Do Warmer Temperatures Really Cause More Snow?
The Kevin Trenberth Effect: Pulling Science Back to the Dark Ages. Part two - The Big Snow Job
Trenberth’s 1999 paper framing the effects of global warming on extreme precipitation declared, “With higher average temperatures in winter expected, more precipitation is likely to fall in the form of rain rather than snow, which will increase both soil moisture and run off, as noted by the IPCC (1996) and found in many models.” The 2001 IPCC 3rd Assessment repeated those expectations stating, “Northern Hemisphere snow cover, permafrost, and sea-ice extent are projected to decrease further.” Soon climate scientists like Dr. Viner proffered alarming scenarios that ‘children would no longer know what snow was’. Similarly in 2008 politicians like RFK Jr. warned DC children would be deprived of the fun of sledding due to global warming. But our climate naturally oscillates and by early February of 2010 Snowmageddon was blanketing the USA’s eastern seaboard with record snows, making global warming predictions the butt of many jokes. The heavy snows didn’t disprove CO2 had caused any warming, but it definitely highlighted failed predictions.
In 2011 Chris Mooney writing for the DeSmog blog noted heavy snowfall had become a “communications nightmare” for global warming theory and urged, “We need to move the public to a place where drawing a warming-snowstorm connection isn’t so challenging”. Kevin Trenberth was already on point. Just two weeks after the 2010 Snowmageddon, Trenberth appeared in a NPR interview flip-flopping to a new climate change framework in which a “Warming Planet Can Mean More Snow”. Now he argued, "The fact that the oceans are warmer now than they were, say, 30 years ago means there's about on average 4 percent more water vapor lurking around over the oceans than there was, say, in the 1970s”. Thus “you can get dumped on with more snow partly as a consequence of global warming," A year later the Union of Concerned Scientists held a press conference asserting global warming was no longer causing less snow, but causing heavier snow. And now, every year as heavy snowstorms approach, Trenberth and his well-groomed media outlets bombard the public, urging them not to be misled by their senses, but trust that cold and snowy days have worsened due to global warming.
Trenberth bases his warmer-earth-more-cold-and-snow alchemy on the Clausius–Clapeyron relation stating, “the water holding capacity of the atmosphere goes up exponentially at a rate of 7% per degree Celsius.” Indeed the Clausius–Clapeyron relation is undeniable physics. The problem is Trenberth misapplies it. First as seen in the graph below from the peer-reviewed paper Weather And Climate Analyses Using Improved Global Water Vapor Observations, there is little evidence of a steady increase in total precipitable water vapor (TPW) ever paralleling rising CO2. The important question Trenberth never asked was, “if TPW has declined since 1998, has there been no warming since 1998?” Indeed in accord with less water vapor, several top climate scientists have reported a global warming hiatus over the same period and the Climate Reference Network reports no warming trend over the USA for the past decade. Furthermore, ocean temperatures were in agreement. Based on Argo data a consensus of scientists reported heat content in the upper 300 meters of the ocean had “increased from 1984 to 1992 followed by a short cooling episode in 1992/93, and then increased from 1994 to 2003/2004, followed by flattening or a decrease.” Note the decline in water vapor from 1992 to 1994 and the decline since 1998 coincides with those ocean temperatures. All things considered, the uptick in heavier snow since 2009 cannot be explained by Trenberth’s new normal “warmer and wetter” assumption.
“Old school” scientists seek to understand causes of extreme events by examining changes in atmospheric circulation and other contributing weather dynamics. In contrast Trenberth does not want scientists to use the standard null hypothesis to test if CO2 warming was a contributing factor. He simply assumes CO2 must be and accuses other researchers of erroneously accepting the standard null hypothesis indicating no effect from rising CO2 (type 2 errors). Based on pure assumptions, he wants to allot some portion of every extreme event to rising CO2, even when an no anthropogenic signal emerges from standard scientific analyses and modeling experiments, as discussed in part 1. According to Trenberth, due to the dominating effects of natural variability, CO2-driven climate models do a very poor job of simulating large changes in atmospheric circulation. While one model run will force large changes, the next model run will not. To side step that problem, instead of asking if there have been trends in atmospheric and oceanic circulation changes that produced snowfall extremes, Trenberth wants researchers to simply ask, “Was it [snowfall] related to higher than normal SSTs off the coast or farther afield” and then assume those higher temperatures were partly due to rising CO2. But that’s bad science. Higher than normal sea surface temperatures often have no connection to any theoretical CO2 heating. Warmer sea surface temperatures associated with a storm can be solely caused by a redistribution of warm water during an El Nino event. A shift in the North Atlantic Oscillation, or a shift in the jet stream can reduce wind fields and warm sea surfaces because weaker winds ventilate less heat and reduce evaporative cooling. Elsewhere shifts in atmospheric circulation can reduce cloudiness and increase solar heating.
Trenberth has reported that 70% of the moisture involved in a storm is typically in place at the beginning of the storm, suggesting global warming has increased the available moisture. But again observations do not support Trenberth’s simplistic “warmer and wetter” attributions. For example in the 2011 Groundhog Day Blizzard the amount of available water vapor was far below normal as seen in the diagram posted by meteorologist Joseph D’Aleo at WUWT. So another question Trenberth’s attribution studies must ask, “where does the moisture come from for an extreme snow event when a region is not “warmer and wetter?”
Still there are many useful questions that can be asked to determine if the affects of climate change have exceeded the boundaries of natural variability. For example, do similar extreme snowfalls happen independently of sea surface temperatures that are warmer or cooler than normal? That question is easily answered from a historical perspective that encompasses just 100 to 150 years. Historical extremes like the Great Blizzard of 1888 dropped very similar amounts of snow on America’s northeast, despite a very different climate background with colder ocean temperatures from the Little Ice Age and extensive Arctic sea ice. Comparing the Great Blizzard of 1888 with Snowmaggedon, higher than normal SST temperatures do not appear to be a critical factor.
To separate natural weather dynamics from climate change scientists must also establish why snowfall varies greatly over small timeframes; timeframes that are too short for CO2 to hypothetically alter ocean temperatures. As anyone having lived in New England knows, during any given winter the depth of snowfall is totally dependent on 2 crucial factors: 1) how fast the storm moves along the coast and 2) how far from the coast the storm travels. Unquestionably slow moving storms cause the most extreme precipitation events, rain or snow. For the American east coast, colder than normal temperatures south of Greenland encourage more frequent blocking ridges of high pressure, and those blocks cause storms to slow down and even stall. These “Greenland blocks” were also responsible for Superstorm Sandy’s sudden shift back towards the coast.
Greenland blocks are more common during negative phases of the North Atlantic Oscillation (NAO), a phase that has coincided with the recent rise in heavy snowstorms. So we must also ask if global warming has affected a shift to the negative phase of the North Atlantic Oscillation (NAO)? But previous research had suggested increased CO2 promoted a more positive NAO during the latter decades of the 20th century. Within a framework of a single year or a few decades, shifts in the NAO are often associated changes in snowfall. But if we ask if climate change altered trends in a given NAO phase, researchers report in the paper Need for Caution in Interpreting Extreme Weather Statistics, “no significant changes either in the mean or in the entire PDFs [Probability Density Functions]” of the NAO index over the last 140 years.
As illustrated in the diagram below, the positions of cold air masses on land and warm air masses over the ocean determine where precipitation falls as snow or rain. For example during the Blizzard of 2013, despite being surrounded by warm ocean waters Nantucket Island received the least amount of snowfall (6.3 inches) while further west Providence Rhode Island (18 inches) and Hartford, Connecticut (22.8 inches) surrounded by a colder air mass received record snow. For snow to form, moist warm air must be raised to an altitude where temperatures are below freezing, with an optimal snow forming temperature hovering around -12 degrees C (10F). Typically a cold air mass (or mountains) forces the rise in altitude. According to the Clausius-Clapeyron relation, air at 31 degree F can only hold a given amount of moisture, no matter how greatly the global average temperature varies. The critical factor that determines how much snow will accumulate is the temperature of the air nearer the ground. If lower air layers are warmer than 0°C (32F), the snow will melt as it falls forming rain, freezing rain or sleet. Only where the entire air column is below freezing do we get snow. If the storm track moves too far out to sea, or if the cold air mass is to far inland, the warm air mass gets less lift, and much less snow forms. Thus to attribute the cause of extreme snowfall a scientist must also ask, “what was the position of the storm track?” And how much cold air was in place?
Trenberth cavalierly suggests that it’s always cold enough to snow in winter, but that that is misleading. For blizzards to occur sufficient cold air must already be in place and that is not a given. Dips in the jet stream and storm tracks across North America pull cold Arctic air southward along the storms trailing edge. To produce Snowmageddon blizzards along the east coast, enough cold air had to reach the southeast and overflow the Appalachian barrier where it is dammed up along the coast (Rauber 2005). The snows that reached Jacksonville Florida in 2015 were the result of a stronger than normal flow of cold air over the Appalachians. Similar to “lake effect snow”, after flowing over the ocean, the cold dry air picked up enough moisture to dust Jacksonville with light snows.
Accordingly the National Snow and Ice Data Center experts tell us, “While it can be too warm to snow, it cannot be too cold to snow. Snow can occur even at incredibly low temperatures as long as there is some source of moisture and some way to lift or cool the air”. In contrast, Mooney relays Trenberth’s message contradicting those experts stating, “Heavy snows mean the temperature is just below freezing, any cooler and the amount would be a lot less.”… “Warmer waters off the coast help elevate winter temperatures and contribute to the greater snow amounts. This is how global warming plays a role.” Why would Trenberth make that up?
Dips in the jet stream and stronger storms capable of pulling an abundance of cold Arctic air equatorward are often associated with the negative phase of the North Atlantic/Arctic Oscillation (AO). Although December 2015 had been mild, when weather forecasters recognized a shift to the AO’s negative phase in early January 2016, they correctly predicted conditions would be just right for the Blizzard of 2016 that buried the mid-Atlantic States in 2 feet of snow 2 weeks later. So to explain contributions of extreme snowfall, scientists must ask how do natural cycles of the North Atlantic/Arctic Oscillation contribute to extremes.
As would be predicted by a shift to more frequent negative phases of the NAO/AO, the USA was experiencing greater incursions of cold Arctic air that promoted both more record low temperatures and greater snowfall, as was the case in the 1960s and 70s. Despite projections by CO2 driven models that the ratio of record high temperatures would exceed record low temperatures by 20 to 1 in 2050, in 2013 and 2014 record low temperatures exceeded record highs. However to counter such contradictory observations, Trenberth pushes another unscientific and non-falsifiable explanation. Suggesting risingCO2 was preventing extreme cold that he claims reduce snowfall, Trenberth submitted,
“below normal temperatures can be fully consistent with climate change but are likely warmer than they otherwise would have been.”
Winter storms are low-pressure systems, or cyclones, that spin in a counter-clockwise direction as they travel across North America. Most winter cyclones in North America are initiated by the curvature of the jet stream as it passes around the Rocky Mountains, or curve northward along the eastern seaboard. The North American topography favors two major storm centers in western North America. One lies just east of the Canadian Rockies where “Alberta Clippers” form. Clippers are fasting moving storms. Typically they will not produce record heavy snowfall because the moisture supply flowing into northern North America is relatively low and the Clippers’ swift passage does not allow for sustained snow accumulation. However Clippers can evolve into major storms over the Great Lakes or eastern seaboard where moisture is available or when they align with storms initiated by the subtropical jet stream. The other storm center lies just east of Colorado. These storms often gather more moisture from the Gulf of Mexico and are slower moving. Typically these storms deliver heavier snowfall. Of importance to east coast snowfall, either storm type will pull cold Arctic air southward and eastward toward the coast, setting the stage for greater snowfall totals from the next storm. In fact it was an Alberta Clipper that set the stage for the east coast Blizzard of 2015.
In general as illustrated below, there are 3 air masses that interact with a winter storm. 1) The cooler air that was left in place from a previous storm. This cooler air mass forces the approaching warm air to rise to altitudes where water vapor can turn to snow. 2) Warm moist air from the Gulf of Mexico or tropical Atlantic that is pulled northward by the storm’s leading edge. 3) The cold dry Arctic air pulled southward along the storms trailing edge. Mild warm conditions generated from the warm air mass typically precede a blizzard, and often catch people ill prepared for the bitter cold that follows. The most famous incident was the January 1888 School Children’s Blizzard that swept through the Great Plains. It was so named because of the 235 people who were killed, many were children who headed to school “lightly dressed because temperatures had been gradually rising to just above freezing as warm moist air was pulled up from the Gulf of Mexico. However within a few hours temperatures dropped to -29 degrees C, as the cold Arctic air advanced. Due to this counter-clockwise circulation pattern, strong storms can reverse a region’s normal latitudinal temperature gradient, temporarily making it warmer in the north and colder in the south.
When storms track along the east coast, they intensify due to the sharp contrast between warm Atlantic temperatures and cold land temperatures. The sharp contrast favors “explosive cyclogenesis”, a phenomenon that is most common along the Gulf Stream and along the Kuroshio Current and promotes extreme snowfalls in New England and Japan respectively. In addition to the land-sea contrast, there is also a steep temperature gradient over the Atlantic due to the warm Gulf Stream. Along the coast of North Carolina in February, coastal waters are typically 10 degrees C (50 F), while just 130 kilometers to the east, Gulf Stream waters register 22 degrees C (72 F) Reddy 1994. In addition to the heat and moisture evaporating from warm Gulf Stream waters, winter storms travelling up the coast will pull warm moist tropical air northward in what is called the “warm conveyor” as illustrated below in the Washington Post illustration of the 2015 blizzard. Notice the head of the storm’s “comma” shape is an area of extreme snowfall, where the storm had pulled warm and moist air northward and westward which then rose over the colder air already in place from previous storms.
Nevertheless ignoring all the potent weather dynamics that naturally drive anomalously warmer sea surface temperatures ahead of a storm, Trenberth emailed his favorite media outlets Joe Romm, Chris Mooney and others to assert, “At present sea surface temperatures are more the 2 degrees F above normal over huge expanses (1000 miles) off the east coast and water vapor in the atmosphere is about 10% higher as a result. About half of this can be attributed to climate change.”
Was this 50% contribution ever scientifically tested and peer reviewed? Did Trenberth determine “how much warmth was transported northward on the warm conveyor side of the storm?” Did Trenberth ask how much warmth was picked up from the Gulf Stream and carried westward to cooler coastal waters? Did the storm temporarily reverse the latitudinal temperature gradient? Trenberth’s untested opinion of a 50% contribution attributed to rising CO2 was simply an opinion. It was an opinion pushed to satisfy the “need to move the public to a place where drawing a warming-snowstorm connection isn’t so challenging” and thus protect the global warming theory.
More yellow journalism followed a few weeks later in Mooney’s “What the massive snowfall in Boston tells us about global warming”. Keeping the focus on global warming Mooney reported, “sea surface temperatures off the coast of New England are flashing red”. Michael Mann added to the global warming meme reporting, “Sea surface temperatures off the coast of New England right now are at record levels, 11.5C (21F) warmer than normal in some locations.” But Mooney, Mann and Trenberth were not interested in discussing the details of those fleeting warm anomalies. They never considered the warm conveyor delivered above normal warmth northwards and then dragged that warmth and Gulf Stream warmth westward. They never tell us how fleeting those warm anomalies were. Yet for the month of February 2015 temperatures on land and sea were all several degrees colder than normal as seen in the illustration by CBSBoston’s chief meteorologist. It was extreme cold that intensified the storm. And despite below normal sea surface temperatures and thus below normal water vapor, the storm gathered enough moisture and Boston experienced record-breaking snows.
Trenberth has now revised his 1999 framework. Despite the record cold that reduces water vapor, he still argues global warming causes more snow in winter. He maintains warming will still cause more rain and reduced snow in the fall and spring. But again the evidence contradicts his claims. Although Trenberth focuses public attention on a decreasing trend in spring snow extent, like the winter, there has also been an increasing trend in autumn snow extent as seen in the graph below from Rutgers Global Snow Lab.
So why does Trenberth persist in claiming extreme snowfalls are due to a warmer and wetter world. Trenberth betrays his intentions when he writes, “The main way climate change is perceived is through changes in extremes because those are outside the bounds of previous weather. Climate change from human influences is difficult to perceive and detect because natural weather-related variability is large. Even with a significant climate change, most of the time, the weather is within previous bounds.” So Trenberth has organized a media campaign to not only overturn the null hypothesis, but to reverse our understanding of the difference between climate and weather. He wants you believe every extreme weather event is worsened by CO2, whether or not there is any evidence.