Hydrochemical and isotopic investigation of atmospheric precipitation in Beijing, China

Abstract

Precipitation water samples were collected at an urban site in Beijing in a hydrological cycle (July 2008-July 2009), and analyzed for TDS, total alkalinity, total hardness, free CO₂, soluble SiO₂, bromide, sulfide, phosphate, major ions (K(+), Na(+), Ca(2+), Mg(2+), NH₄(+), HCO₃(-), Cl(-), SO₄(2-), NO₃(-)), trace elements (CO₃(2-), Mn, Sr(2+), Fe(2+), Fe(3+), Al, F(-), NO₂(-)), stable isotopes ((2)H and (18)O), and radioactive isotope ((3)H). In addition, available published hydrochemical and isotopic data of precipitation of Beijing in the past were also collected and conjointly analyzed. Most of the parameters of samples tested varied considerably in the hydrological cycle. In general, HCO₃(-) and SO₄(2-), and Ca(2+) and NH₄(+) are the dominant anions and cations, respectively. Using Na(+) as an indicator of marine origin, and Al for the terrestrial inputs, the proportions of major elements from sea salt and terrestrial sources were estimated by using the combination of statistical analysis methods and analogy method. More than 70.1% of Cl(-), 98.1% of Ca(2+), and 93.6% of K(+) were non-sea-salt origin, while more than 98.4% of Na(+) was from marine sources. The LMWL (Local Meteoric Water Line) was obtained with an equation of δ(2)H=7.0181δ(18)O+3.5231 (‰, R(2)=0.86, n=36), which was similar to GMWL (Global Meteoric Water Line). δ(2)H, δ(18)O and Δ-excess changed radically with month and season, but had no apparent seasonal effect, precipitation amount effect, and temperature effect. The annual mean values of Δ-excess for 1979 (16.5‰) and 1980 (16.3‰) were much bigger than that for 2007 (7.2‰), 2008 (2.1‰) and 2009 (4.5‰). The composition of (2)H and (18)O was probably intrinsically determined by the sources of water vapor and the physical and chemical processes occurred along the migration paths of water vapor from sources to the aimed precipitation area. Temporal change of (3)H was only influenced by the nuclear testing in the early 1960s worldwide and the natural yield of (3)H in the upper atmosphere intrinsically. (3)H had nearly approached to the natural levels, which would bring difficulty if not invalidation to groundwater dating using (3)H technique.