The first data release(DR1) of the LAMOST regular survey

The Large sky Area Multi-Object Fiber Spectroscopic Telescope（LAMOST） general survey is a spectroscopic survey that will eventually cover approximately half of the celestial sphere and collect 10 million spectra of stars, galaxies and QSOs. Objects in both the pilot survey and the first year regular survey are included in the LAMOST DR1. The pilot survey started in October 2011 and ended in June 2012, and the data have been released to the public as the LAMOST Pilot Data Release in August 2012. The regular survey started in September 2012, and completed its first year of operation in June 2013. The LAMOST DR1 includes a total of 1202 plates containing 2 955 336 spectra, of which 1 790 879 spectra have observed signalto-noise ratio（SNR） ≥ 10. All data with SNR ≥ 2 are formally released as LAMOST DR1 under the LAMOST data policy. This data release contains a total of 2 204 696 spectra, of which 1 944 329 are stellar spectra, 12 082 are galaxy spectra and 5017 are quasars. The DR1 not only includes spectra, but also three stellar catalogs with measured parameters： late A,FGK-type stars with high quality spectra（1 061 918 entries）, A-type stars（100 073 entries）, and M-type stars（121 522 entries）. This paper introduces the survey design, the observational and instrumental limitations, data reduction and analysis, and some caveats. A description of the FITS structure of spectral files and parameter catalogs is also provided.

The design and implementation of a ROACH2+GPU based correlator on the Tianlai dish array

A digital correlator is a crucial element in a modern radio telescope.In this paper,we describe a scalable design for the correlator system of the Tianlai pathfinder array,which is an experiment dedicated to testing key technologies for conducting a 21cm intensity mapping survey.The correlator implements the FX design,which firstly performs a fast Fourier transform(FFT) including polyphase filter bank(PFB) computation using a Collaboration for Astronomy Signal Processing and Electronics Research(CASPER) Reconfigurable Open Architecture Computing Hardware-2(ROACH2) board,then computes cross-correlations by employing Graphics Processing Units(GPUs).The design has been tested both in laboratory and in actual observation.

Radio frequency interference detection based on the AC-UNet model

Radio frequency interference(RFI)is a serious issue in radio astronomy.This paper proposes a U-Net network model with atrous convolution to detect RFI.Using the ability of convolutional neural networks to extract image features of RFI,and learning RFI distribution patterns,the detection model of the RFI is established.We use observational data containing real RFIs obtained by the Tianlai telescope to train the model so that the model can detect RFI.Calculate the probability of a data point being RFI pixel by pixel,and set a threshold.At the same time the dropout layer was added to avoid overfitting problems.If the predicted probability of a data point exceeds the threshold,it is considered that there is RFI,and if the predicted probability of a data point does not exceed the threshold,then it is considered that there is no RFI,so that the part of the image with RFI is flagged.Experimental results show that this approach can achieve satisfactory accuracy in the detection of radio observation images with a small amount of RFI.

Cosmological constraint on Brans-Dicke Model

We combine new Cosmic Microwave Background （CMB） data from Planck with Baryon Acoustic Oscillation （BAO） data to constrain the Brans-Dicke （BD） the- ory, in which the gravitational constant G evolves with time. Observations of type Ia supernovae （SNela） provide another important set of cosmological data, as they may be regarded as standard candles after some empirical corrections. However, in theories that include modified gravity like the BD theory, there is some risk and complication when using the SNIa data because their luminosity may depend on G. In this paper, we assume a power law relation between the SNIa luminosity and G, but treat the power index as a free parameter. We then test whether the differ- ence in distances measured with SNIa data and BAO data can be reduced in such a model. We also constrain the BD theory and cosmological parameters by making a global fit with the CMB, BAO and SNIa data set. For the CMB＋BAO＋SNIa data set, we find 0.08 × 10-2 〈 ζ 〈 0.33 ×10-2 at the 68% confidence level （CL） and -0.01 × 10-2 〈 ζ〈 0.43 × 10-2 at the 95% CL, where ζ is related to the BD parameter w by ζ = In（1 ＋ 1/w）.

Semi-numerical simulation of reionization with semi-analytical modeling of galaxy formation

In a semi-numerical model of reionization, the evolution of ionization fraction is approximately simulated by the criterion of ionizing photon to baryon ratio. We incorporate a semi-analytical model of galaxy formation based on the Millennium II N-body simulation into the semi-numerical modeling of reionization. The semianalytical model is used to predict the production of ionizing photons, then we use the semi-numerical method to model the reionization process. Such an approach allows more detailed modeling of the reionization, and also connects observations of galaxies at low and high redshifts to the reionization history. The galaxy formation model we use was designed to match the low-z observations, and it also fits the high redshift luminosity function reasonably well, but its prediction about star formation falls below the observed value, and we find that it also underpredicts the stellar ionizing photon production rate, hence the reionization cannot be completed at z ~ 6. We also consider simple modifications of the model with more top heavy initial mass functions, which can allow the reionization to occur at earlier epochs. The incorporation of the semi-analytical model may also affect the topology of the HI regions during the epoch of reionization, and the neutral regions produced by our simulations with the semi-analytical model, which appeared less poriferous than the simple halo-based models.

On fitting the full spectrum of luminous red galaxies by using ULySS and STARLIGHT

We select a sample of quiescent luminous red galaxies （LRGs） from the Sloan Digital Sky Survey Data Release 7 with a high signal-to-noise ratio （S/N） to study the consistency of fitting the full spectrum by using different packages, mainly, ULySS and STARLIGHT. The spectrum of each galaxy in the sample is fitted by the full spectrum fitting packages ULySS and STARLIGHT. We find： （1） for spec- tra with higher S/Ns, the ages of stellar populations obtained from ULySS are slightly older than those from STARLIGHT, and metallicities derived from ULySS are slightly richer than those from STARLIGHT. In general, both packages can give roughly con- sistent fitting results. （2） For low S/N spectra, it is possible that the fitting by ULySS can become trapped at some local minimum in the parameter space during execution and thus may give unreliable results, but STARLIGHT can still give reliable results. Based on the fitting results of LRGs, we further analyze their star formation history and the relation between their age and velocity dispersion, and find that they agree well with conclusions from previous works.

Primordial non-Gaussianity from LAMOST surveys

The primordial non-Gaussianity （PNG） in the matter density perturbation is a very powerful probe of the physics of the very early Universe. The local PNG can induce a distinct scale-dependent bias on the large scale structure distribution of galaxies and quasars, which could be used for constraining it. We study the detection limits of PNG from the surveys of the LAMOST telescope. The cases of the main galaxy survey, the luminous red galaxy （LRG） survey, and the quasar survey of dif- ferent magnitude limits are considered. We find that the Mainl sample （i.e. the main galaxy survey which is one magnitude deeper than the SDSS main galaxy survey, or r 〈 18.8） could only provide a very weak constraint on PNG. For the Main2 sample （r 〈 19.8） and the LRG survey, the 2or （95.5%） limits on the PNG parameter fNL are ｜ fNL｜〈 145 and ｜ fNL｜ 〈 114 respectively, which are comparable to the current limit from cosmic microwave background （CMB） data. The quasar survey could provide a much more stringent constraint, and we find that the 2a limit for ｜fNL｜ is between 50 and 103, depending on the magnitude limit of the survey. With Planck-like priors on cosmological parameters, the quasar survey with g 〈 21.65 would improve the constraint to I fNLI 〈 43 （2or）. We also discuss the possibility of further tightening the constraint by using the relative bias method proposed by Seljak.