## Where communities thrive

• Join over 1.5M+ people
• Join over 100K+ communities
• Free without limits
##### Activity
• 01:28
alex-s-gardner commented #68
• 01:28
alex-s-gardner closed #68
• 01:07

mliukis on new_granule_format

Update autoRIFT deployment to p… (compare)

• Nov 28 22:48
jhkennedy review_requested #74
• Nov 28 22:48
jhkennedy review_requested #74
• Nov 28 22:48
jhkennedy opened #74
• Nov 28 22:45
jhkennedy closed #63
• Nov 28 22:45
jhkennedy commented #63
• Nov 28 17:58
jhkennedy edited #15
• Nov 22 01:23
mliukis edited #15
• Nov 22 00:20
mliukis edited #15
• Nov 22 00:20
mliukis edited #15
• Nov 22 00:19
mliukis edited #15
• Nov 22 00:18
mliukis edited #15
• Nov 22 00:18
mliukis edited #15
• Nov 22 00:17
mliukis assigned #15
• Nov 22 00:17
mliukis assigned #15
• Nov 22 00:17
mliukis assigned #15
• Nov 22 00:17
mliukis opened #15
• Nov 16 23:56

mliukis on new_granule_format

Use fixed size arrays for strin… (compare)

Alex Gardner
@alex-s-gardner
Fantastic... now we have a place for the its_live and autoRIFT community to interact with the team
tedscambos
@tedscambos
yay.
Alex Gardner
@alex-s-gardner
ITS_LIVE grounded ice sheet elevation change from 1985-2020 is now available on the ITS_LIVE website: https://its-live.jpl.nasa.gov
nochwat
@nochwat
Hello community. I am a PhD student with Ted Scambos and Mike Willis. I am on a wild goose chase to derive glacier velocities from RADARSAT-1 data from ~2000-2008. I have been using PyCorr with ASTER images, but I understand that to process SAR data you must pre-pocess the data to apply terrain/geolocation corrections. I have been in touch with Mark Fahnestock, my supervisors, and ASF about how to do this, I have also reached out to Rick Forster. To no avail... Can anyone help give me guidance on the pre-processing that needs to occur? Thanks! - Naomi
Joseph H Kennedy
@jhkennedy
Hi @nochwat I'm asking around ASF if anyone has any current RTC/TC recipies for RADARSAT-1. I believe you can do it wither either SNAP or GAMMA (GAMMA typically gives a better product, but is ). What areas are you looking at in particular? Alaska?
12 replies
wbarnhart-usgs
@wbarnhart-usgs
I’m exploring using autoRIFT for earthquake analysis. Could you clarify what the output units are for the geocoded displacement maps? The Github page mentions velocity, but I wanted to check if it was actually velocity or displacement. If velocity, how is the time range determined?
Denis Felikson
@dfelikson
Hi ITS_LIVE community! I'm running notebooks/ITS_LIVE_ipyLeaflet_global_example.ipynb from https://github.com/nasa-jpl/itslive.git and getting an xarray error: TypeError: open_dataset() got an unexpected keyword argument 'storage_options'. It looks like Mark Fahnestock is the author of the notebook and I can contact Mark directly but wondering if anyone else has seen this error.
26 replies
Alex Gardner
@alex-s-gardner
Check out the new notebook written my Mark and Luise that makes accessing the ITS_LIVE velocity data incredibly simple.
7 replies
Alex Gardner
@alex-s-gardner
Link to the new notebook: https://github.com/nasa-jpl/its_live
We've also posted the link to the the ITS_LIVE overview talk given at this year's AGU in New Orleans: https://www.youtube.com/watch?v=iNmL0DJSmXQ
Nick Holschuh
@nholschuh

Hello! I have a quick question and for some reason I can't seem to figure it out...

I can see how to download (both through the NSIDC's website and programmatically) the annual Greenland velocity mosaics and the scene pair velocities, but I'm struggling to find the monthly ice sheet mosaics. Is there an obvious place that I am missing (or piece of code) that we can use to download them?

2 replies
Alex Gardner
@alex-s-gardner

For Julia nerds and the Julia-curious, ITS_LIVE has a new Julia package for exploring the data:

You can plot data for your glacier with 4 lines of code:

julia> Using Pkg; Pkg.add(“ItsLive”)
julia> lat = [59.925]; lon = [-140.620]
julia> C = ItsLive.getvar(lat, lon, ["mid_date","date_dt","v","satellite_img1"], ItsLive.catalog())
julia> p = ItsLive.plotbysensor(C,"v")

S. B.
@SCB12_gitlab

Hi everyone - I hope I'm in the right place to ask questions about the ITS_LIVE Antarctic monthly elevation data! I'd really appreciate some help with these questions.

1) Could you possibly state the difference between the downloaded files (a) 'NETCDF: "ANT_G1920V01_GroundedIceHeight":h' and (b) 'NETCDF: "ANT_G1920V01_GroundedIceHeight":dh'? And why does (a) appear to show significantly less spatial coverage than (b)?

2) What does each 'Band' correspond to when I click identify results in QGIS? Does it correspond to data from one individual month - therefore starting from January 1985? (Please see in the above photos to the right hand side of the screen to see what I'm referring to)

3) Is there a recommended way in which to extract the data on QGIS? For instance, for extracting the data at points every 200m down a plotted line, it gives this (c) when extracting values from (b). I'm using 'Points along geometry' in the QGIS toolbox, and it appears to only extract 'angle' values as shown below. I used this toolbox for extracting velocities from the ITS_LIVE velocity dataset, and it worked - is there another recommended tool to use?

Thank you!

3 replies
Johan Nilsson
@nilssonj
Hi S.B! Hopefully, I can help you with some of the questions. Starting with (1a): The "h" variable is the surface elevation of the land ice portion of the ice sheet from the COP DEM, and is used to provide a reference for the "dh" variable. The "dh" variable is the time variable elevation change of the ice sheet. The difference in spatial coverage depended on what mission is included in the solution at any given time step in the data set. From 1985-1990 only the Geosat mission is available and only covers up to 72 degrees in latitude. The other included mission has better spatial coverage to the south pole and have been interpolated to fill those areas. (2) I don't use QGIS that much but the bands should the the monthly time steps starting from Jan 1985. (3) Unfortunately, I can't help you with question number (3) as I don't work in QGIS. But maybe some other person in the team can help you with that question? I hope my answer helped you a little bit at least.
S. B.
@SCB12_gitlab

@nilssonj @alex-s-gardner @tlohde --> Thanks for your help so far, I've definitely made progress! I hope its ok to ask a few more questions?

1. @nilssonj I understand the difference now between the two variables, and have read the 2021 paper on this elevation dataset a few more times - but I'm curious as to why the "dh" file only shows the spatial coverage of the Geosat (a), and not other missions on QGIS (looking at figure 1 from the paper)? It doesn't stop me extracting data from blank areas as shown in (b) however.
1. @tlohde thank you, that's certainly greatly improved my methodology for the extracting data from the ITS_LIVE velocity dataset (I have tried and tested and it works!)

2. @alex-s-gardner that's perfect, I've downloaded, tried and tested the tool, and it works! But could you please clarify something for me?

(b) https://postimg.cc/v1JPFS0L : data taken from a pixel at the margin of Pine Island Glacier from the "dh" file, with the "h" file overlaid.

With this tool, it gives me the results in (b) which I can extract into excel for plotting. But if you'd take a look at the graph towards the bottom of the screen, what does the y-axis refer to? For instance, in 2004 it is ~25.0m above what variable? Or in 2020 it is -13.2m below what variable?

Many thanks!

(Also, apologies as I would have sent this message in the above reply thread - but there appeared no easy way to format my reply as compared to when I post a new thread)

2 replies
Qi Liang
@magicacer
Hi ITS_LIVE community. I have a small question about the annual Antarctic velocity mosaic product. I found that the velocity at West Ice Shelf from this product is unusual, especially during 2013-2016. There are distinct patterns on the mosaic. I was wondering if anyone noticed that?
By the way, is there any news about the Antarctic velocity mosaic for 2019, 2020 and 2021?
Many thanks.
4 replies
tlohde
@tlohde

Hello ITSLIVE team. A question about uncertainty.

#### Objective

Create seasonal (or monthly, or annual) velocity profiles along a centreline.

#### Outline process, and work done so far

• Draw centreline, and add vertices every ~500 m

#### Step 1: averaging spatially - for a single time slice

• To minimise the effects of point sampling, and to help generate a smooth profile. I am averaging (mean) both the $v_x$ and $v_y$ components within a 3x3 window centered on each centreline vertex, and calculating the resultant velocity from these.
• The uncertainty/confidence in $\bar{v_x}$ and $\bar{v_y}$ can be described by their respective window standard deviations/standard errors.
• this is a spatially variable measure of uncertainty that is not equal to the the supplied vx/y_error. Moreover, it does not take the error associated with that particular time slice.
• create new xarray.dataset with dimensions: mid_date, and distance (along centreline)
• variables: $\bar{v_x}, \bar{v_y}$, std($v_x$). std($v_y$)
• and the supplied vx_error, vy_error, date_dt

#### Step 2: averaging temporally

• filter on date_dt (< 60 days for seasonal, for example)
• Calculate error weighted averages
• the supplied vx_error and vy_error terms are functions of time, and do not vary spatially.
• this does not capture the variability within each 3x3 window.

#### Problem

I have two measures of uncertainty one, measuring the uncertainty across space, the other time.

#### Question:

How best to combine these? Do they to be combined?

Happy to provide more detail if required. And I hope the above is intelligible. Thank you, ed

4 replies
Wang Yuzhe
@WangYuzhe

Hi ITS_LIVE team,
Thank you very much for publish the great ITS_LIVE glacier velocities product.
I'm Yuzhe Wang from Shandong Normal University, China. We're using the surface velocities at East Rongbuk Glacier (28.05N, 86.93E) in Everest to compute the basal firction parameters inverted by an ice flow model.

Although the 240m HMA product provided the annual velocities from 1985, it is still too coarse for individual glacier modelling. We tried to use the 120m time-averaged product which is relatively appropriate for our study. However, we have no idea about the time information.

Could you please, by any chance, help me confirm which period do you choose to compute the averaged velocities on East Ronbuk Glacier?

Thank you very much.

Best wishes,

Yuzhe

2 replies
Alex Gardner
@alex-s-gardner

In case anyone missed the announcement of V2 data:

The NASA ITS_LIVE (https://its-live.jpl.nasa.gov) team has analyzed over a petabyte of radar and optical satellite data to find slight distortions between image-pairs caused by surface motion, creating a detailed record of flow for every glacier on Earth. The data is easily accessed through an interactive widget that generates publishable quality figures within seconds:

Take a minute to discover a new glacier surge, watch glaciers slow in winter across the planet, or tour the Antarctic "doomsday" glaciers of Pine Island and Thwaites.

## Web App

Video tutorial

smcgeeak
@smcgeeak

I just learned of this great resource and am getting familiar with it, but I have a few questions:

1) I assume that the annual velocities take into account leap years, correct? For example, the annual velocity for non-leap years would be for a period of 365 days, while the annual velocity for leap years would be for a period of 366 days. Is this correct?

2) I assume the Its_Live online map and downloaded data are referenced to the WGS84 datum. Is that correct?

3) When I enter a lat/long coordinate into the Global Glacier Velocity Point Data Access map, I enter the coordinate as decimal degrees to 7 decimal places. It then tells me that I have entered an invalid value and that the valid value should be to only 1 decimal place. But then when I export and download the data for the point that I entered (to 7 decimal places), the file name of the downloaded data shows the lat/long to 4 decimal places. It appears that the Its_Live map app is rounding off my 7-decimal coordinate values to 4 decimal places and then returning the data for the 120-meter grid cell in which the 4-decimal place coordinate is located. Is this correct?

Joshua Kelly
@jkelly17_gitlab
Hey there. I am trying to download Image pair velocity data but it seems the AWS server for the Version 1 data is down? I have no trouble downloading Version 2 data. Here's an example granule URL that shows the error upon opening: https://its-live-data.s3.amazonaws.com/velocity_image_pair/landsat/v00.0/32608/LE07_L1TP_058019_20160707_20160901_01_T1_X_LC08_L1TP_058019_20160629_20170221_01_T1_G0240V01_P058.nc
5 replies
Is there any other way to access these data? Thanks!
James
Hello, I am a student researching glacier velocities over time. I am looking to use the IT_LIVE notebook provided in the GitHub to plot and collect some data but I am running into a KeyError for data_epsg, which I have entered below. Does anyone have a workaround for this? Thanks!
File ~/notebooks/datacube_tools.py:83, in DATACUBETOOLS.find_datacube_catalog_entry_for_point(self, point_xy, point_epsg_str)
79         break
81 if cubefeature:
82     # find point x and y in cube native epsg if not already in that projection
---> 83     if point_epsg_str == cubefeature["properties"]["data_epsg"].split(":")[-1]:
84         point_cubexy = point_xy
85     else:

KeyError: 'data_epsg'
smcgeeak
@smcgeeak
Here's an idea that I hope can be implemented for CSV export files: please include a field/column that identifies which satellite a data point is for. Currently, the CSV export file contains only two columns: date, and v. From this, it's impossible to determine if the data point is from Landsat 8, Sentinel 2, etc. I believe there are many use cases where the user needs to pull out the data for a particular satellite. Including a third column (satellite) in the CSV export file would allow for this.
7 replies
smcgeeak
@smcgeeak
Is the Its-Live widget at https://itslive-dashboard.labs.nsidc.org/ undergoing updates/maintenance? I've been trying to access it for the past week and it often fails to load, returning a bad_gateway error. Other times it will load but will not let me enter a point on the map either by clicking on the map or by entering a lat/long coordinate. I have tried three different browsers (Firefox, Edge, and Chrome), clearing the browser cache, and rebooting the computer but nothing has helped. It is currently not usable for me. I wonder if it's just me, or is it not working for everyone.
Oliver King
@oliver01king

Hi there! I am an undergrad student at Lancaster University currently working on my dissertation. Part of this involves looking to analyse velocity data for Upernavik Isstrom (NW Greenland), which ITS_LIVE seems perfect for. However, i'm having some issues working with the data, specifically getting velocity rasters loaded into ArcGIS (3.0).

I have tried creating velocity rasters using both the 'Make NetCDF Raster Layer' and 'Make Multidimensional Raster Layer' tools, but both to no avail. The former tool does create a layer but doesn't display anything, regardless of symbology choices. The latter tool does display a layer, but with very coarse detail - only velocities of 240, 480, 960, and 1920 m/y are shown. No attempt to modify symbology changes this. See example:

Oddly, QGIS displays the layers as expected. See example:

I was wondering if anyone has any experience successfully loading velocity rasters into ArcGIS, or could shed any light on this situation?

Thank you!

4 replies
smcgeeak
@smcgeeak
As I work more with the Its_Live CSV export files, and to expand on my previous comment, it's become apparent to me that all data fields should be added to the CSV export file. For example, these fields should be included in the export: date, v, v_error, vx, vy, location, satellite. To illustrate why this is needed, I'm currently evaluating the Its_Live data against differential GPS data. When I download a CSV file for a point location, all I get is the date and v fields. In order to get the v_error, vx, and vy fields, I have to download three additional CSV files; one each for v_error, vx, and vy, and then I have to manually combine 4 individual CSV files into one in order to have all the fields available for analysis. But then I still don't have all the information I need, as I also need the satellite field. Currently, the only way to get that data from Its_Live is to go into the map all, enter a point location, select to plot by satellite, display the plot, and then visually compare the plot against the CSV file to be able to figure out which satellite corresponds to each date in the CSV file. This is extremely tedious and time consuming. It would make things 1000% easier, quicker, and less error-prone if all data fields could be included in the CSV export files.
4 replies
Laurane Charrier
@LauraneC_gitlab
Hi ITS_LIVE community.
I am looking at scene pair velocities computed with Sentinel-1 images.
I was wondering if some information about the acquisition of these images is available somewhere (ascending or descending orbit, orbit number, incidence angle, conversion matrix) ?
In the preprint of "Processing methodology for the ITS_LIVE Sentinel-1 ice velocity product", it is written that two outputs should be included:
M11 [int16]: velocity conversion matrix element (1st row, 1st column)
M12 [int16]: velocity conversion matrix element (1st row, 2nd column)
However, I did not find these two outputs in the datacube v02 covering my study area... Do you know where I can find this information ? or is this information not yet available ?
Thanks !
7 replies
kiki
@kwallick
Hello! I think I have a simple question for the its_live widget. The ESRI basemap, what year is it valid for? I haven't been able to figure this out from the documentation provided.
3 replies
Whyjay Zheng
@whyjz
Hi all! A small question: For the scene-pair velocity files in ITS_LIVE v2, there are no vx_err and vy_err attributes but v_error instead. What exactly does v_error represent? Standard deviation of a speed measurement or uncertainty of that under a certain confidence level?
Thanks a lot!
4 replies
Oliver King
@oliver01king
Hello! I was just wondering if there is an updated release schedule for the 2019, 2020, and 2021 "Yearly and Composite Glacier Velocities" available at https://nsidc.org/apps/itslive/ (specifically for Greenland)? The webpage has stated "Composites for 2019, 2020 and 2021 will be available in the next weeks" for quite some time now. I would love to include these in my dissertation, but the deadline is fast approaching and so I was just wondering when these might be released? Thank you! :)
5 replies
Will Kochtitzky
@willkochtitzky
I am interested in trying to extract the velocities for lots of glaciers in a loop at specific points. I know these points as they are along flux gates that I have already drawn.
I found your Jupyter Notebook here: https://github.com/nasa-jpl/its_live
I could get those working locally so that I can click on a point and then export the csv file with all those data. But I want to loop through points that I already have defined and get back the data frame that you export to the csv file. I am not immediately seeing how to do this in altering the code above. Does anyone have tips on how to do this?
3 replies
Oliver King
@oliver01king
Hello all! Just thought I would just share one of my favourite ITS_LIVE figures for those interested. Its showing velocity at points along the centerline of Upernavik Isstrøm glacier 2 :)
VDC
@vdevauxchupin

Hello,

I'd like to share some code I wrote for downloading datacubes on custom date/spatial extent (for both yearly and sub-yearly datacubes), tracking specific pixel displacements, and calculating surface strain rates.
Most of it is a GUI, you should be able to run the notebooks without having to dive in the code or understand all of it.

I hope it is useful to some people, let me know if I can help in any way with it ! Just follow the instructions and it should work :)
Also I hope I'm not trespassing any rule by sharing homemade code here. If I am, please let me know and I will delete my message/link. Thank you !