Jose-Luis Jimenez & Ingrid
Marie Ulbrich talk with ScienceWatch.com and answer a
few questions about this month's Fast Breaking Paper Paper in
the field of Geosciences.
Article Title: Interpretation of organic components
from Positive Matrix Factorization of aerosol mass
spectrometric data
Authors: Ulbrich, IM;Canagaratna, MR;Zhang,
Q;Worsnop, DR;Jimenez, JL
Journal: ATMOS CHEM PHYS, Volume: 9, Issue: 9, Page: 2891-2918,
Year: 2009
* Cooperat Inst Res Environm Sci, Boulder, CO USA.
* Cooperat Inst Res Environm Sci, Boulder, CO USA. (addresses have been
truncated.)
Would you summarize the significance of your
paper in layman's terms?
The atmosphere is full of tiny particles, too small to see, yet they are a
deadly pollutant that kills millions of people every year and also impacts
climate change. These particles contain thousands of different "organic"
chemical compounds, i.e., those composed primarily of carbon, oxygen,
hydrogen, and nitrogen. There are too many compounds to measure and analyze
individually.
In our paper, we used a mathematical method to simplify the analysis of
atmospheric organic particles into a few classes, such as those coming from
cars, forest fires, etc. This type of analysis is very powerful but also
quite tricky, and in this study we investigated the power and limitations
of the method thoroughly. We also presented results from a study in
Pittsburgh which showed the influence of pollution emissions from the city,
with an even larger influence of particles transported regionally.
Why do you think your paper is highly cited? Does
it describe a new discovery, methodology, or synthesis of
knowledge?
Our paper is being highly cited for three main reasons. First, we describe
in detail an analysis method that many other people are finding useful.
Second, we also developed and shared publicly a piece of software for this
data analysis and a database of chemical fingerprints of different types of
atmospheric organic particles, and lots of people are now using both of
those.
Coauthor:
Ingrid Marie Ulbrich
"The atmosphere is full of tiny particles, too small to
see, but they are a deadly pollutant that kills millions of
people every year and also impact climate change..."
Finally, we present new results about the sources of a type of particle
material called "oxygenated organic aerosols" (OOA). OOA is mostly formed
from the reactions of gases in the atmosphere, which produce "sticky"
molecules that form particles. In our paper, we show that we can
distinguish an "old" fraction which has been in the atmosphere for several
days from a younger fraction which had probably been formed within the past
day. This had only been shown once before, for a study in Zurich.
How did you become involved in this research, and
were there any problems along the way?
Jose-Luis Jimenez:
I started working on advanced instrumentation for atmospheric particles in
1999. It quickly became clear that our technique ("Aerosol Mass
Spectrometry") produced a lot of information about organic particles, but
it was difficult to quantify and learn what it could tell us about the
sources of these particles.
I was aware of the type of mathematical techniques known as "multivariate
analysis" which had been applied successfully to similar problems, and, in
2004, I began working with Dr. Qi Zhang (then a postdoctoral researcher in
my group, and now a professor at the University of California, Davis) in
applying those techniques to our data. We produced some initial results
which attracted a lot of attention, and, after Qi moved on, Ingrid Ulbrich
undertook the task of applying a new type of technique (known as "Positive
Matrix Factorization") which we knew would be more powerful than what we
had done before.
Ingrid Ulbrich:
I learned about these factorization techniques while working at a job I
held after my undergraduate work. It was amazing that math could help you
find commonalities about the sources of particles that were measured just
in one place. I wanted to apply this type of technique to data that had
higher information content (chemical and time resolution) to try to learn
more about sources and transformations of pollution in the atmosphere, and,
for that reason, I decided to do my Ph.D. studies with Jose-Luis Jimenez at
the University of Colorado, Boulder.
Where do you see your research leading in the
future?
We are continuing to improve the mathematical techniques which we use for
our analyses, to apply them to atmospheric particles from other locations,
and also to apply them to data from other instruments. Much has been
learned recently about the sources of atmospheric organic particles, but
there is still a lot that we don't know, and looking at more places and
using better techniques should allow us to make more rapid progress.
Do you foresee any social or political
implications for your research?
We need to control particle concentrations in the atmosphere because they
have severe health effects on people and also because of their impact on
climate. However, it is expensive to control particle sources, e.g., to
install filters on school buses or to reduce emissions from power plants.
There are also tradeoffs between reducing particle concentrations and other
societal goals: e.g., suppressing forest fires improves the air we breathe,
but is bad for many forests because it allows dead branches, etc. to
accumulate that will eventually cause an extremely intense fire some years
down the road which will be harder for the forest to recover from.
In order to use resources efficiently to reduce pollution in the air that
people breathe, manage our impacts on climate, and manage natural resources
such as forests, we need better information about where the atmospheric
particles are coming from. Then we can put the most resources and effort to
reduce the impacts of the sources that contribute a lot of pollution. Our
work represents a step in that direction.
Jose-Luis Jimenez,
Associate Professor
Fellow of the Cooperative Institute for Research in the Environmental
Sciences (CIRES)
Department of Chemistry & Biochemistry
University of Colorado
Boulder, CO, USA Web
Ingrid Marie Ulbrich
Doctoral candidate, Chemistry
Department of Chemistry & Biochemistry
University of Colorado
Boulder, CO, USA