This study evaluates how eDNA information could be used within Environmental Impact Assessment (EIA). We developed an original model to simulate the conditions for which an eDNA signal detects, or does not detect, an impact on a targeted (receptor) species in a given project area. The simulation has four consecutive steps. First, a deterministic model simulated the dynamics of the receptor population and their eDNA fragment concentrations in the environment. Second, random distributions of receptor organisms and eDNA fragment quantities at steady-state were simulated within the project area. Then Simple Random Samplings were performed for both the receptor and eDNA. Third, post-sampling processes (eDNA extraction, amplification, analysis) were simulated to estimate the detection probability of the species from sample plan characteristics (size of sampling unit, number of samples collected). Fourth, we simulated an impact by modifying the growth, mortality and mobility (null, passive and active) parameters of the receptor species, then determined if an impact was detected. Detection probability curves were estimated for a range of sample volumes fitted with a Weibull cumulative distribution function. An F-like statistic compared detection curves before and after impact. Twelve scenarios were simulated. A statistically significant impact was detected with eDNA when receptor species growth rate was halved, but only in cases of null or passive mobility. When the receptor experienced both reduced growth and increased mortality rates, an impact was detected in all three mobility cases (null, passive and active). Our results suggest that an impact could be detected using eDNA if both the population dynamics of the receptor and the dynamics of DNA shed into the environment are known. These results indicate that caution should be exercised with eDNA data for EIA, but the proposed framework provides a valuable starting point to improve interpretation of indirect observation methods such as eDNA.