Books & Arts

Comments about the book review: A vast machine: Computer Models, Climate Data, and the politics of Clobal Warming by Paul N. Edwards

Embracing an uncertain future A history of climate modelling shows that forecasts that knowledge uncertainty will be the way forward, argues Myles Allen.
Following is a discussion about the bookreview in "Books & Arts" article in Nature Vol. 466 1 July 2010 by Myles Allen
In the last paragraph I explain my own opinion.


Embracing an uncertain future

The article starts with the following text:
Many people find climate models puzzling
I do not know if that is true. What is true that there is a big difference between weather predictions and climate predictions.
Next we read:
A Vast Machine describes how the disciplines of statistical climatology, weather forecasing and theoretical meteorology evolved into modern climate science.
No comments, except that weather predictions and climate predictions are two complete different "animals". Next we read:
Rather than centring on observations and ice-core records, which many others have discussed, the book focuses on computer modelling.
IMO for climate predictions (long time scale) those ice-core records have to be included. For weather predictions (short time span) not. Next we read:
The book explores the nature of climate simulations and controversies over inconsistencies between models and observations.
This can only be an issue related to the past. That means if you predict the present (based on historic observations) it should match the present observations. Next we read:
To those who favor hard data over simulations Edwards points out that there is no such thing as a model-free observation.
IMO if I measure the parameters: temperature, pressure, wind direction, rain fall etc, at my position, now, those are all model-free observations. Next we read:
Every piece of data rests upon some theoretical model of the measurement system,
This is not true. As written above the measuring process of physical parameters itself, is independent of models. Models become imporant if those same physical parameters are required for different places and time as measured. The sentence continues:
and the assumptions that underlie models can be as important as the accuracy of the data
Each of those three "wordings" are important. IMO the concept "uncertainty" does not belong in this list. Next we read:
By incorporating every-higher resolution and more detailed representation of processes uncertainty arising from model specification will eventually be eliminated, leaving only observational uncertainty and chaotic variability to limit the skill of climate forecast.
Weather and climate is principaly a description of a physical process which incorporates the transport (flow) of air water and heat. The better we understand those physical processes the better we can predict the weather on short time scales and the climate on larger time scales. The physical processes are described by mathematical equations and the total collection of those equations we call a model. In order to predict the future: past and present observations are needed. The more and the more accurate the better i.e. the more accurate predictions can be made.
Concepts like uncertainty and chaos are of no importance. For weather and climate predictions computer accuracy is no issue (As long as solar movement around the Subn is not included). Boundary conditions i.e. the shape and composition of the landscape are important.
Next we read:
We can never be sure that the model we converge on is the right one, rather than one that has merely been tuned to fit the data from previous decades.
I doubt if all of that is true. Ofcourse we can never be sure if our model is the right one. It is easy possible that different physical equations are possible, which better describe the physical processes involved and inreturn give more accurate predictions. That is what is called: progress in science. Next we read:
In the 1950s, Edward Lorenz, the meteorologist and pioneer of chaos theory, realized that chaos sets a hard limit in weather forecasting: we must accept that we are unable to predict the timing of storms a month a head.
Such a limit does not exist. The whole issue is accuracy. Accuracy of place and time. The problem is chaos is typical a mathematical issue. Chaos is not a physical issue i.e. weather conditions, climate conditions and solar system behavior. It is closely related to fractals, which is also a mathematical issue.
Next we read:
Yet the current limitations of climate models are much less clear - most of their uncertainty arises from missing information, such as poorly known drivers of climate change.
I do not known if the word drivers comes from the book. It is a poorly selected word. The issue is that climate evolution is a physical process. The question is what are the most important parameters that are involved (i.e. influence the climate) and how. It is not the movement of a wing of a butterfly. Human behaviour can be of importance. The question is how.
Whereas some uncertainties, such as the timing of volcanic eruptions, are irreducible, many of these unknows will respond to more data, better models and a stronger signal of emerging climate change.
The last remark is a guess and does not belong here. No body knows to what extend it will be possible to predict volcanic eruptions, the issue is what is their influence for the weather. The same issue are meteor impacts.
Improvements are possible but intrinsic uncertainties will always remain.
Uncertainties of what ? Next we read:
Over the coming decade, systems for forecasting climate that treat uncertainty as an additional prognostic variable will become the norm.
Uncertainties of what ? I do not agree with this. The whole chalenge is to decrease uncertainties and to improve accuracy. I agree that the whole of weather and climate science is a very difficult subject. Next we read:
Our aim in climate modelling should be to convert unknown unknowns into known unknowns, not to pretend that we can eliminate them altogether.
This is a nice, but rather misty sentence. IMO the object of climate science is first to improve weather models (i.e. short range predictions) secondly to study if their are long range weather changes and identify their causes.
At the end of the article is written:
A Vast Machine puts the whole affair into historical context and should be compulsory reading for anyone who now feels empowered to pontificate on how climate science should be done.
Anyway climate science should be done by a group of only scientists independent of any political party or single issue organization. Those climate scientists should try to answer the question how the climate will evolve assuming that all current processes continue undisturbed. This is the same methodology as the club van Rome. A separate task would be to answer what if questions.


Feedback

None


Created: 21 July 2010

Back to my home page Index
Back to Nature comments Nature Index