Does DESI DR2 Challenge ΛCDM Paradigm?

Abstract Although the debate about the systematic errors of DESI DR1 is still open, recent DESI DR2 is consistent with DESI DR1 and further strengthens the results of DESI DR1. In our analysis, both the LRG1 point at z eff = 0.510 and the LRG3+ELG1 point at z eff = 0.934 are in tension with the ΛCDM-anchored value of Ω m inferred from Planck and the Type Ia supernovae compilations Pantheon + , Union3, and DES-SN5YR. For luminous red galaxy 1 (LRG1) the tensions are 2.42 σ , 1.91 σ , 2.19 σ , and 2.99 σ , respectively; for LRG3+emission line galaxy 1 (ELG1) they are 2.60 σ , 2.24 σ , 2.51 σ , and 2.96 σ , respectively. From low to high redshift bins, DESI DR2 shows improved consistency relative to DESI DR1: the Ω m tension decreases from 2.20 σ to 1.84 σ . However, DESI DR2 alone does not provide decisive evidence against the ΛCDM model, and the apparent signal is largely driven by specific tracers, LRG1 and LRG2. In the ω 0 ω a CDM analysis, including all tracers yields a posterior mean with ω 0 &gt; −1, which aligns with scenarios of dynamical dark energy as a potential explanation and suggests that the DESI DR2 challenges the ΛCDM paradigm. While removing LRG1 and/or LRG2 fully restores ΛCDM concordance (i.e., ω 0 → −1), we also find <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:msubsup> <mml:mi>ω</mml:mi> <mml:mn>0</mml:mn> <mml:mrow> <mml:mspace width="0.1em"/> <mml:mtext>(LRG1)</mml:mtext> </mml:mrow> </mml:msubsup> <mml:mo>&gt;</mml:mo> <mml:msubsup> <mml:mi>ω</mml:mi> <mml:mn>0</mml:mn> <mml:mrow> <mml:mtext>(LRG2)</mml:mtext> <mml:mspace width="0.1em"/> </mml:mrow> </mml:msubsup> </mml:math> , indicating LRG1 drives the apparent dynamical dark energy trend more strongly. Model selection using the natural log Bayes factor <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" overflow="scroll"> <mml:mi>ln</mml:mi> <mml:mi mathvariant="normal">BF</mml:mi> <mml:mo>≡</mml:mo> <mml:mi>ln</mml:mi> <mml:mo stretchy="false">(</mml:mo> <mml:msub> <mml:mrow> <mml:mi class="MJX-tex-calligraphic" mathvariant="script">Z</mml:mi> </mml:mrow> <mml:mrow> <mml:mi mathvariant="normal">Λ</mml:mi> <mml:mi mathvariant="normal">CDM</mml:mi> </mml:mrow> </mml:msub> <mml:mo>/</mml:mo> <mml:msub> <mml:mrow> <mml:mi class="MJX-tex-calligraphic" mathvariant="script">Z</mml:mi> </mml:mrow> <mml:mrow> <mml:msub> <mml:mrow> <mml:mi>ω</mml:mi> </mml:mrow> <mml:mrow> <mml:mn>0</mml:mn> </mml:mrow> </mml:msub> <mml:msub> <mml:mrow> <mml:mi>ω</mml:mi> </mml:mrow> <mml:mrow> <mml:mi>a</mml:mi> </mml:mrow> </mml:msub> <mml:mi mathvariant="normal">CDM</mml:mi> </mml:mrow> </mml:msub> <mml:mo stretchy="false">)</mml:mo> </mml:math> shows weak evidence for ΛCDM when LRG1, LRG2, or both are removed, and it is inconclusive for the full sample; thus, the data do not require the extra ω a freedom, and the apparent ω 0 &gt; −1 preference should be interpreted cautiously as a manifestation of the ω 0 – ω a degeneracy under limited per tracer information.