The space-time distribution of sulfate deposition in the northeastern United States

• Published in: Atmospheric environment 1985, Vol.19 (11), p.1829-1845
• Main Author: Bilonick, Richard A.
• Format: Article
• Language: English
• Subjects: acid deposition; acid precipitation; Applied sciences; Atmospheric pollution; deposition; distribution; duration; Exact sciences and technology; kriging; mapping; Pollutants physicochemistry study: properties, effects, reactions, transport and distribution; Pollution; spatial statistics; Sulfate deposition; sulfates; U.S. Geological Survey
• ISSN: 0004-6981
• DOI: 10.1016/0004-6981(85)90009-5

The space-time semi-variograms of sulfate deposition data collected by the U.S. Geological Survey during the period 1965–1979 in New York State and during the winter of 1980–1981 over the north central and north eastern U.S. follow well-defined patterns. These patterns were estimated and used to produce sulfate deposition maps. The spatially and temporally weighted average level of sulfate deposition was about 50 kg ha −1 per year for 1965–1979. Individual yearly 80-km square block estimates had estimation standard deviations of about 6–9 kg ha −1 per year (or about 15 % relative). The space-time semi-variogram exhibited a well-defined seasonal component in time and an isotropic spatial component. Three-dimensional kriging allowed estimates to be based on data simultaneously dispersed in time and space. Sulfate deposition in the winter of 1980–1981 ranged from about 5 to 12 kg ha −1 per year for the more northern areas, 12 to 19 kg ha −1 per year for the central region and from 19 to 26 kg ha −1 per year for the more southern areas. The estimation standard deviations were roughly uniform within the bounds of the network and were about 2–4 kg ha −1 per year for 80-km square blocks. Replicate samples helped provide an improved estimate of spatial variability at short distances. The semi-variogram for the 1980–1981 data exhibited a strong anisotropy which indicated a drift in mean sulfate deposition in the north-south direction. This drift component was taken into account when estimating sulfate deposition by using the method of universal kriging. The reasonableness of extrapolation for sulfate deposition for blocks as far as 300–400 km outside the boundary of the network is discussed.