Individuals inhabiting different local habitats respond differently to environmental fluctuations either by resisting changes or by moving across local habitats to maximize their fitness. Consequently, adaptation, population dynamics and their interactions with community and ecosystem dynamics should be manifested at landscape scales. Understanding these processes is a fundamental issue in ecology and evolutionary biology, but is extremely challenging because of (1) the difficulty of accurately estimating population sizes of species with wide distributions from a limited number of samples, and (2) the challenge of assessing where individuals are born and grow from samples collected at a limited sampling location. In this talk, we will present our recent approach integrating population genomic analysis with isotopic analysis to address the above challenges using an anadromous salmonid (Satsuki-masu) inhabiting a large watershed (Nagara River system) as a model. For estimating population size, we first detected certain numbers of siblings among randomly collected salmon individuals by the pedigree analysis based on the genome-wide SNPs. Then, the effective number of breeders (Nb) at a whole-watershed in certain years was successfully estimated by using a sibship assignment methods implemented in COLONY, a program for parentage and sibship inference. Second, we measured high-resolution strontium isotope records from otoliths of individual salmon and estimated their birthplaces from the large watershed using a state-space model. Finally, by integrating the two estimates (i.e., population size and individual birthplaces), the Nb of individual breeding habitats was estimated and visualized in the large watershed, allowing us to develop the integrative approach to estimate/visualize the meta-population dynamics at landscape scale. We would like to conclude the presentation by sharing ideas on how we can extend this approach to further understand meta-ecosystem dynamics driven by migratory animals.