Support the Arctic Sea Ice Forum and Blog

Author Topic: Buoy data 23melt  (Read 2703 times)

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Buoy data 23melt
« on: March 12, 2023, 10:00:59 PM »
2 co-located groups of buoys still have active profilers and temp strips:

itp136, top5 and simb3-565600  Beaufort

itp137, top6 and simb3-566570  Beaufort

4 new simb3's were also recently deployed in the Beaufort on mar5:
4560560, 4563570, 5651330 and 5658110.
 

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #1 on: March 12, 2023, 10:07:37 PM »
Looking at itp137 we can see a warm area at 50m depth from feb13 to mar12. Interesting to see where that is.
Animation is from jan1-mar12 and shows it near the Mclure Strait.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #2 on: March 14, 2023, 03:38:21 PM »
update on simb3-560560

Quote
Deployment:
Ice Thickness   142 cm
Snow Thickness 8 cm

1. location
2. Buoy temperature difference over 2cm, purple line is possible freeboard
3. Buoy temperature.
Ice in the borehole is still cooling down since deployment

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #3 on: March 20, 2023, 09:48:00 PM »
update on simb3 651330
Quote
Deployment:
Ice Thickness      107 cm
Snow Thickness   10 cm

1. location
2. Buoy temperature. Air temp overlaid as dashed black line linked to right hand scale.
3. Buoy temperature difference over 2cm.

Today's thickening was only 1cm instead of the usual 2cm, perhaps due to the warmer air temps.
« Last Edit: March 20, 2023, 10:00:24 PM by uniquorn »

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #4 on: March 21, 2023, 08:26:02 PM »
6 simb3's in the Beaufort are a good opportunity to test the proxy insolation idea using top of buoy temp minus air temp. Sunshine should heat the top of the buoy so here we use an average of the top 10 sensors. The buoys are ordered in the chart by latitude, locations are from earlier today and were used for the worldview images to show a rough representation of cloud cover over each buoy.

Quote
-150.178192,72.775776  651330
-148.171312,73.94332   560560
-147.131792,74.864168  563570
-146.204464,75.926208  658110
-132.814392,78.48168   565600
-130.014392,78.935504  566570

https://go.nasa.gov/3FFLVmb

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #5 on: March 22, 2023, 04:55:02 PM »
Thickness of the 6 SIMB3's since mar5. 563570 trying to catch up from 0.78m

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #6 on: March 23, 2023, 03:42:09 PM »
Looking at the thinnest floe, 563570 we can see the effect of the recent warmer temperatures.

1. Full range temps from -35C to -1.5C (coldest air temp to roughly ocean temp)
2. Close up on ice bottom using -7C to -1.5C shows a slight warming at ice bottom today.
3. 2cm temp difference shows less cooling from above.
Interesting to see when the next thickening will be.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #7 on: March 23, 2023, 03:46:24 PM »
Note that nullschool gfs is modelling almost 10C colder than the buoy temp in that location.

edit: correction UTC temp was -20.5C so a bit nearer at -8.75C colder.
« Last Edit: March 24, 2023, 02:07:56 PM by uniquorn »

HapHazard

  • Grease ice
  • Posts: 802
  • Chillin' on Cold Mountain.
    • View Profile
  • Liked: 278
  • Likes Given: 5200
Re: Buoy data 23melt
« Reply #8 on: March 23, 2023, 08:28:13 PM »
Note that nullschool gfs is modelling almost 10C colder than the buoy temp in that location.

Yet another reason why buoy data is so important, and why we need so, so many more "in situ" type data collection. Models aren't always that accurate.

Thanks for all your hard work in this area, uni.
If I call you out but go no further, the reason is Brandolini's law.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #9 on: March 24, 2023, 10:33:45 AM »
Thank you. Nullshool caught up a bit today, modelling only 1.4C colder.

563570 not sure about thickening at -13.8C.

edit: time wasn't UTC so GFS was quite close at -14.1
« Last Edit: March 24, 2023, 12:41:40 PM by uniquorn »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Buoy data 23melt
« Reply #10 on: March 24, 2023, 01:51:26 PM »
Quote
we need so so many more "in situ" type data collection. Models aren't accurate. Nullshool caught up a bit today; GFS was quite close
It sounds like actual buoy data is not currently being used in forecasts -- ok it's tricky to use since they move around and have quiet failures. But what the heck are they using (if anything)? Bering Strait land stations are not even in the Arctic Circle at 65° 40' N. Is Russia still coordinating?

The Arctic Ocean is ~5% of the planet with no other instrumentation -- about the size of Africa. Buoys seem more a source of embarrassment more than anchoring of predictions. Modelers can get away with saying anything eg DMI 80ºN, who's to know?

It matters because melting and freezing of ice is quite sensitive to actual temperatures but rather unresponsive to modeled temperatures.

AO weather prediction kinda reminds me of ENSO. For literally decades, we were told El Nino tele-predicts rainfall in the western US. But after yet another three months of atmospheric rivers in yet another La Nina, we are told to memory-hole those papers, El Nino never accounted for more than a meagre portion of rainfall variation to begin with.

If peer review failed to catch thousands of cases of previous error (cf. Beaufort "Gyre"), is it doing better today?

https://www.cnn.com/2023/03/21/us/california-weather-atmospheric-river-drought-climate/index.html
« Last Edit: March 24, 2023, 02:00:54 PM by A-Team »

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #11 on: March 24, 2023, 02:00:26 PM »
It's possible that these new buoys have only just been registered with meteo. I'll do another comparison next week.

Here are the 6 SIMB3 air temps. Will take a look at 651330 next

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #12 on: March 24, 2023, 05:49:18 PM »
651330 update. Near surface ice and snow have warmed to current air temp of -9.5C. GFS on nullschool reporting -15C

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #13 on: April 06, 2023, 11:47:42 PM »
SIMB3 buoys update.
1. High contrast terra modis of today's buoy locations on yesterday's worldview. The ringed buoy hasn't reported since mar29.  https://go.nasa.gov/3mcvUgS
2. air temps
3. ice thickness
4. difference between top of buoy temp and air temp.
« Last Edit: April 06, 2023, 11:53:45 PM by uniquorn »

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #14 on: April 07, 2023, 12:02:08 AM »
651330 temp contours, air temp and dtc10 temp linked to right hand scale.

nullschool air temp = -21.2C , buoy=-22.56C (at roughly 1m)

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #15 on: April 07, 2023, 03:19:47 PM »
float6903584 in the Barents looking well mixed down to 200m.
« Last Edit: April 07, 2023, 09:22:14 PM by uniquorn »

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Buoy data 23melt
« Reply #16 on: April 07, 2023, 05:51:24 PM »
Quote
the Barents looking well mixed down to 200m
-- Overall depth something like 300m and well-mixed all the way to the bottom?

-- are these literal or potential (ie adjusted) temperatures, salinity, pressures?

-- do journal graphics still show well-mixed Barents (Atlantification) stopping ~50 km short of the FJL islands and is only slowly creeping north?

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #17 on: April 07, 2023, 08:44:57 PM »
Quote
-- Overall depth something like 300m and well-mixed all the way to the bottom?

Some of the depths shown are less than 200m. I only look at <200m because it is of most interest to effects at surface and I prefer to see some detail in the water column.

Quote
-- are these literal or potential (ie adjusted) temperatures, salinity, pressures?

Charts posted above look at in situ temperature and salinity (I hope)
https://www.rdocumentation.org/packages/oce/versions/1.7-10/topics/plotProfile

Quote

    Item(s) plotted on the x axis, either a vector of length equal to that of pressure in the data slot, or a text code from the list below.

        "salinity" Profile of salinity.

        "temperature" Profile of in-situ temperature.

        "density" Profile of density
relevant code:
Quote
  ctd <- as.ctd(unlist(prof1[["salinity"]]),
                unlist(prof1[["temperature"]]),
                unlist(prof1[["pressure"]]))

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #18 on: April 07, 2023, 09:20:07 PM »
Quote
Charts posted above look at in situ temperature and salinity (I hope)
or not

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #19 on: April 08, 2023, 09:25:13 PM »
Was a bit confused by those charts but narrowing down the x axes and removing density makes them clearer. 6903564 surfaced again yesterday with cycle 409.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #20 on: April 10, 2023, 05:02:49 PM »
Another look at the Barents since march this time adding data from 6903588 which seems to be drifting at roughly 200m depth against the established common currents towards the gap west of FJL. See if that continues.

6903564 and 6903588 drift tracks.
3D salinity since march
3D temps
6903588 profiles.



uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #21 on: April 14, 2023, 02:18:41 PM »
SIMB3 buoys update.
1. air temps
2. ice thickness
3. difference between top of buoy temp and air temp (proxy insolation)

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #22 on: April 18, 2023, 03:46:15 PM »
A big change in the profiles as float 6903564 drifts into the Barents MIZ south of FJL.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #23 on: April 18, 2023, 10:48:44 PM »
mercator modelling a 'warm' plume pushing north at 80m depth
https://data.marine.copernicus.eu/-/f2e1mdjbjm

and some upanddownwelling at float 6903564 location
https://data.marine.copernicus.eu/-/8b7sm6r6dg

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Buoy data 23melt
« Reply #24 on: April 19, 2023, 12:16:44 AM »
Quote
mercator modeling a 'warm' plume pushing north at 80m depth
Upper corner says Mercator Pro. Is this free/different from regular Mercator?

Does Mercator subsume data from all buoys, drifters and moorings currently in the area or can those be used to check its modeling?

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #25 on: April 19, 2023, 10:30:16 AM »
I think you get 'pro' status when logging into cmems. I haven't paid anything.

The data shown above is from the Mercator dataset  GLOBAL_ANALYSISFORECAST_PHY_001_024
Quote
The Operational Mercator global ocean analysis and forecast system at 1/12 degree is providing 10 days of 3D global ocean forecasts updated daily. The time series is aggregated in time in order to reach a two full year’s time series sliding window.

This product includes daily and monthly mean files of temperature, salinity, currents, sea level, mixed layer depth and ice parameters from the top to the bottom over the global ocean. It also includes hourly mean surface fields for sea level height, temperature and currents. The global ocean output files are displayed with a 1/12 degree horizontal resolution with regular longitude/latitude equirectangular projection.

50 vertical levels are ranging from 0 to 5500 meters.

This product also delivers a special dataset for surface current which also includes wave and tidal drift called SMOC (Surface merged Ocean Current).

DOI (product):
https://doi.org/10.48670/moi-00016

There are many others to choose from. The advantages of using cmems for Mercator are the 50 vertical levels(there appears to be interpolation in between) selectable min/max values, far better resolution right down to pixellation, ease of overlaying different products and save animations as gif or mp4 with short link to settings.
Maybe it will 'catch on' one day.

On the few occasions I've checked buoy data against Mercator they agree well so I think they are integrated but how can I find out?

From https://hpc.niasra.uow.edu.au/ckan/dataset/global-analysis-forecast-phy-001-024/resource/2d66a089-fe71-47ea-8245-6e1f1d469f59

Deep Integro-Difference Equation Models for Spatio-Temporal Forecasting
Andrew Zammit-Mangiona, Christopher K. Wikleb
https://arxiv.org/pdf/1910.13524.pdf

Quote
Abstract
Integro-difference equation (IDE) models describe the conditional dependence between the spa-
tial process at a future time point and the process at the present time point through an integral
operator. Nonlinearity or temporal dependence in the dynamics is often captured by allowing
the operator parameters to vary temporally, or by re-fitting a model with a temporally-invariant
linear operator in a sliding window. Both procedures tend to be excellent for prediction pur-
poses over small time horizons, but are generally time-consuming and, crucially, do not provide
a global prior model for the temporally-varying dynamics that is realistic. Here, we tackle these
two issues by using a deep convolution neural network (CNN) in a hierarchical statistical IDE
framework, where the CNN is designed to extract process dynamics from the process’ most re-
cent behaviour. Once the CNN is fitted, probabilistic forecasting can be done extremely quickly
online using an ensemble Kalman filter with no requirement for repeated parameter estimation.
We conduct an experiment where we train the model using 13 years of daily sea-surface tem-
perature data in the North Atlantic Ocean. Forecasts are seen to be accurate and calibrated.
A key advantage of our approach is that the CNN provides a global prior model for the dy-
namics that is realistic, interpretable, and computationally efficient. We show the versatility
of the approach by successfully producing 10-minute nowcasts of weather radar reflectivities in
Sydney using the same model that was trained on daily sea-surface temperature data in the
North Atlantic Ocean

No surprise a convoluted neural network turns up in the forecasting process but no mention of buoy data for verification. Was just a quick search though.

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Buoy data 23melt
« Reply #26 on: April 19, 2023, 04:22:02 PM »
Quote
no mention of buoy data for verification. Was just a quick search
Thx! Given the models, do we still need the buoys then? It is a big effort/expense to put them out there.

Seems like upper 2-3 m of ice-related is all that is missing (but that is modelled too).

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #27 on: April 19, 2023, 09:58:23 PM »
..

A-Team

  • Young ice
  • Posts: 2977
    • View Profile
  • Liked: 944
  • Likes Given: 35
Re: Buoy data 23melt
« Reply #28 on: April 20, 2023, 12:15:05 PM »
Disruption of international cooperation on climate change science matters would be very unfortunate. Note though voyages of both IB Odin and RV Polarstern already avoided Russian waters in earlier years because of bureaucratic delays and uncertainties in obtaining permits.

We've been following the Arctic methane studies of Igor Semiletov and Natalia Shakhova for years here. Their most recent paper still shows joint appointments between Alaskan and Russian academic institutions.

Natalia Shakhova 2,3,4
Igor Semiletov 3,4,5

2 Department of Chemistry, Moscow State University, Moscow, Russia
3 Pacific Oceanological Institute, Russian Academy, Vladivostok
4 International Arctic Research Center, University of Alaska Fairbanks 
5 Tomsk State University, Tomsk, Russia

Enhanced Transport of Dissolved Methane From Chukchi Sea to Central Arctic Ocean
https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2022GB007368

So far, the Russian full-text server Sci-Hub provided by Alexandra Elbakyan has not been affected. There would really be an uproar if cooperation there were interrupted.

88,343,822 documents provided so far
125,616 articles provided per hour

The US military-industrial complex pollutes more than 171 countries
Stephen Semler Dec 7, 2021
https://stephensemler.substack.com/p/the-us-military-industrial-complex

Approximately 336 million metric tons of CO2e, based on the latest available data; 56 million metric tons from the military itself (2018 figures) and about 280 million metric tons of CO2e from military industry (2017 figures). Situated amongst the 196 countries the World Bank ranked by greenhouse gas emissions in 2018, the US military-industrial complex came in at 25th-highest, polluting more than 171 countries.

Local interference (not jamming) has been noted at 5.405 Ghz which affects meteorological C-band radar satellites though not so far over Arctic ice.
« Last Edit: April 20, 2023, 12:51:22 PM by A-Team »

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #29 on: April 22, 2023, 05:29:15 PM »
Quote
still need buoys?

Hopefully we are just getting started.
SIMB3 temp profiles, -15C to -1.5C seems a good range for this time of year in the Beaufort

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #30 on: May 02, 2023, 12:07:13 AM »
SIMB3 buoys update.
1. air temps
2. ice thickness
3. difference between top of buoy temp and air temp (proxy insolation)

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #31 on: May 12, 2023, 01:33:32 PM »
SIMB3 buoys update.
1. air temps. First peak above zero on may9 20:00 at 1.19C
2. ice thickness. Early signs of slowing on 2 buoys.
3. difference between top of buoy temp and air temp (very very rough proxy insolation)

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #32 on: May 12, 2023, 09:36:19 PM »
Wondering if a 12hr moving avg would be better for proxy insolation as that would offset the highest insolation temps against the radiative cooling at 'night'.

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #33 on: May 18, 2023, 04:06:08 PM »
563570 ice bottom beginning to warm after the recent positive air temps.

1. buoy temps  -15C to +0.5C
2. 2cm temp difference. Negative heat flux slowing. First signs of positive heat flux down to 0.9m.
3. drift track

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #34 on: May 19, 2023, 09:22:24 PM »
Continuing from https://forum.arctic-sea-ice.net/index.php/topic,3941.msg368203.html#msg368203
651330 water temps down to 2.9m are similar to mar9

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #35 on: May 21, 2023, 01:12:03 PM »
Follow up on 651330. Still looks like warming from below to me.
« Last Edit: May 21, 2023, 01:34:33 PM by uniquorn »

uniquorn

  • First-year ice
  • Posts: 5115
    • View Profile
  • Liked: 2158
  • Likes Given: 388
Re: Buoy data 23melt
« Reply #36 on: May 21, 2023, 09:21:05 PM »
6903587 has been drifting near the entrance to the Barents Sea for 2 years
Quote
Last station date  20/05/2023
Last Surface Data 2.9 dbar 5.312℃ 35.015 PSU