Hearings - Testimony
Full Committee Hearing
The Role of Science in Environmental Policy-Making
Wednesday, September 28, 2005
Dr. William Gray
Department of Atmospheric Science, Colorado State University

Mr. Chairman and Members of the Committee, I am William M. Gray, a Professor of Atmospheric Science at Colorado State University in Fort Collins, Colorado. I have been studying and forecasting weather and climate for over 50 years (see my attached Vitae). My specialty has been tropical meteorology and tropical cyclones. I have made Atlantic basin seasonal hurricane forecasts for the last 22 years.

Over the last 20 years, I have been dismayed over the bogus science and media-hype associated with the nuclear winter and the human-induced global warming hypotheses. My innate sense of how the atmosphere-ocean functions does not allow me to accept either of these scenarios. Observations and theory do not support these ideas. The nuclear winter hypothesis did not recognize that the globe's hydrologic cycle operates on a time scale of 8-10 days and that nuclear- spawned dust material would be quickly rained out of the atmosphere. The human-induced global warming scenarios have a major flaw in that they accept the view that an increase in the global hydrologic cycle will cause enhanced upper-tropospheric water vapor gain and a suppression of outgoing long wave radiation (OLR) to space. The opposite is true. Global Climate Models (GCMs) are also not able to realistically predict the ocean's deep water circulation which is fundamental to any understanding of global temperature change.

As a boy, growing up here in Washington, DC, I remember the many articles on the large global warming that had occurred between 1900 and 1940. No one understood or knew if this warming would continue. Then the warming abated, and a weak global cooling trend set in from the mid-1940s to the early 1970s. The global warming talk ceased and speculation about a coming ice age came into vogue. I anticipate that the trend of the last few decades of global warming will come to an end, and in a few years we will start to see a weak cooling trend similar to that which occurred from the mid-1940s to the early 1970s.

I would like to present a different view on the likelihood of human-induced global warming and also provide evidence that global hurricane activity has not increased as the globe has warmed in recent decades. There is no significant correlation between global warming and global hurricane activity.


Although initially generated by honest scientific questions, this topic has long ago advanced into the political arena and taken on a life of its own. It has been extended and grossly exaggerated and misused by those wishing to make gains from the exploitation of ignorance on this subject. This includes many governments of western countries, the media, and scientists who were willing to bend their objectivity to obtain government grants for research. It is unfortunate that most of the resources for climate research come from the federal government. When a national government takes a political position on a scientific topic, the wise meteorologist or climatologist either joins the crowd or keeps his/her mouth shut. Scientists can be punished if they do not accept the current views of their funding agents. An honest and objective scientific debate cannot be held in such a political environment.

I have closely followed the greenhouse gas warming arguments. From what I have learned of how the atmosphere functions in over 50 years of study and forecasting, I have been unable to convince myself that a doubling of human-induced greenhouse gases can lead to anything but quite small and likely insignificant amounts of global warming (~ 0.2-0.3 degree C).

Most geophysical systems react to forced imbalances by developing responses which oppose and weaken the initial forced imbalance; hence, a negative feedback response. Recently proposed human-induced global warming scenarios go counter to the foregoing in hypothesizing a positive feedback effect. They assume that a stronger hydrologic cycle (due to increased anthropogenic greenhouse gases) will cause additional upper-level atmospheric water vapor. This increased vapor results in a reduction of OLR loss to space and causes additional warming (Fig. 1). This positive water vapor feedback assumption allows the small initial warming due to human-induced greenhouse gases to be unrealistically multiplied 8-10 times. This is where much of the global modeling is in error. As anthropogenic greenhouse gases increase it does not follow that upper-level water vapor will increase. If it does not, little global warming will result. Observation of middle tropospheric water vapor over the last few decades shows that water vapor has in fact been undergoing a small decrease. The assumed positive water vapor feedback as programmed into the GCM models is not occurring. Energy budget studies indicate that if atmospheric water vapor and the rate of condensation were held fixed, a doubling of carbon dioxide would cause only a small (~ 0.2 - 0.3 degree C) global warming. This can be contrasted to the 2-5o C warming projected in the models.

The other primary physical limitations of the GCM simulations are their inability (as yet) to properly treat the global ocean deep circulation. This requires the need to model ocean salinity variations. Climate change cannot be objectively discussed without a realistic treatment of the ocean.

Figure 1. Illustration of the relative magnitude of the suppression of outgoing long-wave radiation (OLR) by water vapor and CO2. The global models assume that as CO2 doubles, water vapor increases and causes more suppression of OLR and warming. The opposite is true.

Skillful initial value GCM climate prediction is not possible and probably never will be. This is due to the complex nature of the atmosphere/ocean system and the inability of numerical models to realistically represent this physical complexity. Realistic features currently cannot be forecast more than a week or two into the future (see Figs. 2 and 3). Imperfect representations of the highly non-linear parameters of the atmosphere-ocean system tend to quickly degrade (the so-called butterfly influence) into unrealistic flow states upon long period integration. Short-range prediction is possible up to a week or 10 days into the future because there tends to be conservatism in the initial momentum fields which can be extrapolated for short periods. But beyond about 1-2 weeks, the multiple unknown and non-linear energy-moisture exchanges within the earth system become dominant. Model results soon decay in chaos. Numerical climate models cannot now and likely never will be able to be accurately forecast more than a few weeks into the future. If skillful GCM climate forecasts were possible, we would be eager to follow their predictions. Currently, GCMs do not make seasonal or yearly forecasts. How can we trust climate forecasts 50 and 100 years into the future (that can’t be verified in our lifetime) when they are not able to make shorter seasonal or yearly forecasts that could be verified? They know that they dare not issue shorter forecasts because they are aware that they have little or no skill.

Figure 2. Illustration of atmosphere-land ocean modeling complexity. It is impossible to write computer code to represent such complexity and then realistically integrate hundreds of thousands of time steps into the future.

Besides the physical uncertainty concerning how to represent the complexity of the atmosphere-ocean system in quantitative terms, climate models have become too complex for any one person or team to understand. Due to the great complexity of the GCM system, the true reasons for success or failure often cannot be determined. These models have been developed by teams of specialists who concentrate on different parts of their model. No one person is able to understand the whole GCM simulation. Most model developers are talented and skilled technicians. However, few have ever given real-world weather briefings or made operational weather forecasts.

Figure 3. Illustration of the two methods of climate prediction. The top diagram shows how numerical climate prediction is made and loses skill rapidly. It does not use past data. The bottom diagram shows how statistical prediction is based on past data and can utilize associations that are not physically understood.

The potential for climate modeling mischief and false scares from incorrect climate model scenarios is enormous. Numerical modeling output gives an air of authenticity which is not warranted by the input physics and long periods of integration. How many more climate scares are we to see from climate models which are not able to realistically predict past and future climate changes let alone future decadal or century changes?

Many of my older meteorological colleagues are very skeptical of these anthropogenic global warming scenarios. But we are seldom asked for any input. Despite my 50 years of meteorology experience and my many years of involvement in seasonal hurricane and climate prediction, I have never been asked for input on any of the International Panels on Climate Change (IPCC) reports. They know my views and do not wish to have to deal with them. Many other experienced but skeptical meteorologists and climatologists are also ignored. I find that the summary page conclusions of the IPCC reports frequently do not agree with the extensive factual material contained within them. In fact, the summary conclusions of many of the IPCC reports give the impression they were written before the research is done.

It is disappointing that more atmospheric scientists have not spoken out about the reality of human-induced global warming and the reliability of the GCM simulations. It is also mystifying to me how the global warming advocates are able to get away with the argument that extreme weather events have become more prevalent in recent years and that they likely have a human-induced component. Such assertions are factually wrong.

There is nothing we humans can do to prevent natural climate change, which I believe nearly all the recent global temperature rise is due too. We have no choice but to adapt to future climate changes. Restricting human-induced greenhouse gas emissions now, on the basis of their assumed influence on global warming, is not a viable economic option, even if it were politically possible. China and India would never restrict their growing fossil fuel usage. Restricting greenhouse gas emissions would have little or no effect on global temperature. We need to keep the western world economies vibrant if for no other reason than to be able to afford the needed large technical research funding that will be required to develop future non-fossil fuel energy sources.

I am convinced that in 15-20 years, we will look back on this period of global warming hysteria as we now look back on so many other popular, and trendy, scientific ideas -- such as the generally accepted Eugenic theories of the 1920s and 1930s that have now been discredited. There are so many other more important problems in the world which need our immediate attention. We should not be distracted by a false threat that is mostly just due to natural changes in climate.


The Atlantic has large multi-decadal variations in major (category 3-4-5) hurricane activity. These variations are observed to result from multi-decadal variations in the North Atlantic Thermohaline Circulation (THC) - Fig. 4. When the THC is strong, it causes the North Atlantic to have warm or positive Sea Surface Temperature Anomalies (SSTA) and when the THC is weak, cold SSTAs prevail. Figure 5 shows these North Atlantic SSTAs over the last century with a projection for the next 15 years.

We observe that there are significantly more Atlantic basin major hurricanes when the THC is strong than when it is weak. Figure 6 shows the sum of tracks of Atlantic major hurricane tracks during a 20-year period when the THC was strong (left) versus an 18-year period when it was weak (right). Note the large differences. Figure 7 gives an illustration of how fortunate peninsula Florida was in terms of landfalling hurricanes during the period of 1966-2003 in comparison with the earlier period of 1932-1965. The varying strength of the Atlantic THC is partly responsible for these differences. Luck also played a role. There were many intense hurricanes just off the Florida coast during the later period that did not come ashore (i.e., Hurricane Floyd, 1999).

Figure 4. Idealized Atlantic Thermohaline Circulation (THC) that becomes stronger and weaker on multi-decadal time periods. More major hurricanes form in the Atlantic when it is stronger than when it is weaker.

Figure 5. Last century Sea Surface Temperature Anomaly (SSTA) in the North Atlantic showing multi-decadal periods of warm and cold anomalies and a projection of these SSTAs to 2020. More major hurricanes form when SSTAs are positive and fewer when they are negative.

Figure 6. Tracks of major hurricanes in 20 years (1950-1969) when the thermohaline circulation was strong and the North Atlantic had positive SSTAs (left) and in 18 years (1970-1987) when the thermohaline circulation was weak and the North Atlantic had negative SSTAs (right).

Figure 7. Comparison of Florida peninsula landfalling major hurricanes in a 33 year period (1933-1965 -- 11 landfalling major hurricanes) and in a later 38 year period (1966-2003 -- 1 landfalling major hurricane).

Recent major hurricanes Katrina and Rita and last year's four U.S. landfalling major hurricanes have spawned an abundance of questions concerning the role that global warming might be playing in these events. The ideas that global warming was the cause for these last two years of greater hurricane activity has been greatly enhanced by two recent papers presenting data to show that global tropical cyclones have become more intense in recent years. They tie this increased hurricane activity to global warming. These papers are:

a) Kerry Emanuel, 4 August 2005: Increasing destructiveness of tropical cyclones over the past 30 years. Nature, 436, 686-688.

b) P.J. Webster, G.J. Holland, J. Currie and P. Chang, 16 September 2005: Changes in tropical cyclone number, duration, and intensity in a warming environment. Science, 309, 1844-1846.

The near universal reference to these two papers over the last two weeks by most major media outlets is helping to establish a belief among the general public and scientists not involved in tropical cyclone studies that global hurricane intensity has been rising and that global warming is primarily responsible. This conclusion is not valid. The authors have improperly handled their data sets and their findings should not be accepted. These papers require a response from a few of us who study hurricanes. I feel I have an obligation to make formal comments on these papers (to the editors of the journals), which I will do in another week or two.


There always has been, and there probably always will be, problems in assigning a representative maximum surface wind to a hurricane. As technology advances and the methods of determining a hurricane’s maximum winds change, different values of maximum winds will be assigned to hurricanes than would have been assigned in previous years.

With the availability of new aircraft deployed inertial dropwindsondes and the new step-frequency surface wind measurement instruments, it is being established that Atlantic hurricane surface winds are sometimes stronger than were previously determined from wind values extrapolated from aircraft altitude. Saffir/Simpson category numbers in the Atlantic due to these changes in measurement techniques have risen slightly in recent years. Although most of the comparative differences in the 38 major hurricanes of the last 10 years in the Atlantic basin (1995-2004) vs. the 14 major hurricanes of the prior 10 years (1985-1994) is thought to represent real variability, a small part of this difference may be due to the assignment of a Category 3 or Category 4 status to a hurricane which in earlier years might have received a one category lower designation.


Despite what many in the atmospheric modeling community may believe, there is no physical basis for assuming that global tropical cyclone intensity or frequency is necessarily related to global temperature. As the ocean surface warms, so does the upper air to maintain conditionally unstable lapse-rates and global rainfall rates at their required values. Although there has been a general warming of the globe and an increase of SSTs in recent decades, observations do not show increases in tropical cyclone frequency or intensity.


The NOAA reanalysis of global mean temperature difference over the last two 10-year periods have shown that the mean annual global surface temperature has risen 0.39 degree C from the 10-year periods of 1985-1994 to 1995-2004. This is a substantial increase in global temperature (rate of 3.9 oC per century). Table 1 shows the number of measured major hurricanes around the globe (excluding the Atlantic). Major hurricanes have not gone up in the more recent 10-year period when SSTAs have warmed considerably.

Table 1. Comparison of observed major (Cat. 3-4-5) hurricanes-typhoons in all global basins (except the Atlantic) in the two most recent 10 year periods of 1985-94 and 1995-2004. The summertime sea surface temperature increases between these two 10-year periods are shown in the right column.

The Atlantic has seen a very large increase in major hurricanes during the last 10-year period in comparison to the previous 10-year period (38 between 1995-2004 vs. 11 during 1985-1994). The large last decade increase is a result of multi-decadal fluctuations in the Atlantic Ocean thermohaline circulation (THC). Changes in salinity are believed to be the driving mechanism. These multi-decadal changes have also been termed the Atlantic Multi-Decadal Oscillation (AMO). Even when the large increase in Atlantic major hurricane activity is added to the non-Atlantic global total of major hurricanes, there is no significant global difference (208 vs. 218) in the numbers of major hurricanes between the two periods.

Comparison of Atlantic hurricane activity between the last 15-year active period (1990-2004) with the activity during the active 15-year period of 1950-1964.

There have been hurricane periods in the Atlantic in the past which have been just as active as the current period. A comparison of the last 15 years of hurricane activity with an earlier 15-year period from 1950-64 shows no significant difference in the more intense major hurricanes (Table 2). Note that there has actually been a slight decrease in major hurricane numbers in the most recent 15 years. The number of weak tropical Named Storms (NS) rose by over 50 percent, however. This is a reflection of the availability of the satellite in the later period. It would not have been possible that a hurricane, particularly a major hurricane, escaped detection in the earlier period. But many weaker systems far out in the Atlantic undoubtedly went undetected before satellite observations.

Table 2. Comparison of Atlantic tropical cyclones of various intensities between 1950-1964 and the recent 15 year period of 1990-2004.

Change in Intensity Measurement Technology of the Northwest (NW) Pacific and Comparison of Earlier and Later Periods

This most active of the tropical cyclone basins had aircraft reconnaissance flights during the period 1945-1986 but has not had aircraft reconnaissance since. The satellite has been the only tool to track NW Pacific typhoons since 1987.

There was an anomaly in the measurement of typhoon intensity in the 14-year period of 1973-1986 when the Atkinson-Holliday (1977) technique for typhoon maximum wind and minimum sea-level pressure (MSLP) was used. This technique is now known to have significantly underestimated the maximum winds of the typhoons in comparison with their central pressures. This has been verified by a combination of satellite-aircraft data from the Atlantic and pre-1973 NW Pacific aircraft-measured wind and MSLP. Table 3 shows the official average of the annual number of super typhoons in the West Pacific (equivalent to the number of category 3-4-5 or major hurricanes of the Atlantic). Note that between 1950-1972 and over the last 18 years, this number of super-typhoons has averaged about five per year while during the Atkinson-Holliday period of 1973-1986 it was less than half this number. Weaker storm numbers during the 1973-1986 period were the same. If we disregard this anomalous 1973-1986 period and compare annual frequency of super-typhoon activity between 1950-1972 versus 1987-2004 we see little difference despite the recent global warming trend.

Table 3. Comparison of the annual average of super- typhoon activity in three multi-decadal periods in the western North Pacific. The middle period (1973-1986) used the Atkinson-Holliday (1977) intensity scheme. This reported maximum wind values that were too low.


I fully subscribe to the view expressed by Max Mayfield, Director of the NOAA National Hurricane Center when he stated last week before the Senate Committee of Commerce, Science and Transportation Sub-Committee: "We believe this heightened period of hurricane activity will continue due to multi-decadal variance, as tropical cyclone activity in the Atlantic is cyclical. The 1940s through the 1960s experienced an above average number of hurricanes, while the 1970s into the mid-1990s averaged fewer hurricanes. The current period of heightened activity could last another 10-20 years. The increased activity since 1995 is due to natural fluctuations/cycles of hurricane activity, driven by the Atlantic Ocean itself along with the atmosphere above it and not enhanced substantially by global warming. The natural cycles are quite large with an average 3-4 major hurricanes a year in active periods and only about 1-2 major hurricanes annually during quiet periods, with each period lasting 25-40 years".

I also subscribe to the views expressed in the new paper titled "Hurricanes and Global Warming" which will soon be published in the Bulletin of the American Meteorological Society. This paper is authored by [Roger Pielke, Jr., Director, Center for Science and Technology, U. of Colorado; Christopher Landsea, Director of Research, NOAA National Hurricane Center, Miami, FL; Max Mayfield, Director, National Hurricane Center, Miami, FL; James Laver, Director, NOAA National Climate Center, Washington, DC; and Richard Pasch, Hurricane Specialist, NOAA National Hurricane Center, Miami, FL] and makes the following statements: "Since 1995 there has been an increase in frequency and in particular the intensity of hurricanes in the Atlantic. But the changes of the past decade are not so large as to clearly indicate that anything is going on other than the multi-decadal variability that has been well documented since at least 1900 (Gray et al. 1997; Landsea et al. 1999; Goldenberg et al. 2001)"......


"Globally there has been no increase in tropical cyclone frequency over at least the past several decades (Lander and Guard 1998, Elsner and Kocher 2000). In addition to a lack of theory for future changes in storm frequencies, the few global modeling results are contradictory (Henderson-Sellers et al. 1998; IPCC 2001)"


Analysis of global tropical cyclone activity of all intensities does not support the hypothesis that there has been a significant increase in tropical cyclone frequency-intensity associated with global temperature rise.


Atkinson, G.D. and C.R. Holliday, 1977: Tropical cyclone minimum sea level pressure/maximum sustained wind relationship for the western North Pacific. Mon. Wea. Rev., 105, 421-427.

Elsner, J.B and B. Kocher, 2000: Global tropical cyclone activity: A link to the North Atlantic Oscillation. Geophysical Research Letters, 27, 129-132.

Goldenberg, S.B., C.W. Landsea, A.M. Mestas-Nunez and W.M. Gray, 2001: The recent increase in Atlantic hurricane activity: Causes and implications. Science, 293, 474-479.

Gray, W.M., J.D. Sheaffer and C.W. Landsea, 1997: Climate trends associated with multidecadal variability of Atlantic hurricane activity. “Hurricanes: Climate and Socioeconomic Impacts.” H.F. Diaz and R.S. Pulwarty, Eds., Springer-Verlag, New York, 15-53.

Henderson-Sellers, A., H. Zhang, G. Berz, K. Emanuel, W. Gray, C. Landsea, G. Holland, J. Lighthill, S-L. Shieh, P. Webster and K. McGuffie, 1998: Tropical cyclones and global climate change: a post-IPCC assessment. Bulletin of the American Meteorological Society, 79, 9-38.

Lander, M.A. and C.P. Guard, 1998: A look at global tropical cyclone activity during 1995: Contrasting high Atlantic activity with low activity in other basins. Mon. Wea. Rev., 126, 1163-1173.

Landsea, C.W., R.A., Pielke, Jr., A.M. Mestas-Nunez and J.A. Knaff, 1999: Atlantic basin hurricanes: Indices of climate changes. Climate Change, 42, 89-129.

Majority Office
410 Dirksen Senate Office Bldg.Washington, DC 20510-6175
phone: 202-224-6176
Minority Office
456 Dirksen Senate Office Bldg.Washington, DC 20510-6175
phone: 202-224-8832