We propose a new rate-distortion based algorithm for compressing 3D surface geometry represented using triangular normal meshes. We apply the Estimation-Quantization (EQ) algorithm to compress normal mesh wavelet coefficients. The EQ algorithm models the wavelet coefficients as a Gaussian random field with slowly varying standard deviation that depends on the local neighborhood and uses rate-distortion optimal scalar quantizers. We achieve gains of 0.5 to 1 dB with the EQ algorithm compared to the recently proposed zerotree compression for normal meshes.