1: \begin{abstract}

2: Kuramoto oscillator networks have the special property that their trajectories

3: are constrained to lie on the (at most) 3D orbits of the M\"obius group acting

4: on the state space $T^N$ (the $N$-fold torus).  This result has been used to

5: explain the existence of the $N-3$ constants of motion discovered by Watanabe

6: and Strogatz for Kuramoto oscillator networks.  In this work we investigate

7: geometric consequences of this M\"obius group action.  The dynamics of Kuramoto

8: phase models can be further reduced to 2D reduced group orbits, which have a

9: natural geometry equivalent to the unit disk $\D$ with the hyperbolic metric.

10: We show that in this metric the original Kuramoto phase model (with order

11: parameter $Z_1$ equal to the centroid of the oscillator configuration of points

12: on the unit circle) is a gradient flow and the model with order parameter

13: $iZ_1$ (corresponding to cosine phase coupling) is a completely integrable

14: Hamiltonian flow.  We give necessary and sufficient conditions for general

15: Kuramoto phase models to be gradient or Hamiltonian flows in this metric.  This

16: allows us to identify several new infinite families of hyperbolic gradient or

17: Hamiltonian Kuramoto oscillator networks which therefore have simple dynamics

18: with respect to this geometry.  We prove that for the $Z_1$ model, a generic 2D

19: reduced group orbit has a unique fixed point corresponding to the hyperbolic

20: barycenter of the oscillator configuration, and therefore the dynamics are

21: equivalent on different generic reduced group orbits.  This is not always the

22: case for more general hyperbolic gradient or Hamiltonian flows; the reduced

23: group orbits  may have multiple fixed points, which also may bifurcate as the

24: reduced group orbits vary.  

25: \end{abstract}