RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Web Content Display

Web Content Display (CoastWatch User Forum)

The NOAA CoastWatch Forum provides information in several categories. The information is relevant to CoastWatch products and software (https://coastwatch.noaa.gov) and vary in technical details. For questions and issues not addressed here, start a thread in the appropriate category or contact the CoastWatch HelpDesk (coastwatch.info@noaa.gov). Please note our Privacy Policy (https://www.noaa.gov/privacy.html) and that this is a public forum. All information is voluntarily and anonymous submissions are accepted pending review prior to posting.

Forums

RSS (Opens New Window)

Brittany Huntington, modified 4 Years ago.

Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 6 Join Date: 6/12/17 Recent Posts

Is my understanding correct: If I want to use a cumulative mean dataset (i.e. Sea Surface Temperature, Coral Reef Watch, CoralTemp - Cumulative Mean 2009-2018) a single value mean SST value is produced for each 5km grid cell for a given time range that I specify between 2009 and 2018?

Melanie Abecassis, modified 4 Years ago.

RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 65 Join Date: 6/12/17 Recent Posts

Hey Britt, no. I've already computed the average over that whole time period for another project. So you can't specify a data range with this dataset. You will get one value per pixel which is the average value from 2009-2018.
If you need a different time period, you can either, compute the average in R using the monthly data for example, or if it's a cumulative mean over a period that's gonna be of interest to a few people, let me know what the time range is and I can work on it tomorrow. If you guys are all using the same period of reference for a big project or several projects, it's worth it to have it for the global data and on ERDDAP.

Brittany Huntington, modified 4 Years ago.

RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 6 Join Date: 6/12/17 Recent Posts

That was my initial guess...one cumulative mean value from 2009-2018. Which is jolly good. I would use that.

But riddle me this...I was stumped as to why the following error code popped up when I dropped the time window specification from the rxtractagon code. And digging into the metadata, it appears time is indeed part of the dataset---albeit one value.

Start	Stride	Stop	Size	Spacing
time (UTC)				1	(just one value

sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord)
[1] "Ranges not given for all of the dataset dimensions"
[1] "Coordinates given: "
[1] "longitude" "latitude"
[1] "Dataset Coordinates: "
[1] "time" "latitude" "longitude"
Error: Execution halted

Clearly I will be booking some office hours tmrw ;)

I'll see if entering those start and stop times helps me move along.

Fun fun!

On Wed, Oct 28, 2020 at 2:32 PM Melanie Abecassis <VLab.Notifications@noaa.gov> wrote:

Hey Britt, no. I've already computed the average over that whole time period for another project. So you can't specify a data range with this dataset. You will get one value per pixel which is the average value from 2009-2018.
If you need a different time period, you can either, compute the average in R using the monthly data for example, or if it's a cumulative mean over a period that's gonna be of interest to a few people, let me know what the time range is and I can work on it tomorrow. If you guys are all using the same period of reference for a big project or several projects, it's worth it to have it for the global data and on ERDDAP.

--
Melanie Abecassis CoastWatch Virtual Lab Forum https://vlab.noaa.gov/web/coastwatch/coastwatch-knowledge-base/forum/-/message_boards/view_message/12350014 VLab.Notifications@noaa.gov

Brittany Huntington, PhD
JIMAR Ecosystem Sciences Division Manager
NOAA/NMFS/PIFSC/ESD/Brittany Huntington
1845 Wasp Blvd., Building 176
Honolulu HI 96818
Ph: (808) 725-5438
http://www.soest.hawaii.edu/jimar

Melanie Abecassis, modified 4 Years ago.

RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 65 Join Date: 6/12/17 Recent Posts

Huh. Interesting!
try this: tcoord=c('2009-01-01T12:00:00Z','2009-01-01T12:00:00Z')
sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord,tcoord=tcoord)

Brittany Huntington, modified 4 Years ago.

RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 6 Join Date: 6/12/17 Recent Posts

Ok...tried that. Threw me a new error.

> sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord, tcoord=tcoord)
[1] "dimension coordinate out of bounds"
[1] "dimension name: time"
[1] "given coordinate bounds 2009-01-01 12:00:00 2009-01-01 12:00:00"
[1] "ERDDAP datasets bounds 1985-01-01 12:00:00 1985-01-01 12:00:00"
Error in checkBounds(dataCoordList, dimargs, cross_dateline_180) :
Coordinates out of dataset bounds - see messages above

So I just used those time bounds. Viola! Data extracted.

***but those dates are bogus yes....this is actually 2009-2018 cumulative mean?

On we go....

On Wed, Oct 28, 2020 at 2:50 PM Melanie Abecassis <VLab.Notifications@noaa.gov> wrote:

Huh. Interesting!
try this: tcoord=c('2009-01-01T12:00:00Z','2009-01-01T12:00:00Z')
sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord,tcoord=tcoord)

--
Melanie Abecassis CoastWatch Virtual Lab Forum https://vlab.noaa.gov/web/coastwatch/coastwatch-knowledge-base/forum/-/message_boards/view_message/12350087 VLab.Notifications@noaa.gov

Melanie Abecassis, modified 4 Years ago.

RE: Quick Q on Cumulative mean datasets --- need to set a data range?

Youngling Posts: 65 Join Date: 6/12/17 Recent Posts

Hum.... yeah it should be, but I'll double check tomorrow. It is weird.... stay tuned.

On Wed, Oct 28, 2020 at 9:01 PM Brittany Huntington <VLab.Notifications@noaa.gov> wrote:

Ok...tried that. Threw me a new error.

> sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord, tcoord=tcoord)
[1] "dimension coordinate out of bounds"
[1] "dimension name: time"
[1] "given coordinate bounds 2009-01-01 12:00:00 2009-01-01 12:00:00"
[1] "ERDDAP datasets bounds 1985-01-01 12:00:00 1985-01-01 12:00:00"
Error in checkBounds(dataCoordList, dimargs, cross_dateline_180) :
Coordinates out of dataset bounds - see messages above

So I just used those time bounds. Viola! Data extracted.

***but those dates are bogus yes....this is actually 2009-2018 cumulative mean?

On we go....

On Wed, Oct 28, 2020 at 2:50 PM Melanie Abecassis <VLab.Notifications@noaa.gov> wrote:
Huh. Interesting!
try this: tcoord=c('2009-01-01T12:00:00Z','2009-01-01T12:00:00Z')
sst <- rxtractogon (dataInfo, parameter=parameter, xcoord=xcoord, ycoord=ycoord,tcoord=tcoord)

--
Melanie Abecassis CoastWatch Virtual Lab Forum https://vlab.noaa.gov/web/coastwatch/coastwatch-knowledge-base/forum/-/message_boards/view_message/12350087 VLab.Notifications@noaa.gov

--
Brittany Huntington, PhD
JIMAR Ecosystem Sciences Division Manager
NOAA/NMFS/PIFSC/ESD/Brittany Huntington
1845 Wasp Blvd., Building 176
Honolulu HI 96818
Ph: (808) 725-5438
http://www.soest.hawaii.edu/jimar

--
Brittany Huntington CoastWatch Virtual Lab Forum http://vlab.noaa.gov/web/coastwatch/coastwatch-knowledge-base/forum/-/message_boards/view_message/12350107 VLab.Notifications@noaa.gov

________________________________________________________
Melanie Abecassis
OceanWatch Operations Manager

https://oceanwatch.pifsc.noaa.gov/

NOAA CoastWatch/OceanWatch/PolarWatch

NOAA/NESDIS/STAR
Satellite Oceanography and Climatology Division
NOAA Center for Weather and Climate Prediction (NCWCP) 
5830 University Research Court
College Park, MD 20740, USA
Earth System Science Interdisciplinary Center (ESSIC)/
Cooperative Institute for Satellite Earth System Studies (CISESS)
University of Maryland
__________________________________________________________