WAIS Divide Ice Core δ15N(NO3−) Variability Over the Last Glacial-Interglacial Cycle: Controlling Factors and Implications for Retrieving Surface Mass Balance Using Ice Core δ15N(NO3−) Records

Abstract

Ice core nitrate can serve as a valuable tracer for past atmospheric nitrogen oxide (NO_x) and oxidant concentrations. However, photolysis-driven postdepositional processing can alter ice core nitrate signals and complicate their interpretation. We present a new nitrogen isotope (δ¹⁵N) record of nitrate measured in the West Antarctic Ice Sheet (WAIS) Divide ice core covering the last glacial period and Holocene. The glacial δ¹⁵N(NO₃⁻) is substantially higher than the Holocene δ¹⁵N(NO₃⁻), with a glacial-interglacial difference of (26.6 ± 5.7) ‰. All samples exhibit a strong negative correlation between δ¹⁵N(NO₃⁻) and the snow accumulation rate, suggesting that postdepositional processing is the primary driver of the δ¹⁵N(NO₃⁻) variability. Photochemical model calculations indicate that changes in the degree of postdepositional processing can fully explain the observed glacial-interglacial δ¹⁵N(NO₃⁻) difference, with 12.7% and 31.8% of nitrate mass loss during the Holocene and glacial climate, respectively. Comparison with the Greenland GISP2 δ¹⁵N(NO₃⁻) record indicates that the glacial level of postdepositional modification was higher at WAIS Divide despite its snow accumulation rate was twice that at Summit, Greenland. This is due to the higher light-absorbing impurity contents in the GISP2 ice core that reduced postdepositional losses. We also assess the ability of using δ¹⁵N(NO₃⁻) for past surface mass balance (SMB) reconstruction at WAIS Divide. We find this proxy is reliable for regions/periods with snow accumulation rates lower than 150 kg m⁻² a⁻¹. While exceeding this threshold, the assumption of a constant δ¹⁵N of initially deposited nitrate is no longer valid, leading to significant bias in SMB reconstruction such as WAIS Divide.