Statistics – Computation
Scientific paper
Dec 2001
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2001agufmgp22b..06j&link_type=abstract
American Geophysical Union, Fall Meeting 2001, abstract #GP22B-06
Statistics
Computation
1507 Core Processes (8115), 1510 Dynamo Theories, 1521 Paleointensity, 1522 Paleomagnetic Secular Variation, 1555 Time Variations: Diurnal To Secular
Scientific paper
The past decade has seen enormous advances in computational simulations of planetary dynamos. Three-dimensional numerical simulations of Earth's dynamo have received much attention, not least because they predict several of the observed large-scale aspects of the geomagnetic field. Numerical models still have some way to go before they adequately represent the physical state of Earth's outer core, most notably its very low viscosity which leads to an exceedingly small and numerically challenging Ekman number. Yet, the wealth of observations of the geomagnetic field over timescales ranging from months to millions of years enable us to pose some strategic tests for current and future geodynamo models. Over historical time periods (the past four hundred years or so), measurements provide almost global coverage of the geomagnetic field, but with decreasing data quality and spatial distribution as one goes from the present-day back to the 16th century. Over archeomagnetic time scales (a few thousands of years), the spatial and temporal resolution of the records decreases, but, it is still possible to track changes in the geomagnetic field structure at 100-year intervals for the past 3 kyr globally. Of particular interest is the persistence of, and changes in, the non-dipolar structure of the field. At paleomagnetic time scales (10,000 to millions of years), the temporal resolution of the magnetic field record again decreases. Thus geomagnetic field observations decrease in temporal and spatial resolution as one goes further back in time. Consequently, different comparisons of dynamo models with data are required, depending on the geomagnetic phenomena and time scales under investigation. Where data comprise adequate global distributions of time series of magnetic field direction or, better still, the full field vector, the evolution of field morphology inferred from observations and predicted by simulations can be compared. Over longer time periods observations require statistical comparisons - both global and local - with dynamo simulations. For example, the mean (time-averaged field) and variance (due to temporal, i.e., secular, variation) of the observations, the statistical properties of reversals and excursions, and the relationship between paleointensity and secular variation (in direction) should be predicted by realistic numerical simulations. Benchmarks of models from several groups are now available and it should become possible to test key predictions (averaged over several dipole diffusion times) against an accepted ensemble of robust geo-, archeo- and paleomagnetic observations.
Courtillot Vincent
Johnson Clifton L.
No associations
LandOfFree
Comparing Numerical Planetary Dynamos with Observations: Constraints from Paleomagnetism 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 Comparing Numerical Planetary Dynamos with Observations: Constraints from Paleomagnetism, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Comparing Numerical Planetary Dynamos with Observations: Constraints from Paleomagnetism will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-1240545