Jump to content

donsutherland1

Members
  • Posts

    19,824
  • Joined

Everything posted by donsutherland1

  1. On November 18, 2016 Arctic Sea Ice Extent (JAXA) was 8,320,714 square kilometers. That was a decline of 97,385 square kilometers from the previous day. It is also 948,831 square kilometers below the previous record minimum for the date of 9,269,545 square kilometers, which was set in 2012. The biggest 1-day declines in the October 1-December 31 timeframe: 1. 97,385 sq. km., 11/18/2016 (exceeded the biggest 2-day decline for this timeframe). 2. 54,064 sq. km., 12/25/2011 3. 53,292 sq. km. 12/1/2007 4. 51,274 sq. km. 12/17/2011 5. 48,440 sq. km. 10/25/2009 The November 17, 2016 decline of 46,717 square kilometers ranked 6th biggest for this timeframe. The 11/17-18/2016 2-day decline of 144,102 square kilometers exceeded the previous record of 81,804 square kilometers, which was established on 12/1-2/2007.
  2. It was working this morning. Perhaps there's a temporary issue that is sending visitors to the site's twitter stream.
  3. You can find the data here: http://ocean.dmi.dk/arctic/meant80n.uk.php
  4. It does. To date, 2016 has set daily minimum records on 46% of days this year.
  5. The 9/6 Arctic sea ice extent figure on JAXA was 4,022,615 square kilometers. That was a decrease of 22,855 square kilometers from 9/5. It would represent the 2nd lowest minimum figure on record.
  6. 9/5 JAXA figure: 4,045,470 square kilometers, down 10,836 kilometers. This would be the second lowest minimum figure on record.
  7. 9/4 JAXA figure: 4,056,306 square kilometers. This was a slight increase from 9/3, but it is not likely the minimum for this year.
  8. Arctic sea ice extent falls to 2nd lowest on record... The 9/3 figure on JAXA was 4,054,179 square kilometers. That would ranks as the 2nd lowest minimum figure on record. Only 2007 (4,065,739 square kilometers) was lower. The 5-year average decline in sea ice extent from 9/3 would produce a minimum figure of 3,852,090 square kilometers. The minimum decline (2002-15) would result in a figure of 3,950,344 square kilometers. The maximum decline (2002-15) would result in a minimum figure of 3,647,151 square kilometers. A minimum extent under 4,000,000 square kilometers appears very likely (> 90%) and a figure just under 3,900,000 square kilometers appears possible.
  9. The 9/2 figure on JAXA was 4,090,129 square kilometers. That would rank as the 3rd lowest minimum figure on record. Only 2012 (3,177,455 square kilometers) and 2007 (4,065,739 square kilometers) were lower. The 5-year average decline in sea ice extent from 9/2 would produce a minimum figure of 3,875,040 square kilometers. The minimum decline (2002-15) would result in a figure of 4,000,554 square kilometers. The maximum decline (2002-15) would result in a minimum figure of 3,610,482 square kilometers. A minimum extent under 4,000,000 square kilometers appears very likely and a figure just under 3,900,000 square kilometers appears possible.
  10. The 9/1 figure on JAXA was 4,168,394 square kilometers. That would rank as the 3rd lowest minimum figure on record. Only 2012 (3,177,455 square kilometers) and 2007 (4,065,739 square kilometers) were lower. The 5-year average decline in sea ice extent from 9/1 would produce a minimum figure of 3,931,250 square kilometers. The minimum decline (2002-15) would result in a figure of 4,050,385 square kilometers. The maximum decline (2002-15) would result in a minimum figure of 3,663,584 square kilometers. A minimum extent under 4,000,000 square kilometers appears likely and a figure just under 3,900,000 square kilometers appears possible.
  11. The 8/31 figure on JAXA was 4,242,650 square kilometers. That would rank as the 3rd lowest minimum figure on record. Only 2012 (3,177,455 square kilometers) and 2007 (4,065,739 square kilometers) were lower. The 5-year average decline in sea ice extent from 8/31 would produce a minimum figure of 3,979,208 square kilometers. The minimum decline (2002-15) would result in a figure of 4,092,669 square kilometers. The maximum decline (2002-15) would result in a minimum figure of 3,740,828 square kilometers.
  12. That's the way it appears. That's why I still think there's a reasonable prospect that the minimum will be under 4 million square kilometers.
  13. Now that August is ending, a closer look at the Arctic sea ice extent figure is in order. The 8/30 figure (JAXA) was 4,302,421 square kilometers. If that were the minimum, it would rank as the 5th lowest on record. However, it is all but certainly not the minimum. A number of scenarios for the minimum figure: Smallest decline from 8/30 to the minimum (2002-15): 4,140,726 square kilometers (would rank 3rd lowest) Largest decline from 8/30 to the minimum (2002-15): 3,793,311 square kilometers (would rank 2nd lowest) Average decline from 8/30 to the minimum (2011-15): 4,007,579 square kilometers All said it appears very likely that 2016 will see the 2nd or 3rd lowest Arctic sea ice extent minimum on record. There remains a reasonable possibility of a minimum figure below 4 million square kilometers for only the 2nd time on record (JAXA). 2012 is currently the only such case.
  14. http://www.nature.com/ngeo/journal/v5/n10/pdf/ngeo1580.pdf
  15. All the RATPAC data can be found here: http://www1.ncdc.noaa.gov/pub/data/ratpac/
  16. At this time, I don't think there's strong reason to avoid using RATPAC. It will be interesting to see if RATPAC picks up on the El Niño's impact in the lower and middle troposphere before UAH (especially UAH v.6.0).
  17. Thanks for the kind words. There's no risk of such an outcome.
  18. But would UAH (especially 6.0) and RSS look different if the problem related to cloud-affected radiances were fixed, especially as the lower and middle tropospheric warming trend would be increased by an estimated 20%-30%?
  19. Although there are 7 satellites, don't each of the satellites depend on a microwave radiometer?
  20. Just because a data set requires more work to complete the product doesn't mean that it isn't reliable. The measure of reliability is whether it fits other measured data sets. RATPAC does. In stark contrast, the satellite estimates have been diverging from the measured data. That raises several questions: 1. How can the possible spurious cooling be explained? 2. What fixes will be applied to address that problem? 3. With regard to UAH v.6.0, which has a linear cooling trend relative to v.5.6 (in addition to the surface data sets) with a very high coefficient of correlation, where is the peer-reviewed paper to address that possible issue and lay out the adjustments? The paper to which I provided a link addresses one of the problems related to such cooling (and the difference in temperatures is large). The proposed fix in the paper has not been applied to the satellite data sets (or at least no papers or web searches indicated such a fix was being made). It's been 3 1/2 months since v.6.0 has been released and no paper has been submitted for peer review. Why? Usually papers precede methodology changes, not the other way around, so that the proposed changes are reviewed in a robust fashion before being implemented. In sum, the data divergence and unaddressed issues (including the brightness issue found in the paper) argue that perhaps there is greater reason to question the satellite-derived estimates than RATPAC. Given the above, IMO, the satellites are part of the mix. Issues exist and improvements are almost certainly necessary given those issues. At this point in time, I don't believe one can say that the satellites are qualitatively superior to RATPAC, much less the surface data sets.
  21. This is a key point. On a year-to-year basis, there might be some differences (though at least this year, the data closely fits the surface data sets). Over timeframes of 10 years or longer, RATPAC has closely matched the surface data sets. IMO, the issue of possible spurious cooling on the satellite data sets, all of which require very complex calculations to try to account for factors such as diurnal drift, among others, is something that likely requires a closer examination, especially as satellites do not directly measure temperatures. That the satellite data has diverged from actual measurements (surface + RATPAC) suggests at least an examination of that issue is warranted. In fact, the accuracy of the present approach to satellite temperature estimates has already been raised in at least one academic paper with at least one major fix suggested: http://www.theguardian.com/environment/climate-consensus-97-per-cent/2014/nov/07/new-study-disputes-satellite-temperature-estimates http://link.springer.com/article/10.1007%2Fs00382-013-1958-7
  22. Its resolution notwithstanding, it has provided surface trends that are representative of those on the major surface data sets.
  23. The latest CFSv2 ENSO forecast is indicating that the ENSO Region 1+2 anomaly will be at or above +1.00°C throughout meteorological winter (December-February). It is also indicating this region's anomaly should have peaked, will fall, and then again increase. A secondary peak is actually not uncommon during strong or super El Niño events.The 1972-73, 1982-83, and 1997-98 events all had secondary peaks in this region. The 1997-98 event had a tertiary peak. Since 1950, only three meteorological winters saw 2 or more months have Region 1+2 anomalies of +1.00°C or above: 1972-73, 1982-83, and 1997-98. The latter two featured all three months with such anomalies. Given the modeling, it is very likely that August will have a Region 1+2 anomaly of at least +1.00°C. Since 1950, such anomalies have occurred in 1951, 1957, 1965, 1972, 1976, 1982, 1983, 1987, 1997, 1998, 2008, 2009, and 2014. However, only 3 of those 13 cases met the threshold being signaled on the CFSv2. By October, ENSO Region 1+2 anomalies of +1.5°C or above, filtered out almost all of those cases, leaving only 1972, 1982, 1987, and 1997. Hence, by October, we should have a strong signal as to whether this current CFSv2 forecast has a reasonable chance of verifying. In November, using the same anomaly threshold as October, only 1972, 1982, and 1997 remained. Such anomalies during meteorological winter would have implications for North American temperature anomalies and possibly snowfall along the East Coast. Should the strong PDO+ persist, the PNA+ could also predominate (1982-83 and 1997-98) leading to another warmer than normal winter in the Pacific Northwest. All of that is still far out, but the scenario currently shown on the CFSv2 would typically favor warmth across much of North America. In the more immediate future, one has typically seen El Niño events with Region 1+2 anomalies at or above +1.00°C produce cool anomalies in large parts of the East in August. Summer 2009 was an exception. Warmth has sometimes occurred in the Pacific Northwest and western half of the U.S. In the end, it's still too soon to be sure about the upcoming winter. But if the CFSv2 is right, a cool autumn could yield to a warmer to much warmer than normal December across a wide swath of the U.S. Given where things stand, one should be aware of this possible scenario, but not yet lock it in. Much can still change in Region 1+2. FOLLOW THE LATEST...
  24. If one takes a look at the overnight 1/16 0z and 6z operational GFS runs, one finds both are very cold in the days 7-15 timeframe. Both also feature a strong EPO-. Both also contain a number of analog dates near moderate or larger snowstorms for some parts of the eastern half of North America. But if one looks more closely, there is actually a rather dramatic change between the 0z and 6z runs. The latter shows the development of meaningful Atlantic blocking to coincide with the strong EPO-. Indeed, if one goes to the 11-day objective analogs from each of these runs, one finds the following: 0z Run: Average AO: +0.125 AO > 0: 60% of analogs AO of -1 or below: 30% of analogs AO of +1 or above: 20% of analogs 6z Run: Average AO: -1.128 AO < 0: 80% of analogs AO of -1 or below: 70% of analogs AO of +1 or above 10% of analogs If one checks out the GFS ensemble forecast for the AO, one also finds that the many members are now favoring a negative AO in the extended range: If this forecast verifies and dual Atlantic and Pacific blocking develop, that would increase prospects for a cold February in the eastern third to half of the CONUS, along with southern Ontario and Quebec. The latter cold anomalies would depend on the magnitude of blocking. If the blocking becomes too strong, then the cold anomalies could be driven southward. Such dual blocking would also allow for potentially more opportunities for larger snowfalls. The Great Lakes region, northern Mid-Atlantic (e.g., Philadelphia) and southern New England areas remain on track for above normal seasonal snowfall, as the December outcomes were consistent with such seasons. The pattern ahead looks promising, particularly for the 1/25-2/15 period. If blocking develops, things could also become more favorable further south in the Mid-Atlantic region, including the Baltimore and Washington, DC areas. The takeaway is that the theme of a growing probability of a cold outcome in parts of the eastern CONUS has been sustained in the overnight guidance. In fact, even as it is still outside its skillful range, the CFSv2 has recently shifted from featuring widespread February warmth to an increasingly colder idea for the eastern third to half of North America.
×
×
  • Create New...