| Atmospheric oxidation capacity is an important indicator of the self-cleaning ability of the troposphere,and is of great importance to the migration and transformation rate of atmospheric trace components.In the context of global warming,it is of great scientific significance to explore the future trend of atmospheric oxidation and its feedback effect on climate.The paleoclimate proxy recorded the evolution of historical atmospheric oxidation can provide new cognition for understanding the mechanism of atmospheric oxidation change.The concentration and oxygen massindependent fractionation signal of nitrate preserved in ice core is one of the most promising indicators of atmospheric oxidation.Atmospheric reactive nitrogen(NOx=NO+NO2)plays an important role in the regeneration of atmospheric oxidizing free radicals and the formation of tropospheric ozone,and nitrate is the final product of atmospheric NOx.Therefore,ice core nitrate can be used to infer the historical variation of atmospheric NOx abundance.The oxygen oxygen mass-independent fractionation signal of nitrate originates from the contribution of ozone during NOx cycle and its oxidation to nitrate,thus can be used to reflect the relative changes of atmospheric ozone and peroxic radical abundance.The main factors affecting the important atmosphere oxidized component ozone include the emission of ozone precursors such as NOx,VOC and methane and the tropospheric-stratospheric exchange controlled by the Brewer-Dobson(BD)circulation,which may also affect tropospheric ozone photolysis rate by impacting the distribution of stratospheric ozone,which in turn affects the atmosphere oxidating capacity.Oxygen isotope of ice cores nitrate have shown that the enhanced BD circulation during the glacial period was an important driving factor impacting on the atmosphere oxidation capacity but chemistry-cliamate model have shown opposite sign in the modelled strength of the BD circulation during the glacial period,and the enhnaced reactive bromine chemistry from sea ice in glacial period may also partly contribute the observed oxygen isotope changes in ice core nitarte.Therefore,the changing direction of the BD circulation during the glacial period and the effect of reactive bromine chemistry on the oxygen isotope of ice core nitrate are still inconclusive.Besides,although ice core nitrate and its isotopic records hold ample information regarding the historical variation of atmospheric NOx and its sources,the post-depositional effect of snow nitrate would hamper such interpretations.Importantly,the impact of postdepositional processing on the preservation of snowpack nitrate and its isotopes at high accumulation rate areas is still under debate.In this study,we took the typical high accumulation rate site Summit,Greenland as an example to quantitatively explore the effect of post-depositional processing on snow nitare at present day by using numerical model simulations and cros-validations using field observations.Based on these modeling and observation results,we constructed an Inverse model which aims to correct the effects of post-depositional processing in ice core nitrate.We reported the first ice core nitrate isotopic records at West Antarctica WAIS Divide covering the last glacial period to Holocene,and using the constructed Inverse model to correct bipolar ice core records including WAIS Divide and GISP2.With the help of global climate models(ICECAP),we estimated the changes of atmospheric NOx abundance in the Northern and Southern hemispheres during glacial and interglacial periods and quantitatively assessed the relative contributions from sea ice reactive bromine chemistry and BD circulation on ice core nitrate oxygen isotopes.We also provided multiple evidences proving that an enhanced BD circulation happened in glacial period.The main conclusions are listed below:1.We reported the first annual atmospheric nitrate isotopic record at Summit,Greenland,and combined with the published nitrate isotope data at Summit to provide evidences for the existence of post-depositional effect therein.Our compiled dataset revealed a systematically enriched trend in δ15N(NO3-)from atmosphre to surface snow and then to snowpack.Such enrichment inδ15N(NO3-)was consistent with the effect of the photo-driven postdepositional processing.Snowpack and atmospheric ⊿170(NO3-)displayed a generally similar seasonal pattern which indicated that the atmospheric⊿17O(NO3-)signal was well preserved at Summit under present conditions,but the ⊿170 and δ180 of primary nitrate would be affected by post-depositional processes to different degrees.2.We used the TRANSITS model to quantitatively evaluate the impact of postdepositional processing on snowpack nitrate and its isotopes at Summit,Greenland.Model results suggested that the active post-depositional processing at Summit would cause significant nitrate mass loss(maximal loss fraction of 21%)and alter its isotopes.The simulated seasonal variation ofδ15N(NO3-)in snowpack was consistent with the observations and reproduced the observed spring δ15N(NO3-)in snowpack and the δ15N(NO3-)enrichment in snowpack comparing to surface snow very well.Post-depositional processing also leaded to a seasonal change in ⊿170(NO3-)of about 2.1‰,mainly caused by the locally reformed nitrate which possessed ⊿17O that was distinctly different from the primary nitrate.The decrease in ⊿17O caused by cage effect was negligible at Summit.The annual net loss fraction of primary nitrate at Summit was calculated to be 4%.These results indicated that postdepositional processing would alter the seasonality of snowpack nitrate isotopes while most of the primary nitrate signal got preserved at annual scale under present Summit conditions.3.We developed an Inverse model based on TRANSITS model which was designed to directly use the preserved snowpack/ice core nitrate to deduce the information about primary nitrate.The model can be used to correct the impact of post-depositional processing on the preserved nitrate signals recorded in ice core.Two existing sets of atmosphere and snow nitrate isotope data at Summit and Dome C were used to evaluate the performance of the Inverse Model.The results suggested that for high snow accumulation rates sites,Inverse model could properly recover the annual varitions of atmospheric nitrate isotopes and accurately correct the shift in primary nitrate isotopes caused by postdepositional processing.For sites with relatively low accumulation rates such as Dome C,the model cannot recover the observed seasonal variation in atmosphere,which was caused by the limited snowpack data resolution that was used as initial condition for the model.However,the model properly reproduced the observed annual average value,suggesting that Inverse model can be used to correct the ice core nitrate records.4.We report the first ice core nitrate isotope record at West Antarctica,WAIS Divide covering the last glacial-interglacial period.The δ15N(NO3-)data in glacial period was significantly higher than the Holocenen data and displayed high correlations with snow accumulation rate,indicating that the variation of the WAIS Divide δ15N(NO3-)was primarily controlled by the postdepositional processing.The glacial ⊿17O(NO3-)was lower than Holocene in WASI Divide ice core,which was in contrary with the observarions from GISP2 and Vostok ice core.However,we found a similar two-regimes⊿17O(NO3-)-δ18O(H2O)relations that changed under different climate conditions,similar to the previous founding for GISP2 nitrate records.5.We applied the Inverse model to correct the bipolar ice core nitrate data.The revised ice core data suggested that the glacial primary nitrate flux in the Northern hemisphere decreased to half of it in Holocene,while for Southern hemisphere it remainde relatively stable.The atmospheric NOx abundance in the Northern hemisphere decreased 26%in the glacial period compared to Hololcene,while for Southern hemisphere it slightly increased,which was likely caused by the different icecap area in both hemispheres that controlled the primary emission of NOx.The inferred changes in atmospheric NOx abundance showed good agreement with the ICECAP model,and the δ15N may reflected the relative changes of NOx sources.The revised ⊿17O of primary nitrate increased in glacial for both hemispheres,and the very high ⊿17O(>40‰)in southern hemisphere likely suggested that the primary nitrate mainly originated from stratosphere.The calculated global distribution of locally produced nitrate ⊿17O indicated that the enhanced reactive bromine chemistry can not fully account for the observed increase ⊿17O in primary nitrate,thus likely reflected an enhanced BD circulation in glacial period,which was also supported by the similar magnitude of the response of ⊿17O(Fpri)on temperature in bothe hemisphere.At short time scale,i.e.,the rapid climate transition period,comparison of bipolar nitrate ice core ⊿17O also supported an enhance BD circulation with colder climate in the Northern hemisphere. |
PDF Full Text Request