 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
The U.S. National Climate Change Assessment: Do the Climate Models Project a Useful Picture of Regional Climate?
Subcommittee on Oversight and Investigations
Dr. James J. O'Brien
Summary of Dr. O'Brien's Major Points
INTRODUCTION I have been a physical scientist in oceanography and meteorology for 40 years. In my early years, my graduate students and postdocs concentrated primarily on modeling time dependent ocean motions. In the late 1970's and 1980's, we contributed to the physical understanding of El Nino. Namely, how it works and how it can be forecast. In the 1990's, while most other scientists were applying ENSO forecasts to tropical countries, my students and I have concentrated on impacts in the United States. We have written papers on: ENSO and Atlantic Hurricanes ENSO and Tornadoes ENSO and Precipitation ENSO and Temperature ENSO and Wild Fires (In Florida) ENSO and Snowfall ENSO and Excessive Wind Events ENSO and Great Lakes Snow Events ENSO and Freezes in Central Florida In 1999, I accepted the pro bono job as official State of Florida Climatologist. We have been advising the Florida Commissioner of Agriculture on wild fire forecasts, droughts, hurricanes, etc. We provide climate advice to the citizens of Florida for all sectors, but, particularly agriculture, forestry, fisheries, tourism and power generation. In some local circles, I am labeled, Dr. El Nino for my research. Turning now to the National Regional Assessment of Climate Change, I was the Co-Chair for the Southeast Regional Assessment. (My Co-Chair, Dr. Ron Ritschard, recently died at a young age). Our work was funded by NASA, Huntsville, Alabama. In the early beginnings of the National Regional Assessment, the entire U.S. team met and we agreed there would be "ONE" Global CO2 doubling model so everyone referring to future projections would be on the same page. There were two choices: (1) The Hadley, (British model), or (2) The Max Planck (German model). The Hadley model was selected. Subsequently, I attended a meeting of the U.S. National Resource Board Committee on Climate. Many senior scientists were shocked that there was no American models. In due time, a very new recent model, the Canadian Model was added for our use. It was recently computed so no documentation or response of this model was available to the assessment teams. In my opinion, the Hadley model is a state-of-the-art model with poor resolution, inadequate physics - particularly, the ocean component. The Hadley Model gives reasonable projections, but it is still flawed and I am sure that, in due time, will be improved. As better ocean models are improved in climate models, the future changes are greatly reduced. The Canadian's model is flawed, and, in my opinion, should never have been used. My effort to capture the attention of the leaders to recognize this were rejected outright. My team discovered that, initially, the data provided to the team had been incorrectly registered with respect to the geography of the U.S. Since the model has horizontal grid boxes around 500 km on a side, being set off by one grid, really confuses geographic identification. (As an aside, I do not know if this was fixed, but I was told it can't make any difference). I really enjoy learning about climate variability over the United States, such as droughts, floods, freezes, hurricanes, etc. For the citizen for whom climate is important, it is the variability which matters! It is not whether the average temperature will rise 3 - 5ºF in 100 years. The citizen wants to know "Is this winter going to be colder than normal?" and other simple questions. I discovered that we could not find ENSO variability in the ocean model of the Canadian model. I was told it was there, but it makes no difference if it is too small. Mr. Chairman, the variability of climate over most of the United States is primarily controlled by ENSO and other ocean-related phenomena (North Atlantic Oscillation, and the Pacific Decadal Oscillation) and land use changes. I cannot accept a 100 year climate run as useful if it doesn't also include the observed variability in the climate system. What is the climate system? It is the entire atmosphere, ocean, land, ice systems which are heated by the sun. The chemistry of global climate change is completely correct. We have an excellent scientific understanding of how radiatively-active gases such as carbon dioxide, methane and water vapor can delay heat in the climate system. There is an assumption that this extra heat will manifest itself in raising the temperature of the biosphere - that portion of the climate system in which we humans live. The data measured from the actual climate system seems to indicate other processes are dominate, such as stronger mid-latitude storms which are important for distributing the extra heat. In my opinion, even the current models are not capable in calculating the climate system well enough for policymakers to believe in any projection. Can we do it better? I believe we can, but it will take a new effort and considerable investment. We know most of the physics of the climate system. In order to calculate the variability of the system, we need adequate computers. We need the kind of investment in computers that the Congress funds to DOD, NSF, NSA, etc. I propose a "Manhattan Type Project" to estimate future climate variability for our National Security. Any future climate change will probably require trillions of dollars to adjust our culture or mitigate the consequences. My vision is a NEW Institute, outside the government with top management, the best scientists and adequate resources. My estimate is $50M/year for at least 10 years. When I suggest this, OMB folks usually ask me, "Dr. O'Brien, where are we going to find that money?" My answer is, "Give us 2 attack helicopters' monies, and we will be happy for a few years. Give us a fighter jet monies, we will be very happy for a few years. Give us an aircraft carrier monies, and we will never ask for any more resources". The Congress has to decide on the priorities. Do we want to understand the future climate or not? Returning to my belief, that we can do better in modeling climate, I am encouraged that each generation of climate change modeling gets better. The original CO2-doubling model by NASS,GISS under the leadership of Dr. Jim Hansen, estimated around 10ºF surface temperature change by 2050. This was so dramatic because no ocean was included. I remember reading in the Tallahassee Democrat, a story that said, as a result of the GISS model, that sea-level would rise 3 meters or 7 - 10 feet by 2050. The current IPCC estimates a few degrees temperature rise by 2100 and a doubling of the current sea-level rise of around 8 - 10 inches to 20 inches as the worst case by 2100. Certainly, policymakers will react differently to plan for a two feet rise in 100 years vs. 10 feet rise in a generation. Let me provide one more remark on sealevel rise. In order to double sealevel, one would expect to observe an increased rate of rise by 2002. Everyone agrees that the current average rise is about 7 - 10 inches a century, averaged over the globe. However, the experts who have tried to find any acceleration find none. How about global warming in the United States? I will leave this subject to my fellow climatologist, Dr. Tom Karl. I am, however, the State Climatologist of Florida. In Florida, the cities are warming at the rate of about one degree in the entire 20th century. But the rural places are cooling at the rate of more than one degree per century. I have included some graphics in my presentation documenting this for minimum temperature over the entire 20th Century. What is happening? My fellow, State Climatologist, from Colorado, Dr. Roger Pielke, Sr., explains this by land use changes. Dr. Karl has published work showing the cooling in the Southeast United States, but unfortunately, the summary of average temperature in Florida in the last century, in the S.E. Assessment summary, shows Florida warmer than the rural data would dictate. Finally, the ocean part of the global climate system models are very inadequate. The research community is aware that warm and cold ocean currents are very important in predicting the weather even for 10 days. It is critical to model the oceans correctly if a global climate model is expected to work at all. A young French ocean modeler said to me recently, "I hope that the ocean models used by global climate models look like the real ocean before I die!" There are hundreds of
scientists other than climate modelers that have been told the Hadley and
Canadian models are good projections of the future.
This is a shame. When I
joined the U.S. National Assessment Team as Co-Chair of the Southeast Regional
Assessment, a bright young EPA ecologist from Louisiana reported to me that the
number of hurricanes in the Gulf of Mexico were increasing due to global
warming. I was unaware of this.
Consequently, my students and I did a study.
We found, that, in fact, the number of hurricanes have decreased
significantly in the Gulf of Mexico. This
is a published paper. 1998: Are
Gulf Hurricanes Getting Stronger? Bull.
Amer. Meteorol. Soc., 79(7), pp. 1327-1328 (with Bove, M.C., and D. F. Zierden). CONCLUSION Global climate changes will occur. Whether surface temperatures will increase due to radiatively-active emissions is not clear. The Global Climate System must change. In order to address what the nation needs to do, I recommend a large investment in improving the basic understanding by investing in very good global climate system calculations.
|
|||||||||||
|
