A climatology and time series for dissolved nitrate in the Gulf of Maine region

A compilation of data from 1932 to 2011 of dissolved inorganic nitrogen was used to generate spatially interpolated monthly climatologies for three depth layers in the Gulf of Maine region. Residuals from the spatial climatology are used to generate seasonal and annual time series. A bivariate co-kriging analysis was used to interpolate to a regular grid for each year within the historical record, then the resultant kriging error variance was used to determine a weighting value to generate the climatological mean. This interpolation method generates a realistic estimate as it incorporates most of the available nutrient data, geographic extrapolations based on defined spatial covariance functions, and an objective reproducible methodology. Residuals from the interpolated fields show the Gulf of Maine tends to vary on seasonal and annual timescales as a single cohesive unit, whereas intra-seasonal dynamics are more localized. Anomalies during the spring drawdown and fall replenishment are highly dependent on timing of the seasonal progression and less reliant on preceding nutrient conditions. Lagged correlations of residuals suggest seasonal mixing dynamics may be more important than elevated nutrient concentrations at depth to supplying surface waters with available nutrients.