SIMULATION OF THE FORMATION AND EVOLUTION OF THE PERSEID METEOROID STREAM

Four major models of cometary meteoroid ejection are developed and use d to simulate plausible starting conditions for the formation of the P erseid stream. Ln addition to these physical variants, three different choices for initial meteoroid density (100, 800, and 4000 kg m(-3)) a re used to produce a total of 12 distinct initial models. The developm ent and evolution of the stream are simulated for each model by ejecti ng 10(4) test meteoroids at seven distinct mass categories over the fu ll are of 109P's orbit inside 4 AU at each perihelion passage from 59 to 1862 AD. All test meteoroids are followed to their descending nodes for times closest to the recent perihelion passage of 109P (1992). In addition to these integrations, we have also performed long term inte grations over the interval from 5000 to 10(5) years ago using two plau sible sets of starting orbits for 109P over this interval. We find tha t the choice of cone angle and precise cutoff distance for ejection ma ke only minor modifications to the overall structure of the stream as seen from Earth. The assumed density for the meteoroids has a major in fluence on the present activity of the stream as radiation pressure mo ves nodal points further outside Earth's orbit and hence decreases the probability of delivery for lower density meteoroids. The initial eje ction velocities strongly influence the final distributions observed f rom Earth for the first approximate to 5 revolutions after ejection, a t which point planetary perturbations and radiation effects become mor e important to subsequent development. The minimum distance between th e osculating orbit of 109P at the epoch of ejection and the Earth's or bit is the principal determinant of subsequent delivery of meteoroids to the Earth. The best fit to the observed present flux location and p eak strengths are found from models using Jones (1995) ejection veloci ty algorithm with an r(-05) dependence and densities between (0.1 and 0.8 g cm(-3). The recent activity outburst maxims observed for the Per seids from 1989 to present show a systematic shift in location from ye ar to year, which is explained by changing ages of the primary compone nt of the meteoroids malting up the outbursts. Specifically, it is fou nd that from 1988 to 1990 ejecta from 1610 and 1737 are the dominant p opulation, while 1862 and 1610 are the primary material encountered in the outbursts from 1991 to 1994. From 1995 to 1997 the most prevalent populations are ejections from 1479 and 1079. The older populations t end to shift the locations of the maximums to higher solar longitudes. A discrepancy which is present for both the 1993 and 1994 peak locati ons of 1-2 h between the observed and modeled flux profiles is most li kely the result of emissions from 1862, which were observed to have a large component of their velocity out of the cometary orbital plane. T he cause of Perseid activity outbursts is found to be direct planetary gravitational perturbations from Jupiter and Saturn that shift the no des of stream meteoroids inward and allow them to collide with Earth. The last such perturbations was due to Jupiter in 1991, and this effec t combined with the return of 109P in 1992 produced the strong display s from 1991 to 1994. On average, it is found that the Perseids observe d each year in the core portion of the stream left the parent comet (2 5 +/- 10) x 10(3) years ago. From the modeling, the total age of the s tream is estimated to be on the order of 10(5) years. From the simulat ions over the last 2000 years, the progression rate of the node of the stream is estimated at (2.2 +/- 0.2) x 10(-4) degrees/annum.The effec t of terrestrial perturbations has been evaluated from the long-term i ntegrations and found to play only a minor role in the stream's develo pment, producing a 5-10% increase in the stream's nodal and radiant sp read as compared to an identical simulation without the Earth. The pri mary sinks for the stream are found to be hyperbolic ejection due to J upiter land to a smaller degree Saturn) as well as attainment of sungr azing states. Both the relative and absolute contributions of these tw o loss mechanisms to the decay of the stream is found to be highly dep endent on the assumed cometary starting orbits, with as much as 35% of initially released stream meteoroids removed by hyperbolic ejection a fter 10(5) years for the smallest Perseids on some starting orbits to less than 1% removed after the same time for larger meteoroids on othe r potential seed orbits. On average, it requires 40-80 x 10(3) years f or a noticeable fraction of the initial population (>0.1%) to be remov ed by these mechanisms, depending on the chosen starting orbits. (C) 1 998 Academic Press.