[Paper Review] Preliminary results on galactic dark matter from the complete EROS2 microlensing survey

P. Tisserand, A. Milsztajn|arXiv (Cornell University)|Jan 26, 2005

Stellar, planetary, and galactic studies3 citations

TL;DR

This study presents the final results from the complete EROS2 microlensing survey of the Large Magellanic Cloud, analyzing 33 million stars over 6.7 years to search for gravitational microlensing events caused by massive compact halo objects (MACHOs). The analysis reveals only four new candidates, with a derived 95% confidence upper limit of 12% on the MACHO contribution to galactic dark matter for masses between 2×10⁻⁴ and 1 solar masses, strongly disfavoring MACHOs as the dominant dark matter component.

ABSTRACT

The EROS-2 collaboration has conducted a survey of the Magellanic Clouds between July 1996 and February 2003 (6.7 years), in order to search for microlensing phenomena due to putative machos (massive astronomical compact halo objects). About 55 million stars were monitored over 100 deg^2, typically with 400 to 500 measurements per star, simultaneously in two wide passbands. The full EROS-2 Magellanic Cloud data set was processed using the same program chain, and a photometric database was constructed. The lightcurves of the brightest 33 million stars were then searched for the characteristic signature of microlensing phenomena, in the wide timescale range of 1 to 1000 days (corresponding to macho masses ranging from about 0.01 solar mass to a few hundred solar masses). The sensitivity of this new analysis is better by a factor 4. The new EROS-2 microlensing candidates towards the LMC and the corresponding (preliminary) final limits from EROS-2 on the macho content of the halo are presented; they are by far not numerous enough to account for the galactic dark matter. We have also followed, whenever possible, the microlensing candidates that were previously published by the EROS and MACHO collaborations. 1 MACHO and 2 EROS LMC candidates have shown subsequent variability and cannot be considered to be serious candidates.

Motivation & Objective

• To search for gravitational microlensing events in the Large Magellanic Cloud using the complete EROS2 dataset.
• To determine the upper limit on the fraction of galactic dark matter composed of MACHOs via microlensing detection.
• To re-evaluate previously published microlensing candidates using improved photometry and analysis techniques.
• To assess the sensitivity and detection efficiency of the EROS2 survey for MACHO masses from 0.0002 to 100 solar masses.
• To compare the final results with earlier MACHO and EROS1 findings to resolve discrepancies in MACHO abundance estimates.

Proposed method

• The EROS2 survey monitored 55 million stars in the Large and Small Magellanic Clouds using two wide-band CCD mosaics over 100 deg² with 400–500 measurements per star.
• Light curves of the 33 million brightest stars were analyzed for microlensing signatures with timescales between 1 and 1000 days, corresponding to MACHO masses from ~0.01 to several hundred solar masses.
• A multi-stage selection process applied signal-to-noise cuts, best-fit point-source point-lens microlensing model comparison, and rejection of physical backgrounds such as blue bumpers and supernovae.
• Supernovae were rejected using an asymmetric microlensing light curve model with an asymmetry parameter S, rejecting candidates with |S| > 0.3 or those near visible galaxies.
• Detection efficiency was calculated using simulated microlensing light curves, enabling sensitivity calibration and statistical limits at 95% confidence level.
• Optical depth was computed from surviving candidates to estimate the MACHO contribution to the halo, with the final limit derived from the LMC data alone.

Experimental results

Research questions

• RQ1 What is the upper limit on the MACHO fraction of galactic dark matter based on the full EROS2 survey of the Large Magellanic Cloud?
• RQ2 Do previously published microlensing candidates from EROS and MACHO surveys remain valid after re-analysis with improved data and methods?
• RQ3 How does the sensitivity of the EROS2 survey compare to earlier surveys like MACHO and EROS1 in detecting MACHOs?
• RQ4 Can the observed number of microlensing candidates be explained by MACHOs, or do they instead point to other astrophysical backgrounds?
• RQ5 What is the significance of re-observed variability in previously reported candidates, and how does it affect the interpretation of microlensing signals?

Key findings

• The EROS2 survey identified only four new microlensing candidates in the Large Magellanic Cloud, with one (EROS2-LMC#8) showing strong chromatic variation and a plausible lensing scenario involving a 13.5 mag M-dwarf at 300 pc.
• Three additional candidates (EROS2-LMC#9 to #11) have lower signal-to-noise and are consistent with a source magnitude of ~21st mag and timescales of 35–55 days.
• The optical depth derived from the three lower-significance candidates is 1.5 × 10⁻⁸, corresponding to only 3% of the standard spherical halo optical depth.
• A preliminary 95% confidence level upper limit on the MACHO content of the halo is less than 12% for MACHO masses between 2×10⁻⁴ and 1 solar mass.
• All previously published EROS and MACHO microlensing candidates were found to be inconsistent with stable microlensing: 1 EROS1 and 1 EROS2 candidate showed new variability after >6 years, and 3 EROS2 candidates were identified as supernovae due to asymmetry (|S| > 0.3).
• The sensitivity of the EROS2 analysis is improved by a factor of 4 compared to the last EROS2 LMC result and by a factor of 2 compared to the MACHO survey, due to longer baseline and refined data processing.