Alaska At The Limits

by Miceal O'Ronain

(3 August 2002)

Abstract

On June 16, 2002 citing "federal officials" as the source for temperature information, the New York Times published a story about climate change in Alaskan which reported that: "the average temperature has risen about seven degrees over the last 30 years".  The temperature increase report by the New York Times was excessive and was inconsistent with values published by Alaskan climatologists and United States Government agencies. In a response, which was ignored by the New York Times, the Alaska Climate Research Center stated that "the correct warming for Alaska is about 1/3 of the quoted amount for the last climatological mean 1971 to 2000".

On July 11, 2002, the New York Times published a correction which stated that: "according to an assessment by the University of Alaska's Center for Global Change and Arctic System Research, the annual mean temperature has risen 5.4 degrees Fahrenheit over 30 years, not 7 degrees". The Alaska Climate Research Center responded to the correction by stating that: "we still find the value of 5.4F too great by a factor of 2 for the 1971 to 2000 period, the last 30 years".

In an interview, given to an Alaskan news paper, the Director of the Alaska's "Center for Global Change and Arctic System Research", made contradictory statements about the information provided to the New York Times, citing differing times periods for the confusion and appeared to support a temperature increase of "about 2-3 degrees F" for the time interval reported by the New York Times. It will be shown that the two sources referenced by the New York Times are in fact very closely affiliated through the United States Global Change Research Program and that the New York Times, used an invalid statistical technique to extract their Alaskan temperature information from a seriously flawed graphic, published by that federal agency, which reported to show the Annual Average Temperature for Alaska during the 20th Century.  

Introduction

In a story titled "Alaska, No Longer So Frigid, Starts to Crack, Burn and Sag" [1], Timothy Egan of the New York Times, reported on June 16, 2002 that: "the average temperature has risen about seven degrees over the last 30 years" in Alaska.  Citing "federal officials" as the source, the story then when went on to report that: "mean temperatures have risen by 5 degrees in summer and 10 degrees in winter since the 1970's". An average annual temperature in increase of 7F for Alaska since the 1970's would be unprecedented.  What makes this temperature so strange is that on page 91 of the controversial "U.S. Climate Action Report - 2002" [2], issued by the Environmental Protection Agency [3], it was stated that during the 20th Century: " warming over the 48 contiguous states amounted to about 0.6C (about 1F), warming in interior Alaska was as much as 1.6C (about 3F)". While the "interior Alaska" is not the same as an average for the State, an increase of only 3F over the entire 20th Century, is inconsistent with a 7F increase over the last 30 years of the 20th Century.

Since the 1980's, Dr. Sue Ann Bowling has been tracking an unusual Alaskan temperature event known as the Great Pacific Climate Shift which occurred from1976 to 1977 and maintains a graph at "The FANB temperature record" [4]. Dr. Bowling's graph, which averages the annual temperatures for the weather stations at Fairbanks, Anchorage, Nome and Barrow, shows that Alaskan temperatures increased by 2.6F suddenly from 1976 to 1977 and have been relatively stable both before and after that date, for the period from 1954 to 2000. The final blow to the New York Times story occurred on June 24, 2002, when Dr. Wendler and Mr Brian Hartmann &  of the "Alaska Climate Research Center" [5] at the University of Alaska, Fairbanks, posted a statement  "In response to the New York Times Article of 16 June 2002 and Op/Ed of 24 June 2002" [6]. In their response to the New York Times story, the ACRC scientists cited the same stations referenced by Dr. Bowling, and showed that, using linear trend analysis, the average temperature increase for these four reference stations was 2.44F for the period from 1971 to 2000. Pertaining to the New York Times report,  the ACRC stated that: "the correct warming for Alaska is about 1/3 of the quoted amount for the last climatological mean 1971 to 2000"

While ignoring the ACRC Response and faced with mounting questions, the New York Times had to find an Alaskan Climatologist to support their story. Apparently they were unable to locate anyone who would corroborate their 7F temperature increase. On July 11, 2002 the New York Times published a "Correction" [7], in which they acknowledged that the 7F temperature was "misstated" and citing the "Center for Global Change and Arctic System Research" [8] at the University of Alaska, Fairbanks, announced a new value of 5.4F.for the annual mean temperature increase for Alaska:

"A front-page article on June 16 about climate change in Alaska misstated the rise in temperatures there in the last 30 years. (The error was repeated in an editorial on Monday and in the Bob Herbert column on the Op-Ed page of June 24.) According to an assessment by the University of Alaska's Center for Global Change and Arctic System Research, the annual mean temperature has risen 5.4 degrees Fahrenheit over 30 years, not 7 degrees."

The ACRC [5] scientists posted an Update Response on July 12, 2002 in which they stated: "We still find the value of 5.4F too great by a factor of 2 for the 1971 to 2000 period, the last 30 years". In the update response, it was noted that the New York Times might be referring to another time period; however, given that the Times was very explicit on several occasions about the time interval, that appears to be unlikely. On July 19, 2002, the Anchorage Daily News, in their "Voice Of The Times" section carried an interview with Dr. Gunter Weller, Director [9] of the "Center for Global Change and Arctic System Research" [8], which was titled "Global Warming Debate Involves . . . Dueling digits" [10]. Reading the interview several times, I found myself totally confused; however, I suspect that the reporter and Dr. Weller were equally confused. During the interview, several of items of interest immerged:

a)     The CGC "told The New York Times that annual mean temperatures increased 5.4 degrees in Alaska between 1966 and 1995", not the period from 1971 to 2001 reported by the Times.

b)     "Weller says his numbers differed from those of the Climate Research Center because the research he was referring to measured the temperature change from 1966-1995 while the Climate Research Center measurements were from 1971-2000".

c)      According to Weller "the strongest warming trend has shifted from Alaska into Northern Canada and the warming trend for most of Alaska . . . is now about half of the 1966-1995 value, or about 2-3 degrees F."

The temperature increase of 5.4F for period from 1966 to 1995 is not  credible. While 1966 was a cold year, the three years immediately following,  were unusually warm for pre 1976 conditions. Using the four stations  referenced by the ACRC and the CGC in some of their reports, and applying  the same linear trend analysis used by the ACRC, the temperature increase  from 1966 to 1995 was 2.89F. Even if an invalid technique, such as taking the absolute difference of the averaged annual mean temperature is used, the increase is only 4.66F. The only way to obtain a temperature increase of 5.4F for the period from 1966 to 1995 would be to select a pair of unusually cold and warm years and take the absolute temperature difference. The application of any legitimate statistical technique, yields values similar to those reported by Dr. Bowling [4].

The second statement in the interview is equally questionable, because the Great Pacific Climate Shift of 1976 to 1977 had such a profound impact in Alaskan temperatures that unless you use a limited set of data about the inflection point (1976-1977) most statistical techniques will only see the inflection point. Dr. Weller's third statement is the most puzzling of all, because raw temperature records are a matter of historical fact and not subject to change. By implication Dr. Weller's appears to be indicating that (1) Alaska is cooling and (2) he is in agreement with the 2.44F value reported by the ACRC. If this interpretation if correct, the New York Times has lost their star climate expert.

The question remains, exactly where did the New York Times obtain the 7F temperature increase for the last thirty years and was the 5.4F increase obtained from the CGC or other sources?    

The Temperature Data

Before we continue, it should be noted that the Alaskan temperature data used in this report, was obtained from two sources. The "Alaska Climate Summaries" [11], which is part of the "Western Regional Climate Center" [12]. This facility is one of the network of NOAA "National Climatic Data Center, Regional Climate Centers" [13] which offers on-line access to climate data. The second source for Alaskan temperature data was NASA's "Goddard Institute for Space Studies, Surface Temperature: Station Data" [14], who's datasets extend back to the 1930's. Using the temperature values from the "Alaska Climate Summaries" [11], the calculations of the ACRC were verified. Calculations and graphs appearing in this report were created. Microsoft Excel 2002 was used to manage the data, perform calculations and create graphs. 

The Investigation

A search of the climate alarmist literature on the Internet identified several possible sources for the numbers cited by the New York Times. Unfortunately, none of the apocalyptic scenarios quite matched the conditions in the Times story and none of the stories could be matched with a credible scientific source. There were three facts which the New York Times cited about their source:

a)   The time interval for both temperatures was a 32-year period between 1970 and 2001.
b)
  The source of the 7F temperature increase was "federal officials".
c)
   The source of the 5.4F temperature increase was the "Center for Global Change and Arctic System Research".

The 1978 date is significant because it is immediately after the Great Pacific Climate Shift of 1976 to 1977, when sea ice did not yet have time to respond to the sudden increase in temperature. The mid-1960's coincides with unusual cold temperatures, when climate alarmists were selling global cooling and the coming of the next ice age. In the climate alarmist industry, the selection of dates is everything. Also, the Times cites a "federal report" as the source of their information.  So the key question is, what is it about the early 1970's which makes it so very interesting to the climate alarmist industry? And, exactly which "federal report" was the New York Times using?

The source of the 7F temperature was difficult, because it is not reported in the literature. The key clue here was the "federal officials" and the "federal report" sighted in the comments pertaining to sea ice.

Finally we come to the "Center for Global Change and Arctic System Research" (CGC) [8]. According to the home web page of the CGC, their major research project through the end of 1999 was the writing of the report "The Potential Consequences of Climate Variability and Change" [15], for the "Alaska Regional Assessment Group" [17]  of the "U.S. Global Change Research Program" [16], which was issued in December of 1999. The Alaskan report was then included the "US National Assessment of the Potential Consequences of Climate Variability and Change" (USNA) [18] was issued in 2000 and contained a regional section devoted to Alaska, "US National Assessment of the Potential Consequences of Climate Variability and Change Mega-Region: Alaska" [19].

In the last two years, there have been three significant United States Government reports on climate. These were:

a)     "U.S. Climate Action Report - 2002" [2].

b)     "Climate Change Science: An Analysis of Some Key Questions" [21].

c)      "US National Assessment of the Potential Consequences of Climate Variability and Change" [18].

The EPA's "U.S. Climate Action Report" undermines the New York Times story by stating that Alaska has warmed by 3F during the 20th Century. The "Climate Change Science: An Analysis of Some Key Questions" report, by the National Academy of Science, was an executive overview and contained not one word about Alaska. So the only remaining "federal report" is the USNA report. As the CGC report was the prequel the Alaskan section of the USNA report and was specifically referenced by the New York Times as the source for its 5.4F temperature increase, it will be considered as with the USNA report.

The question is, can the 7F and 5.4F temperature increases for Alaska be found in either of the CGC or USNA reports? It should be noted that 5.4F is a very strange number for climate alarmists, because they do not like to be give specific numbers for which they can be held accountable.  Terms such as about the 1960's or about 7F are quite common, but 5.4F caries quite a bit of numerical authority and implies a verifiable source. While 5.4 degrees is a strange number in the Fahrenheit scale, it happens to be exactly 3 degrees in the Celsius scale. Recognizing the predictability of climate alarmists, let's expand the search to three temperature values of 7F, 5.4F and 3C. Let's search the two documents in the order in which they were published, beginning with the CGC report.

Investigation of the CGC Report on Alaska

The CGC's report is published in sections as a series of PDF documents at "The Potential Consequences of Climate Variability and Change" [15]. Searching these PDF documents, on page 6 of "Preface and Executive Summary" [15.1], is contained a figure with the follow description: (Fig. 1) "The Alaskan regional Climate and climate-related trends in Alaska, b) annual mean temperature composite from Anchorage, Fairbanks, Nome and Barrow". The following narrative accompanies the graphic:

Observed Climate Trends

Alaskas enormous size encompasses extreme climatic differences. The southern coastal margin is climatically similar to the Pacific Northwest, although cooler and with longer winters. North of the Alaska Range, the climate is continental, with moderate summers, very cold winters, and annual precipitation of 816 in (2040 cm). North of the Brooks Range, an arctic semi-arid climate prevails, with less than 8 in (20 cm) of annual precipitation and snow on the ground for nine months of the year. There are widespread areas of permafrost and large glaciated areas throughout the state, and extensive sea ice along the western and northern coasts.

Alaska has experienced the largest regional warming of any state in the U.S., with a rise in average temperature of about 5F (3C) since the 1960s and 8F (4.5C) in winter. Records from some regions show a warming of nearly 34F (1.52C) quite suddenly in the late 1970s (Fig. 1). There has been extensive melting of glaciers, thawing of permafrost and reduction of sea-ice. The Alaskan regional warming is part of a larger warming trend throughout the Arctic. The large observed warming has been accompanied by increases in precipitation of roughly 30% between 1968 and 1990. Alaska is also strongly affected by El Nio and the interdecadal Arctic Oscillation, bringing warmer and wetter winters to coastal Alaska in their warm phases, and cooler, drier winters in their cool phases.

Note the statement "with a rise in average temperature of about 5F (3C) since the 1960s". There was a significant rounding error made in converting between Fahrenheit and Celsius. Should the caption read "5.4F (3C)" or "5F (2.78C) "? Fig. 1b, is expressed in degrees Celsius, not degrees Fahrenheit, and both the New York Times and the Anchorage Daily News make reference to 5.4F, it is a reasonable assumption that the correct temperature is 3C, which was converted to 5.4F and then truncated to 5F in the CGC report. The New York Times attributed the 5.4F increase to the last 30 years beginning with the 1970's, Dr. Weller in his Anchorage Daily News interview stated that he was talking about the period from 1966 to 1995, while the CGC report appears to imply the period from 1961 to 1998. Let's examine Fig.1b of the temperatures in which we are interested, using an expanded image with a white background:

Figure 1b:      The Alaskan regional Climate and climate-related trends in Alaska.
Annual mean temperature composite from Anchorage, Fairbanks, Nome and Barrow.

Note: The temperature value for 1993 in Fig.1b is inconsistent with the values calculated from both the NCDC and GISS datasets. The CGC value for 1993 is -2.5C, the NCDC value is -1.97C and the GISS value is -1.90 C.

Temperature change scales of 5.4F (3C) was superposed at 1961 and 1966 and scales of 7F (3.87C) at 1964 and 1971. For 1961, a temperature delta of 5.4F (3C) lies between the minimum and maximum temperatures in range from 1961 to 1998, which is clearly the time interval referenced by the CGC report. However, using minimum and maximum temperatures to calculate the temperature change is not a valid statistical methodology. Using linear regression analysis, for the interval from 1961 to 1998, yields a temperature increase of 3.46F (1.92C).

For the interval from 1966 to 1995, the absolute difference in temperature between the two dates is 4.66F (2.59C) and liner regression analysis of the interval yields a temperature increase of 2.74F (1.52C), which is only half the value reported by Dr. Weller in his news paper interview. Using linear regression analysis, it is impossible to obtain the temperature increase reported by the CGC. In point of fact the only way to obtain a temperature change of 5.4F (3C) from Fig.1b is to take the difference between the maximum and minimum limits.

While an absolute temperature increase of 7F (3.87C) can be derived by using 1964 as the starting point, the New York Times explicitly stated that their time interval was from the 1970's. Using the CGC graph, with 1971 as the minimum temperature point, it is impossible to derive a 7F increase and we have to look else ware for it source. If the annual mean temperature for 1993 had been in agreement with the NCDC and GISS values, the difference 1971 and 1993 would have been 7.07F but the CGC and NYT missed that opportunity for anthropogenic global warming.

Before USNA report is investigated, there is one very interesting feature about Alaskan climate which should be reviewed. In a paper, by Stafford et al.  [20] Fig.2 shows a contour map for the mean annual temperature for the State of Alaska.  On that temperature contour map, locations 100 miles or more north of Anchorage have mean annual temperatures below the freezing point of fresh water 32F (0C). The average mean annual temperature for the entire State of Alaska is below the freezing point of water 32F (0C). As the CGC so correctly observed in a section of their report titled "Cross-Cutting Issues and Challenges of Future Climate Change" [15.2]:

 The Thawing of Alaska

The property of water to change phase at 32F (0C) is probably the most important factor in the observed and expected climate impacts in Alaska where the environment and practically all human activities are so strongly dominated by snow and ice. Much of Alaskas environment is close to the melting point of ice (e.g., Fairbanks has a mean annual temperature of about 28F or 2C) and a relatively small warming of the climate can cause major environmental changes and feedbacks.

In addition to the quality of the station's temperature data, one of the primary reasons that Alaskan climatologists use the widely dispersed stations of Anchorage, Fairbanks, Nome and Barrow as a reference, is that their composite annual mean temperature approximates the average conditions throughout Alaska.

In summary, the New York Times temperature increase of 5.4F is derived from Figure 1B of the CGC report but the value is for the wrong time interval, the CGC used a flawed statistical method to calculate warming. Compared to the results obtained using liner regression analysis, warming is overstated for the time interval by 170% to 200% depending on which stating and ending dates are selected.

Investigation of the USNA Report on Alaska

The USNA section for Alaska is presented in two sections, an "Overview" and a "Foundation" with more details.  The Alaskan Overview is presented in HTML [19.1] and PDF [19.2] formats. The Foundation report contains more details, and is published as a Black-and-White PDF [19.3] document and a Color Figure PDF [19.4] appendix. The CGC Alaskan group is acknowledged as contributing to the Foundation report. Page 5 of the Foundation Black-and-White PDF [19.3] document (page 287 of the printed edition) contains a section titled "CLIMATE VARIABILITY AND CHANGE" and a subsection titled "Observed Climate Trends". The following is the text from the subsection:

Observed Climate Trends

Alaskas recent climate has shown a strong warming trend. General Arctic warming began in the mid-19th century, but has accelerated in the past few decades (Overpeck et al., 1997). Alaska has warmed 4F (2C) since the 1950s on average, with the largest about 7F (4C) in the interior in winter (Chapman and Walsh, 1993; Weller et al., 1998). Local weather records show that the growing season in Alaska lengthened by 13 days since 1950 (Keyser et al., 2000). Much of the recent warming occurred suddenly around 1977, coincident with the most recent of the large-scale Arctic atmosphere and ocean regime shifts (Weller and Anderson, 1998).

Alaska has also grown substantially wetter over the 20th century. The sparse historical record since 1900 shows mixed precipitation trends, with increases of up to 30% in the south, southeast and interior, and smaller decreases in the northwest and over the Bering Sea. The trend to higher precipitation has been stronger recently, a 30% average increase over the region west of 141 degrees longitude (i.e., all of Alaska except the panhandle) between 1968 and 1990 (Groisman and Easterling,1994).

Alaskas recent warming is part of a strong trend observed throughout the circumpolar Arctic, except for one large region of cooling over eastern Canada and Greenland. This broad Arctic warming has been accompanied and corroborated by extensive melting, of glaciers, warming and thawing of permafrost, and retreat and thinning of sea ice (Echelmeyer et al., 1996;Sapiano et al.,1998;Lachenbruch and Marshall,1986;Osterkamp,1994;Osterkamp and Romanovsky, 1996;Wadhams 1990;Cavalieri et al., 1997;Serreze et al.,2000;Krabil et al.,1999; Dowdeswell et al.,2000). Paleoclimatic evidence suggests the Arctic is now warmer than at any time in the past 400 years (Overpeck et al., 1997). The start of Arctic warming in the mid-19th century indicates a contribution from natural factors (Overpeck et al., 1997). Of the stronger high-latitude warming of the past three decades, roughly half can be explained by changes in storm track patterns associated with natural patterns of climate variability, although it is possible that anthropogenic changes in radiative forcing may be shifting these patterns so they tend to favor high-latitude warming (Hurrell, 1995,1996). The remaining share of recent high-latitude warming, roughly half, is broadly consistent with model predictions of the consequences of anthropogenic greenhouse forcing (Serreze et al., 2000). Observations of vegetation and snowcover from satellites, and of the annual fluctuation of atmospheric CO2 concentration, further corroborate the broad warming trend over northern mid to high latitudes. Mean annual snowcover of the Northern Hemisphere decreased 10% from 1972 to 1992 (Groisman and Easterling,1994),while the growing season over northern mid- to high latitudes increased by 7 to 14 days (Myneni et al.,1997).

 

The equivalent subsections in the Overview HTML [19.1] and Overview PDF [19.2] contains the identical text, which follows:

Observed Climate Trends

Alaska has warmed substantially over the 20th century, particularly over the past few decades. Average warming since the 1950s has been 4F (2C). The largest warming, about 7F (4C), has occurred in the interior in winter. The growing season has lengthened by more than 14 days since the 1950s. Some records suggest that much of the recent warming occurred suddenly around 1977. Alaska has also grown wetter recently, with precipitation over most of the state increasing 30% between 1968 and 1990. The observed warming is part of a larger trend through most of the Arctic corroborated by many independent measurements of sea ice, glaciers, permafrost, vegetation, and snow cover. In contrast to other regions, the most severe environmental stresses in Alaska at present are climate-related.

We now have so many conflicting numbers for Alaskan temperature change, that a score card is needed. But just wait; there are more numbers to come. In addition to the narratives, this subsection also contained two graphics which showed "Alaska: 20th Century Annual-average Temperature" (Figure 2, Black-and-White PDF [19.3], page 5) and "Alaska: 20th Century Annual Total Precipitation" (Figure 3, Black-and-White PDF [19.3], page 6). In this report, we will focus on the temperature graphic, the Black-and-White copy for which follows with the PDF image expanded 260% to improve readability:

This graphic is extremely difficult to read. Based on an examination of the graphic and its defects, it appears that it may be the result of a 
(1) poor digital scan from a paper image, 
(2) a poor choice of color to grayscale mapping, 
(3) bad image resizing. or 
(4) all of the above. 

After the image was created in its current form, someone went to the trouble of redrawing the grid lines at the 30F and 36F tick marks. With the exception of where the temperature graph crosses the grid lines, the pixels are all the exact same shade of black. Given the poor quality of the graph, the only way that this could have occurred is through manual editing. Why were only these two grid lines added? The grid lines were added to create the impression that during the 20th Century Alaskan warming was more than 6F. On this graph, there are four dates which would be of interest to climate alarmists: 1956, 1964, 1971 and 1993. We can see why there would be interest in the 1950's or 1960's as a starting reference, but as the subject of this report is the New York Times story, let us focus on the last to dates of interest: 1971 and 1993. With the help of the added grid lines, it would appear that the absolute temperature difference between 1971 and 1993 was tantalizing close to 7F. There are five versions of this graph in the USNA which are documented in the References "Alaska: 20th Century Annual-average Temperature" [20]. The best image quality is at Color Figure PDF [19.4] on appendix page 1.

The graph which the CGC submitted to the USNA in December of 1999 covered the period from 1954 to 1998 and used the NCDC datasets for Fairbanks, Anchorage, Nome and Barrow. The Alaskan report published by the USNA in 2000 covered the period from 1900 to 1999 and also cites the Historical Climate Network, National Climate Data Center as the source. The problem with this citation is that the publicly accessible NCDC datasets for Alaska are fragmentary before 1954. The GISS datasets extend back to the 1930's but do not offer the same level of quality afforded by the NCDC datasets. Before the 1930's, Alaskan temperature data is not easily available online. So actual datasets used to create the "20th Century Annual-average Temperature" for Alaska is a mystery, as are the stations which were used to create the annual-average. The accreditation on the graphic may indicate that it was produced at the National Climate Data Center using unspecified data sources.

The following image was extracted from Color Figure PDF [19.4] appendix page 1, after the PDF image was expanded 260% and digitally enhanced to improve readability. The image was then edited to highlight the suspected area which the New York Times used to derive their 7F temperature increase for Alaska. It is interesting to note that the temperature difference between 1971 and 1997 could have been the source used by the New York Times' 5.4F connection, had the CGC not been so identified.

By counting pixels, it was determined that the absolute difference between the annual-average temperature for 1971 and 1993 was 6.83F. The only way that the New York Times could have reported a temperature increase of 7F the last 30 years, was to perform the above calculation and then roundup the value to 7F. When the New York Times was challenged about the source of their first temperature, they had to come up with a new value. The New York Times would have assumed that the Alaskan CGC was the source of the graphic which they used. As was previously noted, the graphic which was submitted to the USNA, was replaced when it reached Washington. When the New York Times contacted the CGC, the CGC response was the erroneous value they had originally submitted to the USNA, which was 3C or 5.4F. Unfortunately for the New York Times, the temperature change was for a different time period and used the same flawed methodology which the New York Times employed to calculate their original 7F value.

This report might have ended here if it were not for one small detail about the "Alaska: 20th Century Annual-average Temperature" [20] graph.

The annual-average temperature graph for Alaska, used in the "US National Assessment of the Potential Consequences of Climate Variability and Change" is not the annual-average temperature for Alaska

As was previously noted, the average mean annual temperature for the entire State of Alaska is below the freezing point of water 32F (0C). The USNA graph of 20th Century Alaskan annual-average temperature is at least 4F too warm. So the question is, exactly what is being depicted by the graph used by the USNA? The graph is definitely that of Alaska but it is not the annual average temperature. What the graph most likely represents is an unknown number of Alaskan stations, with over representation of stations about or south of Anchorage. The question is, how could the Alaskan CGC allow such an error to occur? The most likely answer is that once the CGC submitted their report to the USNA, their control over the final product was limited. The freezing point of water is so critical to Alaskan climatology, that it is very likely the creators of the graph were not Alaskan climatologists and were unaware of how critical the freezing point issue was. If the caption on the graph had been labelled, " Alaska: 20th Century Composite Annual-average Temperature For 24 Alaskan Stations", there would not have been a problem, but that is not what was published in the report.

As was already discussed, all of the versions of this graph were of extremely poor quality, compared to other graphs in the USNA report. Given the many problems with this graph, the question begs to be asked, were the readability issues deliberate or just the very sloppy application of technology? What is contained in this Alaskan graph which the USNA wouldn't want attention called to?

 Observations about the "Alaska: 20th Century Annual-average Temperature" Graph

a)    The graph is not an "Annual-average Temperature" for Alaska, because it appears to overstate the annual-average temperature by at least 4F.

b)    The graph may be the annual-average for a number of unspecified Alaska weather stations, with over sampling from the Southern Alaska panhandle.

c)     The multiple renditions of the graph were so blurry, that it causes one to suspect a deliberate effort was made to obscure its contents.

d)     Given the problems with Alaskan climatic data before 1950, better citation of database sources should have been made.

e)     Over sampling of the Southern Alaska stations may have masked the warming which is still occurring at Barrow station.

f)       Over the last 100 years, the temperature of Alaska has increased.

g)     During the 20th Century, there were three significant climatic periods in Alaska: 1900-1940 a warm period approximating conditions at the end of the Century; 1941-1975 a cool period with some temperatures returning to near 1900's levels; 1976-1999 a warm period with temperatures slightly exceeding those reached during the 1900-1940 period.

h)      During the 20th Century, over 80% of Alaskan warming occurred before 1940.

i)       The "Great Pacific Climate Shift" which occurred from 1976 to 1977, is clearly visible on the graph as an increase in temperature of over 2.5F.

j)        Most of the Alaskan warming which took place in the last quarter of the 20th Century occurred during the "Great Pacific Climate Shift" and is closely associated with a shift in the Pacific Decadal Oscillation (PDO) which occurred at the same time.

k)     Since 1978, Alaskan temperatures have been stable.

The caption on the graphic also states that: "Average temperatures in Alaska has increased over the 20th Century, with about 4F since the 1950s". Terms such as "about 4F" and "since the 1950's" are deliberately vague. Also, we do not know which stations were included in the graph or the datasets which were used. If we use the four Alaskan reference stations of Fairbanks, Anchorage, Nome and Barrow and the NASA-GISS dataset, which is available back to 1930 for these four stations, the temperature change from 1950 to 1999 can be calculated, using linear regression analysis. The yearly averages conform to the NCDC convention of January to December, not the December to November convention used by the GISS.

 

The results of linear regression analysis for the annual mean composite temperature for Fairbanks, Anchorage, Nome and & Barrow indicate the for the period from 1950 to 1999 the Alaskan temperature has increased by 2.91F (1.61C), not the 4F reported by the USNA. Given the lack of adequate accreditation for the "Alaska: 20th Century Annual-average Temperature" graph and the quality issues which have been identified, the assertion by the USNA that Alaskan temperatures has increased by +4F is open to question.

If the Alaskan temperature data is typical of the quality of the datasets in the "US National Assessment of the Potential Consequences of Climate Variability and Change Mega-Region: Alaska", the entire report should be reviewed

Conclusion 

The opening introduction for the 1960's science fiction television show "The Outer Limits" contains the line: "We can change the focus to a soft blur, or sharpen it to crystal clarity."  Perhaps this old science fiction show was the source of inspiration for the United States Global Change Research Program when they wrote the Alaska section of the "US National Assessment of the Potential Consequences of Climate Variability and Change" report, for science fiction is certainly what the report is! The graphic for the most the most critical element of Alaskan climatology, the temperature, was reduced to a "soft blur", lest anyone read it and understand its meaning. To compound this egregious offence to Alaskan climatology, the USNA alleged that this softly blurred graphic depicted the "Annual-average Temperature" for 20th Century Alaska, which it did not because it was too warm. Using this blurred data from a set of nebulous Alaskan weather stations, the USNA announced, without disclosing their methodology that Alaska's temperature had increased by 4F since the 1950's, a value which appears to be too warm by at least 1F. The staff of the USNA, must have considered their temperature in crease of 4F over 50 years to be modest because the Alaskan "Center for Global Change and Arctic System Research", who had submitted the original Alaskan Assessment to the USNA had reported the Alaskan temperatures had increased by 5F, 5.4F or 3C (pick one) over 40 years. Unlike the USNA, who were more modest in their analysis, the CGC apparently used the minimum and maximum temperature limits to determine the extent of temperature in their report.

Into this morass of conflicting numbers, the New York Times wandered, filled with the zeal of all true believers. Searching for signs and portents of the impending climatic apocalypse, they used the minimum and maximum temperature limits of the USNA Alaskan temperature graphic to mean that Alaskan temperature had increase by 7F over the last thirty years of the 20th Century. It is possible that the New York Times did not view the USNA Alaskan graphic directly but had its contents conveyed to them. One thing is very clear, the graph is the source of the disinformation, because the 7F temperature increase is dependent on the dates of 1971 and 1993. When the New York Times was challenged as to the correctness of the information they reported, there were two options available to them: (1) they could have reported 4F temperature increase which was used in the official USNA Alaskan report, or (2) turned to the CGC, which provide the New York Times with their original 5.4F temperature increase. The New York Times would have had no way of knowing that the second value was just a flawed as the first. The real mystery is why didn't the New York Times use the official USNA value of 4F for Alaska?  I suspect they would have to admit that they had committed a serious error and 4F over 50 years, is not a dramatic as 7F over 30 years.

Alaskan climatology is a very serious issue and it is unfortunate that the USNA did not treat it with the clarity it deserves. . Large sums of money and a significant amount of human effort have gone into the creation of a report whose primary objective is the advancement of a political agenda and not science. Hopefully this omission will be corrected in the future.


References

1.      "Alaska, No Longer So Frigid, Starts to Crack, Burn and Sag", by Timothy Egan, June 16, 2002 (http://www.nytimes.com/2002/06/16/national/16ALAS.html).

2.      "U.S. Climate Action Report - 2002", June 3, 2002 (http://www.epa.gov/globalwarming/publications/car/uscar.pdf), page 91.  

3.      "United States Environmental Protection Agency", (http://www.epa.gov/).

4.      "The FANB temperature record", by Dr. Sue Ann Bowling (http://climate.gi.alaska.edu/Bowling/FANB.html).

5.      "The Alaska Climate Research Center", (http://climate.gi.alaska.edu/).

6.      "In response to the New York Times Article of 16 June 2002 and Op/Ed of 24 June 2002" by Brian Hartmann & Dr
   Gerd Wendler, June 24, 2002 & updated July 12, 2002 (http://climate.gi.alaska.edu/nytimes.html).

7.      "New York Times Correction", July 11, 2002 (http://www.nytimes.com/2002/07/11/pageoneplus/corrections.html).

8.      "Center for Global Change and Arctic System Research",  (http://www.cgc.uaf.edu/).

9.      "Organization of Center for Global Change and Arctic System Research", (http://www.cgc.uaf.edu/organization.html).

10.  "GLOBAL WARMING DEBATE INVOLVES . . . Dueling digits", July 19, 2002 (http://www.adn.com/voice/story/1443249p-1561516c.html).

11.  "Alaska Climate Summaries", (http://www.wrcc.sage.dri.edu/summary/climsmak.html).

12.  "Western Regional Climate Center", (http://www.wrcc.dri.edu/index.html).

13.  "National Climatic Data Center, Regional Climate Centers", (http://lwf.ncdc.noaa.gov/oa/climate/regionalclimatecenters.html).

14.  "Goddard Institute for Space Studies, Surface Temperature: Station Data", (http://www.giss.nasa.gov/data/update/gistemp/station_data/).

15.  "The Potential Consequences of Climate Variability and Change" by Gunter Weller, Patricia Anderson & Bronwen Wang, December 1999 (http://www.besis.uaf.edu/regional-report/regional-report.html).

15.1. "Preface and Executive Summary" (http://www.besis.uaf.edu/regional-report/Preface-Ex-Sum.pdf), page 5.

15.2.        "Cross-Cutting Issues and Challenges of Future Climate Change" (http://www.besis.uaf.edu/regional-report/Issues1.pdf).

16.  "United States Global Change Research Program", (http://www.usgcrp.gov/).  

17.  "Alaska Regional Assessment Group", (http://www.usgcrp.gov/usgcrp/nacc/alaska.htm).

18.  "US National Assessment of the Potential Consequences of Climate Variability and Change", 2000 (http://www.usgcrp.gov/usgcrp/nacc/default.htm).   

19.  "US National Assessment of the Potential Consequences of Climate Variability and Change Mega-Region: Alaska", 2000, by National Assessment Synthesis Team, (http://www.usgcrp.gov/usgcrp/nacc/alaska-mega-region.htm).

19.1.        "Climate Change Impacts on the United States The Potential Consequences of Climate Variability and Change Overview: Alaska", by National Assessment Synthesis Team, (http://www.usgcrp.gov/usgcrp/Library/nationalassessment/overviewalaska.htm).

19.2.        "Climate Change Impacts on the United States The Potential Consequences of Climate Variability and Change Overview: Alaska", by National Assessment Synthesis Team, (http://www.usgcrp.gov/usgcrp/Library/nationalassessment/11AK.pdf).

19.3.        "Potential Consequences Of Climate Variability And Change For Alaska", by Edward A. Parson with Lynne Carter, Patricia Anderson, Bronwen Wang & Gunter Weller, (http://www.usgcrp.gov/usgcrp/Library/nationalassessment/10Alaska.pdf).

19.4.        "Potential Consequences Of Climate Variability And Change For Alaska, Color Figures Appendix", (http://www.usgcrp.gov/usgcrp/Library/nationalassessment/10C.pdf).

20.  "Alaska: 20th Century Annual-average Temperature", this graph appear in all of the National Assessment reports on Alaska [19] with five unique images:  

20.1.        URL: http://www.usgcrp.gov/usgcrp/Library/nationalassessment/images/AKObserved-b.jpg, referenced by [19.1], very small poor quality color image.

20.2.
http://www.usgcrp.gov/usgcrp/Library/nationalassessment/LargerImages/RegionGraphics/Alaska/Observed.jpg referenced by [19.1], medium size color image of low quality.

20.3.   URL: http://www.usgcrp.gov/usgcrp/Library/nationalassessment/11AK.pdf, page 1, referenced by [19.2], medium size color image of low quality.

20.4.   URL: http://www.usgcrp.gov/usgcrp/Library/nationalassessment/10Alaska.pdf, page 5, referenced by [19.3], extremely poor quality black and white image.

20.5.   URL: http://www.usgcrp.gov/usgcrp/Library/nationalassessment/10C.pdf, page 1, referenced by [19.4], best quality color image in set but still of low quality.

21.  "Climate Change Science: An Analysis of Some Key Questions" by National Academy of Sciences, 2001 (http://books.nap.edu/html/climatechange/climatechange.pdf).

22.  Stafford, J.M., Wendler, G., Curtis, J. (2000) "Temperature and precipitation of Alaska: 50 year trend analysis", Theor. Appl. Climatol. 67, 33-44.


Return to `Climate Change Guest Papers' page

Return to `Still Waiting for Greenhouse' main page


FastCounter by bCentral