Talk:Sea ice emissivity modelling

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Article[edit]

This article has been challenged because of conflict-of-interest. Here is a short essay on conflict-of-interest in science as it relates to Wikipedia:

When I write an article in my field that strongly references my own work it is unambiguously a conflict of interest. However, it is my belief that in scientific fields, this should be tolerated and I will present several arguments.

The goal of science is to discern objective facts. Thus, the neutral point of view is built in right from the beginning. How much this is achievable in practice is debatable: few philosophers believe that true objectivity is possible. My job as a scientist is not only to advance the state-of-the-art, but to improve existing work, thus in writing an article it would make sense to reference my own work: should I not have confidence in my own abilities, that I have done a good job? Or, who would be more qualified to write about these topics than someone who specializes in the field and who has worked in the field for many years?

Indeed, when I first started working on sea ice remote sensing, it was apparent that there were many issues in emissivity modelling and my first task was to clarify them. The Wikipedia article, "Sea ice emissivity modelling," represents a large portion of that work. There is little of my own original thought in it although it references two of my own papers. Rather it unifies and summarizes much of what came before. The figure which you have removed comes from one of my (peer-reviewed) papers and accurately diagrams the majority of microwave sea ice emissivity models used up to this point.

Which brings me to my second point: the second part of my job as a scientist is to disseminate my findings and knowledge amongst the general public. I can't think of a better platform for doing this than Wikipedia. Perhaps it is arrogance, but I consider this a vital public service. There is a great deal of interest in issues like climate change of which sea ice is a valuable marker. It is important that even ordinary people can learn how the science is actully done: articles like this, by dealing with the more "nitty gritty," as opposed to the lighter surface treatment, which to my eye looks more like spin, make climate science more accessible.

I am not hiding anything. Anybody who cares to check will find that, yes, the same author who wrote this article and many others has also referenced a lot of his own work. And I won't lie either: Wikipedia is an important tool for promoting my work. Unfortunately, the harsh reality of science today is that you cannot survive without promoting your work. The h-index is a common measure of a scientist's output. It is also unambiguously a measure of popularity, much as you might find on a social media or dating site.

So in this case, it's hard to argue conflict-of-interest. In science, promotion and dissemination go hand-in-hand and disseminating your work is one of the twin responsibilities of a scientist. The question is not whether I am referencing my own work, but rather whether that work is relevant and of high-quality to begin with.Peteymills (talk) 23:36, 22 April 2014 (UTC)

Article issues[edit]

It is unclear how relevant the proposed methods are. The general style of writing should aim for a broader audience. The source should use latest science and source from more literature, and reliable sources. Images need to be optimized if relevant. prokaryotes (talk) 15:57, 22 April 2014 (UTC)

First you say that the issue is COI, now you're saying that the material is outdated. I'm not sure how a model can be outdated: if you want to predict planetary orbits, Newton's laws are still just as relevant today as when they were first coined over 300 years ago. Closer to home, radiative transfer simulations in the atmosphere can predict satellite radiances to better than 1 K accuracy. I won't give you any citations: I know because I've done it and all the data and programs I used are freely available online. RT models for sea ice are not quite as good: they are only valid for certain regimes, namely where the dielectric constant is well determined and scattering is weak. When I first got into the field, I did have my doubts about the validity of these ice emissivity models because they were not well validated empirically. Until quite recently, validation consistly primarily of simulations of idealized ice profiles compared with averaged emissivity measurements. So I did what came naturally: checked them myself for more ideal cases, namely freshwater ice at low frequences. No model is perfect of course, but it turns out that a simple RT model can explain a lot of the variability in microwave signals. You can check my work for yourself here: http://arxiv.org/abs/1202.4095 and here: http://arxiv.org/abs/1207.4488.
In any case, this does not seem like sufficient justification for deletion. The whole point of Wikipedia is: if you don't like the content of an article and think it can be improved, then improve it! Yes, there are other methods of modelling sea ice emissivity and the ones described here are only valid for a narrow range of frequencies. I am familiar with these because I was working on specific of instruments for which they were (somewhat) useful. It was also apparent that the field was not that advanced, certainly nowhere near as mature a field as RT modelling in the atmosphere. In part, this is because sea ice emissivity is quite a complex phenomenon: demonstrating this was one of the purposes of displaying the data-flow diagram which you so kindly removed. Peteymills (talk) 23:48, 22 April 2014 (UTC)
Peteymills, normally it is a COI issue when somebody publishes his own work. Also you can link to journal publications and the related PDF's, but it is not enough to link to your own homepage. The diagram is to big and complex for Wikipedia, you take for granted that everybody calculates the formulas you left in the diagram. However, the article should focus on current understanding and applied modeling, rather than to present a narrow view of an active field of research, ie. following this context and outlining common approaches. What is the name of the model you present under Radiative transfer modelling? This should be made more clear, that this model is a proposal. A good outline of the state of modeling emissivity of sea ice can be found here. If you got a novel algorithm, then this belongs into the article, but this has to be made more clear. But the article could very well be covered at sea ice concentration prokaryotes (talk) 02:26, 23 April 2014 (UTC)
I linked to my homepage because journals charge for the privilege of viewing an article, something which I disapprove of. The model is called a radiative transfer model and I used it in the cited paper as well as in a number of other studies. Is it unique to my work? By no means. MEMLS uses exactly the same method of integrating the ice layers when scattering is neglected as do a number of other ice models. Again, when scattering is neglected, the two models agree to within half-a-Kelvin. The mathematical treatment is often quite different, but I can assure it is the same animal. I do not treat scattering in much depth in the article because the scattering models are not as mature.
I was not able to read the entire paper in the link to what you claim is a review of ice emissivity models, but it is nothing of the sort. When they discuss ice emissivity, they are not talking about detailed models that deal with the ice below the surface. The paper deals with an ice concentration algorithm that looks to be similar (although not identical) to the broad class of algorithms that are covered by the equation in the article on sea ice concentration. Again, sea ice concentration does not equal sea ice emissivity. In ice concentration algorithms we are trying to derive a physical quantity from a radiometer measurement, often using empirical considerations. In sea ice emissivity modelling, we are trying to predict the signal seen by the radiometer based on physical first principles and on the properties of the sea ice. These types of "forward models" can be used in retrieval algorithms, either directly or indirectly (I have written a number of other Wikipedia articles about this) but that is not their only intended purpose.
The article is written in a technical style (as was the other one that you butchered) because it is intended for the more serious reader who wants to delve more deeply into climate and remote-sensing related issues. If the reader wants a less technical (and more shallow) treatment, he or she can go to any of a large number of other articles. This is not a paper encyclopedia: there are no space limitations so articles can span the range from very general to extremely narrow and technical. Indeed, there should be articles for those with a more technical bent and who want more depth--these could potentially be students, especially graduate students, as well as other scientists, but also scientifically literate "lay-people". The amount of content on fluffy pop culture subjects is several times that on scientific and technical subjects. I find it offensive that you're trying to gut the more serious scientific content. Content that we used to think of as being most appropriate to an encyclopedia.
Are you here to actually make a constructive contribution, or do you just want to throw a wrench into things?
Peteymills (talk) 03:43, 23 April 2014 (UTC)
Also, those other two pages you linked to don't seem to be relevant in any way. I'm wondering what kind of credentials do you have in this field? I don't get the impression you have a particularly deep understanding of the science being discussed, i.e., that you are not really qualified to write about this material. Peteymills (talk) 03:53, 23 April 2014 (UTC)
Apparently sea ice emissivity modeling based on brightness temperatures is part of determining sea ice concentrations or volume. I think you should stop editing the article, since you fail to acknowledge the shortcomings and only seem to promote your own research. prokaryotes (talk) 05:34, 23 April 2014 (UTC)

Fixing this article[edit]

So first, I'll come out and say that I'm a bit of an expert in this content area. I've cited at least one of papers here in my own research. I don't think I have a COI, but I want to start discussion on the talk page just to keep things on the level. The main problem with the current article is that it's written as a howto, which runs afoul of the purpose of Wikipedia. So we could reasonably put it up for deletion as inappropriate for the purpose of Wikipedia. The current title is also a bit of a misnomer -- this article is about the microwave properties of sea ice, not about it's general emissivity. However, the subject matter is definitely notable. There exist textbooks with entire chapters on modeling the emissivity of water ice (I even have one at home!).

I would suggest the best way forward is to greatly compact the material currently in this article (for instance, the radiative transfer section should be nuked entirely, as it's just an enumeration of standard plane-parallel radiative transfer in general and not particular to sea ice modeling) and retitle it under the name "electromagnetic properties of ice." An article at that title would be more applicable to a broad range of sciences and subject matter and would be more appropriate for an encyclopedia than this current article. Much of the material here would be appropriate for a microwave subsection of a broader article on EM ice. Ice is also spectrally interesting in the near-infrared where it has some substantial absorption features. And we could talk about how in the optical it's one of the highest albedo natural surfaces.

thoughts on moving forward? Sailsbystars (talk) 02:31, 29 April 2014 (UTC)

Sounds like a good plan, and here is a closer look at the references from the article. prokaryotes (talk) 12:50, 29 April 2014 (UTC)
Additionally we require an article on sea ice volume. prokaryotes (talk) 13:50, 30 April 2014 (UTC)
Noted! I thought we had something on that in an article I worked on a while ago, but apparently not. I've added it to my todo list along with area and extent. Sailsbystars (talk) 02:20, 1 May 2014 (UTC)
Removed the radiative transfer section name and replaced it with electromagnetic properties of ice, as has been suggested by Sailsbystars. Maybe someone can move the article to the right place now? prokaryotes (talk) 23:19, 15 June 2014 (UTC)
Nah, it's worse than that. The whole radiative transfer thing has nothing to do with sea ice or ice. It's a generalized radiative transport formula. It's the sort of thing one might included in the intro to one's dissertation on sea ice perhaps. But there's no reason to have it in a sea ice article specifically. Basically what I was proposing above was gutting this article and rebuilding something of value to the encyclopedia based around the more useful bits (e.g. the effect of brine inclusions, general electromagnetic properties of ice and sea ice). If I have some serious free time (might happen in October), I'll try to fix this article up. Sailsbystars (talk) 02:55, 21 June 2014 (UTC)
I think you should just go ahead. The article as it is appears, on not very much inspection, to be far too specialised. Should it even exist? William M. Connolley (talk) 18:44, 21 June 2014 (UTC)