The recent emergence of publicly shared large-scale functional neuroimaging datasets (e.g., ABCD, HCP, ADNI, UK Biobank) offers many opportunities for robust and generalizable research into population-level variability and disease biomarkers. However, comparing and combining neuroimaging data across disparate projects is non-trivial due to systematic biases that are introduced as a result of differences in acquisition hardware and/or protocols. Therefore, there is an urgent need for advanced data harmonization methods that correct for these biases and enable direct comparisons within and across large-scale neuroimaging datasets. But to validate harmonization methods it is critical to have within-subject data collected under these differing protocols, which currently does not exist. To begin to address this problem, in this Administrative Supplement to the ABCD-USA Consortium: Data Analysis, Informatics and Resource Center we propose to increase the scientific utility and usability of five large-scale neuroimaging datasets (ABCD, HCP-Lifespan, ADNI, UK Biobank, and Baby Connectome Project) by generating a within-subject, cross-project neuroimaging harmonization dataset. This will enable: a) using one dataset as a replication data set for analyses conducted on other datasets; and b) aggregating data across projects in order to generate even larger sample sizes for sophisticated modeling and data-driven analyses, including the ability to have out of sample generalization analyses. In addition, many other investigators are generating datasets using one or more of these imaging protocols, and will wish to harmonize both with the protocol from which they based their MR acquisitions, and with other datasets. We will generate the data to develop and validate critically needed harmonization methods, and make the data publicly available.