A new tool for studying GCs emerged during the last few years with the advent of large-format infrared detectors. Kissler-Patig (2000) pointed out that it should be possible to use near-IR (NIR) photometry of GCs to break the age-metallicity degeneracy inherent in optical colors. The V - K color of an old GC is a measure of the temperature of its red giant branch. This temperature is strongly dependent on metallicity but has little sensitivity to age. The prospect of large NIR imaging surveys is exciting because it offers the potential to accurately measure individual ages and mean metallicities for large numbers of GCs in galaxies out to the distance of the Virgo cluster and beyond.
Kissler-Patig and collaborators, primarily using VLT/ISAAC, have taken the lead in such NIR studies. Puzia et al. (2002) reported a comparison in VIK between the GC systems of the group S0 NGC 3115 and the Virgo E NGC 4365. The GC system of NGC 3115 appeared bimodal in color-color space, while the color distribution of GCs in NGC 4365 looked largely unimodal, with a formal peak at supersolar metallicity and intermediate age (~ 2-5 Gyr). However, the presence of a large subpopulation of intermediate-age GCs was confusing, as the central light of the galaxy itself is uniformly old (Davies et al. 2001). How could a large number of GCs form without an accompanying field star component? This puzzling result precipitated a flurry of activity on the GC system of NGC 4365. Spectroscopy of a subset of the candidate intermediate-age GCs seemed to confirm their young ages, while demonstrating that other young candidates (selected from their optical/NIR colors) were in fact old (Larsen et al. 2003).
However, follow-up spectroscopy with somewhat higher S/N found no evidence for any intermediate-age clusters. Moreover, when considered in combination with new HST/ACS imaging, these data seemed to point instead to three subpopulations of old GCs, with a very centrally-concentrated, intermediate-metallicity subpopulation filling in the gap between the normal metal-poor and metal-rich GCs (Brodie et al. 2005). A similar intermediate-metallicity subpopulation has been discovered by Beasley et al. (2006) in the gE NGC 5128. This interpretation is consistent with wider-field K-band photometry obtained by Larsen, Brodie, & Strader (2005). These authors discussed all of the available data and the details of the error estimates, and showed evidence for systematic errors in optical-NIR SSP models. Nonetheless, based on a comparison of HST/NICMOS H-band photometry of 70 GCs in NGC 4365 and 11 GCs NGC 1399 with SSP models, Kundu et al. (2005) concluded that a large number of intermediate-age (2-8 Gyr) GCs with metallicities up to [Fe/H] = +0.4 are present in NGC 4365. Young metal-poor GCs would also be inferred from the distribution of their data on SSP model grids. Since metal-poor GCs are expected to be old under nearly all formation scenarios, this further suggests that at least some of the offset of metal-rich GCs to young ages may be due to systematic errors in the SSP models (Larsen et al. 2005). Clearly, the last word has not yet been spoken on this intriguing galaxy.
Hempel et al. (2003) studied the early-type galaxies NGC 5846 and NGC 7192 with NIR photometry, reporting that the former galaxy hosts a large population of intermediate-age GCs. Followup U-band photometry in NGC 5846 and NGC 4365 (Hempel & Kissler-Patig 2004) was argued to support the earlier results, although the number of GCs detected was low. Follow-up spectroscopy does not appear to support the presence of a large subpopulation of intermediate-age GCs in NGC 5846 (Puzia et al. 2005). The fraction of GCs with ages formally < 8 Gyr in the observed sample of luminous GCs is 15-20% (T. Puzia, private communication).
Goudfrooij et al. (2001a) presented the best evidence to date of a significant population of intermediate-age GCs in any post-merger galaxy. They used a combination of HST and ground-based optical and NIR photometry of GCs in NGC 1316, a merger remnant E in the Fornax cluster. They found metal-rich GCs that were substantially brighter than normal GCs at that distance and deduced photometric ages of ~ 3 Gyr for these bright red objects. These photometric ages were confirmed by spectroscopy of a small number of the brightest clusters (Goudfrooij et al. 2001b). However, even in this galaxy, the strongest pieces of evidence for an intermediate-age subpopulation are (i) the presence of unusually luminous GCs and (ii) the power-law (rather than log-normal) GC luminosity function of the metal-rich GCs (see also Goudfrooij et al. 2004).
Improvements in NIR SSP models will brighten the prospects for future NIR studies and help this field to reach its full potential. In addition, the launch of the James Webb Space Telescope (JWST) in the next decade will allow the collection of high-quality K-band photometry for many GC systems.