High resolution simulation of recent Arctic and Antarctic stratospheric chemical ozone loss compared to observations

Simulations of polar ozone losses were performed using the three-dimensional high-resolution (1^∘ × 1^∘) chemical transport model MIMOSA-CHIM. Three Arctic winters 1999–2000, 2001–2002, 2002–2003 and three Antarctic winters 2001, 2002, and 2003 were considered for the study. The cumulative ozone loss in the Arctic winter 2002–2003 reached around 35% at 475 K inside the vortex, as compared to more than 60% in 1999–2000. During 1999–2000, denitrification induces a maximum of about 23% extra ozone loss at 475 K as compared to 17% in 2002–2003. Unlike these two colder Arctic winters, the 2001–2002 Arctic was warmer and did not experience much ozone loss. Sensitivity tests showed that the chosen resolution of 1^∘ × 1^∘ provides a better evaluation of ozone loss at the edge of the polar vortex in high solar zenith angle conditions. The simulation results for ozone, ClO, HNO₃, N₂O, and NO _y for winters 1999–2000 and 2002–2003 were compared with measurements on board ER-2 and Geophysica aircraft respectively. Sensitivity tests showed that increasing heating rates calculated by the model by 50% and doubling the PSC (Polar Stratospheric Clouds) particle density (from 5 × 10⁻³ to 10⁻² cm⁻³) refines the agreement with in situ ozone, N₂O and NO _y levels. In this configuration, simulated ClO levels are increased and are in better agreement with observations in January but are overestimated by about 20% in March. The use of the Burkholder et al. (1990) Cl₂O₂ absorption cross-sections slightly increases further ClO levels especially in high solar zenith angle conditions. Comparisons of the modelled ozone values with ozonesonde measurement in the Antarctic winter 2003 and with Polar Ozone and Aerosol Measurement III (POAM III) measurements in the Antarctic winters 2001 and 2002, shows that the simulations underestimate the ozone loss rate at the end of the ozone destruction period. A slightly better agreement is obtained with the use of Burkholder et al. (1990) Cl₂O₂ absorption cross-sections.     

Magnetic record of El Niño Southern Oscillation in Late Pleistocene sediments from Mucubají lake (western Venezuela)

El Niño Southern Oscillation (ENSO) is an internal forcing of the climate system. This event has an actual frequency of 2 to 8 years. Evidence from a paleoclimate proxy database of gray scale (GS), in samples from Pallcacocha lake in Ecuador, indicates that the ENSO had a frequency of 35 to 75 years during the Late Pleistocene. In this work we explored the possible relationship between the ENSO proxies (GS) from Pallcacocha and magnetic parameters from sediments sampled at the Mucubají lake in Mérida, Venezuela (i.e. mass-specific magnetic susceptibility, magnetic remanence S ratio and susceptibilitynormalized saturation isothermal remanent magnetization). After applying a Lanczos bandpass filter to the rock magnetic and the GS data, in order to remove, as much as possible, frequencies associated to any periodic event other than ENSO, we found significant correlations between GS and magnetic parameters for the period between 12450 and 10560 cal. yrs BP. These relationships were obtained using an Adaptive Neuro Fuzzy Inference System (ANFIS), a hybrid algorithm that combines fuzzy logic with neural networks. The results show that the magnetic parameters obtained in Mucubají are able to explain 50.5% of the total variance of the ENSO proxy in a range of 35 to 75 years in Pallcacocha, which is roughly the same percentage of the total variance of the temperature in the Venezuelan Andes, explained by the ENSO at present times. In this way we have inferred a possible influence of the ENSO in the Venezuelan Andes during the Late Pleistocene.