Structural engineering defines damage as changes in material property, boundary conditions, or geometry. The changes in these parameters lead to a change in the measured response. The difference in measurements can be due to actual damage in the member due to crack or corrosion, or it might be due to environmental variability. Environmental variability, such as variation in temperature while making measurements, significantly influences the ability to quantify structural damage in structural health monitoring programs. Performing damage detection without isolating these variations can lead to false damage detection. Hence, a method is required to isolate the effect of these variabilities while detecting damage. Various methods have been developed and analyzed to separate environmental-based effects from damage-induced changes in the measures. Generally, two main approaches have emerged from research activity in this field: (a) statistics-based tools analyzing patterns in the data or computed parameters and (b) methods utilizing the structural model of the system considering environmental as well as damage-based changes of stiffness values. This paper addresses the problem of detecting damage under environmental variability. The proposed method uses the Approximate Bayesian Computation Nested Sampling (ABC-NS) algorithm to detect damage under temperature variations. The paper introduces a new damage index for identifying potentially damaged members. After performing damage localization, we estimate the parameters’ posterior distribution for potentially damaged members using ABC-NS to quantify damage. All the relevant data and codes used in this study will be available at https://github.com/akashyadav0210/ABC_SHM .