Wednesday, May 12, 2010

European Trends in Nighttime Lights

The image above shows the monotonic trend in nighttime lights from 1992-2008. The data series is produced from the Operational Linescan System (OLS) sensor of the United States Defense Meteorological Satellite Program (DMSP). This sensor is very sensitive to light and has been used to create annual images of stable lights. The analysis was conducted using the Earth Trends Modeler by Ananya Baruah and Annalise Erkkinen -- two graduate students at Clark University. Trend monotonicity measures the degree to which data values are consistently increasing (red) or decreasing (blue). More extreme values indicate greater consistency in the trend. Areas that appear white are not necessarily dark. It simply means that the light levels are unchanging. Thus the central cores of Paris and London both appear white in this image.

What is remarkable about this image is the clear evidence of political boundaries. The increases in nighttime lights in Poland are extraordinary and are consistent with the fact that Poland's GDP increased over 500% from 1992 to 2008. However, an explanation based on the economy doesn't always work. For example, the UK and France experienced similar economic growth over this period, but opposite trends in the nighttime lights.

Do you have any insights on the trends evident in this image? If so, please leave a comment. Thanks

Sunday, April 4, 2010

AIRS Carbon Dioxide

An Empirical Orthogonal Teleconnection (EOT) analysis of the AIRS CO2 product (monthly anomalies from 2003-2008) using the Earth Trends Modeler (ETM) very clearly shows the progressive increase in CO2 in the mid-troposphere. Below we see the EOT correlation image and its associated temporal graph.



All areas of the globe show very high correlations with the trend, although they are lower in the boreal and tropical forests. Note that this analysis is based on anomalies so that the seasonal variation inherent in the Keeling curve is not so evident.

An analysis of the Theil-Sen slope (below) of the same series shows the rate of increase. Here we see a clearly lower rate of increase in the tropics with the highest rate in Siberia.

Saturday, April 3, 2010

Has the AMO peaked?

Has the Atlantic Multidecadal Oscillation peaked? An analysis of monthly sea surface temperature anomalies for 1982-2009 using the Empirical Orthogonal Teleconnection (EOT) tool in the Earth Trends Modeler suggests that it may have peaked at some point in 2006 or 2007. This timing would be reasonable given that it was at the lowest part of its cycle around 1974. Here is the graph of EOT2 (in degrees Celsius) along with its correlation image:


Sunday, April 26, 2009

The Atlantic Multidecadal Oscillation - Part II

This post concerns a separation of the linear and non-linear components of the Atlantic Multidecadal Oscillation (AMO) from 1982 to 2007 using the Earth Trends Modeler.


The first step in creating this separation was to remove the effects of other climate teleconnections. In an analysis of sea surface temperature teleconnections using the Empirical Orthogonal Teleconnection procedure in the Earth Trends Modeler (see the previous post), the AMO was the second EOT. Thus to remove the effects of other teleconnections, the Linear Modeling tool in the Earth Trends Modeler was used to create a residual series after removing the effects of EOT's 1 and 3-10.

The next step was to run a linear model using this residual series as the dependent variable and two unidimensional time series -- one being a simple linear series (the Earth Trends Modeler has a simple utility to create this) and a second being a detrended version of EOT2. The graphs below show the AMO as measured by EOT2 along with the detrended AMO.



The two images showing the linear and non-linear components of the AMO at the top of this post are the Partial Correlation images of the linear series and the detrended EOT2.

Many areas of the North Atlantic basin show a strong relationship with both the linear and non-linear components. Both images show a strong relationship with the North Atlantic Subpolar Gyre (the dark red area in the Labrador Sea However, there is one striking difference -- the non-linear pattern shows a structure consistent with the negative phase of the North Atlantic Oscillation (NAO) (click here for the graphic). In fact, using a 7-month mean filter on both the NAO index and the non-linear component of the AMO shows that they are negatively correlated (r = -0.46):

Latif et al. (2007) (Latif, M., C. W. Böning, J. Willebrand, A. Biastoch, F. Alvarez, N. Keenlyside, and H. Pohlmann, 2007: Decadal to multidecadal variability of the Atlantic MOC: Mechanisms and Predictability, in Schmittner, A. J. C. H. Chiang, S. R. Hemming (Eds.) : Ocean Circulation: Mechanisms and Impacts - Past and Future Changes of Meridional Overturning, AGU Monograph 173, American Geophysical Union, 149 -166.) provide evidence of a relationship between low-frequency variability of the NAO and the Atlantic meridional overturning circulation (MOC) with which the AMO is thought to be associated. However, this analysis suggests that an even higher frequency relationship may exist.

Sunday, March 29, 2009

The Atlantic Multidecadal Oscillation - Part I

One of the interesting features of the Earth Trends Modeler is the ability to compute Empirical Orthogonal Teleconnections (EOT). The intention of the technique (described in more detail at the end of the end of this post) is to uncover the major underlying patterns of variability in the analyzed series over space and time.


The first EOT in monthly anomalies in sea surface temperature from 1982-2007 (above) is the familiar El Nino / La Nina (ENSO) phenomenon -- not surprising, since it is unquestionably the dominant pattern of interannual variability in sea surface temperature (SST). However this post concerns the second EOT (below) which presents a less familiar pattern.


The second EOT is the largest pattern of space-time variability in SST anomalies that the technique can find in the residuals from ENSO (i.e., after the effects of ENSO have been removed). The image above shows the spatial pattern and the graph below shows the temporal pattern. Areas with high positive values in the image are strongly associated with the temporal pattern (and vice versa).


The space/time portrait of EOT2 is dramatic. First, most of the world's oceans show warming (i.e., a positive association with increasing anomalies in temperature after the mid 1990's). Second, the warming is most pronounced in the Atlantic. The image below shows the temporal graph of EOT2 along with an index to the Atlantic Multidecadal Oscillation (AMO) superimposed (in red) -- a low frequency Atlantic SST oscillation with what is thought to be a 65-70 year cycle associated with the thermohaline circulation.


Clearly the temporal evolution of EOT2 is a VERY close match to the AMO index (r = 0.71). However, the AMO is also the subject of significant controversy. The issue is the extent to which the pattern of Atlantic warming in recent years is attributable to a natural climate cycle (the AMO) or to global warming. Look for more on this issue in the next posting.

About the technique:

The Empirical Orthogonal Teleconnection technique (as implemented in the Earth Trends Modeler) searches for the single location in space (a pixel in this case) whose temporal profile can best describe the temporal evolution of all other locations. The profile at that location becomes the first EOT. A residual series is then created such that the effects of the first EOT are removed from the original image series. The process is then repeated to find the next EOT, and so on. In our implementation, after all the designated number of EOT’s have been found, a multiple regression is run with the original image series as the dependent variable and the EOT’s as independent variables to create partial R images as a spatial portrait of the EOT.

EOT's are similar to obliquely rotated Principal Components. They are independent in time, but not necessarily in space. They have the advantage that they are easily understood and are associated with a specific location.

Saturday, March 14, 2009

Sea Ice Concentration 1982 - 2007

The image below shows the median trend in monthly sea ice concentation from 1982 - 2007 as calculated by the Earth Trends Modeler. The image series comes from the National Snow and Ice Data Center. This analysis is based on anomalies in sea ice concentration. Thus it automatically accounts for seasonal variation. Most images of changes in sea ice concentration focus on the summer extent. However, sea ice is an important factor for Arctic ecology throughout the year. Thus this portrait provides a broader perspective. The image has been contrast stretched to a range from +/- 0.042% per month which translates to a rate for approximately 5% per decade. However, in some locations the rate exceeds 10% per decade.

Lower Tropospheric Temperatures 1982 - 2007

The image below shows the median trend of monthly lower tropospheric temperatures from 1982-2007 as determined by the Earth Trend Modeler from the Microwave Sounding Unit (MSU) image series processed by Remote Sensing Systems. These data are spatially coarse, but they represent a critical resource in understanding atmospheric dynamics. In the image below we can see that the strongest trends are in the Arctic. The image has been contrast stretched to a range between +/- 0.008 degrees Celsius per month, which translates to almost 2.5 degrees over the period of this series. It is interesting to note that the highest increases in temperature are in the northern hemisphere.


The second image (below) shows the monotonic trend analysis for MSU lower troposphere temperatures. Trend monotonicity measures the degree to which a trend is consistently increasing or decreasing. In the Earth Trends Modeler, trend monotonicity is measured using the Mann-Kendall statistic. Here the picture is quite different – the region with the most consistent increase in lower tropospheric temperatures is the tropics. Although the eastern Pacific is less pronounced, this is probably only a result of persistent ENSO events in that region that reduce the measure of monotonicity.