WINTER FORECAST  2001- 02

I'm not afraid of anything in this world
there is nothing that you can throw at me that I haven't already heard
I'm just trying to find a decent melody
A song that I can sing in my own compan
y  

--U2  "Stuck in a Moment  You can get out of"


Those who have read and used the   Seasonal Forecasts  here at WxRisk  know that I had some good success in the last few years. As WxRisk has grown... so has the pressure and the need to improve and become more comprehensive.  There is also an tendency to  one up the competition as it were by being the "first-est with the Most-est".     If one waits until  middle or late NOV    then the value of the forecast is somewhat diminished but issuing the winter forecast in early September is not a good idea either.

Some forecasters/ Private Weather  Services  use purely analog years... some use mainly SSTs  ( sea surface temperatures) in the ENSO   regions ( El Nino and La Nina regions of the central equatorial Pacific ).... some use their own in house computer models.... and some use mainly the CPC  (Climate Prediction Center of the HPC/ NCEP)  forecast. I do not have a preferred method per se -- I kind of use a lot of things.

THE ANALOG METHOD      

CLIMATE  & SEASONAL MODELS    

WXRISK WINTER 2001-02 FORECAST WEATHER SUMMARY  

WHAT OTHERS ARE SAYING (RESTRICTED)

BASICS

I was unconscious   --half asleep
The water is warm until you discover how deep
I wasn't jumping-- for me it was a fall
Its was a long way down to nothing at all

In the Weather Biz... as it is with most things... Cause and Effect is a BIG deal.  All events occur for a reason; One event leads, causes, or contributes in some form or  another to an event happening -- often at a later point in time.   The Mild winters of the late 80s and early 90s that dominated the Central and Eastern US did NOT Occur  out of happenstance. They occurred  because of two main factors. 

  1. The Frequent and long lasting El Nino  ("Warm  El Nino " events  may be referred to as "+ENSOs"). 
    FACT --  In the 18 years  from 1982 to year 2000 ( 18 x 4 quarters/ per year = 72 quarters) 
    there has been a Moderate or Strong  LA Lina/ El Nino in   30 of the 72 quarters!

FACT: there has   been ONLY 1 year since 1981 where SSTs  have been NEUTRAL   in the ENSO regions that lasted 12 month or more.  ( AMJ 1996 to JFM 1997).  Thus one of the reasons why there have been so few prolonged cold/ snowy winter in the Central and Eastern US has been the constant turmoil in the ENSO regions.

It is often and commonly repeated   "fact"   that +ENSO (warm ENSO events)   or -ENSO   (cold ENSO   events)  strongly support or contribute to Warm  Dry winters across the Northern  and Eastern US... Wet and Cooler winters across the Lower Plains and Deep South... and excessive rainfall along the West coast. I have even seen this silly assertion from professional meteorologists.  FIG 1 shows why some warm ENSOs  event can produce cold snowy and stormy winters and some ENSO events do indeed   support warm dry  winters.

FIGURE 1
FIG.2.7.gif (51807 bytes)  Note the Upper two maps  Upper Left  (JAN- FEB-MAR 1966)   and Upper Right (JAN   FEB  MAR 1969)  were  severe  winters in the Central and Eastern US with warm ENSO events... But in both instances the ENSO  was centered   over the CENTRAL PACIFIC   from the 160 E Longitude ...across the Dateline to 140 W  Longitude and the ENSOs  were WEAK.   The other two cases  --  Lower  Left  (JAN- FEB-MAR 1983)    and Lower  Right (JAN  FEB  MAR 1992)  were VERY mild winters.  Note the ENSOs were either Record strong or moderate and located from S. America coast WEST towards 150  W Longitude.

2.  The second driving Factor as to why there has  been such a dearth of  cold snowy winters in the Central and Eastern US is the well known NAO   ( North American Oscillation or Greenland Block)  and the connection between the recently discovered and new researched  Pacific Decade Oscillation (or PDO). The PDO  is leading to   much  better understanding  of longer term climate patterns and teleconnection tendencies.  In Figure 2... I have  placed he PDO and NAO on the same page  for clarity.

fig07.gif (49035 bytes)  The Top Graph is the PDO.... the RED showing positive phase which enhances  ENSO activity and intensity. In the period of 1910-1945  there was overwhelming period of Positive PDO....  EXCEPT for a brief Negative  PDO phase  during from 1915 to 1925... followed by a Positive Phase PDO from 1933 to 1945.

Similarly  when one looks at the NAO   graph at the bottom from 1910 to 1945...   the phase is also frequently Positive  EXCEPT for one period around 1920-22... followed by a Positive Phase NAO from 1933 to 1945.

Then from 1945 to 1980 the PDO was overwhelming  Blue or Negative   which is a VERY close match to the prolonged Negative NAO period of the 1950s   60s and 70s.   Then even more amazingly   the PDO graph at the far end... since 1980.. shows HUGE positive Values  ( when there was excessive ENSO activity)   and the  NAO was JUST as sharply  Positive!
fig07.gif (49035 bytes)     It is rare when one fines such a STRONG correlation from seemingly distant weather features.

**  Given the Strong Negative Phase of the PDO.... the large pool of warm SSTs in the North Atlantic.... the persistent  Blocking  that has been seen for 12-16 months over  eastern Canada and Greenland... WxRisk the Winter of 2001-02 as one with significant Blocking  mechanisms and patterns that will increase as the winter wears on.   This sort of Split in the ENSO regions was last seen for such a long period in the winters of the late 1970s.

 

 

ENSO FORECASTS

The good folks at CPC  --- and  MANY others have been forecasting  a warm ENSO event for some time. Since DEC  98/ Jan 99  in one form or another in fact.  As recently as early AUG  the folks at CPC       and others sites were forecasting a Moderate ENSO developing in the ENSO 3 and 4 regions by mid or late Winter.   However the CMB AUG 28 aug28.gif (18565 bytes) ensembles showed a markedly different situation over the central equatorial Pacific.  This trend for a  NEUTRAL    event --- what the great Joe D' Aleo calls "La Nada"    has intensified both from new models data and ACTUAL  data from the Pacific region's  ENSO areas 1 2 3 and 4.

Here is the OCT 6th  CMB  SSTs forecast. As you can see  the Winter months are  dead bang Neutral with SOME  increase in SSTs temps in ENSO 3.4   at the end of the winter and Spring 2002.

 cmb SSToct.gif (21854 bytes)      Here is he September  SCRIPPS  Institute ENSO Forecast   figf5.gif (12864 bytes)  figf9.gif (56747 bytes)

I know  some of you are saying.... " Hey DT ... what is ACTUALLY happening ? Never the dab gum computer models...give us some real data!"    The Western portions of the ENSO regions  ( 3 and 4)  are warming   while the Eastern portions of ENSO  ( 1and 2)  near South America   remain  COOL with NO sign of any warming at all.

To see this  SPLIT  in the ENSO regions  click on this LINK   from CPC  which shows the ENSO SSTs all the way back to JUNE 2001.. 

The last time there was such a pronounced Split in the ENSO regions was 1977-78.  To see the Raw Data of e ENSO data  since 1950   click HERE

  YEAR / MONTH ENSO  REGION  1.2
ANOMALIES
ENSO  REGION  3.4
ANOMALIES
1977       2 -.07 0.33
1977      3 -.13 0.40
1977      4 -0.21 0.11
1977       5 -0.39 0.27
1977      6 -0.20 0.36
1977      7 -0.10 0.30
1977       8 -0.54 0.14
1977      9 -0.61 0.32
1977      10 -0.07 0.71
1977       11 -0.03 0.77
1977      12 - 0.35 0.69
1978      1 -0.15 0.83
1978       2 -0.06 0.24
1978      3 -0.94 0.22
1978      4 -0.44 -0.40

This pronounced Split in the ENSO   -- warm SSTs to the west and Cool SSTs to the east--- is UNUSUAL   ENSO configuration  and in the Opinion of WXRISK an  important Clue as to what and How the ENSO is going to develop as well as selecting Analog Years.

At this link --- Click   HERE -- we can see   strong warming over Eastern Pacific  has intensified over  the last 30 days.


I never thought you were a fool
Burt darlking look at you
You've got to get stand up  carry your own weight
These tears are going nowhere fast

WXRISK WINTER 2001-02 FORECAST WEATHER SUMMARY  

OVERALL   PATTERN PART1 OVERALL PATTERN PART 2   TEMPERATURES
map1.gif (85953 bytes) map2.gif (102783 bytes) map3.gif (154394 bytes)

  

PRECIPITATION SNOWFALL
map4.gif (113679 bytes) map5.gif (119680 bytes)

  

OTHER  PREDICTIONS not covered in the Maps

  1. The split in the El Nino regions  ( ENSO)  is a KEY   feature;  Only one other time since 1950 has the western ENSO SSTs  been so warm  and eastern ENSO been so cold. That was 1977-78-79.

  2. This winter will b  first time since the late 70s  that the ENSO regions SSTs will be  "neutral" for a period > 9 months.

  3. The similarity of this year's Hurricane season to last year's  may be a sign that some of the same atmospheric conditions that lead to the blocking patterns last winter are going to repeat.

  4. The overall cold pattern appears to be Increasing in area and Intensity from various seasonal/ Climate models as winter approaches.

  5. Depending on How deep the Mean trough is and where its located   there may be several Low pressures areas that move off the Southeast coast with significant Trough amplification. This may bring SEVERAL  Frost/ Freeze threats to FLORIDA.

  6. Lower Plains may become rather Dry from the second half of the Winter   into April and May.

 


ANALOG METHOD

Perhaps the oldest method of making long range or Seasonal weather forecasts is the Analog method. The Idea is to focus in on a particular  parameter or set of parameters and search the climatological records for a period  that has seen similar  conditions to the selected parameter. For Instance   one may wish to focus on the QBO
(
Quasi-Biennial Oscillation ) which is a crucial High level wind pattern.... or sunspots  or SSTs    or    warm ENSO/ Cold ENSO   etc. etc.    WxRisk is focusing on  the SSTs in the El Nino regions. This analog search is premised on the idea that this winter will have Neutral  SSTs     OR   Neutral SSTs  that develop in the winter into    WEAK  warm SSTs in the region of ENSO 3.4

This table is taken from the CPC site and has been generalized for purposes of Clarity.  For a Detailed  view of the ENSO data  since 1950   click HERE.    From the table Below  there are  12 winters that match in one form or another the SSTs of the Summer /Fall of  2001  ( Neutral  ENSO)    and a predicted  Neutral ENSO for this Winter.    These   seasons are   Highlighted in PURPLE   and are:      1952- 53   1959 -60      1960- 61   1961-62   1962-63   1963 -64   1967-68    1978-79        1980-81   1981-82     1985-86   1996-97

4  OTHER POSSIBLE MATCHES IF A WEAK ENSO DEVELOPS (highlighted in Orange)    1951-52   1976-77   1977-78   1979-80

JFM AMJ JAS OND  

 

1950 C C C C
1951 C W- possibly depending on whether Weak ENSO shows up for OND
1952 winter  1952 / 53
1953 W- W-
1954 C- C
1955 C C- C- C+
1956 C C C C-
1957 W- W- W
1958 W+ W W- W-
1959 W- winter  1959 / 60
1960 winter  1960/  61
1961 winter  1961 / 62
1962 winter  1962 / 63
1963 W- W
1964 C- C
1965 C- W W+
1966 W W- W- not 1966-67 since weak ENSO summer 66
1967   winter  1967 / 68
1968 W-
1969 W W- W- W-
1970 W- C
1971 C C- C- C-
1972 W- W W+
1973 W C- C+
1974 C+ C C- C-
1975 C- C- C C+
1976 C W- possibly depending on whether Weak ENSO shows up for OND
1977 W- possibly depending on whether Weak ENSO shows up for OND
1978 W- winter  1978 / 79
1979 winter  1979 / 80
1980 W- possibly depending on whether Weak ENSO shows up for JFM
1981 winter  1981 / 82
1982 W- W W+
1983 W+ W C-
1984 C- C- C-
1985 C- C- winter  1985 / 86
1986 W- W
1987 W W W+ W
1988 W- C- C+
1989 C+ C- C- Winter  1989 / 90 Not a good match since the 88-89  La Nina was very strong
1990 W- W-
1991 W- W- W W
1992 W+ W+ W- W-
1993 W- W W W-
1994 W W
1995 W C-
1996 C- winter  1996 / 97
1997 W W+ W+
1998 W+ W C- C
1999 C+ C C- C
2000 C C- N C-
2001 C- N


In reviewing  these years    I noticed some interesting things   regarding   Snowfall amounts in the Central and Eastern US.  In Boston only 2 of 11  winters featured  Above Normal Snowfall while 6 of 11 analog seasons saw Below Normal Snowfall.   Albany NY and Hartford CT  saw 2  of 11 analog years   with Above Normal Snowfall... conversely Albany shas 3 seasons with  Below Normal Snowfall while Hartford  has 4 of the 11 seasons with Below Normal snowfall.

NYC, Philly both saw 4 of the 11 seasons with Above Normal Snowfall  while NYC saw 7 seasons with Below Normal Snowfall  while Philly  had 5.  Washington DC  saw 4 of the 11 analog years with above normal snowfall  and 3 below.  

To the south locations in the Lower Middle Atlantic saw equal number of  seasons with   above and below  Normal Snowfall....  of the 11 analog years Charlottesville had  5 above and 5 below... Roanoke  had 3 above and 5 below ... Richmond 4 above abnd 4 below... Raliegh 4 above and 4 Below. 

Out west   Detroit  had only 2 of 11 analog seasons with Above normal snowfall and 6 of 11 Below normal snowfall...   Nashville  had 4 Above and 5 Below...  Omaha  3 Above and 6 Below normal...   KC has 4 Above normal  and 5 Below normal  and   Chicago had 4 of the 11 seasons with Above normal snowfall and 2 below Normal snowfall.

. NORM 52-53 59-60 60-61 61-62 62-63 67-68 78-79 80-81 81-82 85-86 96-97
BOS 41 30 41 63 45 31 45 28 22 62 18 49
ALB 65 46 60 73 63 71 42 64 45 97 63 67
BDL 41 25 46 80 44 55 42 37 17 57 28 51
NYC 25 13 34 55 18 19 14 26 12 33 18 12
PHL 23 16 22 40 27 21 16 40 15 30 16 12
IAD 23 8DC 24DC 40DC 15DC 21DC 31 41 4 30 18 18
CHO 24 15 42 42 54 24 19 31 3 34 2 2
ROA 25 9 45 23 45 20 15 28 9 43 12 10
RIC 15 7 32 14 39 17 13 21 1 21 9 2
RDU 9 0 19 4 15 16 9 18 6 7 1 1
DTW 42 17 48 18 28 30 30 36 38 74 54 25
ORD 38 23 51 41 59 43 42 42 59 30 34 50
BNA 11 1 39 5 8 24 27 28 3 10 3 4
OMA 30 --- 57 20 52 40 13 26 9 24 18 23
KC 23 22 59 13 55 15 18 33 10 29 13 20
. . 52-53 59-60 60-61 61-62 62-63 67-68 78-79 80-81 81-82 85-86 96-97

   4  OTHER POSSIBLE MATCHES IF A WEAK ENSO DEVELOPS (highlighted in Orange)    1951-52   1976-77   1977-78   1979-80 

. NORM 51-52 76-77 77-78 79-80 CITY NORM 51-52 76-77 77-78 79-80
BOS 41 40 47 85 13 RIC 15 14 14 20 39
ALB 65 69 71 92 27 RDU 8 5 4 11 18
BDL 41 49 50 70 16 DTW 42 59 42 62 27
NYC 25 17 25 51 10 ORD 38 66 82 90 35
PHL 23 9 19 55 21 BNA 11 4 22 25 10
IAD 23 10DC 11 28 27 OMA 30 -- 26 29 20
CHO 24 7 13 34 46 KC 23 21 29 17 24
ROA 25 8 14 25 34

 


ne of the ways  that a Forecaster CAN use to help choose  which of the 17 selected Analog seasons may be a better "match"  is to consider the Summer/ Autumn  Hurricane Season.   To be fair  this sort of thing can only take you so far  because there is NO direct relationship between winters in Central and Eastern N. America and Hurricane seasons of the previous Summer/ Autumn.   But there is some insight to be detected....

Hurricane season 2001  has had the following characteristics:  track2001.gif (18245 bytes)

For those so inclined   one may wish to scan the historical record at the Unisys weather site for Hurricane tracks at this URL:  http://weather.unisys.com/hurricane/atlantic/index.html

For instance  in early October 2001... Hurricane IRIS became the 2001  season's only   category 4 hurricane... it intensified in the NW Caribbean  ( which is common enough)  then turned SW into Belize and crossed over into  the Eastern Pacific.   Only a few of the 17 seasons  listed above  --- 1961 1963    1964 1966 and 1996-- featured intense Hurricanes in the NW Caribbean in October   that did NOT directly affect the GOM...
1961 HATTIE  which reached 920 MB  on Oct 27
1963 FLORIA  which reached 940 MB Oct 13
1966 INEZ  which reached 929 MB   Sept 21 - Oct 12
1996 LILI  which reached  960 MB  in the NW Caribbean  that moved NE into the open 
Atlantic Oct 14 -29

IMO  of the 17 selected  Analog years...   the Hurricane seasons that seem closet to the 2001 in terms of what the mean 500 MB patterns  storm tracks  and overall seasonal development  are  the Hurricane seasons  of 1966  1967   1977  1978  1981.

Now with the appearance of a large SUBTROPICAL STORM near Bermuda.... again in the 17 Analog years  only 3 of them saw OCT Subtropical Lows.... 1979...   1981 in early November....  then in  1976  ( again that year keeps coming up)
in Middle September  and in late OCT 2000.

CLIMATE  & SEASONAL MODELS --  
This is a Meteorologically Oriented discussion

You've got to get yourself together
You got stuck in a moment and now you can't get out of it

Don't  say that later will be better 
now you're stuck in a moment
And you can't get out of it

In the Private sector of the Meteorology world Seasonal and Long range forecasters  are used by MANY  companies and organizations.  And in this aspect there is a huge Schism between  various Factors in the the Meteorological community  on the value of long range/ seasonal weather forecasts.   Most Weather Hobbyist/  NWT Meteorologists /  TV Meteorologist hold extended ( 10day plus) and Seasonal weather forecasts  as nearly useless and without value. They hod this opinion based on years of experience and what they were taught in undergrad and graduate school.  However  Forecasters in the   Grain,  Energy, and  Transportation  markets hold  that these forecasts have considerable Value and generate a HUGE amount of  $$$ and interest.

 

In explaining and talking with many meteorologists in various businesses and organizations I have been accustomed to the lack of knowledge  many have regarding the much improved seasonal and climate models. Typically when I ask or talk about the seasonal forecast   many forecasters   think I am referring to the CPC forecast. They are totally unaware of the AGCM   or the NASA models  or the Echam... OCN... CM3...  IRI... LIM .... climate models.

The AGCM  stands for   the Atmospheric General Circulation Model.  Not one model but a 20-member ensemble that uses  20 integration's with different observational initial conditions.  he AGCM is forced by predicted sea surface temperatures from the NCEP coupled ENSO forecast system. All anomalies are with respect to 1979-98 hindcast climatology.    Here is the DIRECT  Link for the  AGCM .

First  lets show how much improved the AGCM has been over the past few years as well as some . The Official  forecast  from CPC last autumn for the winter of 2000 -01  was essentially a climatological normal winter for most of the Plains  Midwest and northeast US.  Viewed from a   whole  ( that is December to March)   that is a correct forecast. However in the Monthly forecast CPC also went essentially with a Climo forecast whereas WxRisk did not. 

From July and August  2000 the AGCM started giving a strong consistent VERY COLD  signal  that the first half of the winter of 2000-01 was going to be   darn cold. Mighty cold.... Record cold. Recall that NOV & DEC 2000 were the two coldest NOV/DEC on record  for most of the nation.    The last time  ANY data was showing anything as cold as the AUG- SEPT - OCT AGCM  of 2000  was way back in late 1993.!

OCT 2000  AGCM SURFACE TEMPS OCT 2000  AGCM 200 MB  ANOMALIES VERIFICATION OF AGCM VS WINTER 2000- 2001

oct2000AGCM temps.gif (28999 bytes) The Oct 200 AGCM clearly was forecasting INTENSE cold early in the winter over the Upper Plains/ Midwest/  and somewhat less cold over the Northeast. BUT NOTE how well the model shows the cold pattern breaking down by FEB/ MARCH 2000 over eastern N. America... which is exactly what happened.

200mbagcm global oct2000.gif (29798 bytes)  The AGCM 200 MB forecast depicts a HUGE   block ( deep red) over Siberia into western Canada during NDJ and to a lessor degree DJF. But the Block is forecasted to be Gone By JFM ( 2nd half of winter) which results in a much milder JFM as seen on the OCT 200 Temps. Again this is extremely GOOD forecast and accurate 00-01agcmT.gif (13656 bytes)  The Upper map was the actual Verification Map used on evaluating the AGCM for DJF...    and the lower map is the NOV  AGCM forecast for the DJF 2000-2001. Again   a VERY good forecast!

 

The AGCM also appears to be doing a great Job during  Major events as well.... Here are the AGCM  from the strong  warm ENSO  (La Nina) . 

Looking at this Precipitation FORECAST map  9798agcm.gif (19767 bytes) the Upper LEFT is the ACTUAL cams  depiction of what precipitation during l DJF 1997/98 was like. The 1 month ... 3 months forecast were pretty good as was the 5 Month.   Likewise the AGCM for Temperatures 9798agcmT.gif (20143 bytes) for the DJF 1997/98 period was outstanding....  at 1 month... 3 months... 5 months.  This is rather  typical considering that climate models ALWAYS do better with temperatures than with Precipitation.

FOR THIS COMING WINTER   the AGCM has been again Remarkably consistent since July 2001.  If anything the depth and size of the cold areas have increased over Eastern N. America as the winter has approached. The AUG AGCM showed a  large area of  Much below normal temps  running from central Canada southeast to the Middle Atlantic coast of the US xstemp_us_currentAUG.gif (15757 bytes)... and note the warm temps over Baffin island  & Greenland -- a classic sign of a   Greenland Blocking Feature.  The 200 MB AUG AGCM xz200mb_currentAUG.gif (29318 bytes)  showed INTENSE blocking over Greenland  as we moved into the Heart of winter (DJF). Note the light blue shade over eastern US extending out into the NE Atlantic ocean... this factor combined with the Very cold temps strongly imply that the Mean trough position could be displaced well to the south.

This trend continued with the SEPT AGCM  ( not shown) and the brand new 6 OCT AGCM xstemp_us_currentOCT.gif (16254 bytes). The cold is widespread and noticeable but not extreme over the Eastern half of the nation... with increasing cold over the Great lakes and New England in the SECOND half of the winter !  Again NOTE the warmth over Greenland and Baffin Island-- classic Negative Phase NAO signature.  At 200 MB  xz200mb_currentOCT.gif (31240 bytes)the pattern remains the same  as we saw in AUG and SEP  just more of it.  The Dark Red over Greenland and the West coast as well as the   below normal heights over  the Eastern  US and into the NW Atlantic   is STRONGLY SUPPORTIVE OF A COLD WINTER OVER THE  EASTERN US. The  2nd half of the winter  sees the  Greenland Block   ( - NAO)  reaches Max intensity as does the Deep  Eastern US trough.   The  Precipitation map shows a very dry and cold winter over the Great Lakes and northern New England.xprecip_us_currentOCT.gif (15703 bytes)

 OTHER  CLIMATE  MODELS

Besides the AGCM there are several other climate and seasonal Models that are available for the Meteorologist's consideration.      Many of these can be found on one of 4  main climate sites on the Internet (  the 3 are CPC... ECPC CDC and IRI ) and they the ECMWF's climate model -- the Echam...  NCEPs' climate model... the CCM3... NASA' climate model... the IRI 's model  the  CCA    the  LIM  model just to name a few.

Last year for winter 2000-01   none of these other models were as close as the AGCM.  The CCM3 CMasst.djf01.500ano.gif (43721 bytes)model issued in NOV 2000 was way off. Note the HUGE ridge of High pressure over the EASTERN US for the DJF period.   The Echam ECasst.djf01.500ano.gif (29813 bytes)  ( ECMWF model) was much closer to reality with a deeper trough over the Eastern US  and more ridging off the West coast.

With regard to THE COMING WINTER... The Echam has been the model that is closest to the AGCMaugecdjf02.500ano.n_america.gif (29019 bytes). The AUG Echam  shows very Strong height anomalies ( blocking HIGH ) over Eastern Canada and Greenland with a fairly strong Ridge along the West coast.  NOTE the below normal heights along the SE coast in DJF.  Moreover the 2nd half of the winter is VERY cold and stormyJfm02.500anoEC.gif (29848 bytes). The AUG  CM3 model  shows a basic ordinary   PNA patternJfm02.500anoCM3.gif (33141 bytes).

The AUG NCEP djf02.500anoNCEP.gif (26045 bytes) climate model  has a weak pattern with minor ridging along the West Coast and normal heights over the Eastern third of the Nation.  The NASA model djf02.500anoNSP.gif (29266 bytes) has a completely Different solution  with a Huge ridge stretching from Coast to coast and Below normal heights over Canada. I disregarded it since its SSTs for the coming winter are Extremely warm ... which I have showed it 100% wrong.

The detailed IRI   model has some good signs as far as the new AGCM   is ; The SEPT  IRI  model shows  Drier than NormaliriDJF02_NAm_pcp.gif (23695 bytes)   conditions over the Southeast  and  Significantly Below Normal temps  over the Northeast quadrant. iriNDJ02_NAm_temp.gif (29498 bytes) The NCEP model for AUG  was still out to lunch while  the Echam continued its prediction of Below normal heights over the  Eastern US.  The later half of the winter is the NASA model is only slightly betterJFM.500anoNSP.gif (29041 bytes)

THE CDC  MODELS

The CDC site -- Climate diagnostic Center   has been running several climate models out to 12 months for about a year now. This is a GREAT site and its useful is showing the connection between getting the SSTs   right  and the 500 MB pattern right.   Any climate model that has a bad  SSTs depiction is EXTREMELY  unlikely  to get the overall 500 MB right and the overall weather pattern right

DEC  JAN  FEB

  MODEL   FORECASTED   SSTs  FOR
DEC JAN FEB
MODEL FORECASTED 
500 MB
PATTERN FOR  DEC JAN FEB
ANALYSIS
CCA  CLIMATE MODEL fcst.sst.CCA.lag2.gif (6626 bytes) 
Extreme cold SSTs in the  ENSO 1.2 ( by S.America) is highly UNLIKELY
fcst.z500.CCA.lag2.gif (18389 bytes)
Ridge off the West coast and over the SE coast large Polar Vortex over Canada
A stormy Pattern yes
but Mild KS to KY to VA/MD;
IRI  CLIMATE MODEL fcst.sst.IRI.lag2.gif (5764 bytes)
weak warm SSTs in ENSO 3.4   weak cool SSTs in Enso 1.2
fcst.z500.IRI.lag2.gif (15153 bytes)
Classic PNA pattern
seems quite reasonable
LIM CLIMATE MODEL fcst.sst.LIM.lag2.gif (5623 bytes)
HUGE large pool of Cold SSTs  across all of the ENSO  regions
fcst.z500.LIM.lag2.gif (14839 bytes)
Negative TNH pattern!   Mean trough over Upper Plains and Midwest
OH please!  Gimme a friggin break!   There is NOT going to be a severe cold La Nina this winter
NCEP  CLIMATE
MODEL
fcst.sst.NCE.lag2.gif (5638 bytes)
weak warm SSTs in ENSO 3.4   weak cool SSTs in Enso 1.2
fcst.z500.NCE.lag2.gif (13840 bytes)
Classic PNA pattern
seems quite reasonable
CONSENSUS fcst.sst.ENS.lag2.gif (5481 bytes)
Meaningless drivel since the  LIM & CCA models show massive differences from the IRI and NCEP
Using a safe consesnsus forecast of these 4 models could not be more useless! .

 

JAN FEB MARCH 2002

  MODEL   FORECASTED
  SSTs  FOR 
JAN FEB   MAR
MODEL FORECASTED
  500MB PATTERN  
JAN  FEB  MAR
ANALYSIS
CCA  CLIMATE MODEL fcst.sst.CCA.lag3.gif (6473 bytes)
same problem as in D-J- F
fcst.z500.CCA.lag3.gif (19086 bytes)
Ridge off the West coast and over the SE coast large Polar Vortex over Canada
very stormy and winte-ish
over northern 1/3. Quite
Mild over lower southern
half 
IRI  CLIMATE MODEL fcst.sst.IRI.lag3.gif (5644 bytes)
weak warm SSTs in ENSO 3.4   weak cool SSTs in Enso 1.2
fcst.z500.IRI.lag3.gif (16675 bytes)
Classic PNA pattern but Much stronger in  in this time period
IF the IRI is right
2nd half of winter is
going to be a world of hurt
LIM CLIMATE MODEL fcst.sst.LIM.lag3.gif (5542 bytes)
same problem as in
D-J- F
fcst.z500.LIM.lag3.gif (16528 bytes)
Ridge off the West coast and over the SE coast large Polar Vortex over Canada
very stormy & winte-ish over
northern 1/3. Quite Mild oveR
lower southern half 
NCEP  CLIMATE
MODEL
fcst.sst.NCE.lag3.gif (5659 bytes)
slowly building weak warm ENSO
     Classic PNA pattern but Much stronger in  in this time period  
CONSENSUS fcst.sst.ENS.lag3.gif (5396 bytes)
Meaningless drivel since the  LIM & CCA
models show massive differences from the IRI and NCEP
.

And if the night runs over
And if the forecast wont last
And if your way should falter
Along the stony pass
It's just a moment
This time will pass

WHAT OTHERS ARE SAYING (RESTRICTED)

   

THE  CPC  SEPTEMBER FORECAST

p03.gif (22618 bytes)p04.gif (22380 bytes)p05.gif (21624 bytes)t03.gif (24140 bytes)t04.gif (24380 bytes)t05.gif (25309 bytes)

UC40TN.GIF (1894 bytes)