Physics
Scientific paper
Nov 2001
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2001geoji.147..393r&link_type=abstract
Geophysical Journal International, Volume 147, Issue 2, pp. 393-402.
Physics
1
Scientific paper
A new approach for inverting several geophysical parameters at the same time from altimeter and marine data by implementing genetic algorithms (GAs) is presented. These original techniques of optimization based on non-deterministic rules simulate the evolution of a population of candidate solutions for a given objective function to minimize. They offer a robust and efficient alternative to gradient techniques for non-linear parameter inversion. Here genetic algorithms are used for solving a discrete gravity problem of data associated with an undersea relief, to retrieve seven parameters at the same time: the elastic thickness, the mean ocean depth, the seamount location (longitude/latitude), its amplitude, radius and density from its observed gravity/geoid signature. This approach was also successfully used to adjust lithosphere parameters in the real case of the Rarotonga seamount [21.2°S 159.8°W] in the Southern Cook Islands region, where GA simulations provided robust estimates of these seven parameters. The GA found very realistic values for the mean ocean depth and the seamount amplitude and the precise geographical location of Rarotonga Island. Moreover, the values of elastic thickness (~14-15km) and seamount density (~2850-2870kgm-3) estimated by the GA are consistent with the ones proposed in earlier studies.
No associations
LandOfFree
Genetic algorithms for geophysical parameter inversion from altimeter data does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Genetic algorithms for geophysical parameter inversion from altimeter data, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Genetic algorithms for geophysical parameter inversion from altimeter data will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1551900