Theory and Joint Probes

LSST will enable transformative improvements in our understanding of dark energy and the nature of gravity on cosmic scales. Realizing optimal constraints on the nature of dark energy from LSST will involve the collaborative effort of experts in observation, theory, simulation, and data analysis.

A crucial advantage of a survey like the LSST survey is that the same data can be used for multiple probes. These probes will form multiple lines of attack at the puzzle of dark energy and provide internal cross checks that could potentially detect unknown systematics. Moreover, when analyzed jointly, these probes will enable self-calibrations of systematics, reduce degeneracies between parameters, and strengthen the constraints on dark energy properties. An example of weak lensing (WL) and Baryon Acoustic Oscillations (BAO) is shown in this image:

1 sigma error contours of the dark energy equation of state (w=p/rho) parameters w0 and wafrom LSST WL shear power spectra (left panel) and joint LSST WL and BAO (right panel). The shaded areas represent the results with statistical errors only. The solid contours correspond to those with the anticipated level of systematic errors, which include the uncertainty in the photometric redshift error distribution and additive and multiplicative errors in the power spectra. The dotted contours are calculated with several times larger systematic uncertainties. The joint WL and BAO results are much less affected by these systematics because of the ability to self-calibrate the systematics. For details, see §15.1.3 of the LSST Science Book.

Dark energy studies will also benefit from the combination and cross-correlation of LSST survey data with precursor and contemporary external datasets. It will be a major undertaking to coordinate the connections of all these multiple efforts into the cosmological constraint analysis pipeline. This includes the formation of a software framework to facilitate a common platform for testing and integrating data analysis, simulation and theoretical prediction codes, and data products.



Conveners:  Jonathan Blazek (Ohio State) [blazek dot 35 at osu dot edu], Elisabeth Krause (SLAC) [lise at slac dot stanford dot edu]