SCI ENTI FIC CORRESPON DENCE

Lascar Volcano Impacts and the origin of life SIR-The point in time at which life could ted over the range of crater sizes capable set to erupt? first appear on Earth through chemical of causing that effect. SIR-Earlier this year, we showed how evolution within a time interval between We obtain short-wavelength infrared observations impact events was considered by K.A. from remote-sensing can be used Maher and D.J. Stevenson (Nature 331, Tw = 1,400f(ttl [(1 kmlDmiJ Ls - to monitor the radiant thermal flux from 612-614; 1988) for assumed chemical or (1 kmlDm.,)Lsr' (5) active volcanoes (L. Glaze et al. Nature prebiotic evolution times of 105_10' years. 338, 144; 1989), using Lascar volcano, We would like to point out an error that for the time between impacts that frustrate north as our example. We showed to an incorrect estimate of this time. life on a global scale. We used this equa­ that the radiant thermal energy flux had In Maher and Stevenson's model, the tion, rather than equation (3), to calculate decreased before a phreato-magmatic number of events per unit time was derived the time intervals between global events. eruption in September 1986, and had for each of four classes of impact effects Dmio and Dma< are 55 and 265 km, 265 and subsequently increased greatly, doubling that were considered to be adverse to the 850 km, and 850 and 2,600 km, for climatic between October 1986 and November evolutionary process. The reciprocal of disturbances, surface sterilization and 1987, our last data point. these values are the time periods between planetary sterilization, respectively. Reports from Paul King and his colleag­ the impact events: T, is the time required When the time period between events ues at Minsal, Toconao, Chile, show that to obliterate the surface by crater overlap; disruptive to chemical and early biotic a dome is now growing in the summit T, is the time to cover the surface by 3 evolution became greater than the time crater of Lascar. When first observed on metres of ejecta; Tw is the time between required to evolve life, life could first have an ascent of the volcano in February 1989, impacts that frustrate life on a global scale appeared on Earth. Maher and Stevenson the had filled the summit crater by causing climatic disruption, steriliza­ assumed that life formed in 10'_10' years to a depth of several tens of metres. The tion of the surface or sterilization of the and calculated the point in time at which lava dome was absent when the crater was entire planet; and Tb is the time taken to impact events became infrequent enough previously observed in January 1987, but a obliterate the ocean bottom by crater for life to form. With our revised equation powerful thermal anomaly on the image overlap. These time periods were given by (5) instead of (3), life could first appear at we acquired in November 1987 suggests it times more recent than those reported by may have been growing then. Although T, = 1.0 X 1011 [{(ttl] [(Dm./1 km)"z- Maher and Stevenson. For example, if there are 50-60 major and potentially (Dm,/l km)O't' yr (1) large single events with energy of the active volcanoes in the central , order of 1034 erg, assumed to have been many of them over 6,000 metres high, no T, = 1.0 X 10" [{(ttl] [(D, ma/1 km)051 - capable of 'sterilizing' the entire planet, other lava eruptions have been docu­ (D"m,/l km)05lr' yr (2) were required to frustrate development of mented. Lascar is clearly in an unusually life, Maher and Stevenson find that life active condition, and warrants further Tw = 250 [{(ttl] [(Dm,/l km)" yr (3) could have first originated between 4,100 monitoring. and 4,300 million years ago, whereas we 10 calculate this time interval to be 3,700- Whereas the observation of this lava Tb = 5.0 X 10 [{(tt'] [(Dma/1 km)"' - dome at Lascar emphasizes the value of 4,000 million years ago. Also, using equa­ (Dm;/l km)02r' yr (4) remote-sensing techniques in monitoring tion (5) instead of equation (3) changes remote volcanoes, we stress that there The proportionality constants in these the sequence of times at which life can first are many current limitations, not least of equations consist of the products of appear uninterrupted by each of the which is the 16-day repeat period of the various constants associated with impact impact effects. Using equation (3), for Landsat used in our studies and parameters and a crater-production example, life appears more recently if it the vagaries in behaviour of individual constant k corresponding to eight times can be interrupted by crater obliteration volcanoes. Future generations of remote­ the crater production rate on the Moon. rather than by planetary sterilization, sensing satellites, such as those forming The function f(t) is an exponential decay whereas, using equation (5), the reverse is part of NASA's Earth Observing System function that allows for the changing flux true. This reordering of the sequence of will offer much greater scope for vol­ of impactors early in planetary history. times at which life can first appear consti­ canological studies. Detailed plans for This function is f(t) = 1 + exp[(t-t jT], tutes a significant difference between the these are already in hand. Initially, where t is the time measured backwards results of Maher and Stevenson and those we believe that these techniques are from the present, to = 3,400 million years presented here. valuable tools for monitoring volcanoes; and T = 70 million years. Dmio and Dma< are VERNE R. OBERBECK prediction of the eruption of an individual the minimum and maximum diameters of Solar System Exploration Branch, volcano will remain difficult, ideally crater sizes contributing to the impact GUY FOGLEMAN involving data from as many different effect being considered. SETllnstitute, sources as possible, both terrestrial and In deriving equation (3), Maher and Mail Stop 239-12, space-borne. Stevenson consider the time interval NASA Ames Research Center, P. W. FRANCIS between impacts that are just large enough Moffett Field, Lunar and Planetary Institute, to frustrate life on a global scale. For any California 94035, Houston, Texas 77058, USA particular global-impact effect, however, USA a range of crater diameters, from those L.S. GlAZE just large enough to cause the effect up to Scientific Correspondence MS183-501, the size just large enough to cause the next Scientific Correspondence is intended Jet Propulsion Laboratory, higher level of destruction, should have to provide a forum in which readers Pasadena, California, USA been used in determining the expected may raise points of a scientific charac­ time interval between these events. That ter. They need not arise out of anything D.A. ROTHERY is, the time interval between events should published in Nature. In any case, pri­ Department of Earth Sciences, be calculated from the product of the ority will be given to letters of less than Open University, surface area of the Earth and the integral 500 words and five references. D Milton Keynes MK7 6AA, UK of the incremental number density evalua- 434 NATURE, VOL 339 . 8 JUNE 1989

© 1989 Nature Publishing Group