a note on four dimensional Shrinking Gradient Ricci Solitons with Constant Scalar Curvature

Chen Wang, Guoqiang Wu · 2026

Read a note on four dimensional Shrinking Gradient Ricci Solitons with Constant Scalar Curvature by Chen Wang, Guoqiang Wu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-26163.txt.

Introduction

Wang and Wu give an alternative proof of a classification result for four-dimensional complete noncompact gradient shrinking Ricci solitons with constant scalar curvature equal to one. An earlier proof by Cheng and Zhou showed such a soliton must be a finite quotient of the product of a plane and a sphere. The new argument reaches the same conclusion by analyzing the asymptotic geometry at infinity. The note offers a different route to a sharp geometric classification rather than a new classification itself.

Catalog record

CREATORChen Wang, Guoqiang Wu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE25266
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2604.26163

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

a note on four dimensional Shrinking Gradient Ricci Solitons with Constant Scalar Curvature Chen Wang, Guoqiang Wu Abstract . Let ( M 4 , g , f ) (M^{4},g,f) be a four-dimensional complete noncompact gradient shrinking Ricci soliton with the equation R ​ i ​ c + ∇ 2 f = 1 2 ​ g Ric+\nabla^{2}f=\frac{1}{2}g . If its scalar curvature is 1 1 , Cheng-Zhou [ 9 ] proved that it is a finite quotient of ℝ 2 × 𝕊 2 \mathbb{R}^{2}\times\mathbb{S}^{2} . In this note we present an alternative proof by analyzing the asymptotic geometry at infinity. 2020 Mathematics Subject Classification 53C21; 53E20 1. Introduction Let (M^{n},g) be an n -dimensional complete gradient Ricci soliton with the potential function f satisfying (1.1) \displaystyle\text{Ric}+\nabla^{2}f=\lambda g for some constant \lambda , where Ric is the Ricci tensor of g and \nabla^{2}f denotes the Hessian of the potential function f . The Ricci soliton is said to be shrinking, steady, or expanding accordingly as \lambda is positive, zero, or negative, respectively. A gradient Ricci soliton is a self-similar solution to the Ricci flow which flows by diffeomorphism and homothety. The study of solitons has become increasingly important in both the study of the Ricci flow introduced by Hamilton [ 11 ] and metric measure theory. Solitons play a direct role as singularity dilations in the Ricci flow proof of uniformization. In [ 17 ] , Perelman introduced the ancient \kappa -solutions, which play an important role in the singularity analysis, and he also proved that suitable blow down limit of ancient \kappa -solutions must be a shrinking gradient Ricci soliton. In [ 18 ] , Perelman proved that any two dimensional non-flat ancient \kappa -soluition must be the standard \mathbb{S}^{2} , and he also classified three-dimensional shrinking gradient Ricci soliton under the assumption of nonnegative curvature and \kappa -noncollapseness. Due to the work of Perelman [ 18 ] , Ni-Wallach [ 16 ] , Cao-Chen-Zhu [ 6 ] , the classification of three-dimensional shrinking gradient Ricci soliton is complete. One particularly important case is shrinking gradient Ricci soliton with constant scalar curvature. About this direction, Professor Huai-Dong Cao conjectured that Ricci shrinker has constant scalar curvature if and only if it is isometric to a finite quotient of \mathbb{R}^{n-k}\times\mathbb{N}^{k} , where \mathbb{N}^{k} is an Einstein manifold with Ric(g_{\mathbb{N}})=\frac{1}{2}g_{\mathbb{N}} . Very recently, Cheng and Zhou [ 9 ] confirmed Cao’s conjecture in dimension n=4 , together with works of Petersen-Wylie [ 19 ] and Fernández-López and García-Río [ 10 ] . For four-dimensional complete shrinking gradient Ricci solitons with constant scalar curvature, by [ 10 ] the possible values of R are \{0,2\lambda,3\lambda,4\lambda\} . Moreover, if R=0 or R=4\lambda , then the soliton is Einstein; If R=3\lambda , they are finite quotient of \mathbb{S}^{3}\times\mathbb{R} [ 10 ] . Cheng and Zhou [ 9 ] handled the most subtle case R=(n-2)\lambda=2\lambda . In this case, the Ricci curvature is non-negative by the work of Fernández-López and García-Río [ 10 ] . In addition, the Riemannian curvature is bounded since Munteanu-Wang [ 14 ] proved four-dimensional complete shrinking gradient Ricci solitons with bounded scalar curvature has bounded Riemannian curvature. However, for n -dimensional ( n\geq 5 ), we cannot even guarantee the non-negativity of Ricci curvature and the boundedness of Riemannian curvature. Moreover, we point out that even if the n -dimensional ( n\geq 5 ) shrinking gradient Ricci soliton has non-negative Ricci curvature and bounded Riemannian curvature, Cao’s Conjecture is still open. During the paper, we always assume that \lambda=\frac{1}{2} . In Cheng-Zhou’s work [ 9 ] , they applied the weighted Laplacian \Delta_{f} to the quantity \textrm{tr}(\text{Ric}^{3}) , the trace of the tensor \textrm{Ric}^{3} , for four-dimensional gradient shrinking Ricci soliton with constant scalar curvature 2\lambda and then derived the following nice inequality \displaystyle\Delta_{f}\left[f(\textrm{tr}(\text{Ric}^{3})-\frac{1}{4})\right]\geq 9f\left[\textrm{tr}(\text{Ric}^{3})-\frac{1}{4}\right]. Using integration by parts, they concluded that \textrm{tr}(\text{Ric}^{3})-\frac{1}{4}=0 over M , implying that the Ricci curvature has rank 2, and thus they obtained the rigidity result. In this short note, we provide a new proof to Cheng-Zhou’s Theorem, and the main Theorem is stated as follows. Theorem 1.1 ( [ 9 ] ) . Suppose (M^{4},g,f) is a four-dimensional shrinking gradient Ricci soliton with R=1 , then it is isometric to a finite quotient of \mathbb{R}^{2}\times\mathbb{S}^{2} . Remark . Our new proof is inspired by studying the asymptotic geometry at infinity. Naber [ 15 ] proved that a shrinking gradient Ricci soliton (M^{n},g,f) converge along the integral curve of f to \mathbb{R}\times\mathbb{N}^{n-1} , where \mathbb{N}^{n-1} is an (n-1) -dimensional shrinking gradient Ricci soliton. In our case, \mathbb{N} is of dimension three, and the three-dimensional shrinking gradient Ricci soliton is completely known. Another important new thing is that some geometric quantities can be calculated clearly under constant scalar curvature condition, which was unknown before. In Section 3, based on the point-picking argument, we prove the Riemannian curvature is bounded. With the curvature bound, we can prove that \lambda_{1}+\lambda_{2}\rightarrow 0 . In Section 4, we prove the key estimate of |\nabla Ric|^{2} , and finish the proof of Theorem 1.1 . 2. Notations and basic formulas on gradient shrinking Ricci solitons In this section, we recall the notations and basic formulas on gradient shrinking Ricci solitons with constant scalar curvature. For details, we refer to [ 5 , 9 , 11 , 19 ] . Let (M,g) be an n -dimensional complete gradient shrinking Ricci soliton satisfying ( 1.1 ). By scaling the metric g , one can normalize \lambda so that \lambda=\frac{1}{2} . In this paper, we always assume \lambda=\frac{1}{2} and the gradient shrinking Ricci soliton equation is as follows, (2.2) \displaystyle\text{Ric}+\nabla^{2}f=\frac{1}{2}g. At first we recall some basic formulas which will be used throughout the paper: (2.3) \displaystyle dR=2Ric(\nabla f,\cdot), (2.4) \displaystyle R+\Delta f=\frac{n}{2}, (2.5) \displaystyle R+|\nabla f|^{2}=f, (2.6) \displaystyle\Delta_{f}R=R-2|Ric|^{2}, (2.7) \displaystyle\Delta_{f}R_{ij}=R_{ij}-2R_{ikjl}R_{kl}. where \Delta_{f}=\Delta-\left\langle\nabla f,\nabla\right\rangle is the weighted Laplacian, and \Delta_{f} acting on the function is self-adjoint on the space of square integrable functions with respect to the weighted measure e^{-f}dv. In general, the weighted Laplacian \Delta_{f} acting on tensors is given by \Delta_{f}=\Delta-\nabla_{\nabla f} . Lemma 2.1 . Let (M^{4},g,f) be a four-dimensional shrinking gradient Ricci soliton with constant scalar curvature 1 . Then (2.8) K_{1\alpha}=\frac{\nabla_{\nabla f}R_{\alpha\alpha}+\lambda_{\alpha}(\frac{1}{2}-\lambda_{\alpha})}{f} for \alpha=2,3,4 . Proof. From the Ricci identity, we have \displaystyle-R(\nabla f,e_{\alpha},\nabla f,e_{\beta}) \displaystyle= \displaystyle-\left(\nabla_{\beta}f_{\alpha k}-\nabla_{k}f_{\alpha\beta}\right)f_{k} \displaystyle= \displaystyle\left(\nabla_{\beta}R_{\alpha k}-\nabla_{k}R_{\alpha\beta}\right)f_{k} \displaystyle= \displaystyle-\nabla_{\nabla f}R_{\alpha\beta}+\nabla_{\beta}(R_{\alpha k}f_{k})-R_{\alpha k}f_{k\beta} \displaystyle= \displaystyle-\nabla_{\nabla f}R_{\alpha\beta}-R_{\alpha k}\left(\frac{1}{2}g_{k\beta}-R_{k\beta}\right) \displaystyle= \displaystyle-\nabla_{\nabla f}R_{\alpha\beta}-\left(\frac{1}{2}R_{\alpha\beta}-\sum_{k=1}^{4}R_{\alpha k}R_{k\beta}\right), where ( 2.3 ) was used in the third equality. Therefore, we see (2.9) R(e_{1},e_{\alpha},e_{1},e_{\beta})=\frac{\nabla_{\nabla f}R_{\alpha\beta}+\left(\frac{1}{2}R_{\alpha\beta}-\sum_{k=1}^{4}R_{\alpha k}R_{k\beta}\right)}{f} due to |\nabla f|^{2}=f . ( 2.8 ) holds by setting \beta=\alpha in ( 2.9 ). This completes the proof of the lemma. ∎ Lemma 2.2 . Suppose (M^{4},g,f) is a shrinking gradient Ricci soliton with R=1 , then Ric\geq 0 . Proof. This fact is known in the literature [ 10 ] , for completeness, we give a proof. Let \lambda_{1}\leq\lambda_{2}\leq\lambda_{3}\leq\lambda_{4} be the eigenvalues of Ricci curvature. By Cauchy-Schwarz inequality, \displaystyle(\lambda_{3}+\lambda_{4})^{2}\leq 2(\lambda_{3}^{2}+\lambda_{4}^{2}), since \lambda_{3}+\lambda_{4}=1-\lambda_{1}-\lambda_{2} and \lambda_{3}^{3}+\lambda_{4}^{2}=\frac{1}{2}-\lambda_{1}^{2}-\lambda_{2}^{2} , we have \displaystyle(1-\lambda_{1}-\lambda_{2})^{2}\leq 2(\frac{1}{2}-\lambda_{1}^{2}-\lambda_{2}^{2}), hence \displaystyle 2(\lambda_{1}^{2}+\lambda_{2}^{2})+(\lambda_{1}+\lambda_{2})^{2}\leq 2(\lambda_{1}+\lambda_{2}), this implies that \lambda_{1}+\lambda_{2} is always nonnegative. Combining with Ric(\nabla f)=\frac{1}{2}dR=0 gives the desired results. ∎ 3. Curvature bound and uniform decay of \lambda_{1}+\lambda_{2} In this section, based on the point-picking argument, we will prove

Read the complete text