The cross correlation of the ABS and ACT maps
AI-generated Key Points
⚠The license of the paper does not allow us to build upon its content and the key points are generated using the paper metadata rather than the full article.
- Comprehensive analysis of cross-correlation between ABS and ACT maps
- Significant correlations among measurements and with Planck satellite data in EE and TE spectra at $\ell<400$ over 1100 deg$^2$
- Consistency of TB, EB, and BB cross-spectra with noise suggests minimal systematic errors
- Importance of using cross-correlations to validate results and identify potential errors
- Crucial for future Chilean experiments exploring $\ell$ range between 30 and 8,000 to enhance reliability and accuracy
Authors: Zack Li, Sigurd Naess, Simone Aiola, David Alonso, John W. Appel, J. Richard Bond, Erminia Calabrese, Steve K. Choi, Kevin T. Crowley, Thomas Essinger-Hileman, Shannon M. Duff, Joanna Dunkley, J. W. Fowler, Patricio Gallardo, Shuay-Pwu Patty Ho, Johannes Hubmayr, Akito Kusaka, Thibaut Louis, Mathew S. Madhavacheril, Jeffrey McMahon, Federico Nati, Michael D. Niemack, Lyman Page, Lucas Parker, Bruce Partridge, Maria Salatino, Jonathan L. Sievers, Cristóbal Sifón, Sara M. Simon, Suzanne T. Staggs, Emilie Storer, Edward J. Wollack
Abstract: One of the most important checks for systematic errors in CMB studies is the cross correlation of maps made by independent experiments. In this paper we report on the cross correlation between maps from the Atacama B-mode Search (ABS) and Atacama Cosmology Telescope (ACT) experiments in both temperature and polarization. These completely different measurements have a clear correlation with each other and with the Planck satellite in both the EE and TE spectra at $\ell<400$ over the roughly $1100$ deg$^2$ common to all three. The TB, EB, and BB cross spectra are consistent with noise. Exploiting such cross-correlations will be important for future experiments operating in Chile that aim to probe the $30<\ell<8,000$ range.
Ask questions about this paper to our AI assistant
You can also chat with multiple papers at once here.
⚠The license of the paper does not allow us to build upon its content and the AI assistant only knows about the paper metadata rather than the full article.
Assess the quality of the AI-generated content by voting
Score: 0
Why do we need votes?
Votes are used to determine whether we need to re-run our summarizing tools. If the count reaches -10, our tools can be restarted.
The previous summary was created more than a year ago and can be re-run (if necessary) by clicking on the Run button below.
⚠The license of this specific paper does not allow us to build upon its content and the summarizing tools will be run using the paper metadata rather than the full article. However, it still does a good job, and you can also try our tools on papers with more open licenses.
Look for similar papers (in beta version)
By clicking on the button above, our algorithm will scan all papers in our database to find the closest based on the contents of the full papers and not just on metadata. Please note that it only works for papers that we have generated summaries for and you can rerun it from time to time to get a more accurate result while our database grows.
Disclaimer: The AI-based summarization tool and virtual assistant provided on this website may not always provide accurate and complete summaries or responses. We encourage you to carefully review and evaluate the generated content to ensure its quality and relevance to your needs.