27 February 2011

Critique my writing, please

In the spirit of inviting folks to interact with my blog and hopefully improving my writing, I'm going to step out of my comfort zone and try something new.

Below, I've posted my first-ever official attempt at science writing that I prepared last fall as a journalism class assignment.  I'd like to invite readers to provide their critical review.  What do you like or dislike?  How is the lead? The structure? The quotes?  Does it keep your attention?  Does it provide sufficient and/or appropriate detail?  Etc. etc.

While this is partially a self-serving attempt to generate feedback, I hope the article (which wasn't published elsewhere) makes for interesting and useful reading.

But first, a little background...

The idea for this experiment struck me after reading a post by Anne Jefferson over at Highly Allochthonous (which I thoroughly enjoyed) about what scientists and journalists can learn from each other regarding how to write effectively about science for a broader, not-necessarily-expert audience.  The inspiration for her post, in turn, was the publication of two news pieces* in a recent issue of Nature that discussed two just-published studies** (in the same issue of Nature) linking a greater probability of extreme weather, specifically above-average rainfall and flooding, to anthropogenically-induced climate change.

(Just to recap, this post is based on a post that was based on two news articles that were based on two studies.  How many degrees of separation is that?)

Rather than trying to paraphrase her entire post, I'll suggest interested readers check it out for themselves.  In brief, though, she points out the distinct (and rather amusing, I think) difference between the opening sentences of the two news pieces--one written by a scientist and one by a science writer--and then segues to some personal observations of what the two camps can learn from each other to improve their writing.

As I remarked in a comment on her post, her comparison:
"Makes me recall my first attempt at science writing…coincidentally about linking climate change to extreme weather. For a class assignment, I chose to write a ‘daily’ piece about a talk that Jerry Meehl (a senior climate scientist at NCAR and IPCC co-author) gave at the Univ. of Wisconsin Madison last fall. The talk was nominally for a general audience so Meehl made a point of replacing a lot of the technical details with explanatory plots, etc. In my write-up, though, I focused a bit too much on the nitty gritty, the “why” as you put it, at the expense of compelling quotes, story structure, etc. I thought I’d done a bang-up job until my prof had a go at it. She liked the level of detail, but reminded me that you have to engage your reader with a good lede and then give them reasons to keep reading. If they lose interest and give up, it won’t much matter that you’ve dutifully included the necessary details/context."
I subsequently rewrote, and, I think, improved the article I'd written about Meehl's talk to make it more lively and reader-friendly. But, I figured there's no better way to test if I'd actually improved it, or for that matter, if it was any good to begin with, than to open it up for critique.

The subject of the article--Meehl's talk last fall--has obviously lost some of its timeliness, but the larger issue of the relation between climate change and the probability of extreme weather certainly has not.  Because this is just an experiment, I've removed the names of my sources (except for Meehl).

Okay, without further delay...

            The next time you go shopping for a winter coat, you might want to consider a new pair of shorts and a sturdy umbrella instead.  Warmer weather is headed your way, a prominent climate scientist said Thursday evening, which means fewer cold days, longer heat waves, and more intense rainstorms.
            “Since 2000, we’ve been setting record high temperatures at twice the pace we’ve been setting record low minimum temperatures,” said Dr. Gerald A. “Jerry” Meehl, a senior scientist at the National Center for Atmospheric Research, to an audience of roughly 60 on Sept. 16 in Laudon Auditorium on the University of Wisconsin – Madison campus.  The same 2-1 trend is holding so far for 2010, and, he said, this ratio could increase to 20-to-1 by mid-century and possibly 50-to-1 by the end of the century.
            Meehl spoke about recent examples of extreme weather conditions that have been catastrophic for regional populations and are thought to be related to climate change.  Such events are expected to become more commonplace, he said, based on predictions from computer climate models.  Meehl and others acknowledged that the accuracy of model predictions is limited by available computing power and by our own understanding of the complex phenomena that determine climate.  But, he said, scientists’ ability to replicate past climates with known historical weather data suggests that the predictive capacity of the models is improving.
            “A lot of the impacts from climate change will be felt through extremes in temperature and precipitation,” Meehl said, citing the recent severe heat waves that struck Russia and Japan and the ongoing flooding in Pakistan.  These weather events, which have been blamed for high death and injury tolls, along with widespread damage to homes and agriculture, are symptomatic of changes occurring due to global warming, he said.
            The average global surface temperature increased by more than seven-tenths of a degree Celsius over the 20th century, Meehl said.  People tend to dismiss such apparently small changes as “inconsequential,” he said.  But, he stressed, “You get a huge, disproportionate increase in the extremes from a relatively small shift in average temperature.”
            Meehl said that predictions based on his own and his colleagues’ research are that heat waves will become more severe, more frequent, and longer-lasting, and that rain events will intensify because warmer air is able to hold more moisture than cooler air.  Specific impacts will vary depending on geography, but the upshot, he said, is that “wet areas get wetter, dry areas get drier, in a future warmer climate.”  This trend may have long-term consequences on power consumption, water resources, and growing seasons he said.
            “The topic of climate change is so important,” said Dr. XYZ, professor emeritus in the Department of Atmospheric and Oceanic Sciences at UW – Madison.  XYZ, who is also a member of the Science Council of the Wisconsin Initiative on Climate Change Impacts, a joint partnership between UW, the Wisconsin State Department of Natural Resources and other state agencies, said that Wisconsin will have to deal with the consequences of climate change “in terms of its natural resources, its forests, its fisheries, all the fauna of the state, tourism, agriculture, all those things.”
            Meehl spent most of the 50-minute lecture discussing the development and results of computer models that are used to simulate past, present, and future climates on a global scale.  He said these models are more detailed than simpler atmospheric models used for weather forecasting in that they also account for climatic effects due to oceans, sea ice, land topography, and the sun, among other factors.  The goal is not to predict day-to-day weather patterns and events, he said, but rather to “predict the statistics of seasonal weather.”
            Professor YYY of the AOS department at UW – Madison, said that he thought Meehl did a “pretty good job” of explaining climate models for the general audience present.  But, he said, many of the details about their limitations were left out, and a lot of “in-the-know skeptics will sometimes come down on these studies,” because of the limitations.
            Meehl conceded that the speed of available super-computers constrains the resolution of current climate models, which limits their accuracy particularly at small, local scales.  Model accuracy is also limited by scientists’ incomplete understanding of the complex balance of small- and large-scale phenomena that produce climates.
            He said, however, that climate models are subjected to “sanity checks,” or comparisons between simulated behavior and actual observations, to evaluate their reliability and validity.  An example of agreement he noted was between model-predicted and observed conditions during the severe 1995 heat wave that struck Chicago.  The ability of the models to accurately recreate past and present trends provides credibility for their predictive power, Meehl said.
            The talk was sponsored by the UW-Madison AOS department as the second annual installment in the Robock Lecture series.  The series is intended to provide “public lectures” addressing issues of “public interest,” according to information available from the department.
            “[The talk] addressed current events.  You don’t just have to be a scientist to notice these things,” said ZZZ, a graduate student in the Environment and Resources program at UW’s Nelson Institute for Environmental Studies.
            ZZZ said that she has “clearly seen the applications” and evidence of the effects on the climate that Meehl discussed, and that what stuck with her was that the climate models “are actually capturing” these effects.
            XYZ said that Meehl’s discussion of the greater frequency with which record high temperatures are being set versus record low temperatures resonated most with him.
            “Anybody that looks at his discussion or hears his discussion…if they come away from that and say nothing’s happening…their eyes are shut,” he said.
            “Records are interesting because people pay attention to records,” Meehl said, adding that the trend has been hard for people to argue against so far.
            Even if this trend continues as predicted, however, it does not mean the end of cold weather altogether, Meehl said.  Extreme cold events will still occur, but the odds are shifted so that there is “a much better chance of setting a record high maximum temperature,” he said.
So, that's it. I appreciate any feedback or comments you might have.  Thanks very much!

And, the articles from Nature:

* 1) Allan, R.P. (2011) Climate change: Human influence on rainfall. Nature, 470, 344-345.
   2) Schiermeier, Q. (2011) Increased flood risk linked to global warming. Nature, 470, 316.

** 1) Min, S.-K. et al. (2011) Human contribution to more-intense precipitation extremes. Nature, 470, 378-381.
     2) Pall, P. et al. (2011) Anthropogenic greenhouse gas contribution to flood risk in England and Wales in autumn 2000. Nature, 470, 382-385.


  1. Tim,
    Here goes...I liked the article: the lead was good and the level of detail was about right. You might have been more specific about the types of factors that computer models lack. And did Meehi make any predictions about when supercomputers would be up to the job of routinely accurately making climate forecasts?

    Tom Mitchell,

  2. Hi Tom, thanks for commenting. I don't recall him making any predictions along those lines. Computing power keeps growing, so I think they should be able to keep increasing the resolution of the model. But accuracy is another issue that depends on the data/science/input that goes into the models, as well as on the available computing power. Another tricky thing about assessing the accuracy of a given model's future climate predictions is that we can't tell if it's accurate until the future arrives!