Construction of higher Chow cycles on cyclic coverings of $\mathbb{P}^1 \times \mathbb{P}^1$, Part II

This paper constructs higher Chow cycles of type $(2,1)$ on a family of degree $N$ abelian covers of $\mathbb{P}^1$ branched over $n+2$ points and proves that for a very general member, these cycles generate a subgroup of the indecomposable part of rank at least $n\cdot \phi(N)$ by computing their images under the transcendental regulator map.

The Gist

Imagine you are an architect trying to understand the hidden "skeleton" of a very complex, multi-layered building. This building isn't made of brick and mortar, but of pure mathematics. Specifically, it's a family of surfaces (think of them as 2D sheets floating in a higher-dimensional space) that are constructed by twisting and folding simpler shapes in a very specific way.

This paper, written by Yusuke Nemoto and Ken Sato, is about finding and counting the indestructible structural beams inside these mathematical buildings.

Here is the breakdown of what they did, using simple analogies:

1. The Building: Cyclic Coverings

Imagine you have a flat sheet of paper (a mathematical plane). Now, imagine you take $N$ copies of this sheet and glue them together in a spiral staircase pattern. If you walk around a specific point on the paper, you don't just return to where you started; you move up to the next "floor" of the staircase. After $N$ loops, you finally return to the start.

In math, this is called a cyclic covering. The authors are studying a specific type of these spiral staircases built over a surface that looks like a product of two lines ($P^1 \times P^1$). They are looking at a "family" of these buildings, where the shape changes slightly depending on two knobs, $\lambda_1$ and $\lambda_2$, that you can turn.

2. The Goal: Finding "Indecomposable" Cycles

Inside these buildings, mathematicians look for special loops or shapes called Higher Chow Cycles.

• Decomposable cycles are like a loop made of two smaller, independent loops tied together. You can take them apart. They are "boring" because they don't tell you anything new about the building's unique structure.
• Indecomposable cycles are like a knot that cannot be untied or separated into smaller pieces. These are the "skeleton beams" of the building. They represent the true, unique complexity of the shape.

The authors want to know: How many of these indestructible knots can we find?

3. The Method: The "Regulator" as a Detective

How do you prove a knot is indestructible without trying to untie it for a thousand years? You need a detector.

In this paper, the authors use a tool called the Transcendental Regulator Map. Think of this as a high-tech scanner or a "lie detector" for mathematical shapes.

• If you put a "decomposable" (fake) knot into the scanner, it reads zero. It says, "This is just two simple loops tied together; nothing special here."
• If you put an "indecomposable" (real) knot into the scanner, it produces a signal (a specific number or function).

The authors construct a whole family of these knots (cycles) and run them through the scanner.

4. The Challenge: The Signal is a Moving Target

The tricky part is that the signal the scanner produces isn't a simple number like "5" or "10." It's a complex, wiggly function that changes as you turn the knobs ($\lambda_1$ and $\lambda_2$).

To prove these signals are real and distinct, the authors had to show that these wiggly functions are independent. If two signals were just copies of each other (or one was a multiple of the other), they wouldn't count as two separate beams. They needed to prove they are all unique.

5. The Solution: The "Jordan-Pochhammer" Filter

This is where the paper gets technical, but the idea is simple. The authors realized that these wiggly signals obey a very strict set of rules, described by a specific mathematical machine called the Jordan-Pochhammer differential operator.

Think of this operator as a specialized sieve or a filter.

• Most random noise gets stuck or passes through randomly.
• But these specific signals, because they come from the geometry of the building, pass through the sieve in a very predictable, structured way.

By applying this filter, the authors could calculate exactly what the signals look like. They found that the signals are generated by a specific formula involving Euler's totient function ($\phi(N)$).

The Big Result

The authors proved that for a "very general" setting (meaning, for almost any random position of the knobs $\lambda_1$ and $\lambda_2$), the number of unique, indestructible beams they found is at least:

$$n \times \phi(N)$$

• $n$ is the number of "twist points" (branch points) on the surface.
• $\phi(N)$ is a number that counts how many ways you can rotate the spiral staircase without it looking the same (related to the number of floors $N$).

Why Does This Matter?

In the world of algebraic geometry, counting these "indecomposable" cycles is like counting the number of fundamental dimensions of a shape's complexity.

• Before this paper, we knew how to find these beams for simpler buildings (with fewer twist points).
• This paper shows that even as the buildings get more complex (more twist points), we can still find a massive number of these unique structural beams.

In summary:
The authors built a mathematical factory to create complex, twisted surfaces. They invented a new scanner (the regulator) and a special filter (the differential operator) to prove that these surfaces are filled with a vast number of unique, unbreakable mathematical knots. They showed that the more complex the surface, the more of these unique knots it contains, providing a deeper understanding of the hidden architecture of the mathematical universe.

Technical Summary

Here is a detailed technical summary of the paper "Construction of Higher Chow Cycles on Cyclic Coverings of $\mathbb{P}^1 \times \mathbb{P}^1$, Part II" by Yusuke Nemoto and Ken Sato.

1. Problem Statement

The paper addresses the construction and rank estimation of higher Chow cycles of type $(2, 1)$ on a specific family of algebraic surfaces.

• Context: In their previous work [6], the authors constructed such cycles on surfaces derived from products of hypergeometric curves with two parameters ($\lambda_1, \lambda_2$) and proved they generate a subgroup of rank $36 \cdot \phi(N)$(where$\phi$ is Euler's totient function).
• Generalization: This paper extends the construction to a more general setting involving $n+2$ branch points. The surfaces are constructed as quotients of products of hypergeometric curves defined by equations of the form:
  $$y^N = (x - \lambda)^A \prod_{j=1}^n (x - c_j)^A$$
  and
  $$y^N = (x - \lambda)^{N-A} \prod_{j=1}^n (x - c_j)^{N-A}$$
  where $c_1, \dots, c_n$ are fixed distinct complex numbers, and $\lambda$ varies.
• Goal: To prove that for a very general parameter $t = (\lambda_1, \lambda_2)$, the constructed cycles generate a subgroup of the indecomposable part of the higher Chow group, $CH_2(S_t, 1)_{ind}$, with a rank of at least $n \cdot \phi(N)$.

2. Methodology

The authors employ a combination of algebraic geometry, topology, and the theory of differential equations. The proof strategy follows the outline of their previous work but adapts to the increased complexity of $n$ branch points.

A. Geometric Construction

1. Surface Family: They define a family of surfaces $S \to T$ over an open subset $T \subset (\mathbb{P}^1 \setminus \{c_1, \dots, c_n, \infty\})^2$. The fiber $S_t$ is a smooth model of the quotient $(C_1 \times C_2)/\mu_N$, where $C_1, C_2$ are the hypergeometric curves mentioned above, and $\mu_N$ acts diagonally.
2. Cycle Construction: They construct explicit higher Chow cycles $\xi^{(i)}_{c_j}$ (for $i \in \mathbb{Z}/N\mathbb{Z}$ and $j=1, \dots, n$).
   • These cycles are formal sums of pairs $(Z, f)$, where $Z$ is a divisor (specifically the strict transform of the diagonal $x=y$) and $f$ is a rational function.
   • The construction involves a finite étale base change to handle $N$-th roots of the parameters.
   • The cycles are supported on the diagonal curve $Z$ and exceptional curves $Q(c_j, c_j)$ arising from the resolution of singularities.

B. Transcendental Regulator Map

To determine the rank of the generated subgroup, the authors compute the image of these cycles under the transcendental regulator map:
$$r: CH_2(S_t, 1) \to H^{2,0}(S_t)^\vee / H_2(S_t, \mathbb{Q})$$

• Since the map factors through the indecomposable part $CH_2(S_t, 1)_{ind}$, showing the images are linearly independent implies the cycles generate a large indecomposable subgroup.
• The image is computed via Levine's regulator formula, which involves integrating a canonical relative 2-form $\omega$ over a topological 2-chain $\Gamma$ associated with the cycle.

C. Topological Chains and Normal Functions

• The authors construct specific topological 2-chains $(K^l_j)^{(i)}$ on the surface $S_t$ using paths $\gamma^l_j$ connecting branch points $c_1$ to $c_j$.
• They define normal functions $\nu_{tr}(\xi)$ associated with the cycles. The value of the regulator pairing $\langle r(\xi), \omega \rangle$ is expressed as an integral over these chains.

D. Differential Equations (Picard-Fuchs)

• Jordan-Pochhammer Operator: The periods of the underlying hypergeometric curves satisfy the Jordan-Pochhammer differential equation. The authors prove that the differential operator $D = (D_{\lambda_1}, D_{\lambda_2})$, constructed from these operators, annihilates the period functions of the surface family $S \to T$.
• Explicit Computation: They explicitly compute the action of $D$ on the normal functions. This involves differentiating the integral representations of the regulator images.
• Key Result: The computation yields a system of local holomorphic functions $F_{i,j}$ satisfying specific differential equations involving Pochhammer symbols and powers of $(c_j - \lambda_k)$.

3. Key Contributions

1. Generalization of Rank Bounds: The paper generalizes the rank bound from the $n=2$ case (previous work) to an arbitrary number of branch points $n$. It establishes a lower bound of $n \cdot \phi(N)$ for the rank of the indecomposable higher Chow group.
2. Explicit Differential System: The authors provide an explicit system of differential equations satisfied by the regulator images. This connects the algebraic cycles directly to the Appell-Lauricella hypergeometric functions and the Jordan-Pochhammer differential operators.
3. Handling Trivial Automorphisms: Unlike previous cases where automorphisms of the parameter space were used to generate cycles, this paper notes that for $n \ge 3$, the automorphism group of $\mathbb{P}^1 \setminus \{c_1, \dots, c_n, \infty\}$ is trivial. The authors successfully construct the necessary cycles without relying on these symmetries, relying instead on the combinatorial structure of the $n$ branch points.

4. Main Results

Theorem 1.1 (Theorem 6.1):
For a very general parameter $t \in T$, the higher Chow cycles $\{(\xi^{(i)}_{c_j})_t\}$ generate a subgroup of $CH_2(S_t, 1)_{ind}$ with rank at least:
$$\text{rank} \ge n \cdot \phi(N)$$
where $\phi(N)$ is Euler's totient function.

Proof Logic:

• The authors compute the image of the cycles under the regulator map.
• They show that the differential operator $D$ applied to these images yields a set of functions that are linearly independent over $\mathbb{Q}(\zeta_N)$.
• Specifically, the functions $D(\nu_{tr}(\xi))$ generate a free $\mathbb{Z}[\zeta_N]$-module of rank $n$, which corresponds to an abelian group rank of $n \cdot \phi(N)$.
• By Lemma 6.2, the evaluation of these normal functions at a very general point $t$ preserves this rank.

5. Significance

• Advancement in Higher Chow Theory: This work provides one of the few explicit constructions of large families of indecomposable higher Chow cycles on higher-dimensional varieties (surfaces) with parameters. It demonstrates that the rank of these groups can grow linearly with the number of branch points ($n$).
• Connection to Special Functions: The paper deepens the link between algebraic cycles and special functions (Appell-Lauricella functions, Jordan-Pochhammer equations). It shows that the transcendental regulator map translates algebraic cycle data into solutions of specific systems of differential equations.
• Methodological Robustness: The techniques developed here (constructing cycles without relying on parameter space automorphisms) are applicable to other families of varieties where symmetry is limited, offering a template for future investigations into the structure of $K$-theory and motivic cohomology.

In summary, Nemoto and Sato successfully extend the theory of higher Chow cycles to a broader class of cyclic coverings, proving that the indecomposable part of the higher Chow group is significantly large (rank $\ge n \cdot \phi(N)$) by leveraging the interplay between topological chains, regulator maps, and the Jordan-Pochhammer differential operators.