SYSTEMATIC MATHEMATICS

Calculus on Manifolds & Geometric Integration

This 94-chapter graduate bridge begins with the question hidden by coordinate formulas: what geometric object is being differentiated or integrated, and how must its components change when coordinates change? Twelve content-sized units build derivative maps, smooth manifolds, tangent and cotangent spaces, vector fields and flows, tensor fields and metrics, exterior calculus, orientation and densities, Stokes, metric differential operators, area/coarea and multiplicity, covariant and moving-domain transport, and eight original reproducible dossiers. Every chapter reads symbols aloud, derives one bounded theorem route, completes a calculation, activates a failure case, and pairs practice with a full bilingual solution.

Before this course: Completed Multivariable Calculus, Vector Calculus, Linear Algebra, Real Analysis, Topology, Measure & Lebesgue Integration, Differential Equations & Dynamical Systems, and Differential Geometry & Manifolds. Differential forms and tensor notation are rebuilt from their definitions, so prior index fluency is not assumed.

COURSE FACTSLevel, chapters, units, prerequisite, and outcome
Chapter 1

A formula needs an object and a domain

Objective: Which object type, orientation, regularity, or overlap condition makes the statement in “A formula needs an object and a domain” coordinate-independent?

Unit 1, chapter 1 starts from the geometric task behind “A formula needs an object and a domain.” The same array of numbers can describe a point, vector, covector, or matrix only after its transformation rule is named. Before calculating, name the space, the point or region, the map, and the kind of object being differentiated or integrated. A coordinate formula is useful only after those objects are fixed.

The precise definition is: A geometric object is specified by its value at each point together with a rule describing how its coordinate representatives change between admissible charts. Read each index as a slot or coordinate label, each differential as a map between specified tangent spaces, and each integral together with its orientation or density. This prevents a list of symbols from replacing the geometric meaning.

The chapter’s usable statement is: If two coordinate lists are related by the prescribed transition rule on every overlap, they represent one global object; arbitrary unrelated lists do not. Its derivation has a visible route: Write both charts, compute the transition map, transform the first list by the correct rule, and compare with the second list point by point. Every coordinate change is checked against the intrinsic object, so agreement in one chart is not mistaken for a global proof.

The worked problem is: In Cartesian and polar coordinates at (1,π/2), distinguish the point coordinates, the radial unit vector, and the differential dr. The calculation first fixes conventions, then performs the local algebra, and finally checks overlap, orientation, units, or a boundary term. The resulting claim is The point is (0,1), e_r=(0,1), and dr evaluates a displacement by its radial component; equal-looking pairs have different types and transformation rules.

The failure test is part of the lesson: Matching numerical entries at one point does not prove that two fields are the same, because their bases or object types may differ. A complete solution therefore states hypotheses, object types, transformation rules, the computed value, and the first conclusion that fails when the boundary is crossed. Numerical agreement alone is labeled evidence, never proof.

If two coordinate lists are related by the prescribed transition rule on every overlap, they represent one global object; arbitrary unrelated lists do not.

Start from the chapter definition and select one chart only as a temporary calculation device. For “A formula needs an object and a domain,” write the domain, codomain, tensor degree or form degree, and regularity before differentiating.

Write both charts, compute the transition map, transform the first list by the correct rule, and compare with the second list point by point. On every overlap, apply the chain rule, determinant rule, alternating rule, or connection transformation law that matches the object. Terms that cancel must be displayed rather than hidden inside “it is invariant.”

Evaluate the stated model and independently test the geometric conclusion. This gives The point is (0,1), e_r=(0,1), and dr evaluates a displacement by its radial component; equal-looking pairs have different types and transformation rules. The proof stops where Matching numerical entries at one point does not prove that two fields are the same, because their bases or object types may differ.

In Cartesian and polar coordinates at (1,π/2), distinguish the point coordinates, the radial unit vector, and the differential dr.

  1. List the manifold or region, coordinates, map, orientation, and all input object types; translate the definition A geometric object is specified by its value at each point together with a rule describing how its coordinate representatives change between admissible charts. into those declared data.
  2. Write both charts, compute the transition map, transform the first list by the correct rule, and compare with the second list point by point. Keep Jacobian factors, signs, index positions, and boundary orientations visible on separate lines.
  3. Compute the local expression, compare on overlaps or against an invariant identity, and report The point is (0,1), e_r=(0,1), and dr evaluates a displacement by its radial component; equal-looking pairs have different types and transformation rules.

Result: The point is (0,1), e_r=(0,1), and dr evaluates a displacement by its radial component; equal-looking pairs have different types and transformation rules.