Drought is one of the most critical abiotic stresses influencing maize yield. Improving maize cultivars with drought tolerance using marker-assisted selection requires a better understanding of its genetic basis. In this study, a doubled haploid (DH) population consisting of 217 lines was created by crossing the inbred lines Han 21 (drought-tolerant) and Ye 478 (drought-sensitive). The population was geno-typed with a 6 K SNP assay and 756 SNP (single nucleotide polymorphism) markers were used to con-struct a linkage map with a length of 1344 cM. Grain yield (GY), ear setting percentage (ESP), and anthesis–silking interval (ASI) were recorded in seven environments under well-watered (WW) and water-stressed (WS) regimes. High phenotypic variation was observed for all traits under both water regimes. Using the LSMEAN (least-squares mean) values from all environments for each trait, 18 QTL were detected, with 9 associated with the WW and 9 with the WS regime. Four chromosome regions, Chr. 3:219.8–223.7 Mb, Chr. 5:191.5–194.7 Mb, Chr. 7:132.2–135.6 Mb, and Chr. 10:88.2–89.4 Mb, har-bored at least 2 QTL in each region, and QTL co-located in a region inherited favorable alleles from the same parent. A set of 64 drought-tolerant BC3F6 lines showed preferential accumulation of the favorable alleles in these four regions, supporting an association between the four regions and maize drought tol-erance. QTL-by-environment interaction analysis revealed 28 edQTL (environment-dependent QTL) asso-ciated with the WS regime and 22 associated with the WW regime for GY, ESP, and ASI. All WS QTL and 55.6％of WW QTL were located in the edQTL regions. The hotspot genomic regions identified in this work will support further fine mapping and marker-assisted breeding of drought-tolerant maize.