VOLCANOS AND EL-NINO - SIGNAL SEPARATION IN NORTHERN-HEMISPHERE WINTER

The frequent coincidence of volcanic forcing with El Nino events disab les the clear assignment of climate anomalies to either volcanic or El Nino forcing. In order to select the signals, a set of four different perpetual January GCM experiments was performed (control, volcano cas e, El Nino case and combined volcano/El Nino case) and studied with ad vanced statistical methods for the Northern Hemisphere winter. The res ults were compared with observations. The signals for the different fo rcings are discussed for three variables (temperature, zonal wind and geopotential height) and five levels (surface, 850hPa, 500hPa, 200hPa and 50hPa). The global El Nino signal can be selected more clearly in the troposphere than in the stratosphere. In contrast, the global volc ano signal is strongest in the stratospheric temperature field. The am plitude of the perturbation for the volcano case is largest in the Atl antic region. The observed effect of local cooling due to the volcanic reduction of shortwave radiation over large land areas (like Asia) in subtropical regions, the observed advective warming over Eurasia and the advective cooling over Greenland are well simulated in the model. The radiative cooling near the surface is important for the volcano si gnal in the subtropics, but it is weak in high latitudes during winter . A statistically significant tropospheric signal of El Nino forcing o ccurs in the subtropics and in the midlatitudes of the North Pacific. The local anomalies in the El Nino forcing region in the tropics, and the warming over North America in middle and high latitudes are simula ted as observed. The combined signal is different from a simple linear combination of the separate signals. It leads to a climate perturbati on stronger than for forcing with El Nino or stratospheric aerosol alo ne and to a somewhat modified pattern.