Thursday, April 12, 2012

3317.txt

date: Wed, 16 Jul 2008 11:47:57 +0100
from: "M.D.Bateman" <M.D.BatemanatXYZxyzffield.ac.uk>
subject: [supranet] [Fwd: Re: Outline Paper]
to: supranetatXYZxyzts.shef.ac.uk

<x-flowed>
HI,

I concur with James - looks like a good framework. My comments for what
they are worth are...


[1] para 1.5 I would like to see the paper move away from this "position
paper" stance which comes across as too much like self promoting
supranet. I would like for the paper to have an aim to review existing
approaches for identifying and quantifying uncertainties in
palaeoclimatic data followed by a discussion of how new approaches might
be applied to resolve some of the previously identified outstanding issues.

[2] as the risk of being accused of self-promoting the following two
references deal with issues associated with non-climatic influences of
proxy record - namely bioturbation....
Bateman, M.D., Boulter, C.H., Carr, A.S., Frederick, C.D., Peter, D.,
Wilder, M. (2007). Detecting Post-depositional sediment disturbance in
sandy deposits using optical luminescence. Quaternary Geochronology 2,
57-64.

Boulter, C., Bateman, M.D., Frederick, C.D. (2007). Developing a
protocol for selecting and dating sandy sites in East Central Texas:
Preliminary results. Quaternary Geochronology 2, 45-50.

[3]paragraph 2.4 comes across at present as quite negative. We must
strive hard with this paper to provide an unbiased review AND also to
provide examples of positive outcomes. In this paragraph I would like
to see a counter-balance of a proxy record where assessments of
uncertainty are pretty well quantified. Again I may be bias but a
chronology case study here might provide this balance.

[4] Not sure about section 4 which again sounds like "positioning"
supranet for research grants rather than providing information to a
readership. I would advocate some sort of summary of where we are at,
the exciting first steps that have been taken with bayesian and then a
paragraph of areas where we can see a need for improvements.

[5] couldn't see the point made anywhere that whilst being more open
about all uncertainties in palaeoclimate data may be painful in the
short-term (and lead people to wonder whether any interpretations are
possible) the longer-term better understanding of uncertainty and
application of statistical approaches will be able to better resolve the
uncertainty and give probabilistic information of different interpretations.


cheers

mark

-------- Original Message --------
Subject: [supranet] Re: Outline Paper
Date: Wed, 16 Jul 2008 15:44:30 +0900
From: James Annan <jdannanatXYZxyzstec.go.jp>
To: supranetatXYZxyzts.shef.ac.uk
References: <4874DF04.3060305atXYZxyzffield.ac.uk>

Hi Tamsin (especially) and all,

I have a few comments and references. Overall, it looks like an
excellent framework to me.

First, I think the term "inverse modelling" will be widely
misinterpreted and needs to be changed and/or carefully used. Many,
certainly in the climate science/numerical modelling readership, would
describe any generic process of using observational data to estimate
climate state variables or hidden model parameters as "inverse
modelling" (ie, effectively inverting whatever forward model is under
consideration). Indeed para 2.1 seems to keep this meaning. But as I
understand it, the term was often being applied in the workshop (and
apparently also in the paper, eg para 3.1) as specifically limited to
simple regression-based calibrations. If you want the distinction I
would be tempted to describe it as process-based models vs purely
statistical models but maybe someone will object to that...

I don't understand paragraph 3.5, but then again I don't know what the
reference is! But Guiot et al 2000 (ec. mod.) explicitly describes his
work as following a Bayesian approach.

I'm not sure where and how (or even if) any of the following should fit
in - maybe around 2.4 - so have not attempted to actually write chunks
of text:

Refs 1 and 2 (below) are possibly useful references that talk about
methods for creating climate fields out of proxy measurements. They only
really discuss the issue of getting from a set of point measurements to
a spatial field though, leaving the proxy-climate calibration to someone
else.

If you want an example of climate state reconstruction with explicit
error estimates (para 2.4), then 3 may be a good example, but it is
based on modern observations and they are looking at ocean transports
rather than temp/precip. [There are also the reanalysis projects (NCEP
is the most famous - I believe the Kalnay et al 1996 paper in BAMS is
the most cited ever in the geosciences - but there are also ERA-40 and
JRA-25) which have reconstructed the time-varying atmospheric state over
the past several decades, but that may be a bit of a stretch in terms of
relevance.] There are also some somewhat relevant references in these
three papers 1-3 which talk about paleoclimate as deduced from blending
observations and modelling in various ways.

Also 4 is a recent example on both stacking, and generating chronology
for, sediment cores, with a rather vague discussion of uncertainty. It
seems that there are already well-established plug-and-play Bayesian
methods that could (presumably) improve on this.

5 is the best PMIP ref, which seems needed in the text.

If anyone can't easily access these papers I can send pdfs.

James

refs:
1:
@article{paul2005csp,
title={{How to combine sparse proxy data and coupled climate models}},
author={Paul, A. and Sch{\"a}fer-Neth, C.},
journal={Quaternary Science Reviews},
volume={24},
number={7-9},
pages={1095--1107},
year={2005},
publisher={Elsevier}
}


2:
@article{schaferneth2005pmm,
title={{Perspectives on mapping the MARGO reconstructions by
variogram analysis/kriging and objective analysis}},
author={Sch{\"a}fer-Neth, C. and Paul, A. and Mulitza, S.},
journal={Quaternary Science Reviews},
volume={24},
number={7-9},
pages={1083--1093},
year={2005},
publisher={Elsevier}
}

3:
Ganachaud and Wunsch (2000). A. Ganachaud and C. Wunsch , Oceanic
meridional overturning circulation, mixing, bottom water formation and
heat transport. Nature 408 (2000), pp. 453�457

4:
Lisiecki, L. E., and M. E. Raymo (2005), A Pliocene-Pleistocene stack of
57 globally distributed benthic d18O records, Paleoceanography, 20,
PA1003, doi:10.1029/2004PA001071

5:
Results of PMIP2 coupled simulations of the Mid-Holocene and Last
Glacial Maximum � Part 1: experiments and large-scale features
P. Braconnot, B. Otto-Bliesner, S. Harrison, S. Joussaume, J.-Y.
Peterchmitt, A. Abe-Ouchi, M. Crucifix, E. Driesschaert, Th. Fichefet,
C. D. Hewitt, M. Kageyama, A. Kitoh, A. La�n�, M.-F. Loutre, O. Marti,
U. Merkel, G. Ramstein, P. Valdes, S. L. Weber, Y. Yu, and Y. Zhao
Clim. Past, 3, 261-277, 2007


--
James D Annan jdannanatXYZxyzstec.go.jp Tel: +81-45-778-5618 (Fax 5707)
Senior Scientist, Frontier Research Centre for Global Change, JAMSTEC
3173-25 Showamachi, Kanazawa-ku, Yokohama City, Kanagawa, 236-0001 Japan
http://www.jamstec.go.jp/frcgc/research/d5/jdannan/


--
Dr. Mark D. Bateman
Reader in Palaeoenvironmental Reconstruction
Sheffield Centre for International Drylands Research
Department of Geography, Winter St.,
University of Sheffield Sheffield S10 2TN
E-mail M.D.BatemanatXYZxyzffield.ac.uk
Tel: (+44) 0114 222 7929
Fax: (+44) 0114 279 7912
SCIDR Website: http://www.shef.ac.uk/scidr/
Dept Website: http://www.shef.ac.uk/geography/staff/bateman_mark.html

</x-flowed>

No comments:

Post a Comment