Changes to El Nino in the face of global warming - The rise of Modoki, or the Central Pacific El Nino (CP-ENSO)

My colleague, Faleh Aldham, and I have been exploring the relationship between sea surface temperature (SST), lower tropospheric temperature (TLT), precipitation, and the El Nino Modoki index (EMI) produced by the Japanese Agency for Marine -Earth Science & Technology (JAMSTEC).

El Nino Modoki is a phenomenon, very similar to the canonical El Nino (or ENSO), with slight but significant locational differences.  Both events occur in the tropical zone of the Pacific Ocean.  The canonical ENSO is measured by a shifting of warm water from the far Western Pacific Ocean over to the far Eastern Pacific Ocean.  This shift typically occurs because of a reduction in Easterly trade winds, which normally drive the warm water towards Indonesia through a friction mechanism.  Every 7-10 years (or so) these winds die down and change direction.  The mass of warm water that typically sits over in the Western Pacific Ocean near Indonesia is moved over to the Eastern side of the Ocean, and bumps into the West coast of the Americas.  This movement of warm water has significant impact on global weather pattern.  As warm water evaporates, it typically creates a warmer atmosphere.  Once the moisture has been absorbed into the atmosphere, it is subject to movement around the globe.  It is carried to number of different places based on wind and pressure movements, and eventually dispersed again as precipitation.  This process is illustrated in the above images.  The first image is sea surface temperature, showing the Modoki phenomenon very clearly.  The second is the lower tropospheric temperature, which follows the SST quite well.  Last image shown is the precipitation, which warrants much further discussion that I will cover in a later post.

As can be gathered from the above description, the movement of a mass of warm water in the Pacific Ocean has important implications for the changing nature of precipitation patterns across the globe.  This is why ENSO is an important phenomenon to study, and particularly why we should be tracking the changing nature of El Nino.

This brings me back to to the discussion of the locational distinction of El Nino Modoki, or the Central Pacific El Nino (CP-ENSO) phenomenon, that has been becoming more common since 1990.  This new form of El Nino looks different from the canonical El Nino in its locational make up.  Rather than shifting warmth from far East to far West in the Pacific Ocean, Modoki is characterized by anomalous sea surface temperatures concentrated in the Central Pacific Ocean, with opposite temperature anomalies occurring on either side of it.  For example, a warm central Pacific Ocean flanked by cool West and East sides (or vice versa).

Modoki has only become part of the conversation in recent years.  The Japanese Agency for Marine-Earth Science & Technology has been collecting sea surface temperatures in a number of places in the Pacific Ocean in order to measure the occurrence of Modoki.  They use a number of sensors in the Central Pacific to gather an average temperature for the central area.  They do the same for areas flanked on either side of the Central Pacific.  They subtract the sum of 0.5 of the Western measure and 0.5 of the Eastern measure from the central measure to create their El Modoki Index (EMI).  This is what we used in our study.

An important question to ask is, why?  Why are we seeing a rising number of Modoki events?  The short answer is, we don’t exactly know yet.  however, there is some significant published evidence backed by several research groups explaining that this may be a result of global warming.  The argument has to do with the weakening of the Easterly winds and the flattening of the oceanic thermocline, both due to increased global temperatures.  This will be discussed in more depth in a later post, as well.

To get to a short description of our analysis — Through a series of normal and Extended Principal Component Analyses (PCA & EPCA, respectively) in both S and T mode, we have found some very high correlations between our three datasets, the EMI, and extreme weather events occurring across the world.  We case study six major weather events occurring between 2000 - 2009, all correlated with an R-value of 0.65 to the existing Modoki index.  These events include drought in Ethiopia & Argentina (Mendoza, specifically) in 2000 (a negative Modoki phase), increased precipitation in Mendoza in 2005 (positive Modoki phase), hurricane Katrina in 2005, extreme flooding and mudslides in Brazil in 2009 (continued positive phase), and the Black Saturday Brushfires in Australia in 2009.  These events are mostly in the extra tropics, however, correlations exist close to the equator as well.  It is important to note that none of these years are recorded as typical El Nino/La Nina years.  At best, 2000 and 2005 were weak La Nina years, and 2009 was a moderate El Nino year.  We will be examining these correlations and weather anomalies in much more detail as we move forward with our analysis.  We will continue to look for other past weather events that may correlate with this Pacific phenomenon, as well.  Some teaser images are above, just to give you an overall view and idea of what we are looking at.  The EMI we are using can be found on the JAMSTEC website, here.

I’ll be posting more of the specifics of our analysis in the near future.  In the meantime, if you’re interested in learning more, feel free to get in touch with me.

Fun with Landsat 7. False color composite of Hong Kong.

Fun with Landsat 7. False color composite of Hong Kong.

Today I’m working on Transition Potential Modeling (TPM).  Transition potential refers to the likelihood of a land cover class to change (or transition) into a different land cover class.  This may be useful in determining what land types, and which existing parts of those land types, are likely to be subject to urbanization, deforestation, or even re-growth.

 The method used to produce the above images involves a Multi-Layer Perceptron (MLP) technique to determine the explanatory skill of each variable that has been included in the model.  More on MLPs to come.

A research question

On the topic of innovation theory — I’ve been diving around this topic so much that I have completely avoided pinning down a real research question.  Today I shall endeavor to do so.  Some thoughts:

What are the structural differences between first tier and second tier cities on the east coast of the USA?

What are the differences and similarities among the major cities in the Bos-Wash megalopolis?  What are the differences and similarities between the minor cities in this same area?

How does the proximity of the minor cities to the major cities vary?  Specifically, what are the differences in access via public transportation?

Is Baltimore a major or minor city on the circuit?  Obvious majors: Boston, NYC, Phila, and Wash DC.

After looking at structural similarities and differences, what social factors can be added?  What kind of limitations exist when looking at social media for this type of analysis?  What types of geographic links, sentiment analysis, and other qualitative notes can be gained from looking at social media?

This last question is a big addition to the previous ones.  It will be next on the agenda, once the first questions are fleshed out a bit more and have some answers pending.  My approach will be a mixture of historical analysis and spatial analysis.  I’ll be reading a lot about the manufacturing histories of these cities, and the historical movements of people.  I will also be constructing a GIS model of this circuit, in order to take a critical look at the transportation systems that connect them.  It’s time to get down to reading.

Innovation theory

I started graduate school this past Fall.  I am working for a Master’s Degree in Science at Clark University in the GISDE program.  Our department acronym stands for Geographic Information Systems for Development and Environment.  We are housed in the IDCE department, or: International Development, Community & Environment.  It’s quite the mix of people, all very driven, and all very friendly.

As for GISDE, we are two full years, and approximately 20 students in each year, amassing to just under 40 in the whole program.  We study traditional GIS in the ArcSuite environment, and also branch out into some Free and Open Source Softwares, such as GeoDa, QGIS, and others.  We  undertake detailed study in both Vector and Raster analysis.  Remote Sensing is also required curriculum.

So far I have been very pleased with the program.  The professors here are extremely knowledgeable and dedicated to both their fields of study and their students.  Clark is a research university, and there is a lot of opportunity and encouragement to take initiative in your personal topic of interest.  Students are encouraged to read current research and to look for the gaps in methodologies or questions that are being explored.  Importantly, we are also encouraged to take the next step and attempt to fill those gaps in.  Many students will publish in academic journals during their time here.  I hope to join that crowd.

One topic has recently bubbled up in my life that I believe touches on many of my interests.  That topic is Innovation Theory.  Specifically, locational and infrastructural variables that may be linked to different levels of innovation in different places.

My current understanding of “Innovation”:

Pushing the boundaries in both the arts and sciences.  Political improvements.  Policy implications. Technological advancements. Progress.  Progressive. Big picture.  Able to see the forest for the trees, and vice versa.  Job creation.  Network creation. Transportation. Cleanliness.  Efficiency.  Better systems.

My current questions on the topic of “Innovation Theory”:

Where is innovation happening & why?  Does it happen naturally?  Are there structural requirements for innovation to exist, or that affect its pace?  Do certain physical, educational, cultural infrastructures encourage more or less innovation?  If so, what are they and what weight do they carry?  What are some key indicators of innovation that can be defined and used as a metric?  Are there variables that should be specific to each individual place that is studied?  How do we account for these in our formulas?  What are the anecdotal variables that may need to be studied in person and through interviews?  What are the variables that can be looked at through publicly available data?  Privately available data?  Not yet recorded data?  If field work were required to collect data, what would that look like?  Who would you talk to and about what?

I will be digging into some literature to start seeing what questions others have asked, and how they have answered them so far.  More on all this soon.

A concept that many people blindly accept is that taxing the rich will create massive job loss.  This assumes that the rich are the job creators in a capitalist society.  Nick Hanauer contests this in a compelling 6 minute presentation.  This is a controversial issue, and has been banned from the TED website for being such.  Please watch, share, comment here or on youtube.  This is a conversation topic greatly worth digging into.

Another one of Worcesters’ s many parks #worcester #maps  (at Boynton Park)

Another one of Worcesters’ s many parks #worcester #maps (at Boynton Park)

Tags: maps worcester

Worth it just to see the variety of features cityscapes have to offer riders.  Click through for video, and check out more Proper / IMGTv videos here.

Worth it just to see the variety of features cityscapes have to offer riders.  Click through for video, and check out more Proper / IMGTv videos here.