Page 363 - DMTH402_COMPLEX_ANALYSIS_AND_DIFFERENTIAL_GEOMETRY
P. 363

Complex Analysis and Differential Geometry




                    Notes          Setting s = 0, (14) follows.
                                   Proof of Proposition 5. We compute:

                                                        1  b             b             b
                                                   E    2  a     Yg (X,X)dt   a    Yg   Y X,X dt   a    Yg   X Y,X dt
                                                            Y
                                                     
                                                         b
                                                         a   g   X Y,X  g   X Y, Y X  dt
                                                              Y
                                                         b
                                                         a    g   Y Y,X  g     Y , X  Y,X  g   X Y, X Y  dt
                                                              X
                                                         b d                                          
                                                          
                                                                                                      dt,
                                                             g   Y,X  g    Y, X  R(X,Y,Y,X) g      Y,  Y 
                                                         a dt   Y        Y  X                  X   X  
                                                          
                                                    ,
                                   where as above  X     and Y = . Now, the first term integrates to   g  Y Y,X   a b   0,  and when
                                                                                    
                                   we set s =  0, the  second term  also vanishes since   X X     0.  Furthermore, the last term
                                                                                    
                                                  Y  .  Hence, we conclude:
                                   becomes   g     Y,  
                                                                                    
                                                                   b    2
                                                             
                                                            E (0)   a        Y  R(X,Y,X,Y) dt.     ...(15)
                                                             
                                   The proposition now follows from (12).
                                   Thus,  E (0)  can be viewed as a quadratic form in the generator Y . The corresponding symmetric
                                         
                                         
                                   bilinear form is called the index form of :
                                                         b
                                                  I(Y,Z)   a    g     Y,   Z  K      g Y,Z  g   ,Y g     ,Z    dt.
                                   It is the Hessian of the functional E, and if E has a local minimum, I is positive semi-definite. We
                                   will also write I(Y) = I(Y, Y).
                                   Definition 6. Let  be a geodesic parametrized by arc length on the Riemannian surface (U, g). A
                                   vector field Y along  is called a Jacobi field, if it satisfies the following differential equation:

                                                                   
                                                                  Y K Y    g   ,Y 
                                                                                0. 
                                                                
                                   Two points (a) and (b) along a geodesic  are called conjugate along  if there is a non-zero
                                   Jacobi field along  which vanishes at those two points.
                                   The Jacobi field equation is a linear system of second-order differential equations. Hence given
                                   initial data specifying the initial value and initial derivative of Y, a unique solution exists along
                                   the entire geodesic .
                                   Proposition 6. Let  be a geodesic on the Riemannian surface (U, g). Then given two vectors
                                                                                                        
                                   Z , Z  T  (a) U,  there is a unique Jacobi field Y along  such that Y(a) = Z , and     Y(a) Z .
                                       2
                                                                                                           2
                                    1
                                                                                             1
                                   In particular, any Jacobi field which is tangent to  is a linear combination of     and  t .   The
                                   significance of Jacobi fields is seen in the following two propositions. We say that  is a variation
                                   of  through geodesics if the curves  t   (t;s)  are geodesics for all s.






          356                               LOVELY PROFESSIONAL UNIVERSITY
   358   359   360   361   362   363   364   365   366   367   368