From Climate Change Reconsidered, a work of the Nongovernmental International Panel on Climate Change
The IPCC claims “snow cover has decreased in most regions, especially in spring,” and “decreases in snowpack have been documented in several regions worldwide based upon annual time series of mountain snow water equivalent and snow death. (IPCC, 2007-I, p. 43). Later in the report, the authors claim “observations show a global-scale decline of snow and ice over many years, especially since 1980 and increasing during the past decade, despite growth in some places and little change in others” (p. 376). Has global warming really caused there to be less snow? We addressed this question regarding polar regions in Chapters 3 and 4 of this report. In this section, we focus (as the IPCC does) on studies conducted in North America.
Brown (2000) employed data from Canada and the United States to reconstruct monthly snow cover properties over mid-latitude (40-60°N) regions of North America back to the early 1900s, finding evidence of what he described as “a general twentieth century increase in North American snow cover extent, with significant increases in winter (December-February) snow water equivalent averaging 3.9% per decade.” This finding is consistent with climate model simulations that indicate increased precipitation in response to global warming, but it covers too little time to tell us much about the cause of increased snow cover.
Moore et al. (2002) studied a longer period of time in their analysis of a 103-meter ice core retrieved from a high elevation site on Mount Logan—Canada’s highest mountain—which is located in the heavily glaciated Saint Elias region of the Yukon. From this deep core, as well as from some shallow coring and snow-pit sampling, they derived a snow accumulation record that extended over three centuries (from 1693 to 2000), which indicated that heavier snow accumulation at their study site was generally associated with warmer tropospheric temperatures over northwestern North America.
Over the first half of their record, there is no significant trend in the snow accumulation data. From 1850 onward, however, there is a positive trend that is significant at the 95 percent confidence level, which indicates that recovery from the cold conditions of the Little Ice Age began in the mid-1800s, well before there was a large increase in the air’s CO2 concentration. This finding is further strengthened by the temperature reconstruction of Esper et al. (2002), which places the start of modern warming at about the same time as that suggested by Moore et al.’s snow data, contradicting the temperature record of Mann et al. (1998, 1999), which puts the beginning of the modern warming trend at about 1910.
Cowles et al. (2002) analyzed snow water equivalent (SWE) data obtained from four different measuring systems—snow courses, snow telemetry, aerial markers, and airborne gamma radiation—at more than 2,000 sites in the 11 westernmost states of the conterminous USA over the period 1910-1998, finding that the long-term SWE trend of the region was negative, indicative of declining winter precipitation. In addition, they report that their results “reinforce more tenuous conclusions made by previous authors,” citing Changnon et al. (1993) and McCabe and Legates (1995), who studied snow course data from 1951-1985 and 1948-1987, respectively, at 275 and 311 sites, and who also found a decreasing trend in SWE at most sites in the Pacific Northwest.
Schwartz and Schmidlin (2002) examined past issues of Storm Data—a publication of the U.S. National Weather Service (NWS)—to compile a blizzard database for the years 1959-2000 for the conterminous United States. This effort resulted in a total of 438 blizzards being identified in the 41-year record, yielding an average of 10.7 blizzards per year; linear regression analysis revealed a statistically significant increase in the annual number of blizzards during the 41-year period. However, the total area affected by blizzards each winter remained relatively constant. If these observations are both correct, average blizzard size must have decreased over the four-decade period. On the other hand, as the researchers note, “it may also be that the NWS is recording smaller, weaker blizzards in recent years that went unrecorded earlier in the period, as occurred also in the official record of tornadoes in the United States.”
In a study of winter weather variability, Woodhouse (2003) generated a tree-ring based reconstruction of SWE for the Gunnison River basin of western Colorado that spans the period 1569-1999. This work revealed, in her words, that “the twentieth century is notable for several periods that lack extreme years.” She reports that “the twentieth century is notable for several periods that contain few or no extreme years, for both low and high SWE extremes.”
Lawson (2003) examined meteorological records for information pertaining to the occurrence and severity of blizzards within the Prairie Ecozone of western Canada. Over the period 1953-1997, no significant trends were found in central and eastern locations. However, there was a significant downward trend in blizzard frequency in the western prairies; Lawson remarks that “this trend is consistent with results found by others that indicate a decrease in cyclone frequency over western Canada.” He also notes that the blizzards that do occur there “exhibit no trend in the severity of their individual weather elements.” These findings, in his words, “serve to illustrate that the changes in extreme weather events anticipated under Climate Change may not always be for the worse.”
Berger et al. (2003) collected a 50-year record (1949/1950 to 1998/1999) of snowfall occurrences using data from a dense network of cooperative station observations covering northwest and central Missouri provided by the Missouri Climate Center. The study looked at long-term trends and interannual variability in snowfall occurrence as related to sea surface temperature variations in the Pacific Ocean basin associated with the El Niño and Southern Oscillation (ENSO) and the North Pacific Oscillation (NPO). These trends and variations were then related to four synoptic-scale flow regimes that produce these snowfalls in the Midwest. The authors found no significant long-term trend in overall snowfall occurrence and a decrease in the number of extreme events (≥10 in, >25 cm) was noted. Two years later, Lupo et al. (2005) assembled a similar 54-year database (1948/1949 to 2002/2003) of snowfall occurrences for southwest Missouri and found “no variability or trends with respect to longer-term climatic variability and/or climate change.”
Bartlett et al. (2005) set out to determine what changes might have occurred in the mean onset date of snow and its yearly duration in North America over the period 1950-2002, based on data for the contiguous United States that come from the 1,062 stations of the U.S. Historical Climatology Network, data for Canada that come from the 3,785 stations of the Canadian Daily Climate Dataset, and data for Alaska that come from the 543 stations of the National Weather Service cooperative network in that state. As a result of their efforts, the three researchers found that “for the period 1961-1990 the mean snow onset date in North America [was] 15 December, with mean snow cover duration of 81 days.” They report there were “no significant trends in either onset or duration from 1950 to 2002.” However, interannual variations of as much as 18 and 15 days in onset and duration, respectively, were present in the data; for both parameters they report that “no net trend was observed.”
We find it significant that from 1950 to 2002, during which time the air’s CO2 concentration rose by 20 percent (from approximately 311 to 373 ppm), there was no net change in either the mean onset or duration of snow cover for the entire continent of North America. To provide some context for this 62-ppm increase in atmospheric CO2 concentration, we note that it is essentially identical to the mean difference between the highs and lows of the three interglacials and glacials reported by Siegenthaler et al. (2005) for the period prior to 430,000 years ago. Surely, one would expect that such a change should have had some effect on North American snow cover—unless, of course, atmospheric CO2 enrichment has very little or no impact on climate.
Julander and Bricco (2006) reported that snowpack data were being consistently used as indicators of global warming, and that individuals doing so should quantify, as best they could, all other influences embedded in their data. That meeting this requirement is no trivial undertaking is indicated by their statement that “snow data may be impacted by site physical changes, vegetation changes, weather modification, pollution, sensor changes, changes in transportation or sampling date, comparisons of snow course to SNOTEL data, changes in measurement personnel or recreational and other factors,” including sensors that “do not come back to zero at the end of the snow season.” In an analysis of 134 sites (some having pertinent data stretching back to at least 1912), they thus selected 15 long-term Utah snow courses representing complete elevational and geographic coverage of the dominant snowpacks within the state and adjusted them for the major known site conditions affecting the data, after which the adjusted data for the period 1990-2005 were “compared to earlier portions of the historic record to determine if there were statistically significant differences in snowpack characteristics, particularly those that could indicate the impacts of global warming.”
Of the 15 sites studied in greatest detail, the two researchers found that seven of them exhibited increased snowpack in recent years, while eight exhibited decreased snow accumulation. They also report that “six of the seven sites with increases have significant vegetative or physical conditions leading us to believe that the impacts associated with this analysis are overstated.” The ultimate conclusion of Julander and Bricco, therefore, was that “any signature of global warming currently present in the snowpack data of Utah is not yet at a level of statistical significance … and will likely be very difficult to isolate from other causes of snowpack decline.”
Changnon and Changnon (2006) analyzed the spatial and temporal distributions of damaging snowstorms and their economic losses by means of property-casualty insurance data pertaining to “highly damaging storm events, classed as catastrophes by the insurance industry, during the 1949-2000 period.” This work indicated, as they describe it, that “the incidence of storms peaked in the 1976-1985 period,” but that snowstorm incidence “exhibited no up or down trend during 1949-2000.” The two researchers concluded their paper by stating that “the temporal frequency of damaging snowstorms during 1949-2000 in the United States does not display any increase over time, indicating that either no climate change effect on cyclonic activity has begun, or if it has begun, altered conditions have not influenced the incidence of snowstorms.”
Evidence supporting Changnon and Changnon’s conclusion can be found in the work of Paul Kocin of The Weather Channel and Louis Uccellini of the National Weather Service (Kocin and Uccellini, 2004; Squires and Lawrimore, 2006). The authors created a scale to classify snowstorms, called the Northeast Snowfall Impact Scale (NESIS), that characterizes and ranks high-impact Northeast snowstorms. These storms typically cover large areas with snowfall accumulations of 10 inches or more. NESIS uses population information in addition to meteorological measurements to help communicate the social and economic impact of snowstorms. Storms are put into five categories with 1 being the smallest (“notable”) and 5 being the largest (“extreme”).
Using the NESIS scale, the National Oceanic and Atmospheric Administration’s National Climatic Data Center created a list of 36 “high-impact snowstorms that affected the Northeast urban corridor,” with the earliest storm occurring in 1956 and the most recent on March 1-3, 2009 (NOAA, 2009). Since population has increased in the Northeast over time, more recent storms rank higher on the NESIS scale even if they are no more severe than storms of the past. Nevertheless, fully half (6) of the highest rated (most severe) snowstorms on this record occurred before 1970, as did 14 of all 36 storms in the record. The three most severe storms occurred in 1993, 1996, and 2003, but the next three worst happened in 1960, 1961, and 1964.
Similarly, the National Weather Service tracks the “biggest snowstorms on record” for several cities, with tables showing the dates of the storms and number of inches of snowfall for each posted on its Web site (NWS, 2009). The table for Washington DC shows 15 snowstorms, with five storms having occurred since 1970, four between 1930 and 1970, and six prior to 1930. The biggest snowstorm ever to hit Washington DC arrived in January 1772, when 36 inches fell in the Washington-Baltimore area. It has been called the Washington-Jefferson snowstorm because it was recorded in both of their diaries. It is unlikely that human activity could have contributed to the severity of that storm, or to any other storm prior to the start of significant anthropogenic greenhouse gas emissions in the 1940s.
The research summarized in this section reveals that there has been no trend toward less snowfall or snow accumulation, or toward more snowstorms, in North America during the second half of the twentieth century. This record contradicts either the claim that warmer temperatures will lead to more snowfall and winter storms, or the claim that North America experienced warmer winter temperatures during the past half-century. In either case, the IPCC’s claims in this regard must be erroneous.
Bartlett, M.G., Chapman, D.S. and Harris, R.N. 2005. Snow effect on North American ground temperatures, 1950-2002. Journal of Geophysical Research 110: F03008, 10.1029/2005JF000293.
Brown, R.D. 2000. Northern hemisphere snow cover variability and change, 1915-97. Journal of Climate 13: 2339-2355.
Changnon, D., McKee, T.B. and Doesken, N.J. 1993. Annual snowpack patterns across the Rockies: Long-term trends and associated 500-mb synoptic patterns. Monthly Weather Review 121: 633-647.
Changnon, S.A. and Changnon, D. 2006. A spatial and temporal analysis of damaging snowstorms in the United States. Natural Hazards 37: 373-389.
Climate Change Reconsidered: Website of the Nongovernmental International Panel on Climate Change. http://www.nipccreport.org/archive/archive.html
Cowles, M.K., Zimmerman, D.L., Christ, A. and McGinnis, D.L. 2002. Combining snow water equivalent data from multiple sources to estimate spatio-temporal trends and compare measurement systems. Journal of Agricultural, Biological, and Environmental Statistics 7: 536-557.
Esper, J., Cook, E.R. and Schweingruber, F.H. 2002. Low-frequency signals in long tree-ring chronologies for reconstructing past temperature variability. Science 295: 2250-2253.
IPCC. 2007-I. Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M. and Miller, H.L. (Eds.) Cambridge University Press, Cambridge, UK.
Julander, R.P. and Bricco, M. 2006. An examination of external influences imbedded in the historical snow data of Utah. In: Proceedings of the Western Snow Conference 2006, pp. 61-72.
Kocin, P. J. and Uccellini, L.W. 2004. A snowfall impact scale derived from Northeast storm snowfall distributions. Bulletin of the American Meteorological Society 85: 177-194.
Lawson, B.D. 2003. Trends in blizzards at selected locations on the Canadian prairies. Natural Hazards 29: 123-138.
Lupo, A.R., Albert, D., Hearst, R., Market, P.S., Adnan Akyuz, F., and Almeyer, C.L. 2005. Interannual variability of snowfall events and snowfall-to-liquid water equivalents in Southwest Missouri. National Weather Digest 29: 13 – 24.
Mann, M.E., Bradley, R.S. and Hughes, M.K. 1998. Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392: 779-787.
Mann, M.E., Bradley, R.S. and Hughes, M.K. 1999. Northern Hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations. Geophysical Research Letters 26: 759-762.
McCabe, A.J. and Legates, S.R. 1995. Relationships between 700hPa height anomalies and 1 April snowpack accumulations in the western USA. International Journal of Climatology 14: 517-530.
Moore, G.W.K., Holdsworth, G. and Alverson, K. 2002. Climate change in the North Pacific region over the past three centuries. Nature 420: 401-403.
NOA. 2009. The Northeast snowfall impact scale (NESIS). U.S. National Oceanic and Atmospheric Administration, National Climatic Data Center. http://www.ncdc.noaa.gov/ snow-and-ice/nesis.php. Last accessed May 6, 2009.
NWS. 2009. Biggest snowstorms on record, Baltimore/Washington. U.S. National Weather Service. http://www.erh.noaa.gov/lwx/Historic_Events/snohist.htm. Last accessed May 6, 2009.
Schwartz, R.M. and Schmidlin, T.W. 2002. Climatology of blizzards in the conterminous United States, 1959-2000. Journal of Climate 15: 1765-1772.
Siegenthaler, U., Stocker, T.F., Monnin, E., Luthi, D., Schwander, J., Stauffer, B., Raynaud, D., Barnola, J.-M., Fischer, H., Masson-Delmotte, V. and Jouzel, J. 2005. Stable carbon cycle-climate relationship during the late Pleistocene. Science 310: 1313-1317.
Squires, M. F. and Lawrimore, J. H. 2006. Development of an operational snowfall impact scale. Presentation at 22nd International Conference on Interactive Information Processing Systems for Meteorology, Oceanography, and Hydrology. Atlanta, GA.
Woodhouse, C.A. 2003. A 431-yr reconstruction of western Colorado snowpack from tree rings. Journal of Climate 16: 1551-1561.