Background Recent progress in understanding the genetic basis of many psychiatric diseases has identified rare and common variants responsible for genetic risk. Integrating epigenomics data from disease-relevant cell types and tissues promises to enhance interpretation of these risk variants and the mechanisms by which they confer disease liability. A better understanding of the role of non-coding regulatory DNA and variation in histone modification in mediating the effects of genetic risk variants promises to give insights into the molecular mechanisms underlying psychiatric disease. With this goal in mind, the PsychENCODE Consortium generated a large-scale epigenomics resource for the human brain to serve as a foundation for integrative genomics in psychiatric research. Methods We performed ChIP-Seq on 17 subjects, 2 brain regions, 2 cell types and 2 histone marks that yielded 136 total ChIP-Seq samples. Of these 129 (n=63 H3K4me3; n=66 H3K27ac) passed quality control. For each combination of brain region, cell type and histone mark, reads from corresponding samples were combined into a consolidated subset of the data. This produced 8 consolidated subsets, corresponding to: ACC-neuronal-H3K4me3, PFC-neuronal-H3K4me3, etc. The entire data set was then processed and analyzed through the annotation, peak calling, covariate and pathway analyses pipeline. Results We have generated the largest cell type-specific dataset of histone-modification in the human brain to date across two brain cortical regions, composed of reference maps from 129 samples collected from 17 healthy subjects. These reference maps consist of ChIP-Seq for two histone marks associated with promoters and active enhancers, H3-trimethyl-lysine 4 (H3K4me3) and H3-acetyl-lysine 27 (H3K27ac) from either neurons, or surrounding non-neuronal cells. Analysis of this data has revealed unexplored insights into cell-, brain region- and to a lesser extent, subject- and sex-specific histone methylation and acetylation landscapes shaping the epigenome of the adult human cerebral cortex. We identify multiple loci with robust region-specific differences separating prefrontal from cingulate neurons, and numerous genome loci with extraordinary high inter-individual variability in H3K27ac and H3K4me3. Finally, we found a significant enrichment of risk variants for schizophrenia, educational attainment and depressive symptoms only in neuronal markers, further indicating the importance of conducting cell type-specific epigenome studies. Discussion Risk variants for schizophrenia, depressive symptoms, neuroticism and educational years are significantly enriched in genomic loci marked specifically in neuronal cells. The difference between neuronal and non-neuronal cells is the major axis of variation in histone modification, yet many genomic loci show substantial epigenetic variability across subjects and brain regions. These findings highlight the utility of this NIMH PsychENCODE-sponsored resource to elucidate epigenomic architectures of regulatory and disease-associated non-coding DNA in the human brain.