Lectures on Vector Calculus Kyle Broder – April 4th 2022 i “If only I could understand the meaning of d2 = 0.” – Henri Cartan Introduction Vector calculus is an extremely beautiful subject and is a subject that is essential to the modern development of calculus. To place the subject in the appropriate context, let us remind ourselves that long ago, we learned the calculus of functions of a single real variable. The theory is segregated into two primary branches: the differential calculus and the integral calculus. The first branch gives us a set of tools for understanding how a function changes, and the rate at which this occurs; the main gadget to come out of this theory is the derivative d. The second branch treats infinite sums and gives us a reasonable theory for computing area and (cid:82) volume; the main apparatus which enables this is given by the integral . The crowning achievement of the subject – the crescendo – is the fundamental theorem of calculus, a bridge between these two branches. In its most familiar incarnation, the fundamental theorem of calculus reads (cid:90) b f(cid:48)(x)dx = f(b)−f(a). a The importance of this theorem cannot be understated. Aftertheelationthatoneexperiencesfromlearningthisresultofanalmostdivineandsuper- naturalnature,welearnthecorrespondinghigher-dimensionaltheory; namely,thecalculusof several real variables f : Rn → R. The subject is (more or less) identical to the one-variable story we learned earlier, with the additional hassle that comes from the book-keeping of indices. The first glimpse of novel theory appears in vector calculus, the calculus of maps f : Rn → Rm. The primary beauty of the subject is not in the presence of new concepts, such as curl, divergence, vector fields, etc. The beauty of the subject stems from the insights one attains concerning the previous (one and multi-variable) theories. We will see the fundamental theorem of calculus in its true, glorious generality: (cid:90) (cid:90) ω = dω, ∂Ω Ω which, in this guise, is unanimously referred to as Stokes’ theorem. In contrast with the expedient belief that one typically walks away with after their first two courses in the subject, Stokes’ theorem elucidates the deeper relationship between the ii INTRODUCTION iii differential calculus and integral calculus. It is not that derivatives and integrals are opposite to one another (whatever opposite means, we will use the more precise word dual from now on), it is that the derivative is dual to the region of integration, and the duality is given by the integral. This duality, and Stokes’ theorem, lead to the notions of cohomology (specifically, de Rham cohomology) – a theory for studying spaces employing the vector fields (or more generally, differential forms) which reside on them. In a first course in calculus, we study functions, learn the relevant differentiation theory, this is followed by Riemann’s theory of integration, and they are merged via the fundamental theorem of calculus. The present treatment of vector calculus will proceed in much the same way: In the first chapter, we will introduce vector fields and differential forms – the main objects of study. We will discuss gradient fields, a particular class of vector fields that can be expressedintermsoffunctions,theassociationbetweenvectorfieldsand1–forms,theexterior derivative, and the wedge product. The differentiation theory is taken up in Chapter 2. In contrast with the one-variable and multi-variable calculus, there are two notions of derivatives of a vector field: curl and diver- gence. These are, in fact, both instances of one notion of derivative – the exterior derivative, which we meet in Chapter 1. To see that both curl and divergence are incarnations of the exterior derivative, we will discuss the Hodge (cid:63)–operator. The ability to represent a vector field in terms of a function is dependent on the properties of the domain of a vector field. This topic is treated in Chapter 3. The integration theory is exhibitted in chapters 3 and 4. Here, we introduce line integrals and surface integrals . These are then related via the various incarnations of Stokes’ theorem (i.e., the fundamental theorem of calculus). There are three incarnations – Green’s theorem , Stokes’ theorem , and the divergence theorem. Before getting into the details of the subject, let us remark that vector calculus is not the final point of this calculus theory. The ideas developed here, namely, the first glimpses of cohomology are extended (and treated more thoroughly) in the subjects of differential geometry, algebraic geometry, and algebraic topology (to name just a couple). Further, the subject of one-variable complex analysis is arguably the (second) most beautiful of all the calculus theories, giving a theory of functions f : C → C. The most beautiful of the calculus theories is the function theory of several complex variables f : Cn → C which is, unfortunately, not as well-known in comparison with the other theories mentioned. iv INTRODUCTION Structure and purpose of the notes The lecture notes are intended to treat the subject thoroughly, i.e., all statements are given proofs (or a proof is referenced for the reader’s convenience). Each section has exercises whose completion is highly encouraged. There are many complete examples, and this is one of the key aspects of the notes. Answers (not solutions) will be given (eventually) at the back of the notes. An index is also given at the back of the book for ease of recalling definitions; theorems, definitions, and remarks are also hyperlinked if they are referenced later. Let me now address the main question that is likely at the forefront of the reader’s mind: Why write another vector calculus book? The answer is twofold: The first is that, the vector calculus literature is, in my mind, divided into two classes. There is the computation-focused treatment, primarily catering to engineering students, which avoids any systematic development of the theory and presents the results in a very physical manner. This is the direction taken, for instance, in J. Stewarts’ Calculus [25], and does a very good job at treating the subject from this lens. At this point, the reader is likely to interject with, well, engineering students are the primary audience. There is a pedagogical drawback to this approach, however, that not only afflicts the pure mathematics students, but also impacts the target audience; namely, the engineering students. The approach taken in Stewarts’ book side-steps the systematic theory of differential forms that has been devel- oped over the last century, in favor of the less intimidating ad-hoc approach involving the grad vector, the cross product, and so on. The price one pays for this is that, despite each individual lecture being rather straightforward and elementary, by the end of the teaching semester, students are left with a vast number of disparate and unrelated collection of facts. On the other side, the high-brow approach of doing everything via forms, the enlightened approach taken, for instance, most notably in H. M. Schey’s Div, Grad, Curl, and all that, the audience is evidently those who have seen the computation-focused approach and wish to achieve enlightenment through the theory of differential forms. This is not a criticism of Schey’s book – the book has the sub-heading: an informal text on vector calculus. Harold M. Edwards – Advanced Calculus – A Differential Forms Approach [6] is a more appropriate illustration of this second class. Here, the theory is put first, but the intended audience is strange, at best. The amount of mathematical maturity required to digest [6] is likely only held by those students who have taken much more advanced courses than vector calculus. The aim of the present book is to give an appropriate middle ground. The intention is to develop the theory properly, such that students walk away from the course with a coherent pictureofthesubject. Further,thematerialispresentedinadigestableway. Thefirstchapter maybelabelednaive,butthereisneveramomentinthetextwhereafalsestatementisgiven. Further, students who wish for a more detailed account of the developments are encouraged to see the appendix. STRUCTURE AND PURPOSE OF THE NOTES v Errors/Typos/Misprints. Typos,errors,andmisprintsarelikelytobeubiquitousthrough- out the notes. If any issues are found, please do not hesitate to inform me of them at [email protected]. Contents Introduction ii Structure and purpose of the notes iv Notation viii Chapter 1. Vector Fields and Differential Forms 1 1.1. Definitions, First Examples, and Remarks on Regularity 1 Exercises 17 1.2. Forms and the Exterior Algebra 22 Exercises 31 Chapter 2. Differentiation Theory 36 2.1. The Curl of a Vector Field 36 Exercises 46 2.2. The Divergence of a Vector Field 50 Exercises 55 2.3. The Hodge–(cid:63) operator 58 Exercises 64 Chapter 3. Integration Theory – Curves 67 3.1. Line Integrals 67 Exercises 78 3.2. Path Dependence of Line Integrals 80 Exercises 88 Chapter 4. Integration Theory – Surfaces and Beyond 93 4.1. Multiple integrals 93 Exercises 103 4.2. Green’s theorem 107 Exercises 114 4.3. Surface integrals 117 Exercises 129 vi CONTENTS vii 4.4. Stokes’ theorem 132 Exercises 140 4.5. The Divergence Theorem and Surface Independence 143 Exercises 151 Chapter 5. The High Road to Vector Calculus 156 5.1. The Differentiation Theory 156 5.2. The Integration Theory 161 Chapter 6. The Hard Road to Vector Calculus 168 6.1. Linear Algebra 168 6.2. Topological Spaces 178 6.3. Smooth Manifolds 184 6.4. Vector Bundles 187 6.5. Riemannian metrics 197 Answers 201 Bibliography 210 Index 212 Notation N – the natural numbers. • N – the natural numbers including zero. • 0 Z – the integers. • R – the real numbers. • C – the complex numbers. • Ω – a region in Rn. • ∂Ω – the boundary of a region Ω. • C – a curve. • S – a surface. • V – a solid surface. • C0(Ω) – the space of continuous functions f : Ω → R. • C1(Ω) – the space of continuously differentiable functions f : Ω → R. • Ck(Ω) – the space of k–times continuously differentiable functions f : Ω → R. • C∞(Ω) – the space of smooth functions f : Ω → R. • Λ0(Ω) – the space of 0–forms on Ω. • Λ1(Ω) – the space of 1–forms on Ω. • Λ2(Ω) – the space of 2–forms on Ω. • Λ3(Ω) – the space of 3–forms on Ω. • ω – the 1–form associated to F. • F Vect(Ω) – the space of vector fields on Ω. • i – the vector (1,0) or (1,0,0). • j – the vector (0,1) or (0,1,0). • k – the vector (0,0,1). • 0 – the zero vector. • × – the cross product. • · – the dot product. • ∧ – the wedge product. • f , ∂ f, ∂f – the partial derivative of f with respect to x. • x x ∂x f , ∂ f, ∂f – the partial derivative of f with respect to y. • y y ∂y f , ∂ f, ∂f – the partial derivative of f with respect to z. • z z ∂z viii