Identification (ID) codes offer a method to optimize the required traffic if the goal of the communication is message identification. This approach can drastically reduce traffic between digital twins and their real-world data sources. While previous investigations focused on noiseless ID, we find that the traffic reduction via ID over noisy channels offers additional benefits than previously considered. Specifically, every message bit that we do not need to transmit can instead be used as a redundant bit by forward error correction (FEC) codes to mitigate the effects of noisy channels. Therefore, reducing the required traffic corresponds to the possibility of increasing reliability in ID over noisy channels that benefit from additional FEC coding. Employing identification codes enables communication with some reliability that would not be possible using traditional message transmission at practical block lengths for low SNR communication channels. We find that errors introduced by ID codes are negligible when compared with the reliability gains introduced by the extra redundancy that can be added to the transmissions. For a fixed block length, employing ID codes enables the usage of lower FEC code rates, which cause a reduction in the error probability that corresponds to an improvement of 4 dB SNR. Furthermore, we propose a method to reduce traffic even further when transmitting multiple tags, by avoiding the transmission of all but one random position for the tag determination of ID codes.