associating SST and bottom depth

101 views
Skip to first unread message

Fowler, Mark

unread,
Aug 27, 2015, 10:44:25 AM8/27/15
to ec...@ocean.mit.edu, fo...@hycom.org

Hello,

 

I’ve been trying to find a spatial data source that includes both sea surface temperature and bottom depth. Between the ECCO, HYCOM and NOAA websites I can find any number of variants with one or the other, but not one with both together. I’m trying to avoid merging separate topographical and SST data sources during remote download processing – repeatability concerns associated with subsampling resolutions, efficiency concerns using a large lookup grid (experimented doing this part post-extraction in R). So I’m trying to find a data source in which the two variables conveniently coexist and will remain the same (across updates) for a given location and date-time, unless a change is simply a consequence of edit revisions. Hoping for global, but North Atlantic would suffice. Anybody know of an option here?

 

Motivation – I’m fitting tracks of various shark species tagged with popoffs, where tag measures of sea surface temperature are reconciled with satellite measures of sea surface temperature to estimate the locations of the animals. The tags also  monitor depth, so I would simply like to eliminate satellite SST candidates associated with bottom depths that are shoaler than the maximum diving depth of an animal for a given day.

 

Mark Fowler
Population Ecology Division
Bedford Inst of Oceanography
Dept Fisheries & Oceans
Dartmouth NS Canada
B2Y 4A2
Tel. (902) 426-3529
Fax (902) 426-9710
Email Mark....@dfo-mpo.gc.ca

 

Alan

unread,
Aug 27, 2015, 11:24:27 AM8/27/15
to HYCOM.org Forum, ec...@ocean.mit.edu, Mark....@dfo-mpo.gc.ca
All the GLBa0.08 90.6 to 91.1 cases use the same bathymetry, which is available as a netCDF ftp://ftp.hycom.org/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc

Let me know if you prefer the GLBu0.08 (interpolated from GLBa0.08) fields.  Its bathymetry is not currently on-line, but I can make it available too.

The reanalysis (19.0 and 19.1) uses a different bathymetry and it is already on-line ftp://ftp.hycom.org/datasets/GLBu0.08/expt_19.1/topo/depth_GLBu0.08_07.nc

Alan.

Patrick Heimbach

unread,
Aug 27, 2015, 1:12:55 PM8/27/15
to Alan, HYCOM.org Forum, Mark....@dfo-mpo.gc.ca, ecco-s...@mit.edu

Hi Alan,

similar for ECCO, the latest product (we call it ECCO version 4, release 1) is available online for download. Please take a look at
http://mit.ecco-group.org/opendap/ecco_for_las/version_4/release1/contents.html

Physical fields are available at
http://mit.ecco-group.org/opendap/ecco_for_las/version_4/release1/nctiles/contents.html
Grid files are in
http://mit.ecco-group.org/opendap/ecco_for_las/version_4/release1/nctiles_grid/contents.html

Some rather technical documentation is in
http://mit.ecco-group.org/opendap/ecco_for_las/version_4/release1/ancillary_data/contents.html
but I point you to a paper that is about to appear in GMD, currently in GMDD at
http://www.geosci-model-dev-discuss.net/8/3653/2015/gmdd-8-3653-2015.html

Please let me know if you have further questions, ideally to
ecco-s...@mit.edu

Cheers
-Patrick
--------------------------------------------------------
Patrick Heimbach, Ph.D. | http://heimbach.wordpress.com

* The University of Texas at Austin *
The Institute for Computational Engineering and Sciences
Institute for Geophysics | Jackson School of Geosciences
201 East 24th Street, POB 4.232 | Austin, TX 78712 | USA
FON: +1-512-232-7694 | Email: heim...@utexas.edu

* Massachusetts Institute of Technology *
Department of Earth, Atmospheric, and Planetary Sciences
77 Massachusetts Ave, 54-1420 | Cambridge MA 02139 | USA
FON: +1-617-253-5259 | Email: heim...@mit.edu
--------------------------------------------------------




Fowler, Mark

unread,
Aug 27, 2015, 3:27:43 PM8/27/15
to Patrick Heimbach, Alan, HYCOM.org Forum, ecco-s...@mit.edu, vardis.m...@jpl.nasa.gov
Alan and Patrick,
 
Thank you for the fast response. Focusing on HYCOM for now, let me know if I’m asking too much below.
 
I'm having difficulty deciphering how to access the products. Part of this might be hyperlinks that don’t work for me (I could not access HYCOM at all 2 days ago, now some links work and some not). Is there some example syntax for this? I'm familiar with downloading oceanographic data in netcdf format from the NOAA ERDDAP site. There I use a program to submit a query with a date and spatial window dimensions (downloading daily packets), and process the download into a usable format (for R and ADMB). But I know the variable names of the nc contents when I do this, and I cannot find  a data dictionary for depth_GLBa0.08_09. I think blkdat.input_91.1 provides much but not all of the data definition for GLBa0.08. But I think these are all category variables that define the objective data, and do not include the objective data itself.
 
For instance if I want the bottom depth and sea surface temperature for a given date (and one time of day if multiple choices) within a spatial window defined by latitudes 30 and 60, longitudes -74 and -50, [or whatever projection scale applies] how do I accomplish it? Below I copy syntax as it would look for NOAA products, if that helps - I highlighted what I believe to be mission-critical lines. In the case of HYCOM I expect I would be doing two extractions (‘server’ below becoming server1 and server2), one for SST and one for bottom depth, and these would be parallel files using the same query, so their results readily merged.
 
                           server = "http://coastwatch.pfeg.noaa.gov/erddap/griddap/jplMURSST.nc?analysed_sst",
    for (d in datesteps) {
        opt<-paste("[(",fmtDate(d),"):1:(",fmtDate(d),")][(",minLat,"):",stride,":(",maxLat,")][(",minLon,"):",stride,":(",maxLon,")]",sep="")
        link <- paste(server, opt, ",mask",opt,sep = "")
        y1 <- date.mdy(d - ifelse(nday == "1day", 1, 3))$year
        d1 <- (d - ifelse(nday == "1day", 1, 3)) - mdy.date(month = 1, day = 1, year = y1) + 1
        y2 <- y1
        d2 <- d1
        filename <- paste("erd", y1, fmtDay(d1), "_", y2, fmtDay(d2), "_sst.nc", sep = "")
        dest <- paste(folder, filename, sep = "/")
print(dest)
#network hiccups can be a problem; use error trapping; but let it stop if it keeps happening (e.g. a server down)
        hiccup=try(download.file(link, dest, mode = "wb"))
        if(class(hiccup)=="try-error") {
        hiccup=try(download.file(link, dest, mode = "wb"))
        }
#convert binary download file to ASCII file for ukfsst
        temp.nc <- open.ncdf(dest)
        m=get.var.ncdf(temp.nc,'mask')
        x=get.var.ncdf(temp.nc,'latitude')
        y=get.var.ncdf(temp.nc,'longitude')
       z=get.var.ncdf(temp.nc,'analysed_sst')
         close.ncdf(temp.nc)
         file.remove(dest)
               filename <- paste("erd", y1, fmtDay(d1), "_", y2, fmtDay(d2), "_sst.xyz", sep = "")
               dest <- paste(folder, filename, sep = "/")
         xyz=data.frame(cbind(expand.grid(x, y), as.vector(z)))
         xym=data.frame(cbind(expand.grid(x, y), as.vector(m)))

Michael McDonald

unread,
Aug 28, 2015, 12:58:40 PM8/28/15
to Fowler, Mark, HYCOM.org Forum
On Thu, Aug 27, 2015 at 2:53 PM, Fowler, Mark <Mark....@dfo-mpo.gc.ca> wrote:
> I'm having difficulty deciphering how to access the products. Part of this
> might be hyperlinks that don’t work for me (I could not access HYCOM at all
> 2 days ago, now some links work and some not). Is there some example syntax
> for this?


Mark,
Can you please elaborate more on the hycom URLs you are having difficulty with?
What access methods are you wanting to utilize to obtain our data?
What output format are you wanting the hycom data in (netcdf)?

--
Michael McDonald
HYCOM.org Administrator

Michael McDonald

unread,
Aug 28, 2015, 2:03:40 PM8/28/15
to Fowler, Mark, forum
Mark,

> I do want to use netcdf.

Ok, good. Netcdf4 outputs are easy to provide via subsetting.


> I believe I'm after two objects in parallel (one for SST, one for bottom depth).

Ok. Sea surface temperature is just querying the water_temp varuable at Level 0.

What time range are you wanting to get? The latest data, a range of data...


> I think GLBa0.08 will have the SST and depth_GLBa0.08_09 the bottom depth, and they are layers of the same grid.
> ftp://ftp.hycom.org/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc

Yes. The bathymetry variable in depth_GLBa0.08_09.nc will give you the
respective depth values.

http://hycom.org/data/glbu0pt08/expt-91pt1

1. start here,
http://ncss.hycom.org/thredds/ncss/grid/GLBu0.08/expt_91.1/dataset.html

2. select the variable (water_temp), time value = single time
(2015-08-28T00:00:00Z), Level = 0, netcdf4 output format

3. then click submit.

Then you can compare/merge the water_temp returned netcdf file with
the depth_GLBa0.08_09.nc file.

If you need data above 80N then you can do the same thing with the
GLBa0.08/expt_91.1 data, http://hycom.org/data/glba0pt08/expt-91pt1

Fowler, Mark

unread,
Aug 31, 2015, 8:34:06 AM8/31/15
to fo...@hycom.org
The time range will vary by tag, and we have over two hundred of these so far between 2005 and 2015. I see from the data.html link you provided that I can specify the date range (and spatial window) in the URL, which I'm familiar with. And as you say, selecting at Level 0 gets the SST. And I download a day at a time with error trapping (just a network hiccup carry on, recurrent failure stop trying). I use R for this, spin through a number of tags at a time. And we do have some Greenland sharks above 80N.

The data.html link solves the SST problem. But I'm still not seeing an equivalent for the bathymetry. How do I pick out the bottom depth? I've got depth_GLBa0.08_09.nc in R right now, and I can figure out how to pull dates and locations, but not the bottom depth. And can it be accessed by URL? I'm thinking about how to efficiently merge the bottom depths and SST values at the same locations, maybe nothing more involved than using the same window in a URL?

Michael McDonald

unread,
Aug 31, 2015, 10:28:44 AM8/31/15
to HYCOM.org Forum, Mark....@dfo-mpo.gc.ca
Mark,

On Monday, August 31, 2015 at 8:34:06 AM UTC-4, Fowler, Mark wrote:
The time range will vary by tag, and we have over two hundred of these so far between 2005 and 2015. I see from the data.html link you provided that I can specify the date range (and spatial window) in the URL, which I'm familiar with. And as you say, selecting at Level 0 gets the SST. And I download a day at a time with error trapping (just a network hiccup carry on, recurrent failure stop trying). I use R for this, spin through a number of tags at a time. And we do have some Greenland sharks above 80N.

If you need data above 80N then you will need to use the GLBa0.08/expt_9x.x which uses a mercator grid between 78°S and 47°N (1/12° equatorial resolution). A bipolar patch is used for regions north of 47°N. The horizontal dimensions of the global grid are 4500 x 3298 grid points resulting in ~7 km spacing on average.


 
 
The data.html link solves the SST problem. But I'm still not seeing an equivalent for the bathymetry. How do I pick out the bottom depth? I've got depth_GLBa0.08_09.nc in R right now, and I can figure out how to pull dates and locations, but not the bottom depth. And can it be accessed by URL?

Yes. Any NetCDF file is accessible via OPeNDAP (start at the top of the THREDDS catalog and click the "* Unaggregated * - All Data" link, then browse to the file you want (just like with the FTP view - except THREDDS only shows NetCDF (*.nc) files).

(or go here to this link)

 

I'm thinking about how to efficiently merge the bottom depths and SST values at the same locations, maybe nothing more involved than using the same window in a URL?

Someone from the forum have any insight on using OPeNDAP/Netcdf data in 'R'... 

Fowler, Mark

unread,
Aug 31, 2015, 12:33:36 PM8/31/15
to Michael McDonald, HYCOM.org Forum

Michael,

 

Now I can access the topography by URL – note this did not work trying to drill down in the catalog, but your hyperlink path did the trick Same applies to data.html – I cannot find a way to get to it without using your hyperlink. Similarly I cannot drill back up from your hyperlinks. Probably something wrong with the system at my end. In any case, I’m operational. Thank you very much.

Michael McDonald

unread,
Aug 31, 2015, 12:53:13 PM8/31/15
to Fowler, Mark, HYCOM.org Forum
On Mon, Aug 31, 2015 at 12:24 PM, Fowler, Mark
<Mark....@dfo-mpo.gc.ca> wrote:
> Now I can access the topography by URL – note this did not work trying to
> drill down in the catalog, but your hyperlink path did the trick Same
> applies to data.html – I cannot find a way to get to it without using your
> hyperlink. Similarly I cannot drill back up from your hyperlinks. Probably
> something wrong with the system at my end. In any case, I’m operational.
> Thank you very much.


You need to browse/traverse the THREDDS catalog until you see a page
with an enumerated listing of "Access:" methods.

e.g.,
http://tds.hycom.org/thredds/catalog/datasets/GLBa0.08/expt_91.1/topo/catalog.html?dataset=datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc

These URLs take you to the respective access portals (OPENDAP,NCSS,
etc), and for the direct methods (FTP, HTTPServer) provide the direct
download link.

e.g., this is the OPENDAP data URL for the bathymetry,

http://tds.hycom.org/thredds/dodsC/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc

Fowler, Mark

unread,
Aug 31, 2015, 5:26:59 PM8/31/15
to Michael McDonald, HYCOM.org Forum
Got it. Until you provided an example hyperlink providing the tds.hycom.org portion of the address, I failed to get there from the home page of hycom.org...[drilled in a little bit]

https://hycom.org/data/glba0pt08/expt-91pt1

The mistake I made was assuming equivalence of the visible Data Server menu of the sidebar and that of collapsed dropdown Data Server menu of the header bar. Same names and I just assumed same contents, without looking. The THREDDS link is provided by the dropdown menu, not the readily visible menu.

Doesn't take much to trip me up.

-----Original Message-----
From: e.m.mc...@gmail.com [mailto:e.m.mc...@gmail.com] On Behalf Of Michael McDonald
Sent: August 31, 2015 1:53 PM
To: Fowler, Mark
Cc: HYCOM.org Forum
Subject: Re: FW: associating SST and bottom depth

Fowler, Mark

unread,
Sep 2, 2015, 9:57:25 AM9/2/15
to Michael McDonald, HYCOM.org Forum
Hi Michael,

I appear to have the address now, but I'm doing something wrong. Anything obvious?

trying URL 'http://ncss.hycom.org/thredds/ncss/grid/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?north=58.5&south=31&west=-57.01047&east=-48.11218&horizStride=1&time_start=2012-09-13T00:00:00Z&time_end=2012-09-31T00:00:00Z&timeStride=1&vertCoord=1&accept=netcdf&var=bathymetry'
Error in download.file(link, dest, mode = "wb") :
cannot open URL 'http://ncss.hycom.org/thredds/ncss/grid/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?north=58.5&south=31&west=-57.01047&east=-48.11218&horizStride=1&time_start=2012-09-13T00:00:00Z&time_end=2012-09-31T00:00:00Z&timeStride=1&vertCoord=1&accept=netcdf&var=bathymetry'
In addition: Warning message:
In download.file(link, dest, mode = "wb") :
cannot open: HTTP status was '400 Bad Request'

#R code if matters
serverdepth="http://ncss.hycom.org/thredds/ncss/grid/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?"
opd<-paste("north=",maxLat,"&south=",minLat,"&west=",minLon,"&east=",maxLon,"&horizStride=1&time_start=2012-09-13T00:00:00Z&time_end=2012-09-31T00:00:00Z&timeStride=1&vertCoord=1&accept=netcdf&var=bathymetry",sep="")
link <- paste(serverdepth, opd, sep = "")
dest <- paste(folder, "depthgrid.nc", sep = "/")
hiccup=try(download.file(link, dest, mode = "wb"))


-----Original Message-----
From: e.m.mc...@gmail.com [mailto:e.m.mc...@gmail.com] On Behalf Of Michael McDonald
Sent: August 31, 2015 1:53 PM
To: Fowler, Mark
Cc: HYCOM.org Forum
Subject: Re: FW: associating SST and bottom depth

Fowler, Mark

unread,
Sep 2, 2015, 9:57:25 AM9/2/15
to Michael McDonald, HYCOM.org Forum
Or maybe a problem with the target? I get ' The website cannot display the page' if I try the same request interactively.

-----Original Message-----
From: e.m.mc...@gmail.com [mailto:e.m.mc...@gmail.com] On Behalf Of Michael McDonald
Sent: August 31, 2015 1:53 PM
To: Fowler, Mark
Cc: HYCOM.org Forum
Subject: Re: FW: associating SST and bottom depth

Michael McDonald

unread,
Sep 2, 2015, 10:11:58 AM9/2/15
to HYCOM.org Forum, michael....@hycom.org, Mark....@dfo-mpo.gc.ca
Never try and "guess" netcdf subset urls. The URL is displayed at the bottom of the interactive form page - which can be copied and pasted to the command line if you want to download via wget.


e.g.,
Why do you want vertcoord=1, there is not even an option for subsetting this with bathymetry as it is 2d.



This is the NCSS form page,
http://ncss.hycom.org/thredds/ncss/grid/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc/dataset.html



when you select the "bathymetry" variable and format "netcdf4" the following URL should be shown at the bottom of the page - see it?


Does R accept OPENDAP data URLs...? It looks like you are trying to insert a "direct download" URL into an OPeNDAP function call.

Fowler, Mark

unread,
Sep 2, 2015, 12:34:35 PM9/2/15
to Michael McDonald, HYCOM.org Forum

I was using the default ‘netcdf’ format (netcdf3), which is equivalent to the NOAA ERDDAP format I’m familiar with, and the NOAA site is OPENDAP. Switched to netcdf4. In either case I can see the box where I expect the URL to be displayed, but not the URL (window pasted below). Clicking submit will download the extraction with netcdf4, but fail with netcdf as described earlier.

 

Without an URL I went to the subset documentation and tried to figure it out, but neither of the ‘disable’ variables are mentioned there. And I just tried the same thing cabbing from the URL you provided. That gets me closer (the more typical HTTP 500 error below) but still not there. This has to do with the OPENDAP? Although the NOAA targets were OPENDAP, I’m not aware what that has to do with the netcdf functions in R.

 

trying URL 'http://ncss.hycom.org/thredds/ncss/grid/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?var=bathymetry&north=58.5&south=31&west=-57.01047&east=-48.11218&disableLLSubset=on&disableProjSubset=on&horizStride=1&time=2012-09-13T00%3A00%3A00Z&accept=netcdf4'

Error in download.file(link, dest, mode = "wb") :

In addition: Warning message:

In download.file(link, dest, mode = "wb") :

  cannot open: HTTP status was '500 Internal Server Error'

 

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 2, 2015 11:12 AM
To: HYCOM.org Forum

Cc: michael....@hycom.org; Fowler, Mark
Subject: Re: FW: associating SST and bottom depth

 

Never try and "guess" netcdf subset urls. The URL is displayed at the bottom of the interactive form page - which can be copied and pasted to the command line if you want to download via wget.

Fowler, Mark

unread,
Sep 2, 2015, 1:19:12 PM9/2/15
to Michael McDonald, HYCOM.org Forum

I can see the URL by switching from Explorer to Chrome, and in Chrome both netcdf and netcdf4 work interactively. But still at a loss how to pass the URL to R’s download.file function correctly (tried both netcdf and netcdf4).

 

trying URL 'http://ncss.hycom.org/thredds/ncss/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?var=bathymetry&north=58.5&south=31&west=-57.01047&east=-48.11218&disableLLSubset=on&disableProjSubset=on&horizStride=1&time=2012-09-13T00%3A00%3A00Z&accept=netcdf'

Error in download.file(link, dest, mode = "wb") :

In addition: Warning message:

In download.file(link, dest, mode = "wb") :

  cannot open: HTTP status was '500 Internal Server Error'

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 2, 2015 11:12 AM
To: HYCOM.org Forum

Cc: michael....@hycom.org; Fowler, Mark
Subject: Re: FW: associating SST and bottom depth

 

Never try and "guess" netcdf subset urls. The URL is displayed at the bottom of the interactive form page - which can be copied and pasted to the command line if you want to download via wget.

Fowler, Mark

unread,
Sep 2, 2015, 1:41:51 PM9/2/15
to Michael McDonald, HYCOM.org Forum

Maybe nothing to do with the file functions. Tried the actual query interactively using Chrome (screenshot below). But this fails with…

 

last (-1) must be >= first (0)

 

 

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 2, 2015 11:12 AM
To: HYCOM.org Forum

Cc: michael....@hycom.org; Fowler, Mark
Subject: Re: FW: associating SST and bottom depth

 

Never try and "guess" netcdf subset urls. The URL is displayed at the bottom of the interactive form page - which can be copied and pasted to the command line if you want to download via wget.

Fowler, Mark

unread,
Sep 3, 2015, 8:49:21 AM9/3/15
to Michael McDonald, HYCOM.org Forum

And I did figure out HYCOM uses the 0-360 longitude scale. Switching to this results in...

 

HTTP Status 500 - Internal Server Error

Status 500 - Internal Server Error

THREDDS Data Server Version 4.6 -- Documentation

Snapshot from Chrome, showing the URL being submitted:

 

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 2, 2015 11:12 AM
To: HYCOM.org Forum

Cc: michael....@hycom.org; Fowler, Mark
Subject: Re: FW: associating SST and bottom depth

 

Never try and "guess" netcdf subset urls. The URL is displayed at the bottom of the interactive form page - which can be copied and pasted to the command line if you want to download via wget.

Michael McDonald

unread,
Sep 3, 2015, 6:11:26 PM9/3/15
to HYCOM.org Forum, michael....@hycom.org, Mark....@dfo-mpo.gc.ca

Fowler, Mark

unread,
Sep 4, 2015, 7:09:33 AM9/4/15
to Michael McDonald, HYCOM.org Forum

Hi Michael,

 

I was using -180 to 180 until my last attempt yesterday, and noticed that scrolling over the ‘Choose Spatial Subset’ map displayed 0-360 longitude coordinates, so switched. BTW I could not actually choose the window using that map, could not figure out a mouse-click combo that would produce a box.

 

The full download worked fine:

 

trying URL 'http://ncss.hycom.org/thredds/ncss/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?var=bathymetry&accept=netcdf'

Content type 'application/x-netcdf' length unknown

opened URL

downloaded 169.8 Mb

 

But subsetting (now back to -180 to 180) does not seem to work… (also tried this without &horizStride=1)

 

trying URL 'http://ncss.hycom.org/thredds/ncss/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?var=bathymetry&north=58.5&south=31&west=-57.01047&east=-48.11218&horizStride=1&accept=netcdf'

Error in download.file(link, dest, mode = "wb") :

In addition: Warning message:

In download.file(link, dest, mode = "wb") :

  cannot open: HTTP status was '500 Internal Server Error'

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 3, 2015 7:11 PM
To: HYCOM.org Forum
Cc: michael....@hycom.org; Fowler, Mark

Subject: Re: FW: associating SST and bottom depth

 

You are likely having a web browser issue.

Fowler, Mark

unread,
Sep 4, 2015, 8:10:27 AM9/4/15
to Michael McDonald, HYCOM.org Forum

Content type 'application/x-netcdf' length unknown

opened URL

downloaded 7076 bytes

 

But if that means we don’t have bottom depth for the originally requested window I’m out of luck. And it doesn’t make sense. There is no spot on earth we don’t have that information.  Are you able to obtain the depths for:

 

'http://ncss.hycom.org/thredds/ncss/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc?var=bathymetry&north=58.5&south=31&west=-57.01047&east=-48.11218&accept=netcdf'

 

From: Michael McDonald [mailto:michael....@hycom.org]

Sent: September 3, 2015 7:11 PM
To: HYCOM.org Forum
Cc: michael....@hycom.org; Fowler, Mark

Subject: Re: FW: associating SST and bottom depth

 

You are likely having a web browser issue.

Michael McDonald

unread,
Sep 4, 2015, 10:34:38 AM9/4/15
to Fowler, Mark, HYCOM.org Forum
There is an issue with NCSS subsetting this file due to the X/Y to
lat/lon conversion.

I would recommend downloading the file locally and subsetting on your client.

ftp://ftp.hycom.org/datasets/GLBa0.08/expt_91.1/topo/depth_GLBa0.08_09.nc

Fowler, Mark

unread,
Sep 4, 2015, 12:02:02 PM9/4/15
to Michael McDonald, HYCOM.org Forum
Okay. The subsetting works for the temperature at depth data, so I'll just keep one permanent topography file and subset that in parallel with the temperatures.

Thank you for the help, I could not have figured it out on my own.

-----Original Message-----
From: e.m.mc...@gmail.com [mailto:e.m.mc...@gmail.com] On Behalf Of Michael McDonald
Sent: September 4, 2015 11:34 AM
To: Fowler, Mark
Cc: HYCOM.org Forum
Subject: Re: FW: associating SST and bottom depth

Reply all
Reply to author
Forward
0 new messages