Testing Gauss–Bonnet gravity with DESI BAO data

Abstract In this work, we observationally constrain two phenomenological f ( G ) gravity models at the background level using Type Ia supernovae from the Pantheon+ (PP) sample, cosmic chronometer (CC) measurements, and the recent Baryon Acoustic Oscillation (BAO) data released by DESI. The analysis is performed using two dataset combinations: (i) PP+CC and (ii) PP+CC+DESI BAO. The modified Friedmann equations are solved numerically for two functional forms of f ( G ) gravity, namely the power-law and exponential models. The cosmological parameters are constrained through Markov Chain Monte Carlo simulations. To evaluate the statistical performance of the models, we compute the corrected Akaike Information Criterion (AIC $$_c)$$ c ) and the Bayesian Information Criterion (BIC). Our results show that both f ( G ) models provide background evolutions consistent with current observational data. While the AIC $$_c$$ c criterion indicates that the models are statistically comparable to the $$\varLambda $$ Λ CDM model, the BIC slightly favors the simpler $$\varLambda $$ Λ CDM scenario due to its smaller number of free parameters. In addition, the exponential model predicts a distinctive future dynamical feature, namely an additional transition from accelerated to decelerated expansion at $$z \approx -0.086,$$ z ≈ - 0.086 , which is absent in both the power-law model and the $$\varLambda $$ Λ CDM framework.