For the noise problem of two-dimensional concentration distribution reconstruction of trace gases in the atmosphere, the method of numerical simulation is adopted to study the influence of noise on the spatial distribution reconstruction of atmospheric trace gases by the differential absorption spectrum tomography. The noise model is established, the algebraic iterative reconstruction algorithm (ART) and simultaneous iterative reconstruction algorithm (SIRT) are researched, and the influence of noise on the reconstruction effects of the two kinds of reconstruction algorithm is also researched respectively. The error index Ea can be reduced from 0.1080 to 0.0182 by reconstruction of ART algorithm, and the error index Ea can be reduced from 0.0476 to 0.0102 by reconstruction of SIRT algorithm. The results show that with the increase of random noise, the abnormal degree of reconstruction image is more and more serious. From the relative error indicators of reconstruction, with the increase of signal-to-noise ratio, reconstruction of relative error is smaller and smaller. Therefore, noise should try to be avoided in field reconstruction experiment.