Saltwater intrusion affects nitrogen, phosphorus and iron transformations under oxic and anoxic conditions: an incubation experiment

Abstract

Low-lying coastal ecosystems are rapidly salinizing due to sea level rise and associated saltwater intrusion (SWI). In agricultural soils, SWI can alter biogeochemical cycling of key nutrients such as nitrogen (N), phosphorus (P), and iron (Fe). The main objective of this study was to quantify the amount of nitrate–N (NO3–N), ammonium–N (NH4–N), soluble reactive P (SRP), and total dissolved iron (TDFe) released from agricultural soils undergoing SWI to determine their potential loss to downstream waterways. Agricultural soils were incubated for 0, 15, and 30 days (under oxic and anoxic conditions) with various salt solution combinations of sodium chloride (NaCl), sodium sulfate (Na2SO4), calcium sulfate (CaSO4), and Instant Ocean® to mimic (1) different ionic constituents of saltwater at different ionic strengths and (2) the presence or absence of gypsum, a soil amendment, through the addition of CaSO4. We also included a set of incubations treated with deionized water as a no ionic strength control (0.00 M). To increase statistical power, we grouped individual salt treatments based on our initial hypotheses at the end of the incubation period (day 30) to determine the effects of high (0.26–0.28 M) and low (0.03–0.04 M) ionic strength on inorganic N release and combinations of Ca and SO42− additions on SRP release to microcosm soil solution. Calcium additions decreased SRP release relative to saltwater that contained only NaCl additions under oxic and anoxic conditions. Additionally, high ionic strength treatments, which were about 7 times the ionic strength of low ionic strength treatments, released two times as much NH4–N to the soil solution, which suggests a non-linear relationship between ionic strength and NH4–N release. At day 30, anoxic microcosm soils treated with Instant Ocean® to simulate 15 parts per thousand seawater (ionic strength 0.26 M) released significantly more NH4–N (by 782 times), SRP (by 6 times), and TDFe (by 197 times) to the soil solution than oxic microcosm soils (P < 0.05). This treatment was designed to reflect a field inundated by brackish seawater for almost a month, which is representative of conditions on farm fields undergoing SWI. Under these anoxic conditions, as much as 22% of bioavailable soil P (as SRP) and 45% of total inorganic N (as NH4–N) could be released to overlying water when inundated with saltwater. Our work indicates that the influx of salts and inundation of SWI-affected farm fields could lead to a large export of N and P from agricultural soils and potentially affect downstream water quality.