Supernova ( SN ) rates are potentially powerful diagnostics of metal enrichment and SN physics , particularly in galaxy clusters with their deep , metal-retaining potentials and relatively simple star-formation histories .
We have carried out a survey for supernovae ( SNe ) in galaxy clusters , at a redshift range 0.5 < z < 0.9 , using the Advanced Camera for Surveys ( ACS ) on the Hubble Space Telescope .
We reimaged a sample of 15 clusters that were previously imaged by ACS , thus obtaining two to three epochs per cluster , in which we discovered five likely cluster SNe , six possible cluster SNe Ia , two hostless SN candidates , and several background and foreground events .
Keck spectra of the host galaxies were obtained to establish cluster membership .
We conducted detailed efficiency simulations , and measured the stellar luminosities of the clusters using Subaru images .
We derive a cluster SN rate of 0.35 Â SNu _ { B } ^ { +0.17 } _ { -0.12 } Â ( statistical ) { \pm 0.13 } Â ( classification ) { \pm 0.01 } Â ( systematic ) [ where SNu _ { B } = SNe ( 100 yr 10 ^ { 10 } { L } _ { B, \sun } ) ^ { -1 } ] and 0.112 Â SNu _ { M } ^ { +0.055 } _ { -0.039 } Â ( statistical ) { \pm 0.042 } Â ( classification ) { \pm 0.005 } Â ( systematic ) [ where SNu _ { M } = SNe ( 100 yr 10 ^ { 10 } { M } _ { \sun } ) ^ { -1 } ] .
As in previous measurements of cluster SN rates , the uncertainties are dominated by small-number statistics .
The SN rate in this redshift bin is consistent with the SN rate in clusters at lower redshifts ( to within the uncertainties ) , and shows that there is , at most , only a slight increase of cluster SN rate with increasing redshift .
The low and fairly constant SNÂ Ia rate out to z \approx 1 implies that the bulk of the iron mass in clusters was already in place by z \approx 1 .
The recently observed doubling of iron abundances in the intracluster medium between z = 1 and 0 , if real , is likely the result of redistribution of existing iron , rather than new production of iron .