To demonstrate global simultaneity in climatic response to volcanic aerosols, we have picked climate events from two regions separated by ~180¿ longitude. These events are the Indian monsoon and the frequency of tropical cyclones in the North Atlantic Ocean, both of which occur during the summer season in the northern hemisphere. The simultaneity of these two regions is shown through several different methods. As a first step, we show there is a strong linear correlation between the deviations from the local averages of these two widely separated climate events. From 1900 to 1986 the correlation is above the 99.75% significance level. In the case of the Indian monsoon, a decrease in solar radaition results in a nonlinear decrease in precipitation. In the case of North Atlantic tropical cyclones, decreased radiation results in decreased occurrence. We divide the Indian monsoon and tropical cyclone seasons, each, into two classes, above or below the local average. We examine the association of this seasonal classification in a &khgr;2 analysis using a two by two contingency table, where we find &khgr;2=13.74 for the period 1900--1986. The probability of achieving this degree of association by chance is less than two parts in ten thousand.

Finally, the casual element of the observed simultaneity is hypothesized to be the presence of low-latitude volcanic aerosols. Over the preiod 1900--1986 there are 19 seasons in which a known, reported low-latitude aerosol could have affected the climate anomalies without interference from preexisting, or concurrent, high-latitude aerosols. Of the 19 seasons, 94% had below normal frequency of tropical cyclones, and in 73% of the seasons the precipitation of the Indian monsoon was below normal. A similar analysis is performed for high-latitude aerosols (≥30 ¿ latitude), where opposite results are expected. In this case there are 13 seasons, 46% of which have above normal frequency of tropical cyclones and 77% with above normal Indian monsoons. The Indian monsoon shows a significant response to high-latitude aerosols, while the tropical cyclone frequency does not. Data prior to 1900 (1871--1899) are shown for completeness. We conclude that there is a significant correlation between the frequency of tropical cyclones in the North Atlantic Ocean and the Indian monsoon and there is little change that thic correlation is due to random events. From the Atlantic to the Indian subcontinent there are simultaneous climatic anomalies in conjunction with the presence of low-latitude aerosols. While it is possible to conceive a complex set of internal teleconnections to bring about this climate pattern, the simplest and most economical hypothesis is to assume that global forcing through the reduction of solar radiation resulting from the presence of stratospheric volcanic aerosols may play a major role in producing the observed climate anomalies. ¿ American Geophysical Union 1987