Tag Archives: Pro

Top 500: Historical overfishing and the recent collapse of coastal ecosystems

I’ve started a long project of identifying my top 500 articles and chapters– the papers that either had a great impact on me or that I keep returning to to illustrate a point. One of those is Jeremey Jackson et al. (2001), Historical overfishing and the recent collapse of coastal ecosystems.

The main argument– that overfishing precedes, predicts, and predisposes the present fragility of ecosystems to modern drivers like pollution– is less interesting than the case studies themselves: kelp forests, coral reefs, seagrass beds, oyster estuaries, and benthic communities. This before-after diagram drives the point home (I colored the changes):


The most depressing line is in the abstract:

Paleoecological, archaeological, and historical data show that time lags of decades to centuries occurred between the onset of overfishing and consequent changes in ecological communities, because unfished species of similar trophic level assumed the ecological roles of overfished species until they too were overfished or died of epidemic diseases related to overcrowding.

Resolving a Hurricane of Questions

Maybe questions of social science never get truly resolved. The first year of my PhD, I remember John Mutter describing the question of creative destruction. Sometimes, the story goes, a disaster can lead to an unexpected silver lining. By destroying outdated infrastructure, or motiving people to work together, or driving a needed influx of aid, a disaster can eventually leave a community better off than beforehand. Mutter described it almost like a philosophical quandary. In the face of all the specifics of institutions, internal perceptions, and international relations, how will we ever know?

For years now, Solomon Hsiang has been producing insights from his LICRICE model, turning hurricanes into exogenous predictors. As these random shocks echo through societies, he’s been picking up everything that falls out. I got to listen to some of it when news reporters would call his office. His work with Jesse Anttila-Hughes turned up the true mortality threat of disasters, typically 10x the lives lost at the time of the event. Jesse dug further, finding how family assets changed, how meals were redistributed, and how young girls are often the most hurt, even those born after the disaster.

Last month, Sol and Amir Jina produced an NBER working paper that steps back from the individual lives affected. Their result is that a single storm produces losses that continue to accumulate for 20 years. People are not only continuing to feel the effects of a hurricane 20 years down the road, but they additional poverty they feel at 10 years is only half of the poverty that they’ll feel in another 10.


Of course, this is an average effect, and an average of 6415 different country results. But that means that for every country that experiences no long-term effect, one experiences twice the poverty.

So, is there creative destruction? It’s very, very unlikely. The most likely situation is “no recovery”: countries will never return to the trend that they were on prior to the event. Things are even more dire under climate change,

For a sense of scale, our estimates suggest that under the “Business as usual” scenario (with a 5% discount rate) the [present discounted value] of lost long-run growth is $855 billion for the United States (5.9% of current GDP), $299 billion for the Philippines (83.3% of current GDP), $1 trillion for South Korea (73% of current GDP), $1.4 trillion for China (12.6% of current GDP), and $4.5 trillion for Japan (101.5% of current GDP).

That’s what we should be willing to pay to avoid these costs. In comparison to the $9.7 trillion that just additional hurricanes are expected to cost, the $2 trillion that Nordhaus (2008) estimates for the cost of a climate policy seems trivial. That’s two big, seemingly unanswerable questions as settled as things get in social science.

Classes Diagram

Sometimes a diagram helps me find order in chaos, and sometimes it’s just a way to stroke my ego. I was recently trying to make sense of my graduate school classes, even as they’re becoming a less and less important share of my gradschool learning. So, I added to a diagram I’d made ten years ago for college.  In the end, I’m not sure which purpose it serves more.


The diagram is arrayed by discipline and year. The disciplines are arranged like a color wheel (and classes colored accordingly), from theoretical math, through progressively more applied sciences, through engineering and out the other end into applied humanities (like music), and finally back to theoretical philosophy. Arrows give a sense of thematic and prerequisite relationships.

Economics, a core of the Sustainable Development program, probably sits around the back-side of the spectrum, between philosophy and math. I squeezed it in on the left, more as a reflection of how I’ve approached it than what it tried to teach.

This is also missing everything I’ve learned from classes I’ve taught. I wish there were a place for Progressive Alternatives from two years ago, or Complexity Science from last year. I guess I need another diagram.

Python SoundTouch Wrapper

SoundTouch is a very useful set of audio manipulation tools, with three powerful features:

  • Adjusting the pitch of a segment, without changing its tempo
  • Adjusting the tempo of a segment, without changing its pitch
  • Detecting the tempo of a segment, using beat detection

I used SoundTouch when I was developing CantoVario under the direction of Diana Dabby and using her algorithms for generating new music from existing music, using Lorenz attractors.  SoundTouch is a C++ library, but CantoVario was in python, so I built a wrapper for it.

Now you can use it too!  PySoundTouch, a python wrapper for the SoundTouch library is available on github!  It’s easy to use, especially with the super-cool AudioReader abstraction that I made with it.

AudioReader provides a single interface to any audio file (currently MP3, WAV, AIF, and AU files are supported).  Here’s an example of using AudioReader with the SoundTouch library:

# Open the file and convert it to have SoundTouch's required 2-byte samples
reader = AudioReader.open(srcpath)
reader2 = ConvertReader(reader, set_raw_width=2)

# Create the SoundTouch object and set the given shift
st = soundtouch.SoundTouch(reader2.sampling_rate(), reader2.channels())

# Create the .WAV file to write the result to
writer = wave.open(dstpath, 'w')

# Read values and feed them into SoundTouch
while True:
    data = reader2.raw_read()
    if not data:

    print len(data)

    while st.ready_count() > 0:

# Flush any remaining samples
waiting = st.waiting_count()
ready = st.ready_count()
flushed = ""

# Add silence until another chunk is pushed out
silence = array('h', [0] * 64)
while st.ready_count() == ready:

# Get all of the additional samples
while st.ready_count() > 0:
    flushed += st.get_samples(4000)


if len(flushed) > 2 * reader2.getnchannels() * waiting:
    flushed = flushed[0:(2 * reader2.getnchannels() * waiting)]


# Clean up

Web Scraping in Python

I’m running a pair of seminars to introduce people to python, for the purpose of extracting data from various online sources.  I still need to write up the content of the seminars, with plenty of examples at from trivial to intermediate.  But first, I wanted to post the diagram I did for myself, to think about how to organize all of this material: a diagram.

How do the elements of python connect to each other, how do they relate to elements on the web, and how do elements on the web related to each other?

Scraping Python Tutorial

Boxes are python elements and ovals are web elements. I aimed to cover everything in brown, touch on items in blue, and at-most mention items in grey.

Risky Business Report released today!

Sol, Amir, and I have been slaving away over a report on the business-case for fighting climate change.  And it was released this morning!  The media outlets give a sense of the highlights:

Today’s report from Risky Business – the project chaired by Steyer, former U.S. Treasury Secretary Hank Paulson, and former NYC Mayor Michael Bloomberg – puts actual numbers on the financial risk the United States faces from unmitigated climate change.
New York Times:
[Quotes our guy:] “the most detailed modeling ever done on the impact of climate change on specific sectors of the U.S. economy.”

Huffington Post:
Parts Of America Will Be ‘Unsuited For Outdoor Activity’ Thanks To Climate Change, Report Finds

Financial Times:
For example, by the last two decades of the century, if greenhouse gas emissions carry on rising unchecked, the net number of heat and cold-related deaths in the US is forecast as likely to be 0.9 per cent to 18.8 per cent higher. But the analysis also shows a one in 20 chance that the number of deaths will rise more than 32.56 per cent, and another one in 20 chance that it will fall by more than 7.77 per cent.

#RiskyBusiness: By end of the century, OR, WA & ID could have more days > 95°F/yr than there are currently in Texas | http://riskybusiness.org/uploads/files/RiskyBusiness_PrintedReport_FINAL_WEB_OPTIMIZED.pdf …

2050: $66b-$106b of US coastal property likely under water, $238b-$507b worth by 2100 #ClimateChange #riskybusiness http://bit.ly/1sANaJj

Also Huff Po:
Higher temperatures will reduce Midwest crop yields by 19 percent by midcentury and by 63 percent by the end of the century.

The region, which has averaged eight days of temperatures over 95 degrees each year, will likely see an additional 17 to 52 of these days by midcentury and up to four months of them by the end of the century. This could lead to 11,000 to 36,000 additional deaths per year.

There’s also some over-the-top praise from the choir– Amir can send you some gems from
Capitalists Take on Climate Change.

Take a look!  Here’s the media report:

And the scientific report (what we actually helped write):