اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدSuperNova Acceleration Probe (SNAP): Investigating Photometric Redshift Optimization
114
0
0.0
(
0
)
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
The aim of this paper is to investigate ways to optimize the accuracy of photometric redshifts for a SNAP like mission. We focus on how the accuracy of the photometric redshifts depends on the magnitude limit and signal-to-noise ratio, wave-length coverage, number of filters and their shapes and observed galaxy type. We use simulated galaxy catalogs constructed to reproduce observed galaxy luminosity functions from GOODS, and derive photometric redshifts using a template fitting method. By using a catalog that resembles real data, we can estimate the expected number density of galaxies for which photometric redshifts can be derived. We find that the accuracy of photometric redshifts is strongly dependent on the signal-to-noise (S/N) (i.e., S/N>10 is needed for accurate photometric redshifts). The accuracy of the photometric redshifts is also dependent on galaxy type, with smaller scatter for earlier type galaxies. Comparing results using different filter sets, we find that including the U-band is important for decreasing the fraction of outliers, i.e., ``catastrophic failures. Using broad overlapping filters with resolution ~4gives better photometric redshifts compared to narrower filters (resolution >~5) with the same integration time. We find that filters with square response curves result in a slightly higher scatter, mainly due to a higher fraction of outliers at faint magnitudes. We also compare a 9-filter set to a 17-filter set, where we assume that the available exposure time per filter in the latter set is half that of the first set. We find that the 9-filter set gives more accurate redshifts for a larger number of objects and reaches higher redshift, while the 17-filter set is gives better results at bright magnitudes.
قيم البحث
اقرأ أيضاً
Knowing the redshift of galaxies is one of the first requirements of many cosmological experiments, and as its impossible to perform spectroscopy for every galaxy being observed, photometric redshift (photo-z) estimations are still of particular inte
rest. Here, we investigate different deep learning methods for obtaining photo-z estimates directly from images, comparing these with traditional machine learning algorithms which make use of magnitudes retrieved through photometry. As well as testing a convolutional neural network (CNN) and inception-module CNN, we introduce a novel mixed-input model which allows for both images and magnitude data to be used in the same model as a way of further improving the estimated redshifts. We also perform benchmarking as a way of demonstrating the performance and scalability of the different algorithms. The data used in the study comes entirely from the Sloan Digital Sky Survey (SDSS) from which 1 million galaxies were used, each having 5-filter (ugriz) images with complete photometry and a spectroscopic redshift which was taken as the ground truth. The mixed-input inception CNN achieved a mean squared error (MSE)=0.009, which was a significant improvement (30%) over the traditional Random Forest (RF), and the model performed even better at lower redshifts achieving a MSE=0.0007 (a 50% improvement over the RF) in the range of z<0.3. This method could be hugely beneficial to upcoming surveys such as the Vera C. Rubin Observatorys Legacy Survey of Space and Time (LSST) which will require vast numbers of photo-z estimates produced as quickly and accurately as possible.
The Supernova / Acceleration Probe (SNAP) is a proposed space-based experiment designed to study the dark energy and alternative explanations of the acceleration of the Universes expansion by performing a series of complementary systematics-controlle
d measurements. We describe a self-consistent reference mission design for building a Type Ia supernova Hubble diagram and for performing a wide-area weak gravitational lensing study. A 2-m wide-field telescope feeds a focal plane consisting of a 0.7 square-degree imager tiled with equal areas of optical CCDs and near infrared sensors, and a high-efficiency low-resolution integral field spectrograph. The SNAP mission will obtain high-signal-to-noise calibrated light-curves and spectra for several thousand supernovae at redshifts between z=0.1 and 1.7. A wide-field survey covering one thousand square degrees resolves ~100 galaxies per square arcminute. If we assume we live in a cosmological-constant-dominated Universe, the matter density, dark energy density, and flatness of space can all be measured with SNAP supernova and weak-lensing measurements to a systematics-limited accuracy of 1%. For a flat universe, the density-to-pressure ratio of dark energy can be similarly measured to 5% for the present value w0 and ~0.1 for the time variation w. The large survey area, depth, spatial resolution, time-sampling, and nine-band optical to NIR photometry will support additional independent and/or complementary dark-energy measurement approaches as well as a broad range of auxiliary science programs. (Abridged)
In the modern galaxy surveys photometric redshifts play a central role in a broad range of studies, from gravitational lensing and dark matter distribution to galaxy evolution. Using a dataset of about 25,000 galaxies from the second data release of
the Kilo Degree Survey (KiDS) we obtain photometric redshifts with five different methods: (i) Random forest, (ii) Multi Layer Perceptron with Quasi Newton Algorithm, (iii) Multi Layer Perceptron with an optimization network based on the Levenberg-Marquardt learning rule, (iv) the Bayesian Photometric Redshift model (or BPZ) and (v) a classical SED template fitting procedure (Le Phare). We show how SED fitting techniques could provide useful information on the galaxy spectral type which can be used to improve the capability of machine learning methods constraining systematic errors and reduce the occurrence of catastrophic outliers. We use such classification to train specialized regression estimators, by demonstrating that such hybrid approach, involving SED fitting and machine learning in a single collaborative framework, is capable to improve the overall prediction accuracy of photometric redshifts.
We investigate six supernova remnant (SNR) candidates --- G51.21+0.11, G52.37-0.70, G53.07+0.49, G53.41+0.03, G53.84-0.75, and the possible shell around G54.1-0.3 --- in the Galactic Plane using newly acquired LOw-Frequency ARray (LOFAR) High-Band An
tenna (HBA) observations, as well as archival Westerbork Synthesis Radio Telescope (WSRT) and Very Large Array Galactic Plane Survey (VGPS) mosaics. We find that G52.37-0.70, G53.84-0.75, and the possible shell around pulsar wind nebula G54.1+0.3 are unlikely to be SNRs, while G53.07+0.49 remains a candidate SNR. G51.21+0.11 has a spectral index of $alpha=-0.7pm0.21$, but lacks X-ray observations and as such requires further investigation to confirm its nature. We confirm one candidate, G53.41+0.03, as a new SNR because it has a shell-like morphology, a radio spectral index of $alpha=-0.6pm0.2$ and it has the X-ray spectral characteristics of a 1000-8000 year old SNR. The X-ray analysis was performed using archival XMM-Newton observations, which show that G53.41+0.03 has strong emission lines and is best characterized by a non-equilibrium ionization model, consistent with an SNR interpretation. Deep Arecibo radio telescope searches for a pulsar associated with G53.41+0.03 resulted in no detection, but place stringent upper limits on the flux density of such a source if it is beamed towards Earth.
We use a sample of 42 supernovae detected with the Advanced Camera for Surveys on-board the Hubble Space Telescope as part of the Great Observatories Origins Deep Survey to measure the rate of core collapse supernovae to z~0.7 and type Ia supernovae
to z~1.6. This significantly increases the redshift range where supernova rates have been estimated from observations. The rate of core collapse supernovae can be used as an independent probe of the cosmic star formation rate. Based on the observations of 17 core collapse supernovae, we measure an increase in the core collapse supernova rate by a factor of 1.6 in the range 0.3<z<0.7, and an overall increase by a factor of 7 to z~0.7 in comparison to the local core collapse supernova rate. The increase in the rate in this redshift range in consistent with recent measurements of the star formation rate derived from UV-luminosity densities and IR datasets. Based on 25 type Ia supernovae, we find a SN Ia rate that is a factor 3-5 higher at z~1 compared to earlier estimates at lower redshifts (z<0.5), implying that the type Ia supernova rate traces a higher star formation rate at redshifts z>1 compared to low redshift. At higher redshift (z>1), we find a suggested decrease in the type Ia rate with redshift. This evolution of the Ia rate with redshift is consistent with a type Ia progenitor model where there is a substantial delay between the formation of the progenitor star and the explosion of the supernova. Assuming that the type Ia progenitor stars have initial main sequence masses 3-8 M_Sun, we find that 5-7% of the available progenitors explode as type Ia supernovae.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
