by Judith Curry
The first 20 years.
The American Meteorological Society is publishing an online monograph The Atmospheric Radiation Measurement (ARM) Program: The first 20 Years. So far, the monograph includes 30 articles/chapters.
One of my earliest blog posts Confidence in Radiation Transfer Models drew heavily from ARM.
I’ve included a selection of these articles below, focusing on the data system and radiation measurements and modeling:
- Introduction. D. D. Turner, R. G. Ellingson [pdf]
- The ARM Data System and Archive. Raymond McCord, Jimmy Voyles [pdf]
- The ARM Data Quality Program. Randy A. Peppler, Kenneth E. Kehoe, Justin W. Monroe, Adam K. Theisen, Sean T. Moore [pdf]
- Spectral Radiation Measurements and Analysis in the ARM Program. E. J. Mlawer, D. D. Turner [pdf]
- Contributions of the ARM Program to Radiative Transfer Modeling for Climate and Weather Applications. Eli J. Mlawer, Michael J. Iacono, Robert Pincus, Howard W. Barker, Lazaros Oreopoulos, David L. Mitchell [pdf]
- ARM Solar and Infrared Broadband and Filter Radiometry
Joseph J. Michalsky, Charles N. Long [pdf]
- ARM’s Progress on Improving Atmospheric Broadband Radiative Fluxes and Heating Rates. Sally A. McFarlane, James H. Mather, Eli J. Mlawer [pdf]
- Aerosol Physical and Optical Properties and Processes in the ARM Program. Allison McComiskey, Richard A. Ferrare [pdf]
- ARM’s Aerosol–Cloud–Precipitation Research (Aerosol Indirect Effects) Graham Feingold, Allison McComiskey [pdf]
- The Impact of ARM on Climate Modeling. David A. Randall, Anthony D. Del Genio, Leo J. Donner, William D. Collins, Stephen A. Klein [pdf]
- ARM-Led Improvements in Aerosols in Climate and Climate Models. Stephen Ghan, Joyce Penner [pdf]
- ARM’s Impact on Numerical Weather Prediction at ECMWF. Maike Ahlgrimm, Richard M. Forbes, Jean-Jacques Morcrette, Roel A. J. Neggers [pdf]
The articles are very readable and provide a wealth of background information on the topic.
The Atmospheric Radiation Measurement (ARM) Program was developed to supply an improved predictive capability, particularly as it relates to the cloud–climate feedback. ARM has made important contributions to the systematic measurement of atmospheric radiation and the atmospheric components that influence radiation. They have done a very good job on data quality control and making the data set accessible. The data have been used to evaluate weather and climate model parameterizations, and there have been some direct benefits from the ARM program in improving climate models (see esp the articles by Randall et al. and Ahlgrimm et al.)
Back when I was in the ‘know’, ARM had an annual budget of about $45M. Assuming that level of funding over 20 years produces a total budget of almost $1B. Has all that money been worth it? Do we know understand and reliably model the cloud-radiation feedback? Well the answer to the latter question is definitely ‘no’.
ARM is an example of use-inspired, top-down science. The ‘recipe’ for ARM science included:
- Long-term measurements of radiation, clouds etc and geographically disparate sites
- Intensive observation periods (IOPs) using aircraft and other observations
- Use of a single-column model strategy to evaluate climate model parameterizations against ARM observations
- Use of cloud-resolving models (limited area, LES) to simulate cloud processes
The recipe generated lots of activity, lots of publications, etc. But the actual insights into the cloud-climate feedback and new parameterizations didn’t automatically emerge from this recipe. New scientific insights didn’t really emerge from this process.
The ‘top down’ nature of the program and the ‘recipe’ left some very good scientists rather cold, and many didn’t last long in the program. But a cadre of young scientists emerged, that were trained by senior investigators, that become very invested in the program and the ‘recipe’ and became the mainstays of the program.
I had funding from ARM for the period 1992-2008 (totalling about $2M). The name of my project was “Towards the Understanding and Parameterization of High Latitude Cloud and Radiation Processes.” I was a member of the Science Steering Committee for the ARM Alaska site, and chaired the committee from 2997-2000. I served on the ARM Executive Committee from 1993-1996. And I served on Science Board for the ARM Climate Reference Facility for a few months in 2008; my appt was rescinded once my funding wasn’t continued.
Here is my 2007 annual report [arm-ar07] and the 2008 proposal that was rejected [armp 08 final]. In effect, I was told that I wasn’t ‘playing the game’, i.e. they felt that I should be organizing field experiments and attending their working group meetings (rather than sending students, postdocs). Apparently they preferred such activities to actually writing papers and getting parameterizations in weather and climate models. Further, they didn’t like my style of parameterization development — I was using their data to evaluate parameterizations (having a theoretical basis), rather than to derive empirical parameterizations that directly used their data.
I was frankly astonished not to be funded, but I had managed to hold on to funding longer than many of the other original scientists that were funded, who presumably also lost their funding since they weren’t playing the science game by ARM’s ‘recipe’. (All this occurred prior to the time (circa 2009) when I became ‘controversial.’) So no sour grapes on my part. After all, my cloud-radiation research didn’t require much funding — the data were publicly available, and the analyses and modeling that Vitaly Khvorostyanov and I were doing could be done on a laptop. Our main research expense was paying page charges to journals for our publications. Our response was to write the text Thermodynamics, Kinetics and Microphysics of Clouds. This book provides a foundation for physically based parameterizations of cloud microphysical processes, which ARM apparently wasn’t too interested in.
The point of this narrative goes back to my post Pasteur’s Quadrant. Use-inspired research, especially the top-down variety, will not be successful in tackling difficult problems without the creativity associated with Bohr’s quadrant. The ARM program is a prime example of large (expensive) top-down use-inspired research program. There is no question that observations are important, but observations/data do not automatically imbue understanding and knowledge. Too much data, and emphasis on the process of collecting and analyzing data can become a substitute for thinking and doing the hard work of actually figuring things out.
So the ARM program has some successes to claim, and we have learned something about the sociology of science from this type of scientific program. In terms of ‘bang per buck’ I will leave it to others to assess whether that $2B was well spent. The money did train a new generation of scientists active in the field of cloud/radiation observations and processes.
But at the end of the day, we are still arguing about the sign of the cloud-radiation feedback and the magnitude of the aerosol-cloud indirect radiative effect, and climate models still have major shortcomings in simulating these processes. These are very challenging problems to be sure. It is time to stimulate some new ideas on how to approach these problems.