This letter presents a wavelength-resolution syn-thetic aperture radar incoherentchange detection method basedon image stack, i.e., there are more than one reference or/andsurveillance image. Considering image stack in statistical hypoth-esis test for change detection is expected to result into a simplemathematical expression for implementation and provide betterchange detection results. As presented in this letter, a statisticalhypothesis test is developed on bivariate Gaussian distributionfor an image stack of two reference images and one surveillanceimage. The requirement for the image stack is three imagesassociated with three measurements with no change between twoof them. A detection method with simple processing scheme isproposed. The method is experimented with 24 CARABAS datasets. The results indicate that high average detection probability,e.g., 96%, with very low false alarm rate, e.g., only 0.19/km2,is obtained with the proposal.