Macmillan Holdings, LLC. winninginvestor, QDnow, and Quick and Dirty Tips are trademarks of Macmillan Holdings, LLC. Scientists use mathematical models for everything from weather forecasts to predicting who will win the next election. But how do these models work and why do they sometimes get it wrong? Everyday Einstein looks at the science behind weather prediction. theoldadam everydayeinstein Because they quite often don't have any windows in their 'weather' office. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57209#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57209#Comments Fri, 11 Jan 2013 02:20:05 GMT Amy everydayeinstein The weather service was established to warn Americans when life-threatening megastorms were approaching, before the era of information-at-your-fingertips. Not so we can complain when it drizzles on our picnic. It is a highly complicated science, with much guesswork due to the complexities of ever-changing conditions. It often seems like the weathermen are "wrong". But when was the last time a hurricane hit a US shoreline unnoticed? Not since the weather service started. They most certainly do thei jobs. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57208#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57208#Comments Fri, 11 Jan 2013 00:37:11 GMT Steve everydayeinstein Or...you could say that weather forecasts are often wrong because you're trying to predict THE WEATHER! It's a natural force, effected by natural processes that humans cannot control. Weathermen often do predict trends and more generalized patterns correctly, but it's downright impossible to predict what natural forces are going to do on a daily basis. You don't need a bunch of math to prove this. It's reality. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57166#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=57166#Comments Wed, 09 Jan 2013 21:31:59 GMT Snow Blind everydayeinstein Haha, I hear you. I'm not sure what has happened but as a young child in the 70's, heck even into the 1980's, I remember weather forecasts being accurate most of the time. Now, it's just a joke. I tend to rely on satellite imagery more so than your average newspaper or internet forecast, and attempt to make my own judgement. Who knows...what could be happening is they'll report snow for example even if it's a 10% chance of flurries. 90% of the time it's not going to happen. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=56028#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=56028#Comments Thu, 29 Nov 2012 01:31:00 GMT Chuck everydayeinstein Factoid. The man who started weather forecasts in England, was the Captain of the 'Beagle,' the ship Darwin traveled on and made his initial observations upon which he based his arguments supporting evolution. Even so-so forecasts save lives and property in huge numbers every year. As with most things, even forecasts that are right only 20% of the time are quite valuable. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=53037#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=53037#Comments Wed, 08 Aug 2012 03:48:42 GMT CHuck everydayeinstein Technically, the models aren't 'wrong' since they are always a percentile chance. :) Models are based upon historical data, good weather data only goes back a bit over 100 years and it's not really good until about 1960 or so. Such modeling works by looking at current conditions, and the historical conditions and the predicting tomorrow based upon what happened historically (which will have varied.) Local terrain can and does greatly affect the weather you experience, as anyone walking near skyscrapers knows. Lakes, rivers and hills similarly affect it. Most people develop a sense over time as to how accurate a forecast for the nearest large city may actually happen in their area. But historical patterns are good only so long as the major patterns remain stable, which is currently not the case in most of the world. Our ability to predict based upon current conditions when there are no precedents is far more limited than when patterns are stable (I could, until about 15 years ago, tell you what tomorrow's weather in my town would be like based upon weather maps or even just the clouds. I am no longer able to do that.) Which leads us to 'climate change.' Climate change is a misnomer. Climate is an expectation based upon historical record, thus, you only know that the climate has changed AFTER a period of time has occurred. For instance, 40 years ago it was common in my area to get 18=24" snowfalls and many below -20F days each winter, while our summers generally were in the 80's with the exception of a few days in late July. This particular pattern stopped being the case a decade and more ago. What does change, are weather patterns,and they may change from one relatively stable state to another, or they my fluctuate quite widely. If they are stable, new predictions will gradually become more accurate. If not, accuracy will improve only as we improve our understanding of how weather works (and weather is far more complex than it might seem!) A conservative scientist, predicting climatary change, ought to err on the side of predicting things will be worse than the actual events turnout, on average, if the forecasts and the models improve, the forecast will become closer to the actual events. Unfortunately, for most of the past decade this has NOT been the case, nearly every year when the hemispheric data comes out (Northern and Southern Hemisphere) the headlines read 'Scientists were surprised to find how rapidly X is happening.' This is a problem because overly optimistic forecasts can leave people unprepared for major disasters, and it is always better to be prepared for the disaster that never arrives than to be unprepared for one that does! Part of this problem is that the general public tends to dismiss ALL forecasts if they are too pessimistic. Long-term preparedness is the exception rather than the rule. Thus forecasts have to balance these two contradictory factors. For a non-weather example, take the Y2K computer issue. Because nothing major happened, a lot of folk thought that it was overblown (and in some cases it was...many computer applications couldn't care less what the date is.) In fact, nothing major happened because literally millions of hard-working programmers worked to fix problems for the 10 years before the event so that nothing could happen! Similarly, you may have noticed that many rivers flood on a semi-regular basis. In almost every case, the news is full of people out trying desperately to construct barriers to protect their property. Ever wonder WHY they are doing this at the last minute, especially when many of them get the same high water event every 3-5 years? Wouldn't it make more sense to build permanent dikes, or better still, not build on the areas that flood? Then there are tornadoes. Tornadoes are not very predictable individually, but it is quite easy to predict statistically where they are most likely to hit--like lightening, tornadoes tend to hit the same locations repeatedly because those locations are dictated by the weather response to the terrain. Every tornado that strikes an inhabited area has the exact same news stories. People who were missed thanking God for not being hit. People hit who survived thanking God to be alive. But we have known how to build tornado-proof buildings fora more than 20 years, and even before that, we had apretty good idea of where they were likely to hit. But so far as I know, there is no place in tornado country which has building regulations preventing people from building in inappropriate locations or building non-tornado resistant buildings. Despite the fact that such building codes would undoubtedly save lives and billions of dollars in property damage. In hurricane areas, we have building codes--they're not particular protective, but they are better than what used to be there. Even such limited rules save lives and property. I have been forecasting political, economic and weather events for some 14 years now ith a fairly igh degree of success (~80%.) The past few years I've concentrated upon climate, and my projections are much less optimistic than those you usually see published--but I have had no good reason to change them for the past 5 years, as each year the new data merely confirms my forecast. Many people around the world live next to the oceans, often very low lying areas (California;s Central Valley, a major food producing area, is only 2-3 meters above sea level.) Increasing storms cause even small sea level changes to be exacerbated (made worse) by wave action, thus even a 1 meter rise (39 inches) puts many at risk, not just people and property, but food production too. My current forecast is that we are likely to see a 1 meter to 10 meter rise between 2013 1nd 2027, since such a change is non-linear (each year the increase will get larger than the year before e.g. a 4" rise one year could be followed by a 6" rise the next and a 10" rise the year after,) the amount of time we have to respond is very short once the change becomes measurable. This is complicated by the fact that 'sea level' is a convenient fiction. Actual altitude of the sea surface various around the globe, changes can and are local as well as global (sea level on the NW Alaska coast has risen at least 24" in the past decade.) There is a lot of media attention given to the rather unimportant argument about who's fault this change is...though most scientists are convinced that human activities have accelerated the change, it really makes little difference, since the time to have changed what we do to have any short-term effect was 50 years ago. Well there may be some short-term project which could slow, stop or reverse the change, such is increasingly unlikely with each passing hour. (I have one idea which might work, but it has yet to be modeled and goes against 'current' thought about climate...since current models do so badly, I'm not impressed with that as an argument not to investigate it, especially since I see no other hope of making major changes in less than a century.) The concept is simple, transfer deep cold Pacific ocean water through giant pipes under Panama to the Atlantic surface, bringing 34 degree F water to cool the 80-90+F surface waters on the Atlantic side. Since the Pacific has a higher sea level than the Atlantic, no pumping is necessary. The idea is to restore the prehistoric currents of 3-7 million years ago before the Panama Isthmus blocked the flow from ocean to ocean, cooling the surface which would reduce the energy available for storms. (Before the Isthmus arose, weather patterns and climate were stable for much longer periods than since that time.) The advantages of the plan are: 1) It could be built in less than 10 years, and we would know if it worked withing months. 2) It's cheap, about 7 billion dollars. 3) It's easily stopped if it makes things worse. 4) It would in any case revitalize the fishing industry off the Atlantic coast of Panama, where the ecosystems are nearly destroyed. 5) It requires only one country to approve it. 6) It could generate a substantial portion of the electricity Panama no generates by burning fossil fuels. 7) along the course of the pipeline (some 80-100 miles) and for several miles on either side, it would be simple to tape the cold water for a/c use in cities, towns and recreational areas...generating a potentially huge revenue flow for Panama. It is unlikely that we will ever be able to perfectly predict the weather more than a few hours in advance (though I could easily be wrong,) it's incredibly complex and we aren't even certain at this time what all of the variables which go into generating weather are. But they include the weather on the Sun, the place in our orbit,the direction we move through interstellar space, cosmic radiation, interstellar dust,bacteria in the air, vegetation, ground type, air pressure, humidity, pollen and dust, pollution and many many other factors, most of which we are either unaware of or have little or knowledge of their values or effects. Your very existence affects the weather! http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=53036#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=53036#Comments Wed, 08 Aug 2012 03:44:09 GMT Haggus everydayeinstein Or....you could disregard most everything they say and just look at the maps yourself. It's less stressful. http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=52980#Comments http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx?commentid=52980#Comments Mon, 06 Aug 2012 14:05:38 GMT Fri, 11 Jan 2013 02:20:05 GMT http://everydayeinstein.quickanddirtytips.com/why-are-weather-forecasts-wrong.aspx feedback@quickanddirtytips.com (Managing Editor) feedback@quickanddirtytips.com (Webmaster) en-us