Synchrony in ecological variables over wide geographic areas suggests that large-scale environmental factors drive the structure and function of ecosystems and override more local-scale environmental variation. Described also as coherence, this phenomenon has been documented broadly in the ecological literature and has recently received increasing attention as scientists attempt to quantify the impacts of global changes on organisms and their habitats. Using a mesic grassland site in North America, we assessed coherence in ecosystem function by quantifying similarity in aboveground net primary production (ANPP) dynamics in 48 permanent sampling locations (PSLs) over a 16-yr period. Our primary objective was to characterize coherence across a broad geographic region (with similar ecosystem structure and function), and we hypothesized that precipitation and a similar fire frequency would strengthen coherence between PSLs. All 48 PSLs at our site (Konza Prairie Biological Station, Manhattan, KS, USA; KPBS) were exposed to a similar regional driver of ANPP (precipitation); however, local drivers (including differences in fire frequency and soil depth at different topographic positions) varied strongly among individual PSLs. For the purpose of this assessment, the watershed-level experimental design of KPBS was considered a model, which represented different fire management strategies across the Great Plains Region. Our analyses revealed a site-level (KPBS) coherence in ANPP dynamics of 0.53 for the period of 1984–1999. Annual fire enhanced coherence among PSLs to 0.76, whereas less frequent fire (fire exclusion or a 4-yr fire return interval) failed to further increase coherence beyond that of the KPBS site level. Soil depth also strongly influenced coherence among PSLs with shallow soils at upland sites showing strong coherence across fire regimes and annually burned uplands closely linked to annual precipitation dynamics. The lack of coherence in ecosystem function in PSLs with deep soils and low fire frequencies suggests that conservation and management efforts will need to be more location specific in such areas where biotic interactions may be more important than regional abiotic drivers.