Analyzing Climate Change using the Earth Trends Modeler

Climate change has become the new norm. While climate models are designed to project how climate will be affected in the future, the Earth Trends Modeler (ETM) in the TerrSet software can tell you how it’s changing now. Earth Trends Modeler is an integrated suite of tools for the analysis of earth observation image time series such as satellite image series and gridded climate data sets. It offers state of the art tools for Earth system scientists while being exceptionally easy to use. Thus it is appropriate not only for front line research, but also for teaching and practical applications in local, state and national environmental management.

  • Linear Modeling

    An analysis of trends in sea surface temperature from 1982 to 2006. The strong monotonic trend of increasing temperature in the Atlantic is seen to be related to the Atlantic Multidecadal Oscillation (AMO) as determined from a temporal regression with four major climate teleconnection indices.

  • EOT El Nino Climate Change Analysis

    An Empirical Orthogonal Teleconnection (EOT) analysis of monthly anomalies in sea surface temperature (SST) from 1982 to 2007. EOT is a regression-based spectral decomposition technique which is similar in intent and character to an obliquely rotated Principal Components Analysis. Each EOT represents a major pattern of variability over time. The graph shows EOT1 which is clearly the El Niño/Southern Oscillation (ENSO) phenomenon while the foreground image shows what is known as the loading image – the correlation (over time) of each pixel location with this temporal pattern. The background image is the loading image for EOT2 – a combination of the Atlantic Multidecadal Oscillation with a global warming signal.

Some Key Features of Earth Trends Modeler for Analyzing Climate Change

  • Provides highly noise-resistant trend measures for both linear and non-linear trends. Earth observation data is commonly very noisy and contains many data gaps. ETM provides highly robust trend measures that can selectively avoid noise as well as gap-filling interpolation procedures to estimate the values for missing data.
  • Although climate change is often characterized in terms of average conditions, changes in the seasonal manifestation of climate are also extremely important. Changes in seasonality can have major impacts on biodiversity and the provision of ecosystem services such as fresh water and agriculture. ETM provides a new analytical tool called Seasonal Trend Analysis that is highly effective in characterizing trends in seasonality in the face of substantial high frequency noise and short-term interannual variability.
  • The climate system also has preferred patterns of variability known as teleconnections. ETM has a wide range of spectral decomposition procedures for the analysis of these kinds of patterns across space and time. These include Principal Components Analysis (PCA), Extended PCA, Multichannel Singular Spectrum Analysis (MSSA), Canonical Correlation Analysis (CCA) and Empirical Orthogonal Teleconnection (EOT) analysis.
  • For the analysis of climate oscillations, the system also includes Wavelet Analysis and Fourier PCA.
  • Imports major climate data formats including NetCDF and HDF. There has been a virtual explosion of freely available data to support climate change and land/earth system change as a result of programs such as NASA’s Earth Observing System (EOS). ETM is supported by the TerrSet software that provides facilities for the import of a wide variety of Earth observation data sets without limitation of size.
  • Does not require learning a complex scripting language. Although the TerrSet system that supports ETM provides full COM-compliant programming access, including Python, ETM itself provides a very rich feature set without the need for end user programming.
  • Comes with a tutorial and global monthly image time series from 1982 to 2010 for sea surface temperature, lower tropospheric temperature and precipitation.