Prestack seismic attributes are efficient tools for hydrocarbon exploration and pore fluid detection with the help of various techniques, such as amplitude variation with offset analysis. Such studies focus mainly on siliciclastics rather than carbonates because detection of fluid effects in carbonate rocks can be masked by their complex pore structure and heterogeneity. Current fluid detection methods from seismic attributes usually rely on a linear background model for P- and S-wave velocities of the water-saturated rocks, and any deviation from this trend is assigned to possible pore fluid changes. This means that the false (or even missed) effect of fluids can be detected in carbonate rocks if inappropriate fluid detection attributes (such as the ones designed for siliciclastics) are used. This is mainly due to the varying pore structure in carbonates, which can make their background model (the model for fully water-saturated rock) mainly nonlinear. I observed that this nonlinearity in the carbonates background model becomes more linear by using P-velocity squared versus the product of the P- and S-velocities crossplot instead of P-velocity versus S-velocity crossplot. Furthermore, I used this proposed crossplot to define a more appropriate background model for my carbonate sequence containing some percentages of gas. I derived a new seismic fluid attribute based on the proposed background model, and I compared the results with various other fluid factors. My results highlight fluid changes more brightly and consistently than existing alternatives for carbonate environments.