Special reports focus on environmental issues and allow a closer examination of sentences that contain references to a certain topic and at least one of the candidates. The first two reports available in this series cover energy policy and global warming.

The sentences can be sorted by semantic orientation (positive or negative), or listed in random order. A link at the end of each sentence points towards the original article containing the sentence, or to the organization's homepage in case the article is no longer available. Moving the mouse pointer over underlined words reveals their semantic category: Positive, Negative, Virtue, Vice, Fail, Complete, PosAffect and NegAffect. The number indicates the weight of this term for computing the semantic orientation.

• Global Warming. A devastating hurricane season and the Russian government's recent decision to ratify the Kyoto Protocol have put climate change back in the limelight. But how important is the issue of global warming in the run for US presidency? An in-depth investigation of the the Web coverage on global warming, greenhouse gases and climate change tries to answer this question.
• Energy Policy. What strategies have presidential candidates suggested to cope with record-level energy prices and increasing domestic demand? Public glossaries of the Energy Information Association, the US Department of Energy, and the California Energy Commission helped compile the lists of terms represented in this section: renewable energy, fossil fuels and nuclear energy.
• Media Map. Correspondence analysis uncovers general patterns in textual Web data. It approximates the original distribution of data points in only two dimensions. The resulting diagram compares the prominence of energy terms (round markers) among selected media sites (rectangular markers). Concepts that appear close to each other have a similar distribution of term frequencies. Please note that term frequencies and not the media's attitudes determine the exact position of the markers. The analysis considered media with at least 50 sentences containing energy-related terms. This explains why the diagram omits some well-known names.