MASK EFFECTS ON COSMOLOGICAL STUDIES WITH WEAK-LENSING PEAK STATISTICS
MASK EFFECTS ON COSMOLOGICAL STUDIES WITH WEAK-LENSING PEAK STATISTICS
复制标题
DOI：
10.1088/0004-637x/784/1/31
复制
发表时间：
2013-04
期刊：
影响因子：
--
通讯作者：
Xiangkun Liu;Qiao Wang;Chuzhong Pan;Z. Fan
中科院分区：
文献类型：
--
作者：
Xiangkun Liu;Qiao Wang;Chuzhong Pan;Z. Fan
通过数值模拟，详细分析了坏数据的去除，即掩模效应对背景星系剪切测量重建的弱透镜会聚场峰值统计量的影响。发现由于掩模的存在，高峰分数被系统地增强；掩蔽面积越大，增强效果越好。在总掩蔽区域约为调查区域13%的情况下，与我们考虑的宇宙学模型中无掩蔽情况的~ 7%相比，具有信号噪声比ν大于或等于3的峰的分数是峰总数的~ 11%。这可能会对具有弱透镜收敛峰值统计的宇宙学研究产生重大影响，如果不适当考虑这些影响，则会在参数约束中引起很大的偏差。即使在9度2的测量区域，（Ωm, σ8）的偏差已经大到无法容忍，接近3σ。值得注意的是，大多数受影响的峰值都接近掩蔽区。因此，在峰值统计中排除这些区域的峰值可以减少偏差效应，但代价是失去可用的调查区域。进一步的研究发现，掩蔽区域周围的峰值数量的增加很大程度上归因于弱透镜收敛重建中可用的星系数量较少，导致比远离掩蔽区域的噪声更高。因此，我们开发了一个模型，在该模型中，我们只排除那些半径大于3 '的非常大的掩模，而将所有其他掩模区域保留在峰值计数统计中。对于剩下的部分，我们分别用不同的噪声水平来处理靠近和远离掩蔽区域的区域。结果表明，该双噪声级模型可以很好地解释峰值统计量的掩模效应，并且在参数拟合中显著减小了宇宙学参数的偏差。
With numerical simulations, we analyze in detail how the bad data removal, i.e., the mask effect, can influence the peak statistics of the weak-lensing convergence field reconstructed from the shear measurement of background galaxies. It is found that high peak fractions are systematically enhanced because of the presence of masks; the larger the masked area is, the higher the enhancement is. In the case where the total masked area is about 13% of the survey area, the fraction of peaks with signal-to-noise ratio ν ⩾ 3 is ∼11% of the total number of peaks, compared with ∼7% of the mask-free case in our considered cosmological model. This can have significant effects on cosmological studies with weak-lensing convergence peak statistics, inducing a large bias in the parameter constraints if the effects are not taken into account properly. Even for a survey area of 9 deg2, the bias in (Ωm, σ8) is already intolerably large and close to 3σ. It is noted that most of the affected peaks are close to the masked regions. Therefore, excluding peaks in those regions in the peak statistics can reduce the bias effect but at the expense of losing usable survey areas. Further investigations find that the enhancement of the number of high peaks around the masked regions can be largely attributed to the smaller number of galaxies usable in the weak-lensing convergence reconstruction, leading to higher noise than that of the areas away from the masks. We thus develop a model in which we exclude only those very large masks with radius larger than 3′ but keep all the other masked regions in peak counting statistics. For the remaining part, we treat the areas close to and away from the masked regions separately with different noise levels. It is shown that this two-noise-level model can account for the mask effect on peak statistics very well, and the bias in cosmological parameters is significantly reduced if this model is applied in the parameter fitting.