As a result of regulatory decisions, atmospheric deposition of most toxic metals and metalloids (MEs) has decreased in Europe over the past few decades. However, little is known about how this reduction translates into exposure at higher trophic levels in the terrestrial environment where temporal trends may be spatially heterogeneous due to local current or legacy sources of emissions (e.g., industry) or long-range transport of elements (e.g., marine transport). The aim of this study was to characterize temporal and spatial trends of exposure to MEs in terrestrial food webs using a predatory bird, the tawny owl Strix aluco, as a biomonitor. Toxic (Al, As, Cd, Hg, Pb) and essential/beneficial (B, Co, Cu, Mn, Se) elemental concentrations were measured in feathers of nest-captured females from 1986 to 2016, extending a previous study published over the time-series 1986-2005 (n=1051), in a breeding population in Norway. A drastic decline over time was shown for the toxic MEs (-97% for Pb,-89% for Cd,-48% for Al, and-43% for As) except Hg. The beneficial elements B, Mn, and Se showed oscillations but an overall decline (-86%,-34%, and-12%, respectively) whereas the essentials Co and Cu did not exhibit significant trends. The distance to potential sources of contamination influenced both the spatial patterns of concentrations in owl feathers and their temporal trends. The accumulation of As, Cd, Co, Mn and Pb was overall higher in the vicinity of sites recorded as polluted, and a greater temporal decrease of As, B, and Cd concentrations was found in the areas of further distance to polluted sites. The decrease of Pb concentrations was sharper further from the coast during the 1980s than in coastal areas, while the opposite was observed for Mn. The levels of Hg and Se were higher in coastal areas, and Hg temporal trends differed according to the distance to the coast. This study highlights the valuable insights provided by long-term survey of wildlife exposure to pollutants and landscape indicators to reveal regional or local patterns and detect unexpected events, data that are crucial for regulation and conservation of ecosystem health.