Can we detect hot/cold spots in the CMB with Minkowski Functionals?

Abstract

In this paper, we investigate the utility of Minkowski Functionals as a probe of cold/hot disk-like structures in the CMB. In order to construct an accurate estimator, we resolve a long-standing issue with the use of Minkowski Functionals as probes of the CMB sky — namely that of systematic differences (``residuals'') when numerical and analytical MF are compared. We show that such residuals are in fact by-products of binning, whereas it was originally attributed to pixelation or masking effects. We then derive a map-independent estimator that encodes the effects of binning, applicable to beyond our present work. Using this residual-free estimator, we show that small disk-like effects (as claimed by Vielva et al. [1,2]) can be detected only when a large sample of such maps are averaged over. In other words, our estimator is noise-dominated for small disk sizes at WMAP resolution. To confirm our suspicion, we apply our estimator to the WMAP7 data to obtain a null result.