SNLS - the CFHTLS Supernova Survey

SNLS (the CFHT Legacy Survey Supernova program) is a large observational program to study supernovae out to redshifts of 0.8 and beyond. It is a major constituent of the CFHTLS Deep Survey, which has been allocated 202 nights of CFHT observing time over the next 5 years with the MegaCam 1 deg x 1 deg CCD mosaic.

The primary goal of the SNLS program is to detect and followup 1000 Type Ia supernovae out to z=0.8. From these data the equation-of-state parameter w of the Universe will be derived to an accuracy of +-0.1, with constraints on the nature of the "dark energy" that causes the acceleration of the expansion of the Universe. Many other science goals (e.g. cosmology from Type II SNe, the cosmic star formation rate from SNeII, GRB's, variable AGN's, ...) will be pursued with the same data.

The survey produces approximately 5 epochs per 18 day dark run, with u*g'r'i'z' at almost every epoch. Primary SN detection is in i' (limiting AB~=25); approximately 20-40 new supernovae can be detected in each dark run. Spectroscopic followup time has been allocated on 8-10m class telescopes to provide redshifts and SN types. The combination of the large commitment of CFHT observing time, areal coverage (4 deg^2), sampling rate, multicolour photometry (to determine interstellar absorption), and allocated spectroscopic followup time is unprecedented.

More information on the survey and the science goals can be found at the links below:

SNLS Ca	- Canadian SNLS web site
Gemini	- Gemini spectroscopy proposal (good overview of science case)
SNLS Fr	- French SNLS web site
Deep Survey	- CFHTLS Deep Survey (coordinator Genevieve Soucail)
MegaCam	- CFHT's MegaCam/MegaPrime environment
CFHTLS	- CFHTLS web site
CFHTLS users	- CFHTLS user registration
CFHTLS email	- CFHTLS email exploders
CFHTLS v2.3	- CFHTLS Project description v2.3
CFHTLS SNe	- Appendix on Supernovae - CFHTLS project description v2.3

The SNLS collaboration between Canada and France is now formalized by an MOU which describes the operation and goals of the SNLS. The collaboration is managed and overseen by a Collaboration Board, the chair of which is Reynald Pain (2 year term). Although the SNLS collaboration is led by the Canadian and French communities, smaller UK, US, and Portugese groups also are a part of it, since they are contributing Keck, Gemini, and VLT time for spectroscopy.

Many Canadian scientists are participating in SNLS, and especially in the spectroscopic followup on Gemini. The Canadian SNLS core survey team consists of:

U. Toronto	Ray Carlberg	Professor
	Andy Howell	PDF
	Tom Merrall	PDF
	Kathy Perrett	PDF
	Mark Sullivan	PDF
U. Victoria	Chris Pritchet	Professor, CFHTLS SN Coordinator
	David Balam	RA
	........	PDF - hiring in progress

Some of the PDF's are or will be involved with LS Wide Survey as well. The Canadian SNLS team is supported by an NSERC CRO grant, and by individual Discovery Grants to RC and CP.

CFHT MegaCam is routinely producing images once every ~4 nights on the Deep fields (u*g'r'i'z', roughly 2h 30m total each epoch). The four 1 deg x 1 deg fields being observed (0226-0430, 1000+0212, 1418+5231, and 2216-1744) are described at this link. Both French and Canadian teams have real-time access to the data (after it reaches Waimea), and each team scans the data for new SN (and other variable) candidates using independent pipelines. (The French team has people in Hawaii to do this; the Canadians rely on a typical bandwidth of ~1 MB/s from Victoria to Hawaii to run remote X sessions on dedicated Waimea computers.) Information on the 2 independent lists of SN candidates can be found at

http://legacy.astro.utoronto.ca (Canada)
http://makiki.cfht.hawaii.edu:872/sne/ (France)

Generally the two lists of candidates agree quite well down to i'=24, and a preliminary analysis shows that our detections are reliable to at least i'=24.5 (because more than 80% of all variables discovered at i'=24.5 turn out to be real). We may consider merging the Canadian and French detection pipelines in the future.

The following table and figure show the disovery statistics of supernova up to the end of October 2003.

An important issue for the LS, and MegaCam users in general, is image quality. The MegaPrime corrector produces a centre-to-edge variation in IQ that is greater than expected; much of the engineering in 03A was used in attempting (unsuccessfully) to solve this problem. However, there is a definite trend towards better image quality now, and this is seen in all filters except u* (for which there is only fragmentary data that the SN team has analyzed).

A detailed breakdown of observing time for the Deep survey by filter, field, and queue run, along with summary statistics for all LS surveys, can be found at

www.astro.uvic.ca/~pritchet/SN/CassiopeiaDec2003/summary.ps

(with an alternate, prettier, link for those with access to the password protected site).

The SNLS has been allocated spectroscopic followup time at the VLT, Gemini N and S, Keck, and Magellan. Initially we have tended to followup moderately bright (i'AB < 23.5) candidates, but we are gradually trying fainter candidates. Gemini "nod-and-shuffle" is proving to be a real boon for the faintest objects, though Gemini overheads are still the highest of any telescope (around 30 min typically).

There is enough spectroscopic time to follow up most good SNe Ia out to z=0.9, and a handful of SNe II and fainter, more distant Ia's.

With only 4 months of LS Deep data in hand, 2/3 of the expected data flow, and NONE of the final Elixir-processed data available (on which we depend for photometry), it is obviously still early days for getting science out of the SNLS data!

Nevertheless, there is much work underway. A meeting of members of the SNLS team was held in Lisbon in early July to discuss progress to date, and to set up the transition from engineering presurvey time to the real SNLS survey. Members of the SNLS team also met in Toronto Nov 17-18, both to discuss technical aspects of the observing and data reduction, and especially plan for the first science papers. The goal is to have SN Ia rate and equation of state analysis for the first data set (2003B, with some 2003A objects added). (The total number of SNLS SNeIa observed by Feb 2004 should already exceed the numbers of SNe ever observed by other teams!) In addition, a collaboration led by R. Ellis (CalTech) is following up some of the brighter objects with Keck, and plans to have a first science paper ready in early 2004 on the UV spectra and metallicities of high z SNeIa.

The discovery of the acceleration of the expansion of the Universe in 1998 was perhaps one of the most remarkable discoveries in cosmology over the past few decades. The large commitment of resources (both observing time and funding) to SNLS represents an outstanding opportunity for Canadian and French scientists to solve many of the mysteries surrounding the "dark energy" that drives this acceleration.

There are many ways in which the SNLS/Deep data can be used for science goals other than those described here. If you are interested in participating in SNLS, or wish further information, please contact the authors at the email addresses below. You may register as a CFHTLS user at the user registration page . Information on email exploders may be found here .

Chris Pritchet	pritchet@uvic.ca
Ray Carlberg	carlberg@astro.utoronto.ca