The History of the East Asia Monsoon

So I went to Washington DC last week for the AGU Chapman Conference on the East Asian monsoon. I found it to be a very rewarding conference, and even learned a bit about navigating around Washington on transit, as I was on a limited budget.

The conference was in AGU headquarters, which is near to Dupont Circle.

Not all that far from the Mall, although I didn't visit this time.

Speaking of scientists . . .

I was speaking during the opening session, which was about climate dynamics (and its role on the changes in monsoonal strengths through geologic history). A major dynamic role has been the rise of the Himalayan mountains and the Tibetan Plateau during the period of interest, and there is still a lot of debate about the importance of these tectonic events on the development of the monsoon. Some of the modeling studies suggest that the mountains only change the specific location of the rainfall, and that monsoon behaviour may occur even if there were no continents at all.

My work was based on analysis of global to regional proxy data sets, and has been summarized in all these places. Unfortunately, due to limited time, after working through the phase space reconstructions, I had to rush through the statistical computation part, and wasn't certain whether any of the message made it to the audience ungarbled. Fortunately, I was able to learn that at least some members of the audience understood the message.

The afternoon sessions were all about paleoceanographic records of the monsoon. Over the past decade, the International Ocean Discovery Program (IODP, formerly ODP and DSDP) has put down a number of boreholes in the Indus and Bengal Fans, and other boreholes in the Huang He fan and the Sea of Japan also provide useful records of at least some parts of the monsoon. The records I studied were generally global in scope--these other records allow for regional variations to be studied.

The next day's sessions dealt with continental environments (a common issue was the change in photosynthetic pathways of plants in response to environmental change during the Miocene) and records of continental erosion. Erosion is important because either rising mountains or increased rainfall will lead to increased erosion.

The last session was on modeling the effects of tectonic uplift as well as changes in the timing of the uplift, because there is still some disagreement about when the Tibetan Plateau was formed. I mean disagreement between it being less than 10 million years ago to more than 40 million years ago, which is a significant difference of opinion for something so recent.

The last portion of the conference was to break up into groups for focussed discussions on topics of interest leading to the testing of several hypotheses proposed at the start of the conference. I started off in the wrong room, so I was  with the tectonic modeling people rather than the climate modeling people, but was still able to ask about whether anyone had successfully had chaos appear in their model output. Results were inconclusive.

For the second group meeting I joined the combined discussion between the climate modelers and the paleoceanographic records group. Over the course of the discussion I eventually managed to come up with a proposal. See if the modelers observe chaos, and see if they can tell which style of chaos they have. Such chaos will be manifested as spatial variability in some climate effects, such as the location of the maximum rainfall. The models may have the type of spatial variability modeled correctly, but the specific timing of variations will be incorrect. That spatial variability will be recorded in the widely spaced paleoceanographic records which already exist. They type of chaos observed in the models will tell us what to look for in the cores; from the cores we can obtain the correct timing of the modeled chaotic spatial variations of the monsoon system.

Exiting the Metro Station at Dupont Circle

I wasn't sure how the last part of the conference would go--early on, many of the old hands were of the opinion that nothing ever comes from these things. But I thought it was pretty rewarding, particularly as it was during these sessions that I came to realize that people felt that whatever I was doing was worthwhile.

Alone in my corner of the world, I had never been sure.

Night flight back to Toronto

Updating "scaling laws of life and death"

It has long been known that there is very long-term cyclical behaviour in a variety of geological records, including mass extinction events, stratigraphic sequences, and climate change. The lengths of these inferred cycles are approximately 60 million years, and 140 million years. Raup and Sepkoski argued for a cycle length of about 28 million years for major extinction events, and proposed the "Death Star" hypothesis, in which a companion star in mutual orbit with the Sun would cause catastrophic meteorite showers resulting in these extinctions.

The Death Star hypothesis has apparently been disproved--at least the Sun's companion has not been discovered despite prudent searching.

In the recent issue of Eos (only available by paid subscription, but your university library probably has it) the headline article suggests that the long-term cyclicity noted above may be present in mantle upwelling events.

The method of analysis makes much use of wavelet analysis, which reminiscent of Fourier analysis but also takes into consideration the portions of the time series in which the periodicities occur.

Wavelet analysis of biological diversity (top) compared with that 

of initiation of large igneous provinces (i.e., really big eruptions) 

over the last 525 million years. From Rampino and Prokoph (2013).

In the figure above, the lighter colours (green to red) represent significant peaks in cyclicity in a) number of marine genera (caused mainly by episodes of sudden loss and more gradual replacement) and b) the timing of the onset of large igneous provinces (which means eruption of huge volumes of magma). The period of the observed cycles is on the scale at left. The method allows us to consider that the cycles may only be active over portions of the records.

As you get closer to the ends of the record, there is a real possibility that the truncation of your data set will affect the data. The longer the period under investigation, the greater the impact on the data near the beginning and end of the record. Consequently, the yellow dashed arcs in both figures above show us the limit of viable interpretation--we only take seriously the portion of the graph in the "bowl" of the yellow arc.

Although the authors do not describe a precise mechanism by which such cyclicity may arise, they note that previous studies suggest that instabilities in the lowest level of the mantle may occur in a repetitive fashion, leading to cyclicity in the emplacement of large igneous provinces (eruptions that may cover > 100,000 sq-km in a geologically short interval).

Alternatively, the formation of supercontinents can act as a cap on the heat produced within the earth, resulting in mantle upwelling within the continent leading to emplacement of huge volumes of magma and rifting. The supercontinent cycle lasts from 300 to 500 million years and is thus too long to be recognizable on the above graphs.

Both graphs above suggest the sudden appearance of an oscillation with a period on the order of 30-35 My beginning a little less than 150 million years ago. That kind of sudden appearance of a new phenomenon in the geological record is an example of innovation in the earth system.

Plot of spectral power of extinctions vs originations. After Kirchner (2002).

In an earlier post (based on the variograms produced by Kirchner, 2002), I argued that extinction was more-or less random, on the basis of the lack of a consistent slope. There were some spectral peaks which could be correlated to the cycles described above--the ca. 28-million-year cyclicity is probably related to the 30-million-year periodicity noted above. An additional oscillation with a period of about 140 million years is also present.

The sudden appearance of cyclicity may be an example of an emergent property. As previously discussed, emergent properties are global-scale behaviours which arise in a manner which is not at all understood from the interaction of local subsystems. Interestingly, small changes at the local scale may bring about structural change at the global scale, which I have previously referred to as innovation.


Kirchner, J. W., 2002. Evolutionary speed limits inferred from the fossil record. Nature, 415: 65-68.

Rampino, M. R. and Prokoph, A., 2013. Are mantle plumes periodic? Eos, Transactions of the American Geophysical Union, 94 (12): 113-114.

Innovation--the new catastrophe

In previous articles I have discussed how global climate (and other systems) may evolve rapidly from one condition to another.

In describing the transition from one state to another I have been using a statistical computational method akin to Markov chains to create a descriptive finite state machine (which I have previously referred to as epsilon machines).

Finite state machines normally consider that there is some sort of impulse or input which drives the transition, and the exact value of the input controls defines which particular successor state follows from a predictive state. For systems like global climate, we do not know what the external forcing, so we cannot create a matrix showing the dependence of state transitions on the forcing. Instead I have calculated the probability of each successor state following a predictive state purely from the observed sequence of states.

For instance, in the chart above, from state A3, the system shifts to A1 20% of the time, to A2 20% of the time, and to A4 60% of the time. Perhaps there were five observed transitions from state A3. The probabilities have been computed on the basis of a single predictive state. There is an argument that the evolution of the system is a function of the entire past history, suggesting we should really consider more than one sequential state for prediction. I have not done this because there are not enough transitions to determine such a matrix.

Conventional thinking is that a state change is driven by some external forcing. I have generally assumed that this is not the case, but that the observed discontinuities arise from a smooth evolution of one or more underlying parameters. Such discontinuities can be referred to as catastrophes.

Today's discussion is a reflection on catastrophe-- in particular, how a system governed by gradual, continuous changes in key parameters can bring about the appearance of sudden, discontinuous change.

One way to visualize how slow, continuous change can bring about a discontinuous response is through the use of a quartic function. In this case, we will use y = x^4 - 1.5 x^2 + a (x - 0.5) + 1. A family of such curves is presented below, for values of a ranging from -1.2 to 1.6.

The general form of the curve is two wells separated by one high point. Place a yellow marble in the right well (with a = -1.2), and consider its position as a gradually varies from -1.2 to 1.6.

At some point, (it is difficult to tell just when, visually), the well at the right will disappear, and the marble will roll down into the well at the left. We can calculate for which value of a this occurs. At each of the minimum and maximum points, the first derivative of the function is zero. The derivative of the expression above is:

y' = 4 x^3 - 3 x + a

We can see by inspection that when a = 1, two solutions degrade into one (at x = 0.5). This point is an inflection point, neither a maximum nor a minimum.

The position of the ball suddenly changes from 0.7 to -1 at a = 1.

The well on the left disappears at a = -1, so if we ran the experiment in reverse, we would see the jump at a = -1. This phenomenon is known as hysteresis. We would describe the change above as irreversible, as restoring the condition of the system to just before the discontinuity does not reverse the change.

If we studied some system with the above behaviour, conventional thinking would have some large input right at around a = 1. In reality, as we see, the jump happens when the system passes the tipping point.

Does climate behave like the mathematical function described above? There is growing agreement that it does, however the probability of such an event in our near future cannot realistically be assessed (Livina's comments to the contrary on page 8 of the linked pdf).