The Peculiar Motions of Early-type Galaxies in Two Distant Regions. IV. The Photometric Fitting Procedure
The Peculiar Motions of Early-type Galaxies in Two Distant Regions. IV. The Photometric Fitting Procedure
复制标题
两个遥远区域的早期型星系的特殊运动。
DOI：
10.1086/312978
复制
发表时间：
1996
期刊：
影响因子：
--
通讯作者：
G. Wegner
中科院分区：
文献类型：
--
作者：
R. Saglia;E. Bertschinger;G. Baggley;D. Burstein;M. Colless;R. Davies;Robert K. McMahan, Jr.;G. Wegner
EFAR 项目是一项对 84 个星团中 736 个候选早期型星系的研究，这些星系团位于武仙座-北冕座和英仙座-鲸鱼座的两个区域，距离 cz ≈ 6000-15,000 km s-1。在本文中，我们描述了一种新的星系光度测量方法，用于推导 EFAR 星系的光度参数。该算法将圆形表面亮度轮廓拟合为两个视宁卷积分量（R1/4 和指数定律）的总和。这种方法使我们能够均匀地拟合 EFAR 星系显示的各种光度剖面，并导出（至少是其中的一个子集）核球和盘参数。对同一物体的多次曝光进行了最佳组合，并开发了可选的天空拟合程序来纠正天空相减错误。分析了广泛的蒙特卡罗模拟，以测试算法的性能并估计随机误差和系统误差的大小。如果拟合轮廓的全局信噪比大于 ≈ 300，则随机误差很小。系统误差可能是由于 (1) 天空减法中的误差、(2) 拟合轮廓的有限径向范围、(3) 由于观察卷积和像素采样而缺乏分辨率、(4) 对边缘看到的非常扁平的物体使用圆形轮廓，以及 (5) 拟合函数与物体轮廓的匹配不佳造成的。当存在小至总光的 20% 的圆盘分量时，广泛使用的简单 R1/4 定律会产生较大的系统误差来拟合光度分布。由于涉及外推法，系统误差的大小无法从 χ2 函数接近最小值的形状确定。相反，我们必须通过一组质量参数来估计它们，并根据我们的模拟进行校准，这些参数考虑了导出总星等所涉及的外推量、天空校正的大小、星系相对于天空的平均表面亮度、轮廓的径向范围、其信噪比、视宁度以及拟合的减少的 χ2。我们制定了一个组合质量参数 Q，它表示拟合的预期精度。如果 Q = 1，预计总星等误差 MTOT 小于 0.05 mag，半光度半径 Re 小于 10%；如果 Q = 2，则预计误差小于 0.15 mag 和 25%； 89% 的 EFAR 星系符合 Q = 1 或 Q = 2。即使 Q = 3，组合基本平面量 FP = log Re-0.3⟨SBe⟩（其中 ⟨SBe⟩ 是平均有效表面亮度）的误差也小于 0.03。因此，MTOT 和 Re 的系统误差对涉及 FP 的距离估计仅具有边际效应。我们表明，最近用于拟合椭圆星系光度剖面的 R1/n 剖面序列相当于（n ≤ 8）R1/4 和指数剖面的子样本，具有适当的尺度长度和盘核球比。这表明早期型星系所表现出的各种光度分布可能是由于盘状成分的存在造成的。
The EFAR project is a study of 736 candidate early-type galaxies in 84 clusters lying in two regions toward Hercules-Corona Borealis and Perseus-Cetus at distances cz ≈ 6000-15,000 km s-1. In this paper we describe a new method of galaxy photometry adopted to derive the photometric parameters of the EFAR galaxies. The algorithm fits the circularized surface brightness profiles as the sum of two seeing-convolved components, an R1/4 and an exponential law. This approach allows us to fit the large variety of luminosity profiles displayed by the EFAR galaxies homogeneously and to derive (for at least a subset of these) bulge and disk parameters. Multiple exposures of the same objects are optimally combined and an optional sky-fitting procedure has been developed to correct for sky-subtraction errors. Extensive Monte Carlo simulations are analyzed to test the performance of the algorithm and estimate the size of random and systematic errors. Random errors are small, provided that the global signal-to-noise ratio of the fitted profiles is larger than ≈ 300. Systematic errors can result from (1) errors in the sky subtraction, (2) the limited radial extent of the fitted profiles, (3) the lack of resolution due to seeing convolution and pixel sampling, (4) the use of circularized profiles for very flattened objects seen edge-on, and (5) a poor match of the fitting functions to the object profiles. Large systematic errors are generated by the widely used simple R1/4 law to fit luminosity profiles when a disk component, as small as 20% of the total light, is present. The size of the systematic errors cannot be determined from the shape of the χ2 function near its minimum because extrapolation is involved. Rather, we must estimate them by a set of quality parameters, calibrated against our simulations, which take into account the amount of extrapolation involved to derive the total magnitudes, the size of the sky correction, the average surface brightness of the galaxy relative to the sky, the radial extent of the profile, its signal-to-noise ratio, the seeing value, and the reduced χ2 of the fit. We formulate a combined quality parameter Q, which indicates the expected precision of the fits. Errors in total magnitudes MTOT less than 0.05 mag and in half-luminosity radii Re less than 10% are expected if Q = 1, and less than 0.15 mag and 25% if Q = 2; 89% of the EFAR galaxies have fits with Q = 1 or Q = 2. The errors on the combined fundamental plane quantity FP = log Re-0.3⟨SBe⟩, where ⟨SBe⟩ is the average effective surface brightness, are smaller than 0.03 even if Q = 3. Thus, systematic errors on MTOT and Re only have a marginal effect on the distance estimates that involve FP. We show that the sequence of R1/n profiles, recently used to fit the luminosity profiles of elliptical galaxies, is equivalent (for n ≤ 8) to a subsample of R1/4 and exponential profiles, with appropriate scale lengths and disk-to-bulge ratios. This suggests that the variety of luminosity profiles shown by early-type galaxies may be due to the presence of a disk component.