# zbMATH — the first resource for mathematics

On orientable real algebraic $$M$$-surfaces. (English. Russian original) Zbl 0936.14039
Math. Notes 62, No. 4, 434-438 (1997); translation from Mat. Zametki 62, No. 4, 520-526 (1997).
From the text: Let $$X$$ be a nonsingular projective real algebraic $$M$$-surface. Suppose that the set of real points $$X(\mathbb{R})$$ is orientable, and the homology group $$H_* (X(\mathbb{C}),\mathbb{Z})$$ is free. Let $$X_1,\dots, X_s$$ be the connected components of $$X(\mathbb{R})$$; they determine homology classes $x_1,\dots, x_s\in H_2 (X(\mathbb{C}), \mathbb{F}_2).$ Theorem 1. Let $$q$$ be the irregularity of $$X$$. Then there exists $$q+1$$ linearly independent relations between the classes $$x_1,\dots, x_s$$.
Note that the number of linearly independent relations between $$x_1,\dots, x_s$$ does not exceed $$q+1$$. In what follows, $$\chi_k$$ denotes the Euler characteristic of $$X_k$$, $$k=1,\dots, s$$.
Theorem 2. Suppose that $$\chi_k\equiv 0\pmod {2^\mu}$$, where $$\mu\geq 1$$ and $$k=1,\dots, s$$. If $$x_{i_1}+\cdots+ x_{i_n}=0$$, then the congruence $$\chi_{i_1}+\cdots+ \chi_{i_n}\equiv 0\pmod {2^{\mu+2}}$$ holds.
If $$H_1(X(\mathbb{C}),\mathbb{Z})=0$$, then there is only one relation between $$x_1,\dots, x_s$$, namely $$x_1+\cdots+ x_s=0$$. In the general case, the description of possible relations is based on the Albanese mapping $$\alpha:X\to A$$. The set of complex points of the Albanese variety $$A(\mathbb{C})$$ is a complex torus with an antiholomorphic involution $$\theta: A(\mathbb{C})\to A(\mathbb{C})$$. Since $$X$$ is an $$M$$-surface, $$A$$ is an $$M$$-variety and the pair $$(A(\mathbb{C}),\theta)$$ is topologically identified with the complex torus $$\mathbb{C}^q/ \mathbb{Z}^q$$ endowed with the complex conjugation involution. The set of real points $$A(\mathbb{R})$$ is the union of $$2^q$$ real tori. Let $$A(\mathbb{R})^{(i)}$$, $$i=1,\dots, q$$, be the union of the connected components of $$A(\mathbb{R})$$ that are contained in the preimage of the curve $$\mathbb{R}/ \mathbb{Z}\subset \mathbb{C}/\mathbb{Z}$$ under the projection $$\pi_i: \mathbb{C}^q/ \mathbb{Z}^q\to \mathbb{C}/\mathbb{Z}$$, where $$\pi_i (z_1,\dots, z_q)= z_i$$. By $$X(\mathbb{R})^{(i)}$$ denote the union of the connected components of $$X(\mathbb{R})$$ that are contained in the preimge of $$A(\mathbb{R})^{(i)}$$ under the Albanese mapping $$\alpha:X\to A$$.
Theorem 3. The fundamental classes $$[X(\mathbb{R})]$$ and $$[X(\mathbb{R})^{(i)}]$$, $$i= 1,\dots, q$$, equal 0 in $$H_2(X(\mathbb{C}), \mathbb{F}_2)$$; the corresponding relations between the classes $$x_1,\dots, x_s$$ form a basis of the system of relations.
Now suppose that the basis of the Albanese mapping $$\alpha:X\to A$$ is a curve. Then there is another method for obtaining the basis of the system of relations between $$x_1,\dots, x_s$$. Let $$C= \alpha(X)$$ be a curve; then $$C$$ has no singular points. The homomorphism $$\alpha_*: H_1(X(\mathbb{C}), \mathbb{F}_2)\to H_1(C(\mathbb{C}), \mathbb{F}_2)$$ is an isomorphism, and the compex conjugation involution acts trivially on the group $$H_1(X(\mathbb{C}), \mathbb{F}_2)$$. Therefore, $$C$$ is an $$M$$-curve, and the number of the connected components of $$C(\mathbb{R})$$ is $$q+1$$; denote these components by $$C_1,\dots, C_{q+1}$$. Let $$X(\mathbb{R})_i$$ be the union of the connected components of $$X(\mathbb{R})$$ that are contained in the preimage of $$C_i$$, where $$i=1,\dots, q+1$$.
Theorem 4. The fundamental classes $$[X(\mathbb{R})_i]$$, $$i= 1,\dots, q+1$$, equal 0 in $$H_2(X(\mathbb{C}), \mathbb{F}_2)$$; the corresponding relations between the classes $$x_1,\dots, x_s$$ form a basis of the system of relations.
##### MSC:
 14P25 Topology of real algebraic varieties 14F25 Classical real and complex (co)homology in algebraic geometry
Full Text:
##### References:
  V. A. Krasnov, ”On homology classes determined by real points of a real variety”,Izv. Akad. Nauk SSSR Ser. Mat. [Math. USSR-Izv.],55, No. 2, 282–302 (1991). · Zbl 0751.14036  V. A. Krasnov, ”The Albanese mapping for real varieties”,Mat. Zametki [Math. Notes],32, No. 3, 365–374 (1982). · Zbl 0515.14028  V. A. Krasnov, ”The Albanese mapping forGM$$\mathbb{Z}$$-varieties,”Mat. Zametki [Math. Notes],35, No. 5, 739–747 (1984).  I. R. Shafarevich (editor),Algebraic Surfaces [in Russian],Trudy Mat. Inst. Steklov [Proc. Steklov Inst. Math.] Vol. 75, (1965). · Zbl 0154.21001  V. A. Krasnov, ”On cohomology classes determined by real points of a real algebraicGM-surface”,Izv. Ross. Akad. Nauk Ser. Mat. [Russian Acad. Sci. Izv. Math.],57, No. 5, 210–221 (1993).  V. A. Krasnov, ”The Harnack-Thom inequalities for mappings of real varieties”,Izv. Akad. Nauk SSSR Ser. Mat. [Math. USSR-Izv.],47, No. 2, 268–297 (1983).  V. A. Krasnov, ”Equivariant cohomology of real algebraic surfaces and its approximation”,Izv. Ross. Akad. Nauk Ser. Mat. [Russian Acad. Sci. Izv. Math.], (to appear). · Zbl 0843.14024  V. A. Rokhlin, ”Congruences modulo 16 in the Hilbert 16th problem”,Funktsional. Anal. i Prilozhen, [Functional Anal. Appl.],6, No. 4, 58–64 (1972). · Zbl 0259.14010 · doi:10.1007/BF01077517  V. V. Nikulin, ”Involutions of integer-valued quadratic forms and their applications to real algebraic geometry”,Izv. Akad. Nauk SSSR Ser. Mat. [Math. USSR-Izv.],47, No. 1, 109–177 (1983).
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.