Large-scale tests of the Dvali-Gabadadze-Porrati model

The self-accelerating braneworld model (DGP) can be tested from measurements of the expansion history of the Universe and the formation of structure. Current constraints on the expansion history from supernova luminosity distances, the cosmic microwave background (CMB), and the Hubble constant exclude the simplest flat DGP model at about $3\sigma$. Even including spatial curvature, the best-fit open DGP model is a marginally poorer fit to the data than flat $\Lambda \mathrm{CDM}$. Moreover, its substantially different expansion history raises serious challenges for the model from structure formation. A dark energy model with the same expansion history would predict a highly significant discrepancy with the baryon oscillation measurement due the high Hubble constant required. For the DGP model to satisfy this constraint new nonlinear phenomena must correct this discrepancy. Likewise the large enhancement of CMB anisotropies at the lowest multipoles due to the ISW effect would require either a cutoff in the initial power or new phenomena at the cross-over scale. A prediction that is robust to both possibilities is that high-redshift galaxies should be substantially correlated with the CMB through the ISW effect. This correlation should provide a sharp test of the DGP model in the future.