Chern numbers of algebraic varieties –
Viewed topologically, the surface of a ball is always a sphere, even when the ball is very deformed: Precise geometric shapes are not important in topology. This is different in algebraic geometry, where objects like the sphere are described by polynomial equations. Professor Dieter Kotschick has recently achieved a breakthrough at the interface of topology and algebraic geometry.
"I was able to solve a problem that was formulated more than 50 years ago by the influential German mathematician Friedrich Hirzebruch" says Kotschick. "Hirzebruch's problem concerns the relation between different mathematical structures. These are so-called algebraic varieties, which are the zero-sets of polynomials, and certain geometric objects called manifolds." Manifolds are smooth topological spaces that can be considered in arbitrary dimensions. The spherical surface of a ball is just a two-dimensional manifold.
In mathematical terminology Hirzebruch's problem was to determine which Chern numbers are topological invariants of complex-algebraic varieties. "I have proved that – except for the obvious ones – no Chern numbers are topologically invariant", says Kotschick. "Thus, these numbers do indeed depend on the algebraic structure of a variety, and are not determined by coarser, so-called topological properties. Put differently: The underlying manifold of an algebraic variety does not determine these invariants."
The solution to Hirzebruch's problem is announced in the current issue of PNAS Early Edition, the online version of PNAS.
"Characteristic numbers of algebraic varieties"
PNAS online, 9 June 2009
Professor Dieter Kotschick, D.Phil.
Faculty for Mathematics, Computer Science and Statistics
Phone: +49 (0) 89 / 2180 - 4444