Seasonal synchronicity and multi-decadal stability of headwater biogeochemistry in the northern temperate zone

Temporal patterns in chemistry of headwater streams reflect responses of water and elemental cycles to perturbations occurring at local to global scales. We evaluated multi-scale temporal patterns in up to 32 y of monthly observations of stream chemistry (ammonium, calcium, dissolved organic carbon, nitrate, total dissolved phosphorus, and sulfate) in 22 reference catchments within the northern temperate zone of North America. Multivariate autoregressive state-space (MARSS) models were applied to quantify patterns at multi-decadal, seasonal, and shorter intervals during a period that encompassed warming climate, seasonal changes in precipitation, and regional declines in atmospheric deposition. Significant long-term trends in solute concentrations within a subset of the catchments were consistent with recovery from atmospheric deposition (e.g., calcium, nitrate, sulfate) and increased precipitation (e.g., dissolved organic carbon). Lack of evidence for multi-decadal trends in most catchments suggests resilience of northern temperate ecosystems or that subtle net effects of simultaneous changes in climate and disturbance regimes do not result in directional trends. Synchronous seasonal oscillations of solute concentrations occurred across many catchments, reflecting shared climate and biotic drivers of seasonality within the northern temperate zone. Despite shared patterns among catchments at a seasonal scale, multi-scale temporal patterns were statistically distinct among even adjacent headwater catchments, implying that local attributes of headwater catchments modify the signals imparted by atmospheric phenomena and regional disturbances. To effectively characterize hydrologic and biogeochemical responses to changing climate and disturbance regimes, catchment monitoring programs could include multiple streams with contributing areas that encompass regional heterogeneity in vegetation, topography, and elevation. Overall, detection of long-term patterns and trends requires monitoring multiple catchments at a frequency that captures periodic variation (e.g., seasonality) and a duration encompassing the perturbations of interest.