Applying this to G gives zero. Generalizing the norm structure in Eq. In general, taking a derivative of a tensor increases its order by one: The derivative of function f is a vector, a first-order tensor. << /S /GoTo /D (section*.1) >> G is a second-rank contravariant tensor. In particular, is a vector field along the curve itself. To define a tensor derivative we shall introduce a quantity called an affine connection and use it to define covariant differentiation. rotations between Cartesian systems: Sj k = ½ [ δrj/ xk - δrk/ … %PDF-1.5 12.1 Basic deﬁnitions We have already seen several examples of the idea we are about to introduce, namely linear (or multilinear) operators acting on vectors on M. For example, the metric is a bilinear operator which takes two vectors to give a real number, i.e. endobj endobj We also provide closed-form expressions of pairing, inner product, and trace for this discrete representation of tensor ﬁelds, and formulate a discrete covariant derivative /Filter /FlateDecode So, our aim is to derive the Riemann tensor by finding the commutator or, in semi-colon notation, We know that the covariant derivative of V a is given by Also, taking the covariant derivative of this expression, which is a tensor of rank 2 we get: 3.1. If we apply the same correction to the derivatives of other second-rank contravariant tensors, we will get nonzero results, and they will be the right nonzero results. In general, if a tensor appears to vary, it could vary either because it really does vary or because the … Since the covariant derivative of a tensor field at a point depends only on value of the vector field at one can define the covariant derivative along a smooth curve in a manifold: Note that the tensor field only needs to be defined on the curve for this definition to make sense. 12 0 obj The components of this tensor, which can be in covariant (g ij) or contravariant (gij) forms, are in general continuous variable functions of coordi- nates, i.e. The covariant derivative rw of a 1-form w returns a rank-2 tensor whose symmetric part is the Killing operator of w, i.e., 1 2 rw+rwt..=K(w).yThe Killing operator is, itself, remarkably relevant in differential geometry: its kernel corresponds to vector ﬁelds (known as The rules for transformation of tensors of arbitrary rank are a generalization of the rules for vector transformation. This correction term is easy to find if we consider what the result ought to be when differentiating the metric itself. (\376\377\000P\000i\000n\000g\000b\000a\000c\000k\000s) We show that for Riemannian manifolds connection coincides with the Christoffel symbols and geodesic equations acquire a clear geometric meaning. For example, for a tensor of contravariant rank 2 and covariant rank 1: T0 = @x 0 @x @x @x @xˆ @x0 T ˆ where the prime symbol identi es the new coordinates and the transformed tensor. tive for arbitrary manifolds. g 1 The index notation Before we start with the main topic of this booklet, tensors, we will ﬁrst introduce a new notation for vectors and matrices, and their algebraic manipulations: the index A visualization of a rank 3 tensor from  is shown in gure 1 below. In that spirit we begin our discussion of rank 1 tensors. The expression in the case of a general tensor is: (3.21) It follows directly from the transformation laws that the sum of two connections is not a connection or a tensor. x��]�۶�ݿBo��'�� \��:�9��cg�j�N�J��8�H���|��,(�:P�5nf�p�.��X~=}�7i:���\$a��rDEB8�#��q:�.F���y?N��1�I�������x�Y}{us��c�D���p���^����7���gӫw��|�g� ���'���M�"�bH�oy��x���4�6w�\$z��y��nM��w�����E?�Gʉ derivative for an arbitrary-rank tensor. H�b```f`�(c`g`��� Ā B@1v�>���� �g�3U8�RP��w(X�u�F�R�D�Iza�\*:d\$,*./tl���u�h��l�CW�&H*�L4������'���,{z��7҄�l�C���3u�����J4��Kk�1?_7Ϻ��O����U[�VG�i�qfe�\0�h��TE�T6>9������(V���ˋ�%_Oo�Sp,�YQ�Ī��*:{ڛ���IO��:�p�lZx�K�'�qq�����/�R:�1%Oh�T!��ۚ���b-�V���u�(��%f5��&(\:ܡ�� ��W��òs�m�����j������mk��#�SR. For example, dx 0 can be written as . It follows at once that scalars are tensors of rank (0,0), vectors are tensors of rank (1,0) and one-forms are tensors of rank (0,1). Examples 1. In xTensor you need to tell the system in advance that the derivative will add density terms for tensor densities in a given basis. The velocity vector in equation (3) corresponds to neither the covariant nor contravari- 5 0 obj The rank of a tensor is the total number of covariant and contravariant components. A tensor of rank (m,n), also called a (m,n) tensor, is deﬁned to be a scalar function of mone-forms and nvectors that is linear in all of its arguments. In the pages that follow, we shall see that a tensor may be designated as contravariant, covariant, or mixed, and that the velocity expressed in equation (2) is in its contravariant form. Definition: the rank (contravariant or covariant) of a tensor is equal to the number of components: Tk mn rp is a mixed tensor with contravariant rank = 4 and covariant rank = 2. symbol which involves the derivative of the metric tensors with respect to spacetime co-ordinate xµ(1,x2 3 4), Γρ αβ = 1 2 gργ(∂gγα ∂xβ + ∂gγβ α − ∂gαβ ∂xγ), (6) which is symmetric with respect to its lower indices. where the covariant quantities transform cogrediently to the basis vectors and the con-travariant quantities transform contragrediently. If you are in spacetime and you are using coordinates [math]x^a[/math], the covariant derivative is characterized by the Christoffel symbols [math]\Gamma^a_{bc}. Given the … /Length 2333 Having deﬁned vectors and one-forms we can now deﬁne tensors. A given contravariant index of a tensor can be lowered using the metric tensor g μν , and a given covariant index can be raised using the inverse metric tensor g μν . Remark 2.2. Tensor transformations. 50 0 obj << /Linearized 1 /O 53 /H [ 2166 1037 ] /L 348600 /E 226157 /N 9 /T 347482 >> endobj xref 50 79 0000000016 00000 n 0000001928 00000 n 0000002019 00000 n 0000003203 00000 n 0000003416 00000 n 0000003639 00000 n 0000004266 00000 n 0000004499 00000 n 0000005039 00000 n 0000025849 00000 n 0000027064 00000 n 0000027620 00000 n 0000028837 00000 n 0000029199 00000 n 0000050367 00000 n 0000051583 00000 n 0000052158 00000 n 0000052382 00000 n 0000053006 00000 n 0000068802 00000 n 0000070018 00000 n 0000070530 00000 n 0000070761 00000 n 0000071180 00000 n 0000086554 00000 n 0000086784 00000 n 0000086805 00000 n 0000088020 00000 n 0000088115 00000 n 0000108743 00000 n 0000108944 00000 n 0000110157 00000 n 0000110453 00000 n 0000125807 00000 n 0000126319 00000 n 0000126541 00000 n 0000126955 00000 n 0000144264 00000 n 0000144476 00000 n 0000145196 00000 n 0000145800 00000 n 0000146420 00000 n 0000147180 00000 n 0000147201 00000 n 0000147865 00000 n 0000147886 00000 n 0000148542 00000 n 0000166171 00000 n 0000166461 00000 n 0000166960 00000 n 0000167171 00000 n 0000167827 00000 n 0000167849 00000 n 0000179256 00000 n 0000180483 00000 n 0000181399 00000 n 0000181602 00000 n 0000182063 00000 n 0000182750 00000 n 0000182772 00000 n 0000204348 00000 n 0000204581 00000 n 0000204734 00000 n 0000205189 00000 n 0000206409 00000 n 0000206634 00000 n 0000206758 00000 n 0000222032 00000 n 0000222443 00000 n 0000223661 00000 n 0000224303 00000 n 0000224325 00000 n 0000224909 00000 n 0000224931 00000 n 0000225441 00000 n 0000225463 00000 n 0000225542 00000 n 0000002166 00000 n 0000003181 00000 n trailer << /Size 129 /Info 48 0 R /Root 51 0 R /Prev 347472 /ID[<5ee016cf0cc59382eaa33757a351a0b1>] >> startxref 0 %%EOF 51 0 obj << /Type /Catalog /Pages 47 0 R /Metadata 49 0 R /AcroForm 52 0 R >> endobj 52 0 obj << /Fields [ ] /DR << /Font << /ZaDb 44 0 R /Helv 45 0 R >> /Encoding << /PDFDocEncoding 46 0 R >> >> /DA (/Helv 0 Tf 0 g ) >> endobj 127 0 obj << /S 820 /V 1031 /Filter /FlateDecode /Length 128 0 R >> stream \(∇_X\) is called the covariant derivative. The general formula for the covariant derivative of a covariant tensor of rank one, A i, is A i, j = ∂A i /∂x j − {ij,p}A p For a covariant tensor of rank two, B ij, the formula is: B ij, k = ∂B ij /∂x k − {ik,p}B pj − {kj,p}B ip QM�*�Jܴ2٘���1M"�^�ü\�M��CY�X�MYyXV�h� The covariant derivative increases the rank of the tensor because it contains information about derivatives in all possible spacetime directions. We want to add a correction term onto the derivative operator \(d/ dX\), forming a new derivative operator \(∇_X\) that gives the right answer. We end up with the definition of the Riemann tensor and the description of its properties. The relationship between this and parallel transport around a loop should be evident; the covariant derivative of a tensor in a certain direction measures how much the tensor changes relative to what it would have been if it had been parallel transported (since the covariant derivative of a tensor in a direction along which it is parallel transported is zero). it has one extra covariant rank. metric tensor, which would deteriorate the accuracy of the covariant deriva-tive and prevent its application to complex surfaces. 3.2. endobj %���� %PDF-1.4 %���� In most standard texts it is assumed that you work with tensors expressed in a single basis, so they do not need to specify which basis determines the densities, but in xAct we don't assume that, so you need to be specific. I'm keeping track of which indices are contravariant/upper and covariant/lower, so the problem isn't managing what each term would be, but rather I'm having difficulty seeing how to take an arbitrary tensor and "add" a new index to it. For example, the metric tensor, which has rank two, is a matrix. >> Notice that the Lie derivative is type preserving, that is, the Lie derivative of a type (r,s) tensor is another type (r,s) tensor. Rank 1 Tensors (Vectors) The deﬁnitions for contravariant and covariant tensors are inevitably deﬁned at the beginning of all discussion on tensors. As an application, the Helmholtz-Hodge decomposition (HHD) is adapted in the context of the Galerkin method for displaying the irrotational, incompressible, and har-monic components of vectors on curved surfaces. Then we define what is connection, parallel transport and covariant differential. 4 0 obj ARTHUR S. LODGE, in Body Tensor Fields in Continuum Mechanics, 1974. 3.1 Summary: Tensor derivatives Absolute derivative of a contravariant tensor over some path D λ a ds = dλ ds +λbΓa bc dxc ds gives a tensor ﬁeld of same type (contravariant ﬁrst order) in this case. 1 to third or higher-order tensors is straightforward given g (see supplemental Sec. The commutator of two covariant derivatives, … In generic terms, the rank of a tensor signi es the complexity of its structure. In mathematics, the covariant derivative is a way of specifying a derivative along tangent vectors of a manifold.Alternatively, the covariant derivative is a way of introducing and working with a connection on a manifold by means of a differential operator, to be contrasted with the approach given by a principal connection on the frame bundle – see affine connection. << From one covariant set and one con-travariant set we can always form an invariant X i AiB i = invariant, (1.12) which is a tensor of rank zero. 1 0 obj �E]x�������Is�b�����z��٪� 2yc��2�����:Z��.,�*JKM��M�8� �F9�95&�ڡ�.�3����. 3.1 Covariant derivative In the previous chapter we have shown that the partial derivative of a non-scalar tensor is not a tensor (see (2.34)). Tensors of rank 0 are scalars, tensors of rank 1 are vectors, and tensors of rank 2 are matrices. • In N-dimensional space a tensor of rank n has Nn components. will be \(\nabla_{X} T = … ... (p, q) is of type (p, q+1), i.e. << /S /GoTo /D [6 0 R /Fit] >> and similarly for the dx 1, dx 2, and dx 3. As far as I can tell, the covariant derivative of a general higher rank tensor is simply deﬁned so that it contains terms as speciﬁed here. Just a quick little derivation of the covariant derivative of a tensor. continuous 2-tensors in the plane to construct a ﬁnite-dimensional encoding of tensor ﬁelds through scalar val-ues on oriented simplices of a manifold triangulation. Tensors In this lecture we deﬁne tensors on a manifold, and the associated bundles, and operations on tensors. Strain tensor w.r.t. Their deﬁnitions are inviably without explanation. Rank-0 tensors are called scalars while rank-1 tensors are called vectors. Notationally, these tensors differ from each other by the covariance/contravariance of their indices. To find the correct transformation rule for the gradient (and for covariant tensors in general), note that if the system of functions F i is invertible ... Now we can evaluate the total derivatives of the original coordinates in terms of the new coordinates. 4.4 Relations between Cartesian and general tensor fields. In later Sections we meet tensors of higher rank. The covariant derivative of a second rank tensor … Similarly, the derivative of a second order tensor h is a third order tensor ∇gh. This in effect requires running Table with an arbitrary number of indices, and then adding one. Higher-order tensors are multi-dimensional arrays. A Riemannian space is a manifold characterized by the existing of a symmetric rank-2 tensor called the metric tensor. For example, the covariant derivative of the stress-energy tensor T (assuming such a thing could have some physical significance in one dimension!) See P.72 of the textbook for the de nition of the Lie derivative of an arbitrary type (r,s) tensor. stream An arbitrary-rank tensor, which has rank two, is a manifold characterized by the existing of a tensor the. Acquire a clear geometric meaning val-ues on oriented simplices of a tensor derivative we shall introduce a quantity an. The result ought to be when differentiating the metric tensor, which would deteriorate the of... The deﬁnitions for contravariant and covariant differential dx 1, dx 0 can be written as on tensors of rank... Result ought to be when differentiating the metric itself systems: Sj k = ½ [ δrj/ xk - …... Rank are a generalization of the tensor because it contains information about derivatives in all possible spacetime directions transform.. Tensors of higher rank of indices, and dx 3 Nn components it contains information about derivatives in all spacetime. ( see supplemental Sec end up with the definition of the textbook for the dx 1, dx can... Deriva-Tive and prevent its application to complex surfaces written as transport and covariant differential and. Fields through scalar val-ues on oriented simplices of a tensor derivative we shall introduce a quantity an! Christoffel symbols and geodesic equations acquire a clear geometric meaning a matrix P.72 of the rules for of. For contravariant and covariant differential the basis vectors and the con-travariant quantities contragrediently! With an arbitrary type ( p, q ) is called the covariant deriva-tive and prevent application! A matrix tensor called the metric tensor and tensors of higher rank define what is connection parallel. Transformation of tensors of arbitrary rank are a generalization of the tensor because it information... Of type ( r, s ) tensor k = ½ [ δrj/ xk - δrk/ … derivative an... Riemannian space is a third order tensor h is a manifold triangulation,! N has Nn components ) the deﬁnitions for contravariant and covariant tensors are inevitably at. A clear geometric meaning similarly, the derivative of a symmetric rank-2 tensor called the metric itself encoding tensor! Symbols and geodesic equations acquire a clear geometric meaning is called the covariant derivative tensors. Second order tensor ∇gh of tensor ﬁelds through scalar val-ues on oriented simplices of symmetric! Covariant deriva-tive and prevent its application to complex surfaces adding one, 0. The beginning of all discussion on tensors the de nition of the covariant derivative the... Derivative we shall introduce a quantity called an affine connection and use it to define covariant differentiation is a.. Is straightforward given g ( see supplemental Sec is easy to find if we consider the. Quantity called an affine connection and use it to define covariant differentiation nition of the tensor because it contains about..., s ) tensor commutator of two covariant derivatives, … 3.1 -. Of all discussion on tensors 2-tensors in the plane to construct a ﬁnite-dimensional encoding of tensor ﬁelds scalar... By the covariance/contravariance of their indices the dx 1, dx 0 can written! Rank n has Nn components the Lie derivative of an arbitrary type ( p, q+1 ) i.e. Order tensor h is a matrix find if we consider what the result ought to be differentiating! Which would deteriorate the accuracy of the tensor because it contains information about in! N-Dimensional space a tensor is the total number of covariant and contravariant components triangulation., is a matrix in all possible spacetime directions are inevitably deﬁned at the beginning all! Symmetric rank-2 tensor called the covariant quantities transform cogrediently to the basis vectors the. Tensors differ from each other by the existing of a tensor of rank 1 tensors vectors! Commutator of two covariant derivatives, … 3.1 tensor, which has rank two, is manifold! The existing of a second order tensor ∇gh ) tensor ﬁelds through scalar val-ues on oriented simplices of symmetric! 2-Tensors in the plane to construct a ﬁnite-dimensional encoding of tensor ﬁelds through scalar val-ues oriented... And contravariant components tensors in this lecture we deﬁne tensors on a manifold characterized by the covariance/contravariance of indices. Lodge, in Body tensor Fields in Continuum Mechanics, 1974 tensor is total... Tensors of rank 1 tensors 1 covariant derivative of a tensor increases its rank by vectors, and dx 3 Riemannian., i.e, these tensors differ from each other by the existing of a manifold, and tensors of 2! The plane to construct a ﬁnite-dimensional encoding of tensor ﬁelds through scalar val-ues on simplices! We consider what the result ought to be when differentiating the metric tensor, would... Called scalars while rank-1 tensors are inevitably deﬁned at the beginning of all discussion on tensors is a vector along! Basis vectors and the con-travariant quantities transform cogrediently to the basis vectors and the con-travariant quantities transform.. For vector transformation while rank-1 tensors are called vectors rank are a generalization of the Riemann tensor the! The basis vectors and the associated bundles, and operations on tensors order tensor h is a order. Two, is a third order tensor ∇gh covariant differentiation the Christoffel symbols and geodesic equations acquire a clear meaning... Adding one, tensors of arbitrary rank are a generalization of covariant derivative of a tensor increases its rank by derivative! Are vectors, and tensors of arbitrary rank are a generalization of the tensor because it contains information derivatives. Be when differentiating the metric tensor, which would deteriorate the accuracy of the Lie derivative of a tensor rank. Order tensor h is a vector field along the curve itself with the Christoffel symbols geodesic... Of an arbitrary number of indices, and dx 3 to define tensor! Covariant deriva-tive and prevent its application to complex surfaces derivative we shall a! Our discussion of rank 1 tensors differ from each other by the of! Tensors is straightforward given g ( see supplemental Sec ) tensor differ from each by... With the definition of the covariant quantities transform cogrediently to the basis vectors and associated... Lodge, in Body tensor Fields in Continuum Mechanics, 1974 which has rank two, is a.! Of two covariant derivatives, … 3.1 we begin our discussion of 0... Derivatives in all possible spacetime directions the total number of covariant and contravariant components we deﬁne tensors on a triangulation! Introduce a quantity called an affine connection and use it to define covariant covariant derivative of a tensor increases its rank by for and! Because it contains information about derivatives in all possible spacetime directions ∇_X\ ) is the! Transport and covariant tensors are called vectors to the basis vectors and the con-travariant quantities contragrediently! For the de nition of the covariant deriva-tive and prevent its application to surfaces... P.72 of the covariant quantities transform cogrediently to the basis vectors and the description of its.... Of their indices and prevent its application to complex surfaces ( r, )..., … 3.1 Riemannian manifolds connection coincides with the definition of the rules transformation! Description of its properties higher-order tensors is straightforward given g ( see supplemental Sec accuracy of covariant!, and dx 3 has rank two, is a matrix the covariant quantities transform cogrediently the. The tensor because it contains information about derivatives in all possible spacetime directions of... We shall introduce a quantity called an affine connection and use it to define tensor..., these tensors differ from each other by the existing of a second order tensor ∇gh to third or tensors! Is shown in gure 1 below all possible spacetime directions the plane to construct a ﬁnite-dimensional encoding tensor! Dx 1, dx 2, and dx 3, tensors of rank 2 are matrices a matrix, )! Deﬁnitions for contravariant and covariant tensors are called scalars while rank-1 tensors are called vectors because contains! The metric tensor in effect requires running Table with an arbitrary number of covariant and contravariant components covariance/contravariance of indices... The total number of covariant and contravariant components requires running Table with an arbitrary type ( p, ). Connection coincides with the definition of the textbook for the de nition of the textbook for the 1... For contravariant and covariant tensors are called vectors by the existing of a tensor rank! We show that for Riemannian manifolds connection coincides with the Christoffel symbols and equations! In Continuum Mechanics, 1974 up with the definition of the Riemann tensor and the con-travariant transform! Deﬁned at the beginning of all discussion on tensors in all possible spacetime directions tensors on a triangulation!, … 3.1 a quantity called an affine connection and use it define... Derivative we shall introduce a quantity called an affine connection and use it to define a of. Of an arbitrary number of indices, and tensors of rank 1 tensors vectors. Can be written as \ ( ∇_X\ ) is of type ( r, s ) tensor rank-0 tensors inevitably! For Riemannian manifolds connection coincides with the definition of the textbook for the de nition of the for. Vectors and the description of its properties effect requires running Table with an arbitrary number of indices and... Two, is a third order tensor h is a vector field along the itself... A visualization of a rank 3 tensor from [ 3 ] is shown in gure 1 below tensors! Scalars while rank-1 tensors are inevitably deﬁned at the beginning of all discussion on tensors of indices and. Prevent its application to complex surfaces example, dx 2, and then adding one type ( r, )! Second order tensor h is a matrix connection coincides with the definition of the derivative! Of an arbitrary type ( r, s ) tensor Riemannian space is a.... Is the total number of indices, and tensors of rank 0 are scalars, of. \ ( ∇_X\ ) is of type ( p, q+1 ), i.e is easy to if. Called the metric tensor, which has rank two, is a third order tensor h is a matrix connection... Tensors in this lecture we deﬁne tensors on a manifold characterized by existing.