GaWx Posted October 25, 2014 Share Posted October 25, 2014 Here's the snow cover percentage on 31 October 2012 from the 20th Century Reanalysis v2. The extent listed in the image is calculated by multiplying the area of each lat-lon cell by the fractional percentage of snow cover. Obviously, the resolution is significantly degraded with these data which may account for some of the differences with respect to the Rutgers dataset. To get an estimate of the current snow cover, I used the field initialized in last night's 00z GDAS. The snow cover field in this model has greater resolution but is binary and thus not completely comparable to the above plots. The discrepancies between these values and those posted by cfbaggett are probably due to the process of assimilating snow cover data into the models. Also, I'm not entirely sure how snow cover is defined for a particular location from the remote sensing observations, but it is likely not consistent with the values derived from percentage cover. snow_cover_gfsw.pngHeavywx/folks, If heavywx's sub 60N 10/25 SCE of only 6.87 msk is accurate, it would be VERY disappointingly low vs. my hope/expectations. As of way back on 10/19, cfbaggett's graph had it already up to 6.70 msk. Furthermore, cfbaggett estimated it was up to 8.5 msk just yesterday, which makes sense considering the continued good snowfall in new areas since then. If it really has stalled and is only at 6.87 now, that would suggest SIA that may not be too much better than mediocre. Could it be that cfbaggett and heavywx are both right due to different methodologies? I wish Rutgers would return as that gives me apples o apples better than other graphs. Edit: heavywx, do you have your sub 60N SCE as of 10/19? Link to comment Share on other sites More sharing options...
mitchnick Posted October 25, 2014 Share Posted October 25, 2014 of all years for the NAOO SIAS to go down (and I believe their info supports the Riutgers site) lol and wth is the delay in getting them back up Link to comment Share on other sites More sharing options...
heavy_wx Posted October 25, 2014 Share Posted October 25, 2014 Heavywx/folks, If this sub 60N 10/25 SCE of only 6.87 msk is accurate, it would be VERY disappointingly low vs. my hope/expectations. As of way back on 10/19, cfbaggett's graph had it already up to 6.70 msk. Furthermore, cfbaggett estimated it was up to 8.5 msk just yesterday, which makes sense considering the continued good snowfall in new areas since then. If it really has stalled and is only at 6.87 now, that would suggest SIA that may not be too much better than mediocre. Could it be that cfbaggett and heavywx are both right due to different methodologies? I wish Rutgers would return as that gives me apples o apples better than other graphs. I would not compare the 6.87 msk from the GDAS with the remotely-sensed values. I think they are probably systemically lower due to the assimilation process and the different definitions of snow cover used. I think if you went through and performed the regression of these data with the DJF AO, the coefficient would probably map lower SCE values to the same AO values as before. Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 of all years for the NAOO SIAS to go down lol and wth is the delay in getting them back up I'm not so sure it has slowed down vs. 10/19 as heavywx's 6.87 implies. I'm waiting on his response. Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 I would not compare the 6.87 msk from the GDAS with the remotely-sensed values. I think they are probably systemically lower due to the assimilation process and the different definitions of snow cover used. I think if you went through and performed the regression of these data with the DJF AO, the coefficient would probably map lower SCE values to the same AO values as before. Heavywx, What was your sub 60N as of 10/19? The answer is obviously very important. Link to comment Share on other sites More sharing options...
heavy_wx Posted October 25, 2014 Share Posted October 25, 2014 Heavywx, What was your sub 60N as of 10/19? The answer is obviously very important. Here's the snow cover from 10/19. Eurasian extent was 5.17 msk. Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 Here's the snow cover from 10/19. Eurasian extent was 5.17 msk. snow_cover_gfsw.png WHEW!!! What a relief. So, you have only 5.17 vs cfbaggett 's 6.7 as of 10/19 and your SCE has grown a whopping 1.70 just since 10/19. Assuming the same ratio, cfbaggett's 10/25 # would be all of the way up to 8.9, which makes much more sense and would mean an absolutely stellar SAI is practically in the bank for 2014 being that there's still more snow to fall over bare areas and it is very cold in much of the region! I have been saying that if cfbaggett's 10/31 SCE were to be 8.5+ or perhaps only 8.0+ that we'd almost definitely be among the elite highest 4 of the last 46 years for total Eurasian SCE Oct. increase: 1976, 2009, 2012 being the others. Also, it would likely be a little better than 1968. The implications of this, along with it hopefully being an actual Niño, are about as good as could be imagined for a strong -AO this DJF based on Oct. Eurasian SCE advance, alone! Edit: cfbaggett's next update will likely get close to going off the top of his chart! Link to comment Share on other sites More sharing options...
mitchnick Posted October 25, 2014 Share Posted October 25, 2014 I'm not so sure it has slowed down vs. 10/19 as heavywx's 6.87 implies. I'm waiting on his response. no Ga, I meant the website! Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 no Ga, I meant the website! Cool! I can't wait to see cfbaggett's next graphical update. 2014 is liable to threaten to go off the top of his chart unless he expands it from its current max. of 9.0! As much attention as the SCE/SAI has been getting, I think it is going to get a whole lot more attention once he releases his next chart here. Link to comment Share on other sites More sharing options...
natg2020 Posted October 25, 2014 Share Posted October 25, 2014 ......I have been saying that if cfbaggett's 10/31 SCE were to be 8.5+ or perhaps only 8.0+ that we'd almost definitely be among the elite highest 4 of the last 46 years for total Eurasian SCE Oct. increase: 1976, 2009, 2012 being the others. .... Do you mean 2012 or 2013? http://climate.rutgers.edu/snowcover/table_rankings.php?ui_set=1#eurasia Here, it is 2013 ranking #4 , not 2012 (ranking #12) Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 Do you mean 2012 or 2013? http://climate.rutgers.edu/snowcover/table_rankings.php?ui_set=1#eurasia Here, it is 2013 ranking #4 , not 2012 (ranking #12) No, I really meant 2012. You are looking at something totally different. The high ranking (4th) for 2013 is based on the average SCE for the entire month. It started very high relative to normal. However, it crapped out near the end of the month, making its SCE increase below average! I'm saying that the Oct. SCE increase of ~15.5 (total Eurasian) for 2012 was 3rd highest. 2013 was way further back to as low as only 8 when you consider the large amount of melting 10/29-31/13 before coming back. Link to comment Share on other sites More sharing options...
natg2020 Posted October 25, 2014 Share Posted October 25, 2014 No, I really meant 2012. You are looking at something totally different. The high ranking (4th) for 2013 is based on the average SCE for the entire month. It started very high relative to normal. However, it crapped out near the end of the month, making its SCE increase only mediocre. I'm saying that the Oct. SCE increase of ~15.5 (total Eurasian) for 2012 was 3rd highest. 2013 was way further back. Thank you! Link to comment Share on other sites More sharing options...
Absolute Humidity Posted October 25, 2014 Share Posted October 25, 2014 Clearly a ton of more snow today than one year ago today. Link to comment Share on other sites More sharing options...
Mike.Ventrice Posted October 25, 2014 Share Posted October 25, 2014 Hi all again, So I weighted my latitudes appropriately and the resulting graph is as follows: Not a dramatic change, if any at all. So that wasn't the issue with my graph. I'm truly puzzled. I agree with the folks in there that there "looks" like a great deal more of snowfall this year than the other years, but it seems to be all in the 45-60N band. Maybe my analysis is very stringent, which is causing the snowfall % to be low. I'm looping through all days, latitudes, and longitudes. Where a grid point has a snow depth > 0, I assign that grid point "1". If the grid point has no value, I assign that point "0". Then I sum all the points per day, and divide by the total number of grid points be analyzed. For the 1.0 degree GFS, 61 degrees of latitude X 181 degrees of longitude = 11041 points total. The most current time step is 1165 snow covered grid points / 11041 points total, yielding 10.5% covered points.Still scratching my head... would love to see an objective analysis from another person here to confirm that my charts are incorrect. Maybe it comes down to GFS initialization verses satellite coverage. If that were the case.. that would be extremely interesting IMO. Link to comment Share on other sites More sharing options...
StudentOfClimatology Posted October 25, 2014 Share Posted October 25, 2014 Perhaps there's an error in the reanalysis data itself? The satellite imagery is pretty clear on this so I'd go to the source to find the problem. Do you have a map of 2012 coded up and ready? Would make this much easier. Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 Mike, It isn't just the current SCE. I see no way that 2014 could be anywhere near as low as 2009 as of 10/1 (whether all of Eurasia or just sub 60N). Just compare the two maps at Rutgers. They're like night and day. http://www.americanwx.com/bb/index.php/topic/44473-and-we-begin/?p=3095614 Link to comment Share on other sites More sharing options...
Mike.Ventrice Posted October 25, 2014 Share Posted October 25, 2014 Mike, It isn't just the current SCE. I see no way that 2014 could be anywhere near as low as 2009 as of 10/1 (whether all of Eurasia or just sub 60N). Just compare the two maps at Rutgers. They're like night and day. http://www.americanwx.com/bb/index.php/topic/44473-and-we-begin/?p=3095614 Could this be another SST dataset issue? Not all SST reanalyses are alike. I wonder if there are systematic differences between the three data sets. Recall my archive years are from the 0.75 era interim. Link to comment Share on other sites More sharing options...
cfbaggett Posted October 25, 2014 Share Posted October 25, 2014 Cool! I can't wait to see cfbaggett's next graphical update. 2014 is liable to threaten to go off the top of his chart unless he expands it from its current max. of 9.0! As much attention as the SCE/SAI has been getting, I think it is going to get a whole lot more attention once he releases his next chart here. I'm very excited too! I'll update the graph as soon as NCEP fixes their satellite issues. I get my data from here: ftp://sidads.colorado.edu/pub/DATASETS/NOAA/G02156/24km/2014/ Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 Could this be another SST dataset issue? Not all SST reanalyses are alike. I wonder if there are systematic differences between the three data sets. Recall my archive years are from the 0.75 era interim. Mike, I'd like to ask you this. Did you do 2014 differently from 2005-2009/2011-3? If you did, I'm wondering if 2014 could actually be correct while all of the other years are too high. When I think about it, 3% SCE sub 60N on 10/1 actually doesn't sound low at all (gut feel..I may not be right). All of the others listed are 3-7% on 10/1. Could those 3-7%ers all be too high? I'm just brainstorming here to try to help you figure out what the heck is going on with your graphs. Link to comment Share on other sites More sharing options...
GaWx Posted October 25, 2014 Share Posted October 25, 2014 Mike, I just noticed another thing. Why is 10/1/2013 at ~6.7 while 10/1/2014, which was very similar, at ~2.7? That can't be per Rutgers site. I'm again wondering if the problem is that the non-2014 years are all too high while 2014 is perhaps ok? Or maybe it is just that 2014 is too low. Bottom line: 2014 starting out with 2009 as the lowest doesn't jibe with Rutgers. On the contrary, it should be near the high together with 2013 at the start (10/1). Link to comment Share on other sites More sharing options...
The_Global_Warmer Posted October 25, 2014 Share Posted October 25, 2014 Does longitude matter in terms of where snow cover is South of 60S? Does snow cover near Asia matter more than SW Russia and vice versa? Link to comment Share on other sites More sharing options...
heavy_wx Posted October 25, 2014 Share Posted October 25, 2014 Hi all again, So I weighted my latitudes appropriately and the resulting graph is as follows: Not a dramatic change, if any at all. So that wasn't the issue with my graph. I'm truly puzzled. I agree with the folks in there that there "looks" like a great deal more of snowfall this year than the other years, but it seems to be all in the 45-60N band. Maybe my analysis is very stringent, which is causing the snowfall % to be low. I'm looping through all days, latitudes, and longitudes. Where a grid point has a snow depth > 0, I assign that grid point "1". If the grid point has no value, I assign that point "0". Then I sum all the points per day, and divide by the total number of grid points be analyzed. For the 1.0 degree GFS, 61 degrees of latitude X 181 degrees of longitude = 11041 points total. The most current time step is 1165 snow covered grid points / 11041 points total, yielding 10.5% covered points.Still scratching my head... would love to see an objective analysis from another person here to confirm that my charts are incorrect. Maybe it comes down to GFS initialization verses satellite coverage. If that were the case.. that would be extremely interesting IMO. Hey Mike, I got the same number of snow-covered points and coverage ratio using the GFS data. In order to get a percentage coverage, I think you need to calculate the ratio of the area of all the cells associated with a snow-covered point and the total area of all the cells. It looks as if you are using NCL which has a function 'area_global_rectilinear_grid(-lat,lon,False) ' in the 'shea_util.ncl' script. For some weird reason, I had to put a minus sign in front of the lattitude array to get a positive area. FYI, It's also more efficient to use the 'mask' function to find the number of snow-covered points than to loop through all the lat/lon points. There does seem to be some disparity between the model data and the satellite obs, for whatever reason. Link to comment Share on other sites More sharing options...
griteater Posted October 25, 2014 Share Posted October 25, 2014 Does longitude matter in terms of where snow cover is South of 60S? Does snow cover near Asia matter more than SW Russia and vice versa? The Cohen SAI paper doesn't point out any distinction...it's just snow cover advance equatorward of 60N in all of Europe and Asia Link to comment Share on other sites More sharing options...
Mallow Posted October 25, 2014 Share Posted October 25, 2014 Hi all again, So I weighted my latitudes appropriately and the resulting graph is as follows: Not a dramatic change, if any at all. So that wasn't the issue with my graph. I'm truly puzzled. I agree with the folks in there that there "looks" like a great deal more of snowfall this year than the other years, but it seems to be all in the 45-60N band. Maybe my analysis is very stringent, which is causing the snowfall % to be low. I'm looping through all days, latitudes, and longitudes. Where a grid point has a snow depth > 0, I assign that grid point "1". If the grid point has no value, I assign that point "0". Then I sum all the points per day, and divide by the total number of grid points be analyzed. For the 1.0 degree GFS, 61 degrees of latitude X 181 degrees of longitude = 11041 points total. The most current time step is 1165 snow covered grid points / 11041 points total, yielding 10.5% covered points.Still scratching my head... would love to see an objective analysis from another person here to confirm that my charts are incorrect. Maybe it comes down to GFS initialization verses satellite coverage. If that were the case.. that would be extremely interesting IMO. Mike, I still think the best/easiest comparison would be to simply plot a map (like the one heavy_wx plotted) of the snowcover used to get a value in 2014, vs. the analogous map for 2012 (say the GFS map for 10/24/2014 vs. the ECMWF interim map for 10/24/2012). That way, any discrepancy should be obvious to spot. Link to comment Share on other sites More sharing options...
The_Global_Warmer Posted October 26, 2014 Share Posted October 26, 2014 Is there any correlation between the PDO index and October Eurasian snow cover. Link to comment Share on other sites More sharing options...
hosj III Posted October 26, 2014 Share Posted October 26, 2014 Is there any correlation between the PDO index and October Eurasian snow cover. Apparently, there is. Link to comment Share on other sites More sharing options...
DTWXRISK Posted October 26, 2014 Share Posted October 26, 2014 GREAT POST ... great images great analysis The deep trough north of the Caspian this month so far is a great October snow signal for Eurasia. This has been the pattern for the top 5 Eurasian snow cover extents in October. A strong ridge in that region is associated with the 5 lowest Eurasian snow cover extents for October. 2014 October 14.gif Top 5 snow cover October composite COMP.png 5 lowest snow cover October composite LOW.png Link to comment Share on other sites More sharing options...
GaWx Posted October 26, 2014 Share Posted October 26, 2014 Does anyone have any clue as to when the Natice/Rutgers daily snowcover updates will resume? It doesn't appear it will be tonight as it looks like another blank map will be released (for 10/25). They haven't updated since 10/19. Could we be looking at several more weeks of this? When is the satellite data expected to be back to 100%? Link to comment Share on other sites More sharing options...
Harry Posted October 26, 2014 Share Posted October 26, 2014 The Oct 2012 map seems a bit off. I for one do not remember having snow cover on Oct 31 2012; Nov 7th yes but that was from the son of Sandy. I don't think Sandy's snowfall was nearly that expansive. Maybe I'm missing something? Oh it is off. There was no snow here either which that shows. Link to comment Share on other sites More sharing options...
bluewave Posted October 26, 2014 Share Posted October 26, 2014 GREAT POST ... great images great analysis Thanks, Dave. The recent pattern of more blocking in October across Greenland and the Arctic since 2002 seems to favor this teleconnection pattern leading to greater Eurasian snow cover. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.