Mapping the patterns of internal migration in Latvia using spatial autocorrelation methods

Over the past decade, the dynamics of population change and redistribution have been a subject of academic discussion, particularly in thriving metropolitan regions of major cities and in non-metropolitan regions and rural peripheries that are experiencing depopulation. Internal migration has replaced fertility and mortality as the primary demographic process that shapes the spatial distribution of populations across regions within countries. This is a global phenomenon, though its intensity and patterns vary significantly across different national contexts. While a rich comparative literature has examined the intensity, composition, and flows of population movement, the spatial patterns of internal migration are less well understood. Spatial autocorrelation methods can reveal spatial patterns; however, their application to internal migration research remains limited, particularly in examining the temporal dynamics of spatial clustering. While recent studies have begun applying these methods to migration analysis, few have investigated how spatial autocorrelation patterns evolve over time. This study aims to explore the spatial patterns of internal migration using selected indicators of spatial autocorrelation, with Latvia as a case study. The analysis covers three census periods (2000, 2011, and 2021) and employs a hexagonal grid spatial framework at both national and metropolitan scales.