ABSTRACT Human complex traits are jointly influenced by genetic and environmental risk factors, whose exact contributions are often subject to extensive debate. Detailed environmental risk factors are not often available, which makes it hard to jointly assess the genetic and environmental contributions. Yet, the emergence of large- scale national biobanks as well international genetic studies offers a great opportunity to make up for this knowledge gap. In particular, as study participants come from diverse locations, geospatial information of the study participants can be used as a proxy for environmental exposure. Models that incorporate geospatial information of study participants will lead to improved power for association analysis and more accurate heritability estimates. In this application, we propose to develop a Spatial MIxed Linear Effect model (SMILE) for improved association analysis and heritability estimation and Spatial Meta-Analysis Regression Test (SMART) for more powerful meta-analyses of genetic association studies. We will apply them to UK Biobank, MarketScan insurance billing database, TOPMed sequence data, and various large consortia studies on smoking/drinking addictions, lipids levels, and diabetes. To achieve the proposed research aims, we assembled a strong research team with complementary expertise from statistical genetics, addiction genetics, lung function genetics, biomedical informatics, and environmental epidemiology. Methods and tools developed from this study will open up new avenues for analyzing national biobanks such as UK Biobank and All of Us cohorts, and global consortium studies. The results from this study will help elucidate the genetic architecture of complex traits with significant