Kirk Mellish's blog has moved. Update your bookmarks to this link .

I will use this page to do periodic updates. I will cover insights, and long-range outlooks and misc. things I don't have time on-air to talk about or which are too complex for radio.

FEBRUARY 4TH

Last month was for the nation as a whole, the coldest and snowiest January in 5 years. Next week, nationally speaking, looks to be the warmest pre-Valentines Day week in 10 years. Here at home, using only Hartsfield data December was 3.2 degrees warmer than normal with 15% more precip than normal. January was 1.2 degrees warmer than normal and 43% drier than normal. Still looks like the long-advertised delayed "January Thaw" will give us a little Spring tease by the weekend and next week.

-------------------------------------------------------------------------------------------------------

JAN. 31ST: Its 11:50 pm Saturday, do you know where your weather forecaster is? At the ole update desk typing this for you. Baring any last minute, and always possible, major changes in future computer guidance, looks like nothing but drizzle and a flurry to worry about around here and then a brief intense cold snap. The models had been in agreement on a blizzard for the TN and Ohio Valleys into the NE states early next week. But since then the various equations have shown the 3 jet streams will not phase into a monster after all, or wont do so until New England states and SE Canada. If this is indeed the case, I will remember it as one of the more dramatic busts of a consensus forecast I can remember in over a decade. Not that the models ever agreed on the details, but on the big picture, they were all screaming the same message about a possible historic storm starting just North of Georgia. The single wave Gulf low is now projected further South and weaker with less moisture, and is projected to not strengthen or re-develop until somewhere around the VA Capes. Polar air will send the freeze line back down into FLA by the 4th behind the departing system. Meanwhile the ENSO El Nino Southern Oscillation after a brief "hiccup" toward El Nino is back in La Ninas arms again in weak status. You will recall that my Oct and Dec outlook for the coming winter allowed for a neutral to weak La Nina expectation. Based on MEI measures the weak La Nina is ranked closest to the 1974 analog year, one which saw La Nina last well into 1976. You can web search the government weather services views of what they feel this means to the weather in the months ahead. Of course, other factors will play a role, including the SSW event. The stratospheric warming usually is a pre-cursor signal to colder weather East of the Rockies with about a 3 week lag. So after the brief cold wave, we get a delayed January thaw-- in Feb.-- before colder readings return, combined with the La Nina and MJO, the nation probably faces an interesting Mid-Feb to early March weather wise.

-------------------------------------------------------------------------------------------------------

JAN. 30th: At mid-morning I see a forecasters worst nightmare. Instead of the numerical weather prediction models trending toward one solution, the latest model guidance is diverging away from each other and even away from their previous output. They run the gamut on what to expect here and even up in the Northern states where there HAD long been consensus on a blizzard with 50-70% the power or past storms like 93, 50, 96 etc. BUT now even that has vanished. This is why I don't change my forecast with every model run, or make foolish predictions like 5 inches for Atlanta when the projected system is still days away. Model mayhem will continue. If you've been reading along since Sunday you know this was to be expected. It's just too soon to make a hard and fast call, as ANY call made now (dry, rain, heavy snow, 3 flakes, sunny) would have to be adjusted as we get closer to the event, if there is to be one at all! Taken at face value the trend in at least the GFS American model would point to a lot of nothing beyond a brief cold wave. And since this model and all the others keep changing every day, it is silly to panic and too soon to just write it off. That's why I look at data almost non-stop all day long, and forecast every day--not just once a week.

-------------------------------------------------------------------------------------------------------

JAN. 29TH: Not a lot new to add this evening most of below still applies. The expected storm, at its height, encompassing 1-2 million square miles will probably be about 60% as disruptive as the March 93 storm (FOR THE NATION NOT GEORGIA) with OH, PA, West VA, NY and adjacent areas being hardest hit by a foot or more and blizzard conditions, while severe weather will threaten near the Gulf Coast to FLA and the Eastern Carolinas and maybe SE GA. A snowflake could be seen in New Orleans and the Florida Pan Handle before all is said and done. It does not look like a lot of wintry mix or freezing rain with this system, more of a sharp rain or snow divide or rain changing over to or ending as snow. Not the best scenario for significant snow if you are in Atlanta. But the storm track will be more critical than usual, and its always critical. So how this plays out remains to be seen. Snow will probably break out in AR, West TN/KY Sunday night as the low forms South of TX in the Gulf then to near the mouth of the MS River onto South of Mobile before curving NE up the Fall Line/Piedmont then North to near DC to SE PA. The storm looks to explode once it is NE Past GA. But none of this is written in stone as I wrote below. I am sure model madness will continue which is why its pointless to try to pin point snow locations or amounts except where we think the storm will mature, but that's North of us as of now. So it is WAY too soon to panic, and way too soon to completely write it off.

----------------------------------------------------------------------------------------------------------------------------------

JAN 28/29: The potential snowstorm I've been blogging about since Sunday that some models initially showed for Friday has now been moved to Monday or Tuesday. All 3 Jet streams combine (jet coupling and ageostrophic adjustments) for a major planetary wave amplification over the Eastern USA next week could mean storm with 70% of the intensity of the Blizzard of 93-- DO NOT read that as Kirk is forecasting a blizzard in Atlanta that's not what I wrote. Because the projected system (does not exist in the real world right now!) also is similar to other HISTORIC Nor Easters like storms in 2000, 2006, 1978, 1996, and 1950 etc. that did NOT slam Atlanta. PRELIMINARY signs point to AR, MS, AL, TN, KY, NC, SC, VA with some ice for GA Mountains and a dusting or less for Atlanta but this could change. One thing worth noting is that there is now consensus that some "system" will develop over the weekend or Monday and it will probably be a monster for somebody, as it tracks somewhere across the South and East USA early next week. Virtually every model out there shows some similar low pressure and frontal system but of course track, timing, moisture and intensity all differ. The general idea is that a Gulf of Mexico low forms and track East near the Gulf Coast then turns up the East Coast as it cranks up while at the same time a new arctic front advances toward the developing system as a new upper level long wave trough forms East of the MS River. The low tracks from South TX to South of New Orleans to Mobile-Tallahassee then EITHER Jax FL North or NE inland over GA to New England. There is some energy shown on Sat imagery in the Southern Pacific trailed by a kicker system North of Hawaii plus vorticity impulses in the Northern Jet Stream so there is support in the obs data for something brewing. But keep in mind NO low exists right now. Its all a figment of the computers imagination if you will, a result of the equations. And at 5-6 days a way a lot can change. Sure this COULD be a big snow for somebody in Dixie and/or the East Coast, but its way-way-way too soon to make a realistic estimate as to where. I will say that the general pattern PREDICTED by the various algorithms is favorable in the general sense. BUT it is NOT yet the classic Atlanta snowstorm pattern as the thermal profile is far from ideal, the track uncertain and not ideal, and we don't yet know if this will be a so-called "Miller A" or "Miller B" storm, a suppressed "fish" storm, or a "Lakes Cutter" or "Appalachian runner". IF there is a storm at all. I think there WILL be a low, given all the model and data support, and the model run to run persistence and consensus for a couple days in a row now and multiple model runs. BUT the devil is in the detail and there is NO consensus on that at this point, and at least FOR NOW, I lean toward mostly rain, with at least some snow flurries on the back end as the polar plunge comes into the SE once again. This is because I favor more of an inland route up the Piedmont for 3 reasons. Cold dense air is not in place ahead of the storm but would have to move in just in time-this only rarely occurs. Model error this season has been too far SE with storm tracks so I adjust NW. Thirdly as the split flow jet energy phases into one long-wave trof, the system goes from neutral to negative tilt, so it gets drawn North toward the dropping Arctic air core and arctic short-wave favoring more of the climatological Chattanooga express path. Of course, I could be wrong and it wont phase, or will phase later further East and North, or "bomb" out (bombogenesis) with enough dynamic cooling, lift and frontogentic forcing and backlash to be a biggie here. It could even never phase and just stay weak with little action for anyone as it stays a flat wave and heads out into the Atlantic. Regardless, most models favor a discharge of cAK air from Canada to the Keys between Feb 1-4. As is always the case, in the days ahead, the models will do their typical cha-cha-cha of giving and taking away and swapping roles etc. The weatherman says somethings on the move, but I am not willing to howl at the moon at this juncture. The situation bears watching, and I will, as I have since I first saw the snow threat last Sunday afternoon.

----------------------------------------------------------------------------------------------------------------------------------

JAN 27th: SSW event continues, could be of historic magnitude. Meanwhile the 12Z Euro and American GFS both hint at a Gulf Low snow chance early next week. However, not well supported by the ensemble means. But I'll keep an eye on it and update if it starts looking real.

----------------------------------------------------------------------------------------------------------------------------------

JAN 25th: An extended period of unsettled weather with more frequent fronts and/or storms looks to be setting up for the next 2-3 weeks with see-saw temps above and below normal alternating. There have been hints from some of the numerical equations of a snow scare this Friday. One model showed an inch or more of snow across the Mountains and Northern Suburbs but 12 hours later the GFS is dry. Clearly this transition pattern is unstable and so we have model mayhem and I suspect both the actual weather and weather forecasts will be rather changeable during this period. There is also a "Stratospheric Warming" episode underway at 10-30 mb and research has shown this often leads to an eventual discharge of arctic air into the Lower 48. Models also project a -AO/-NAO pattern heading into early next month. This combination suggests the potential for a cold wave for the middle and/or end of February. I have not heard this elsewhere on the radio and not seen any of this mentioned on any TV station.

----------------------------------------------------------------------------------------------------------------------------------

JAN 21ST: A return to normalcy. At least largely uneventful weather with no extremes in sight for now. The worst of winter is probably behind us now, but not guaranteed. A normal or seasonal high is now 52 and a low of 33. The terms normal or seasonal are meteorological terms that have been around since at least the 1970s when I first started studying weather. The terms should therefore not be confused with standard English or mathematics. As such, "normal" does not equate to a numerical mean or average and the standard deviations of statistics, but instead references a standard 30 year norm. In many years Atlanta will never see a day where the high never breaks 32 or where the minimum is in the single digits or teens. Likewise since the normal high is 52 and 33 if the forecast is a high much below this it can be said to be unseasonably cold and if its much above this unseasonably warm. This applies to all seasons based on the current meteorological normal. If you were reading my blogs over the past week you see the only place ever forecast to get significant snow was well North of most of Metro Atlanta and that worked out right. I did get to see snow showers and snow flurries (in Kennesaw) on two seperate days out of 4 which is sure rare here and I love to see it. Last year Atlanta had 1.3 inches of snow on Jan 19th. By the way, we are now PAST the dead of winter and the normal high and low now go up from here into the heart of Summer as we start gaining daylight with longer days going forward.

----------------------------------------------------------------------------------------------------------------------------------

JAN. 19TH: Now some of the models show a surface low forming on the front I talked about yesterday, the Euro model shows explosive cyclogenesis (bombogenesis) as a strong vorticity max reaches the bottom of the upper level troff (sic) over GA/Carolinas. There could be snow flurries even in Northern Florida with a freeze for the citrus groves. Highest threat for significant accumulation along and North of a Rome to Athens to Augusta line. This is a close call for ATL as to getting more than a dusting. Because of the model mayhem and the "developing overhead" scenario- as opposed to having a system we can track moving into our area- this means it could be an hour to hour minute to minute weather situation with more NOWcasting than FOREcasting. That said, this still does NOT look like the classic set up for significant ATL snow. So if we end up with a couple inches it would be a first of its kind that I know about, which is why I currently expect only 0.4 of an inch OR LESS. But keep your ear on the radio.

JAN. 18TH: Even though it didn't last its always fun to see snow flakes in Atlanta (or anywhere) makes the kid in me smile. Well, the computer model wars continue with the GFS (American Global Forecast System) model and the SREF showing flakes possible Monday and/or Tuesday as far south as the FLA state line and Valdosta GA with minor accumulation as far south as Columbus. Not my forecast yet as newer runs of the GFS have backed off, and the NAM and ECMWF models are drier. This is NOT a snow storm system, no surface low. I've never seen a significant snow in Atlanta without a surface storm low. What is projected by the GFS is a strong short-wave disturbance in the jet stream comes down from the North into the base of the long-wave troff bringing enough lift and moisture with another shot of polar air. So as always, its an ongoing process of evaluation of a never ending stream of observed (actual data) and prognosticated (figment of numerical equations simulated future state of the atmosphere) computer guidance to reach a conclusion to forecast which may be on target or way off base or somewhere in between. And as if that were not enough, the ECMWF and the GFS switch roles for a possible winter event around the 24th give or take a couple days. Oh my, I have a headache already. This job sucks ;) I'll have one eye on the weather and another on football championship games today. Stay tuned.

JAN 16TH: Extreme weather patterns sometimes beget more extreme weather, and numerical simulations of the atmosphere perform more poorly the more the weather deviates from the norm or long-term average. Hence, 5-10 day forecasts are prone to more flip-flopping than usual and 1-3 day forecasts can suddenly change with little advanced notice. Keep this in mind now and for the future. In recent days the American model has been trending more moist with the next cold front and upper-air disturbance Sunday raising a red flag for a bigger threat of some freezing rain or sleet changing to just light rain preceded and followed by snow flurries. Also, the trends in a usually reliable model- the European- has been more ominous for snow Monday/Tuesday. This means I have myself more of a working Friday and a working weekend ahead than normal. I hope I have time for football Sunday. There are 3 dozen computer models to evaluate and thousands of parameters to calculate in making a forecast. The data comes in continuously because the weather is in constant flux. Therefore weather forecasting is an ongoing process, not a static snap shot in time that fixes in place for history one day at a time or even one forecast at a time. My forecast this hour may be different in one hour or 3 hours or tomorrow. At least some part or parts of it. There is also not, not just ONE forecast. The "they are saying..." I hear that all the time and it's invalid, because there is no THEY, and "they" do not all have the same forecast TV/Radio/National Weather Service/computer internet/hand-held device. "They" can all be different. And your I-Phone, blackberry, or computer forecasts are automated robots with no meteorologist involved. And of course 95% of your TV/radio forecasts are delivered by non-meteorologists even if the person gets called one. So beware. Yesterday I monitored one popular .com weather site and it changed 6 times between 5 and 9 am! I adjusted my forecast just once and it was a change from a high of 39 to a high of 36. So don't waste your time with those "other" forecasts they are free and easy for a reason. Stick with me, I am not always right, but at least I am a degreed and certified meteorologist who will try to give you the earliest warning about POSSIBILITIES and THREATS but WITHOUT trying to ALARM OR HYPE or cry wolf. Sorry for the long diatribe, didn't mean to go off on a tangent. I only meant to say the computer model guidance is shifting, so STAY TUNED in case I have to make forecast changes of the wintry variety in the hours and days ahead.

JAN. 13TH: Coldest temps in 6 years on the way (based on Hartsfield only) get ready now, winterize car: battery-anti-freeze-windshield fluid not just water, ready furnace, turn-off water to the outside, wrap exposed pipes, shut-down sprinkler systems, check well-water systems and emergency generators, let faucets drip in sinks with outside wall exposure, find the winter coat and dress in layers, extra socks, shirts, hat and gloves, scarves. Your fingers, toes, ears and nose are most susceptible to frostbite so cover them. Coaches and athletic directors should give 10 minute indoor warming periods every 30 minutes for athletes training or practicing outside if not canceling outdoor activities. Make sure space-heaters are working properly and beware the danger of deadly fire and deadly fumes from some of these. Check on the wellness of the elderly and infirm. Shelters should be available for the homeless, gas and electric shut-offs should be suspended. Extra busses and trains should run to reduce exposure wait times. Any outside water smaller than a lake will freeze, fish ponds-waterfalls-birdbaths. Let your pets stay inside. Put out food and water for birds and other wildlife, redo water when it freezes. From late Thursday to late Saturday on the order of 60 hours below freezing that's a pipe-bursting cold spell which will include single digits in the Mountains and maybe some Atlanta suburbs with teens common elsewhere, wind chill factors around zero before the cold wave eases. The modern era record low in Atlanta is 8 below zero, minus 8 in January 1985. As for snow or ice nothing substantial in the offing any time soon. There have been hints from some of the numerical algorithms of something Sunday/Monday and again late this month but too inconsistent to get excited about for now. The next 5 days looks like the troff axis is too far East so I'll keep it out of my forecast. I'll keep an eye on it for any surprises.

JAN 11: The nations ice box door is swinging open and Chicago and New York City face the coldest temps in about 13 years. For Atlanta we should easily see a day or 2 where temps struggle to get out of the 30s during the day even with sun with lows in the teens and 20s AT BEST and the wind chill factor on one or more days will be brutal by Southern standards, as the frost line goes down to North-Central Florida threatening citrus groves. The next 15 days average colder and drier than normal on average. It appears the upper level wind flow pattern is NOT highly favorable for any significant snow or ice as the troff is neutral to negatively tilted instead of giving us a split flow pattern which is more favorable for snow or ice. Still, the various equations are known to not handle shallow arctic air masses well, so things could change, therefore I wont yet change the alert level but it remains only at the lowest level on a 3-1 scale. NOAA has declared LA NINA is back, other such late blooming La Nina years were 1945, 46, 68, and 06. In those years more cold weather struck in February after the Jan thaw, then March was warmer than normal. Precipitation was near-normal to a little less than normal.

JAN 8TH: SET CONDITION SQ (storm quotient) 3 ALERT LEVEL 3 pattern is FAVORABLE but NOT probable... there is NO disagreement in the dozens of computer models that a major pattern change to dramatically colder is on the way for the next 10-15 days and more than one model Euro, Canadian, American hint at a couple of chances for ice or snow in Atlanta. Too early for specifics but one threat comes Jan 14th give or take a couple days, and the other Jan 19th give or take a few days. The UPPER LEVEL winds aloft flow pattern of the Jet Stream is forecast to become positive for winter weather. In the South it is always best to have cold air established and firmly in place in order to get ice or snow. The cases where we need the moisture and cold temps to come together just in time almost never work out. So step one is getting a polar front to move through Dixie and stall on the Gulf Coast or in the Gulf and off the East Coast. Then the upper level troff is positively tilted over the Center of the country leaning back toward CA and AZ to the MS River and the energy aloft comes out of that and cyclogenesis occurs along the baroclinic zone (front) in TX or the Gulf and it sends overrunning precip over the cold dome to produce winter precip. This is being shown by many models. But keep in mind with no cold air in place yet and NO storm in existence this is still in the "pattern recognition" stage of synoptic meteorology hence no specifics can be given. It looks like for at least parts of the nation, if not here, the coldest air mass in a decade to 15 years.

JAN 6TH: Update on the Jan 1 entry...major signals for a big pattern change continue, major stratospheric warming event, MJO active Kelvin wave and split flow pattern in the Pacific Kona Low and Gulf of Alaska low, +PNA, projection of Hudson Bay vortex displaced Southward all point to the Polar Plunge Arctic express pattern with air mass drainage from Canada to Mexico and Cuba which could spell record low temps in much of Dixie with the change starting slowly the next 3-5 days and accelerating Jan 13-19th give or take a few days. No clear cut snow or ice storms show up yet, but this is a pattern that sets the table for such a thing so stay tuned, and remember where you heard (or read) it all first.

JAN 1ST: Happy New year. Interesting weather and challenging forecasting days ahead. Risk of Severe weather, heavy rain and some winter precip in Georgia with below-normal temps on average between now and Mid-January. More weather news headlines in the making for much of the nation West Coast to Rockies short-term Rockies East Medium and long-term. Big cold wave possible with frost to Northern FLA possible with a window of opportunity for sleet or freezing rain in Dixie between Jan 7 and 15, then Jan thaw by the 17th or so, but possible renewed chill after the 22nd to end the month. All the major models agree a deep jet stream troff forms over the Eastern USA with a ridge out West draining air mass from the Yukon Territory. Interesting stat note: in August of 1983 there was record heat in the Southeast U.S. followed 18 months later by record cold in Jan 1985. We had record 100 degree heat in August of 2007, if history repeats that makes this month about 18 months later and maybe the coming cold snap will include some records tied or broken.

DEC 17TH: Atlanta temps have been colder than Normal since October until now. WEATHER OUTLOOK FOR GEORGIA next 30 days average with temps near-normal and precip normal to above-normal. The next 2-3 weeks will feature a battle between warm spells and cold waves. A couple computer models show that Christmas Day ONLY 6 states will have NO snow on the ground or in the air LA, MS, AL, GA, FL, and SC. IF current trends continue this month will end with GA temps having averaged below-normal. But Jan-March would turn warmer than normal with precip. near-average to below-normal. There are preliminary signs of a new LA Nina forming in the Pacific, IF SO the rest of this winter would be on a par with last winter.

DEC 11th: SNOW in San Antonio and Houston TX first time since 1944, snow in Lake Charles LA, Lafayette LA and New Orleans! Baton Rouge LA THUNDERSNOW 3 inches. While nothing like this is expected here in Atlanta as of now, whenever the weather is doing crazy things under a closed low you never let your guard down. I can not RULE OUT a dusting to 2 inches in parts of North GA tonight including Atlanta. Stay tuned in case I have to break-in on the radio.

DEC 9TH: Stay tuned, complex weather pattern ahead, and the various numerical equations are not handling it well. For example, one model showed 1-3 inches of snow for our NW suburbs Thursday, then just 6 hours later the new run showed none. One day it had highs Thursday only in the 40s now it has a high in the low 60s. The other models are swinging back and forth also. Either way, after a badly needed soaking rain next week looks drier and much warmer, the following week temps could go above normal ironically in time for the official start of winter. Though for meteorologists winter starts Dec 1st. So we will keep a sharp weather eye on Thursday and Friday for any last minute surprises.

DEC 6TH: My thanks to ATLANTA MAGAZINE, and its readers for being voted BEST OF Atlanta in Weather Forecasting! A big thank you.

DECEMBER 1ST: MY OFFICIAL WINTER FORECAST FOR GEORGIA DEC 1 - FEB 28:... Changeability, volatility and variability are the key watchwords for this winter. Due to the neutral ENSO state and contradictory signals confidence in this forecast is lower than normal. I think temps and precip., including snowfall, will be plus or minus 10% of average. Not good for the drought. I still see a "front-loaded, bookends" winter--coldest at the start, thaw in the middle, but a chill returns in early spring. A day of severe thunderstorms possible in Jan or Feb. The odds of snow look about normal. Chances for a major snowstorm however, look less than average. Odds of a significant ice storm on the other hand look greater than average. Key analog years are 1950, 55, 59, 66, 74, 85, 89 and 2000. There are a host of secondary analogs. Here are the National Maps of the outlook.

NOV. 30TH:Snow flurries are possible even in South Metro Atlanta Monday but no impact on roads if we do see some. Best chance I-20 North, best chance for accumulation on grassy areas etc North of a Rome to Toccoa Line, an inch or 2 possible in the higher elevations of the NE Ga Mountains. It will stay colder than normal for the coming week. Beyond that, more storm threats and see-saw temps.

NOV 28TH:...Some computer model winter forecasts: CFC temps near-normal and drier than normal. CAS warmer than normal precip normal to slight dry. UKMET cold and dry. ECMWF warm and dry. IRI warmer and drier than normal. HURRICANE season ends Sunday it was an active season with 16 named storms 4th most in history, 8 hurricanes 5th most in history, 5 major hurricanes, and for the FIRST TIME IN HISTORY 6 storms hit the USA. An inch or 2 snow possible in the NE GA Mountains Sun or Monday, flurries not out of the question in ATL Monday.

NOV 27th:...I guess all the other stations are jumping on my old man winter bandwagon now. Welcome aboard! There is now computer model consensus on the cold wave, and that it will eventually be cold enough for snow. But there is no consensus on IF it will snow, or if so, how much would fall. I do not currently see us getting a snow storm, but I would not be shocked by at least snow flurries or a snow shower somewhere in the late Sunday-Tuesday time frame even in Atlanta.

NOV 21ST:...One problem with most of the free point and click forecasts you can get on the internet on your computer or telephone or other handheld device is they are automated roboforecasts not the forecast of an actual meteorologist who can adjust the raw computer output, that is why phantom snow pops up in the 7-10 day forecast from now through winter just to disappear hours later or the next day then it shows up again. Dont waste your time, these free weather sites are worth every penny. Still looks like much below normal temps for GA Dec 4-Dec 10, cold enough to keep an eye out for any moisture that may come along. Stay tuned.

NOV 21ST: COLD WAVE WINTER ALERT DELTA FOXTROT TANGO THIS IS NOT A DRILL... POTENTIAL for some of the coldest early December weather in at least 5-8 years if not 20 years!! One model Dec 3rd ATL high of just 34 and a dusting of snow. NOT A LOCK YET BUT NO sustainable warmth in sight the next 3 weeks even if it doesn't go to the extreme. Potential for wintry weather in parts of the Southern USA growing by Dec 5th. At least more models and actual measured data are showing a favorable PATTERN in the atmosphere that is January-like and could include a storm system. This is your 3rd exclusive alert on this at this site from WSB-AM as I've been dropping hints on the radio and in my blog, so stay tuned for more.

NOV 20th: Have not been posting much as I had a family emergency and I am working on the winter forecast. Meanwhile looks like the post from Nov 8 sure worked out well. It still looks like the average of the next 30 days is for below-norm temps but it looks like something is brewing and the weatherman says somethings on the move. Lyrics from the music group Jethro Tull and Storm Watch album. Looks like the last week of this month and first week of next month the weather gets interesting, see-saw temps and storm systems. A risk of severe weather and a flake. Final winter forecast on the radio early December.

NOV 8TH: Stormier more active weather pattern across the country the rest of this month will spell a step-down temp pattern to much colder for Georgia between now and Thanksgiving week, signs of a cold December also. November is one of our mini-severe weather months along with Jan/Feb that fall outside the primary spring severe weather season.

NOV 6TH: Hurricane season way above the normal average of 10, Tropical Storm Paloma forms East of Honduras heading toward Cayman Islands and Cuba expected to reach at least CAT II status. 30-90 DAY OUTLOOK FOR GEORGIA temps and precip near-average.

OCT 31ST: Indian Summer for Atlanta. Interesting weather signs around the world. 4th coldest Oct on Record in Fairbanks AK, early snow in Alaska, coldest start to winter in Fairbanks in 16 years. Record lows in Florida, rare snow in South America, historic snow in Northern Spain, first October snow in London in 70 years!

OCT 28TH: MY PRELIMINARY WINTER OUTLOOK FOR ATLANTA... early indications point to a "front-loaded, bookends winter"... cold start, cold end warm in the middle...coldest relative to normal the first part of winter rather than the second part, a delayed Spring possible, odds of a solidly warm winter look small. Taken as a whole Dec.-Feb look more or less normal or near-average with a slight tilt toward slightly cool and slightly dry, with snowfall half an inch below average. Beware "near-normal" as nature sometimes tries to regress to the mean by way of wild swings which increases the chance of extremes along the way for volatility this winter. By the numbers... TEMPS -0.17 degrees F. PRECIP. -0.46 inches or around 95% of normal. Snowfall 1.5 inches. INPUT: ENSO neutral, sunspot minimum, negative PDO, positive AMO, neutral NAO, negative PNA, weak La Nina possible, second year Nina, active Gulf-centered hurricane season, negative QBO, Pacific and Atlantic ocean temp patterns, ENSO indices, Summer weather pattern, early season global snow cover, October temp pattern in USA, global atmospheric angular momentum, stratospheric trends above North Pole, Southern Hemisphere winter behavior (our Summer). These yield a series of analog years from the past that were similar and assumes the ensuing weather this winter will progress in a like manner as the average of those past years. I should point out that while I was almost dead on in my prediction last winter, I doubt I will be this time. Because this year there are contradictory indications, and at least as of now, there is NO strong signal of what is to come for the low sun season. IF that changes in November, for example if a LA NINA or EL NINO starts or appears more likely this will alter my outlook. But for now my preliminary winter outlook has LOW CONFIDENCE. My final official winter forecast will be issued on the radio in early December.

OCT 26TH: A real accordion temp pattern between now and Nov 8th including first freeze for some areas Wed./Thur. then moderating temps in early November before the next cold snap. Next 10 days average dry and below normal in temp then mostly dry but with above normal temps the following 5 days. A secondary Southern Jet Stream storm track is indicated to form by model consensus for the early weeks of November and this could lead to a spell of above-normal rainfall in about 15 days or so. Global trends continue to point to a stormy winter in the USA. Hear my FIRST LOOK at winter in Atlanta this Tuesday Oct 28th at 6:20 and 7:20 A.M. ONLY on News Talk 750 WSB.

OCT 19TH: The quiet weather will not last-- more active weather pattern, more wet and more cold maybe including strong thunderstorms followed by frost and freeze expected in the period between October 23rd and November 4th. Listen for my first look at THE COMING WINTER IN GEORGIA on the radio on Atlantas Morning News 5a-8:30a Tuesday Oct 28th.

OCT 10TH: Now that Summer is over I can grade my Summer forecast which called for a near-normal Summer, and it might be the best Summer forecast I've issued since I started doing them on WSB, a B+/A minus. June-August temps were just +0.7F warmer than normal and rainfall was only 14% below normal. Much less extreme than the previous Summer. JUNE was +3F warmer than normal and 84% drier than average. JULY was -0.2F cooler than normal and 40% wetter than normal. AUGUST was -0.7 cooler than normal and +2% wetter than normal. While the consensus of everyone else was we were in for a replay of last years brutal hot and dry, I said it would be much closer to normal and not even in the same ballpark as last Summer. And that is exactly the way it played out with one wet and cool month, one slightly wet/cool month, and only one hot and dry month, and NONE of those weeks of tripple digit temps. Now, back to the future MEDIUM RANGE FORECAST: The next 15 days looks like temps and rainfall will average out near-normal with a warm spell and weak cold snap in the mix. First big snowstorm of the season for the Rockies this weekend. Caribbean and/or Gulf tropical development possible next week.

SEPT. 22nd: Tropical development or a Nor'Easter possible later this week with possible hit by weekend or so anywhere from SC Northward. The 60 DAY OUTLOOK for Georgia calls for temps to average near-normal with rainfall above-normal.

Sept 19th: Tropics should perk up again by early October if not before. A "Quick look" at THE COMING WINTER IN GEORGIA...one degree cooler than normal with precipitation 75-95% of normal (average of Dec. Jan. Feb.) Current analog years are 1976, 1985, 1989, 1996, 2000 and 2005. I'll have updates as research continues. I currently expect an ENSO neutral to weak El Nino winter. A preliminary winter outlook will be out early November. My official winter forecast usually comes out in early December.

SEPT. 5: Updates on tropical weather available on-air. I am going on vacation, something planned long ago. The station requires that I give them all my vacation weeks for the year ahead-- in January. Plus, many months are blacked out for broadcast reasons, meaning I cant take a vaction during those times. So I am forced to take most of my vacation in concentrated doses in just a few months of the year, not my choice. Thanks for understanding the way radio works.

HANNA...expected to reach land around or North of Wilmington NC Saturday as a strong TS or weak HUR (no-skill intensity forecast), but its a large storm so the affects will still at least brush the GA Coast with tropical storm force winds as far inland as Raleigh and Richmond on the way to New Jersey. IKE...heads into the Bahamas Sunday-Tuesday, after that all points are on the table. TPC shows it SE of Miami 2pm Tue with 126 MPH winds (no skill intensity forecast) large track errors exist at 5 days. I favor an eventual path somewhere up the East Coast on or off shore crawling from South FL to the Carolinas next week, but can't rule out entry into the Gulf. I think IKE will weaken before or as it reaches the USA, assuming it ever does. JOESEPHINE too far away to worry about.

Sept 3rd: I first warned August 23rd that there would be a hurricane threat to the SE or Gulf by the end of the month at the latest and maybe 2 threats. Statistically Savannah is long overdue for a big direct hit. But direct hits from Jacksonville FL to Savannah are rare. There have only been 3 direct hits on Savannah in the last 107 years! The last bad one was 1940 and before that in 1824 and 1898. A weak storm in 1947 was seeded with dry ice by a B-17 as part of a weather modification program called project cirrus. The last storm to hit in the Greater Savannah area was a weak David in 1979. Hugo brushed the area in 1989. So as I've been saying for several days now, based on history and climatology I expect HANNA would hit either FL or more likely, closer to NC than Georgia. With IKE and Josephine and maybe Kyle out there the expected MJO burst is active. Hurricane season peaks September 10th and is cut in half by October 15th. After we get through the current string of tropical systems, things should calm down until next month. ADVISORY UPDATE ON HANNA: I would advise everyone to calm down and keep your powder dry on Hanna for now, don't be complacent but neither panic nor hype are called for (hype never is, but reporters and anchors often feed it), Hanna is a weak and struggling for her life entity. The steering mechanisms (multiple) are so complex, that the intensity forecast is lower confidence than normal, which is low to begin with, and track confidence is average until it establishes definitive movement. But there is now considerable model consensus on a track somewhere around Wilmington-Cape Fear NC Saturday. The NO-SKILL intensity forecast is a minimal hurricane but that could could change. I think we should wait to see if Hanna can bulk up and decide a definite direction of movement before getting worked up about the current outlook. The models keep shifting the track farther to the North following what I've been saying all along. Minimal impact on GA unless something big changes. I also would NOT write off Hanna either.

SEPT 1 8pm: HURRICANE HANNA.... The little to no skill intensity models show it becoming an 80-105 MPH hurricane by Thursday Night. As for a HANNA track, the consensus forecast is toward South of Savannah GA sometime Friday as a CAT 1 or 2 98 MPH, but this would be a vary rare path. Charleston or even farther North is the way I lean for now. The track spread in the Global equations is high and all areas from Daytona Beach FL (if not all of FLA) to Cape Hatteras NC are at risk. POTENTIAL HURRICANE IKE...the Canadian and European models suggest this could be a threat to the Caribbean in 6 days and the Gulf in 10 days. JOSEPHINE also looks to form and another possible behind that. I will not concentrate on IKE or the other systems for now and mostly update on Hanna.

August 31st: 8pm GUSTAV: outer rain bands and thunderstorms hitting the Gulf Coast now, tropical storm conditions to increase overnight. New Orleans looks to escape the worst, the worst will impact a relatively small area of South Central LA. The computer model track guidance is in good agreement past 48 hours with a landfall South of Houma in Terrebone, Lafourche and St. Mary Parishes BUT THEN they spread out indicating slower forward motion and a variety of directions, including back SW or a fall apart and stall scenario over LA and TX becoming a remnant low in 120 hours with flooding Tuesday and Wednesday., BUT confidence is moderate to high for a Central LA strike. We should not be surprised if the eye strikes land as far East as New Orleans or as far West as Western Louisiana. The GFDL and LBAR models have performed the best so far. I expect Gustav to remain a major hurricane when it hits shore Monday between 9am tomorrow morning and 1pm as a CAT 3 storm sustained winds of 115-127 MPH near the eye around Lafourche Parish, St Mary Parish and Terrebonne Parish South of Houma then the eye will head just South of or over Morgan City LA by Monday evening on its way across Lafayette LA. Hurricane force wind gusts may extend as far as Mobile Bay AL. Tropical storm force winds are possible as far West as Port Arthur TX and as far East as Pensacola FL. If the slow down onshore occurs as forecast, some places could have hurricane conditions for up to 8 hours. At the risk of sounding like a broken record, I am going to emphasize the LACK OF SKILL in forecasting intensity, in the end game Gustav may strike as a CAT 2 or 1 storm or a 4. The info here is based on the not so certain assumption the official forecast of a 3 is right. Hurricane winds are expected along the entire coastline of AL, MS, LA. Gustav will have an extremely dangerous storm surge of 10-20 feet (based on storm surge and SLOSH model output) and battering waves near and just East of where the 30 mile wide eye hits land--so from at least around Terrebonne Parish to to MS including Houma, Grand Isle, St Mary and New Iberia Parishes and SW as well as Gulfport, Biloxi and possibly New Orleans. The latest storm surge model shows storm surge in New Orleans to NOT be enough to breach the levees if the model is correct in predicting a surge more like a CAT 3 storm. If the storm surge here is that of a CAT 3 or weaker, and the Army Corps is right that the levees can take a CAT 3 surge, then the levees should hold. If it is stronger than forecast, the levees will likely breach in spots. But with the daytime models shifting GUSTAV a little more West again, it looks like New Orleans may only get a Cat 3 or 2 level surge. Also, if Gustav is in a weakening phase as it goes inland, both the wind and storm surge will be less than normal for a CAT 3 storm. Gustav is similar to BETSY in 1965 which caused a billion dollars in damage (10 billion today's dollars), breached the levees and killed 76 people in LA. Gustav looks a cross between Betsy, Rita and Katrina. If the track shifts East 60-90 miles Gustav will be at least as bad as Betsy and Katrina. Remember, there is little to no skill in intensity forecasting so it could be much stronger or much weaker. But at this time cooler waters, upwelling, and greater wind shear ahead of Gustav prior to landfall and some dry air are expected to prevent Gustav from being any more powerful than a CAT 3 which is still bad enough to overwhelm New Orleans IF it comes close enough, even if New Orleans is spared the worst as expected, other areas will not be. The current forecast is for devastation 75 miles inland and 75 miles wide where homes and businesses and industrial buildings will be destroyed or heavily damaged, mobile homes gone, trees flattened, electricity and water lost for days or weeks along the eye path. 6-12 inches of rain with some 20 inch totals and isolated tornadoes are also expected from AL to SE TX. The flooding in LA and East TX is expected to shift into AR later in the week but may stay in TX. The cone of uncertainty for where the eye hits runs from Eastern MS to Port Lavaca TX. Tropical storm conditions are expected as far East as the FL Pan Handle and West to Houston. Hurricane force winds extend out 65 miles from the center while tropical storm force winds extend out almost 220 miles from the eye. Tropical storm force winds at landfall are forecast to extend out 360 miles from the eye, making it only 20% smaller in diameter than Katrina if correct. Tropical storm conditions should commence on the LA coast tonight and New Orleans after Midnight. Hurricane conditions will reach the coast of LA tomorrow morning and possibly even New Orleans late morning. While LA is currently expected to take the brunt of the storm, interests from the Western FL Pan Handle to the Upper TX coast should monitor for any last minute change in direction of Gustav. Your not safe until you know your safe. GAS UP the car before markets reopen Tuesday. Oil and Natural Gas production to take big hit with Gustav, plus speculation fears of Hanna and possible IKE and others to come. Extensive damage and extended downtime to rigs in the Gulf and refineries on land with repairs needed, and even pipes to shore at risk. POOR LA has been clocked worse than any other state in the past 5 decades, with NC #2 and FL #3. If you have not been reading my posts daily since last week, you missed a lot and should read previous posts for some things I am not repeating that are still relevant. I try to update twice a day.HANNA cant spend too much time here, but as of now the range for Hanna is anywhere from South FL to NC as a CAT 1, but confidence in path and strength is LOW. There is a potential IKE behind Hanna, but lets get past Gustav before we worry to much about Hanna or Ike.

August 30 2pm: The official NHC forecast track has edged slightly more toward the TX/LA border, something I've been saying since Wednesday morning when others were saying New Orleans. But there's every reason to expect future projections could change in any direction. The margin of error for landfall runs from MS to Baffin Bay TX. It is still possible the eye will not move inland on LA but turn West along the coast as its forward motion is forecast to slow down considerably. This would increase the flooding threat to South LA and SE TX. The storm could ramp up to a 5 but is currently forecast to weaken to a Cat 4 or 3 before the center makes landfall on the Central or Western coast of LA Tuesday, but the storm affects will start to impact the Gulf Coast on Monday, baring any changes in speed or direction. A CAT 3 storm IF close enough can still overwhelm New Orleans like Betsy in 1965. I think once we see how Gustav navigates the area between Cuba and the Yucatan of Mexico the models will stablize and forecast confidence on the track will go up by tomorrow. As of now New Orleans is still NOT expected to get a direct hit, but is expecting tropical storm force winds and 9 inches or more rain. New Orleans is still a below sea-level bowl and the new CAT 3 rated levees are untested. Expect serious impact from the storm as far inland as Northern LA and adjacent states, with severe impact as far as Central LA and adjacent states, and extreme impact for the Southern Third of LA and maybe SE TX. Again, forecast wind shear, cooler waters ahead, a slowing down and westward bend of path, and upwelling of cooler water just off shore all could lead to a weakening hurricane at landfall and as detailed in previous posts the life cycle phase at landfall has implications. At any given category 1-5, a weakening storm has less impact than normal, a growing storm has more. Think of it as the difference between a baseball bater taking something off at contact vs. following through with the whole body including a snap of the wrist. As of this mornings data Gustav is forecast to strike land from Caillou Bay LA to West of Morgan City to Vermillion Bay LA (near 92 degrees West) 3am to 8am Tuesday. Places getting hurricane force winds include Cameron Parish, Grand Isle, Morgan City, New Iberia, Houma, Lafayette and adjacent areas. A storm surge of 13-18 feet with large battering waves and tornadoes are expected near and East of where the eye hits. Impacts from Gustav will start on the Gulf Coast as early as Monday. The current forecast could be a worst case scenario for Central LA to SE TX. Katrina was on a wost-case trajectory for New Orleans at ANY strength, Gustav currently is NOT. But I dont mean to downplay the threat to New Orleans for serious to severe impact which is still real. As of now, Gustav is expected to be 20% smaller in SIZE as Katrina. It still looks to have major disruptive impact and possible damage to the Oil and Natural Gas industry in the Gulf and along the coast. It is pointless to focus too much on the line track forecast and eyelandfall, as the storm is over 300 miles across. Atlanta could feels some side-effects from Gustav next week, depending on the final track and strength of Gus. Even is Gustav spares New Orleans the worst, a new threat of a hit by Hanna is still on the board. Read the earlier posts for info I have left out of this post which are still relevant. On HANNA... expected to become a hurricane but to do loops and or stalls, so its a non-issue for the USA the next 4 days, but residents from FL to VA and even the Eastern Half of the Gulf Coast should monitor its progress as all bets are off with this storm due to computer model mayhem. Some models strike Hanna in South FLA Thursday as a CAT 3 and the FL Pan Handle Friday or Saturday, but it could just as easy be East FL or SC-VA. As if this were not all bad enough, a strong tropical wave just off Africa is proged by some models to become a TS or HUR (IKE) and reach the USA Coast around or after Sept 9th either South FLA or the Carolinas. GAS up your car by Tuesday, Monday is a market day off, but these 3 storms will probably lead to oil price specualtion that will spike gasoline prices. While the Atlantic is active the Pacific is near record quiet with no tropical cyclones at his point for only the second time in 40 years-1980.

August 28th: As of 5pm Thursday Gustav is fighting mid-level dry air, wind shear, and the Islands of the Caribbean preventing intensification. If it stays away from land, it could reach CAT II status Friday, otherwise it will be weaker. I expect it to reach the SE Gulf off the West tip of Cuba by Sunday night. If in this position, it could then intensify over the 83-88 degree water (to a depth of 2,000 feet in the Loop Current) to a Cat 3 or 4 storm. Virtually all available intensity model guidance, dynamic and statistical, indicate Gustav will eventually become a major class dangerous hurricane. Models are in good agreement on the path upto the Yucatan channel West of Cuba, but diverge after that both in forward speed and path. So there is still a great deal of uncertainty in the predicted track from Sunday onward. The latest models give a window from the FL Pan Handle all the way to Mexico, but Mexico is less likely. It looks like it will be very disruptive of Oil and Natural Gas production in the Gulf with damage to rigs and/or refineries possible depending on strength and location at landfall. It also matters if the storm is getting stronger or weakening at landfall as surface winds can be less than the listed Category if it is weakening as it reaches land. I think the track models will come into better agreement in a couple days. Remember, intensity forecasts are poor and the track error grows with time, reaching an average of 300 miles in 5 days with about a 33% chance of an even larger error. Therefore, much of Mexico and the U.S. still has an equal threat this far out. But the model consensus puts the highest threat from AL to the TX/LA border. Whatever strength and where ever it ends up in the USA if ever, would not be until Monday or Tuesday at the earliest. Being that far out, while landfall on the USA seems likely, it could rapidly intensify even to a Cat 4 or 5 but then weaken to a 2 or less before striking. NOAA Hunter aircraft are dropping 60 data bouys in the Gulf to help the forecast. Gustav is forecast to have tropical storm force winds out to 75 miles from the center-- a small to moderate sized storm, compared to Katrina which had those winds out 200 miles from the core. I know everyone is concerned about New Orleans. Another direct hit there this soon would defy the law of averages (but nature can has and will somtimes do that), as statistics there show the average interval between hits is 12.5 years. For a CAT 3 like Katrina (2005) its 31 years, CAT 4 65 years, a CAT 5 170 years. The average length of time between any point on the map being brushed by the outskirts of a hurricane is once every 3.8 years. The 5pm official NHC (TPC) forecast is for landfall between 2pm Monday and 2pm Tuesday SE of HOUMA LA SW of PORT SULPHUR LA as a CAT 3 with sustained winds of 115 mph. Guess what, it looks like Gustavs sister Hanna could threaten FL or the East Coast sometime next week as a hurricane. Be sure and read my previous two blogs for info I haven't repeated in this post.

August 27th5pm: Gustav is still expected to become a hurricane again, and reach at least a category 2 as it heads into the South and Central Gulf. But some of the models show a slow down or even stall, which has 3 big implications. One, that it may not strike the U.S. until the middle of next week, and two, that we can't know how strong it will be if and when it hits land because the slower speed can first help it intensify, but then give it time to weaken again for a variety of reasons. So just because it is expected to reach CAT 2 or 3 and maybe higher in the days ahead, that does NOT guarantee it will be the same strength when it hits land, since we dont yet have a likely scenario of when that would be! The 3rd implication of a slower movement and path is a longer period of time for disruption of oil and natural gas production in the Gulf, even if they are not severely damaged, just because of the evacuation time and down production time waiting to get the storm out of the way. We get about 25% of U.S. crude and 15% of U.S. nat gas comes from this area. The past 3 official track forecasts have varied a little from Central LA to Eastern LA Monday or Tuesday. The offical intensity forecasts, which have a poor track record, now predicts a peak at 115 MPH with gusts to 138 MPH, a median of the various model output. They point out there is still much uncertainty, and it is too early to tell what if any impact Gustav will have and where it will have it. Their latest cone of uncertainty includes Cancun and Cozumel Mexico and the West Coast of Florida all the way to NE Texas. I would still include MOST of the TX coast all the way through the FL Pan Handle. The model choices include into the Southern Yukatan or up the East Coast of FL offshore. I reject those outlier models, but that still leaves most of the Mexican and U.S. Gulf Coast open for business. I still lean toward a track closer to TX than FL but with low confidence. Thursday the hurricane hunter aircraft will drop 20 special expendable data bouys into the gulf to measure water temps to help with intensity forecasts. See my previous post below for more.

August 27th: 5am: Hurricane GUSTAV figures to become a dangerous CAT 3 or higher hurricane into the Southern Gulf/NW Caribbean Sunday and Monday. The model forecasts vary from a CAT 3 to CAT 5 from a 111 to a 160 MPH storm. However, the intensity will depend on its interaction with the storm-killing Island Mountains of the Caribbean and the storm-high-octane fuel of the Loop Current and warm Gulf of Mexico Loop Current Eddies that have high heat energy content for explosive growth like Charley in 04, and Rita and Katrina in 05. But remember there is little to no skill for strength forecasts beyond 48 hours, and only slight skill in shorter time frames. The NHC forecast is for a 121 MPH storm with gusts to 144. The models have better agreement than earlier in a W/NW track but the future track is still uncertain, with the various equations split between a track up toward Southeast FLA or a West track into Central America. The outliers of the far West and East look least likely. But a track anywhere from the FL Pan Handle to somewhere in Mexico is still on the table Monday or Tuesday. The current hurricane center forecast splits the difference sending the center to the Central Gulf of Mexico South of LA Monday, with the "cone of uncertainty" from near the West Coast of FL to halfway to Mexico/TX . Remember, 2/3rds of the time the cone of uncertainty accurately captures the margin of error, leaving one-third of the time that it does not--meaning 1/3 of the time we can expect the actual path to be OUTSIDE of the "cone of uncertainty". Right now I favor a path of the center more toward the Upper TX Coast Monday or Tuesday, but with just average confidence in my track. By the way, the European ECMWF shows another hurricane (Hanna) hitting FLA/GA from the MIA-FT. Lauderdale area the end of next week! I'll do 2 updates a day as time allows.

August 24th: FAY tropical depression stalls out weakening to remant low may finally bring beneficial rain (too much?) to North Georgia including Lake Lanier basin during the next 2-5 days. Still watching for GUSTAV to form as I first warned you Friday morning on Atlantas Morning News might happen the next threat to the SE USA, East Coast more likely but Gulf is possible. ACTIVE tropics next 3 weeks.

August 23rd: FAY still a threat even as SOME GLOBAL PREDICITON EQUATIONS (UKMET, ECMWF, CMC, NOGAPS) SHOW MAJOR HURRICANE off the SE USA NEXT WEEK AT THIS TIME, (however, none of the tropical waves in the Atlantic show any signs of this for now, but the pattern looks favorable in general the next 3 weeks!), ALSO SOME COMPUTERS SUGGEST REMNANTS OF FAY COULD BRING VERY HEAVY RAIN TO NW GEORGIA AND ATL MON-WED I HAVE MY DOUBTS BUT WILL KEEP AN EYE ON IT......FAY center went ashore at Cape Romano FL early Tuesday morning and has caused at least 8 tornado reports and has produced measured wind gusts of 67-78mph TS FAY will bring more flooding rain and gusty winds to FL with isolated tornados possible. I expect FAY to bend back West across Northern FLA including the Pan Handle Fri./Sat into AL/MS Tue./Wed. eventually weakening to a depression rain storm as it heads NW then NE. An upper level low over TX and the MS Valley should eventually steer the remnant moisture NE toward the TN Valley with just insignificant side effects for ATL unless something changes, better chance it gives us better rain next week. There's a 60% chance it gets back into the Gulf of Mexico by Sat., if that happened it would have to be watched for re-strengthening. There are computer indications the next hurricane threat to the SE Coast or Gulf Coast could come by the end of this month at the latest, maybe 2 threats between now and then. The last CAT 1 storm to hit NE FL was Dora in 1964. Savannah is statistically long-overdue for a hit. The last hur direct hit in SAV was 1947 (storm was seeded with dry ice by a B-17 as part of experiment) and before that 1940 and prior to that 1898. As far as strength keep in mind, the meteorological community as a whole (and our computer simulations) has/have little or no skill at forecasting hurricane or tropical storm intensity changes, that is if they will get stronger or weaker, and if so, how fast and how much they will weaken or get stronger. Impact FAY will have on Atlanta is little, but Extreme South/SW GA and the Coast could see flooding with gusty winds, rough surf, beach errosion and rip currents for the GA/SC/FL Coast. My vacation was canceled so I can cover the future of FAY. A CAT II or stronger hit on JAX FL by the way is a once in 400 year event.

New research published in NATURE forecasts a temporary cooling for Europe and North America over the next decade caused by a change in Atlantic and Tropical Pacific Ocean Currents projected to blunt the present warming or even lead to slightly cooler air temps for about 10 years according to a new computer model run by the Leibniz Institue of Marine Sciences in Germany.

AUGUST 12th: Last August weeks of mid and upper 90s and a string of 100s, that will not happen this year! The tropics come alive the next 4 weeks with the latest MJO moving East and tropical waves moving West. At least 2 or 3 threats of a tropical storm or hurricane anywhere East of the MS River maybe as early as this weekend and next week.

August 4th: Not the brutal hot summer as bad or worse than last year that so many others predicted. SO FAR anyway, the Summer is playing out like my long-range forecast which said this Summer would NOT be even close to last year but would be closer to normal. Using Hartsfield JUNE was hot and dry 3F above-normal with just 16% of normal rainfall. JULY was slightly cooler and wetter than average. -0.2F below-normal with rainfall 40% above-normal. MY 30-DAY OUTLOOK calls for above-normal temps with rainfall near-average to a little above-average. After the current heat wave, relief is in the outlook between August 9th and 17th.

JULY 17TH: RANDOM NOTES from the AMS DENVER CONFERENCE...the following are quotes from professional presentors at the conference: a good site on climate change www.aip.org/history/climate. 53% of AMS seal holding members disagree with the AMS Climate Statement. 78% of those blame urbanization. From 1950 to the present there has been an 8.2 million increase in the U.S. population adding to the impact of expected 50 year weather extremes. 40% of our $10 trillion economy is affected by climate and weather events annually and the weather variablility can produce year to year swings in the nations economy of plus or minus 100 billion. Weather forecast services save U.S. agriculture alone $820 million a year. Climate models are verified against the past and current climate to give real-world testing down to the season. Projections for the future are run with and without the fingerprint of human activity, with and without changes in the sun and volcanoes inorder to see the signals of each and all. Due to uncertainty projections give a range of future outcomes not just a single sky is falling projection or everything will be ok projection. There are differing levels of confidence in various ranges. Half of all CO2 emissions have taken place since 1983, 75% since 1961. It takes 1000s of years for CO2 to leave the atmosphere naturally. Even some former skeptics now believe in GW. Scientists who believe in man-made global warming do NOT believe they have all the answers or understand everything. The "Al Gore" sea level rises are NUTS, and the gulf stream is not going to suddenly stop or reverse, though it could change over the next 50,000 years, the projected sea level rise of scientists is over the course of 100 years about 12 inches. Model projections of Arctic sea ice at the end of the century range from near zero ice to little change. There is not an "even split" among scientists on global warming, there is a worldwide consensus not a sharp divide. The latest IPCC used 72% new scientists who were not in previous reports-- from 130 countries and 800 authors using 15 different GCMs Global Circulation Models which a