viz for GLOS or do we want to translate to %, similar to the link that @ColinNBrooks provided.

kknee commented 6 years ago

@ColinNBrooks @kkoch @tslawecki anyone have an opinion on tenths vs. percent? We're harvesting the data onto EDS staging, but before we promote to production and put in the portal I'd like to nail down the legend.

ColinNBrooks commented 6 years ago

I like percent categories because I find them easy to interpret. NOAA GLERL uses some categories that I'm used to seeing when interpreting their GLSEA ice mapping product (<10%, 10-39%, 40-69%, 70-89%, 90-99%, 100%) as displayed at https://coastwatch.glerl.noaa.gov/glsea/cur/glsea_cur.png If we want to try 10% categories to start with, that would work, but I suspect that GLERL is on to something with they way they summarize the NIC ice cover data.

kknee commented 6 years ago

GLCFS Ice Concentration is on dev if anyone wants to check it out. It will go live in the next data catalog update:

image

ColinNBrooks commented 6 years ago

Couple of quick things - the legend goes from 0.00 to 1.00 in 0.11 fractions rather than a % legend - do we actually want percent there? If so I'd multiply the values by 100 (and maybe change them to 10% size instead of 0.11).

Also, when I used the virtual buoy function, I got a response that the ice concentration was 0.95 meters... wrong units!

image

kknee commented 6 years ago

@kkoch is there a GN record for the National Ice Center data - I didn't see anything obvious when I searched for 'ice', but we have 42 restuls? If not, can you make one?

kkoch commented 6 years ago

There is not. I will go do so now.

kkoch commented 6 years ago

GN record > http://data.glos.us/metadata/srv/eng/main.home?uuid=f37586ca-a895-4803-8903-85a5c8a69037

kknee commented 6 years ago

@kkoch that link is throwing a 500 error 'operation not allowed'

while you look into that, can you post the layer title, abstract and keywords that you used?

kkoch commented 6 years ago

Sorry. Fixed.

Title: National Ice Center (NIC) Great Lakes Ice Analysis Products Abstract: The SIGRID-3 vector archive format is one of the World Meteorological Organization (WMO) standards for archiving digital ice charts in the Global Digital Sea Ice Data Bank (GDSIDB). The WMO ice chart archiving formats are the Sea Ice Grid (SIGRID) format developed in 1981 and formalized in 1989 and its successor SIGRID-2. The National Ice Center digital Ice Analysis charts (Hemispheric, Regional and Daily are encoded in SIGRID-3 and have two main components: the shapefile containing the Ice Analysis ice information (ice polygons and related attributes) and the metadata describing the Ice Analysis data under the SIGRID-3 format. The National Ice Center legacy ice data set contains hemispheric spatial data files from 1972 through 2004 inclusive. Regional Spatial data files of the traditional "seas"(ex Beaufort Sea) have been available since 1997. Legacy data are available in various formats from the National Snow and Ice Data Center in Boulder, Colorado ( http://nsidc.org/ ). Legacy Great Lakes charts are available from the Great Lakes Environmental Research Laboratory in Ann Arbor, Michigan ( http://www.glerl.noaa.gov/ ). Currently (October, 2006), spatial data files are available for each hemisphere, and for each sub-regional area, of which there are approximately 62. The ice analyses for each sea or sub-sea are also available in a jpeg format. They represent the ice conditions for the week in which they are published. Data for the analyses can go back 96 hours from when they are completed. They are dated with the week they are published. They are based on an analysis and integration of all available data on ice conditions, including weather and oceanographic information, visual observations from shore, ship and aircraft, airborne radar, satellite imagery and climatological information. The Regional Ice Analyses describe areas of differing ice conditions using the WMO ice observing standards, and indicate where significant changes in the amount, stage of development and form of the predominant ice types have occurred. The Regional Ice Analyses for all of the Northern and Southern Hemisphere Seas are produced every other week. The analyses for the Bering, Chukchi, Beaufort, Arctic Basin (High Arctic), Northern East Greenland, Barents, Kara and White Seas and the Cook Inlet are produced at least weekly. The Ross Sea is produced weekly during Austral Summer navigation season. The Chesapeake and Delaware Bay areas are produced weekly when there is ice. The analyses for the Great Lakes (when there is ice and in collaboration with the Canadian Ice Service), the Bering Sea, the Cook Inlet and the Chukchi and Beaufort Seas are produced twice per week when there is an ice edge present. The National Ice Center also produced daily ice edges in text format for both the Northern and Southern Hemispheres. Keywords: sea ice, ice type, ice age, ice concentration, form of ice, floe size, sea ice concentration, ice types, sea ice age, ice floes, NIC, National Ice Center, Great Lakes Categories: models, climatology

tslawecki commented 6 years ago

@kknee, what's left to make the NIC data live in the portal? It'd be convenient to show to GLOS Tuesday AM (4/17).